Professional Documents
Culture Documents
ANESTHESIOLOGY
COMPLICATIONS IN
ANESTHESIOLOGY
Emilio B. Lobato, MD
P R O F E S S O R O F A N E S T H E S I O LO G Y
U N I V E R S I T Y O F F LO R I DA CO L L E G E O F M E D I C I N E ;
C H I E F, A N E S T H E S I O LO G Y S E R V I C E
N O RT H F LO R I DA / S O U T H G E O R G I A V E T E R A N S H E A LT H S Y S T E M
M A LCO M B . R A N DA L L V ET E R A N S A F FA I R S M E D I C A L C E N T E R
G A I N E S V I L L E , F LO R I DA
N i k o l a u s G r a v e n s te i n , M D
TH E J EROM E H. MODELL, M D P ROF ESSOR AN D CHAI R MAN OF
A N E S T H E S I O LO G Y
P ROF ESSOR OF N EU ROSU RGERY
U N I V E R S I T Y O F F LO R I DA CO L L E G E O F M E D I C I N E
G A I N E S V I L L E , F LO R I DA
R o b e r t R . K i r b y, M D
P R O F E S S O R E M E R I T U S O F A N E S T H E S I O LO G Y
U N I V E R S I T Y O F F LO R I DA CO L L E G E O F M E D I C I N E
G A I N E S V I L L E , F LO R I DA
Acquisitions Editor: Brian Brown
Developmental Editor: Brigitte Wilke
Managing Editor: Nicole T. Dernoski
Marketing Manager: Angela Panetta
Project Manager: Fran Gunning
Design Coordinator: Stephen Druding
Manufacturing Coordinator: Kathleen Brown
Production Services: Laserwords Private Limited, Chennai, India
All rights reserved. This book is protected by copyright. No part of this book may be reproduced
in any form by any means, including photocopying, or utilized by any information storage and
retrieval system without written permission from the copyright owner, except for brief
quotations embodied in critical articles and reviews. Materials appearing in this book prepared
by individuals as part of their official duties as U.S. government employees are not covered by the
above-mentioned copyright.
Care has been taken to confirm the accuracy of the information presented and to describe generally
accepted practices. However, the authors, editors, and publisher are not responsible for errors or
omissions or for any consequences from application of the information in this book and make
no warranty, expressed or implied, with respect to the currency, completeness, or accuracy of the
contents of the publication. Application of this information in a particular situation remains the
professional responsibility of the practitioner.
The authors, editors, and publisher have exerted every effort to ensure that drug selection and
dosage set forth in this text are in accordance with current recommendations and practice at the
time of publication. However, in view of ongoing research, changes in government regulations,
and the constant flow of information relating to drug therapy and drug reactions, the reader is
urged to check the package insert for each drug for any change in indications and dosage and for
added warnings and precautions. This is particularly important when the recommended agent is
a new or infrequently employed drug.
Some drugs and medical devices presented in the publication have Food and Drug
Administration (FDA) clearance for limited use in restricted research settings. It is the
responsibility of the health care provider to ascertain the FDA status of each drug or device
planned for use in their clinical practice.
To purchase additional copies of this book, call our customer service department at (800)
638-3030 or fax orders to (301) 223-2320. International customers should call (301) 223-2300.
Visit Lippincott Williams & Wilkins on the Internet: at LWW.com. Lippincott Williams &
Wilkins customer service representatives are available from 8:30 am to 6 pm, EST.
10 9 8 7 6 5 4 3 2 1
Many people are responsible for this book through their education
and training of us, their pupils. Three of them stand
apart from the crowd.
vii
viii CONTENTS
hen the idea of another text addressing understanding of how such heat loss occurs, what can be
W
Complications in Anesthesiology was first done to minimize it, and how the related problems can be
entertained, the main consideration was managed.
why this book should be written. After all, Ten more chapters appear in the current book than
the practice of anesthesia and periopera- in the previous text, and some which appear to be
tive medicine is safer in the 21st century less relevant today (dental and salivary gland compli-
than in any other period in history. Nevertheless, despite cations, epidemiologic methods in anesthesia) have been
advances in prevention, diagnosis, and treatment, com- decreased or eliminated. The latter subject is present but
plications still occur, and medical malpractice following is distributed among several chapters rather than in stand
such complications continues to be a problem. Moreover, alone format. Other subjects have been added because
some complications continue to challenge physicians de- of their obvious relevance in todays world (bioterror-
spite a reasonable understanding of the pathophysiology ism, herbal remedies, and over-the-counter drugs). Each
(e.g., perioperative myocardial infarction), whereas oth- reader undoubtedly can come up with an additional sub-
ers previously considered uncommon (e.g., postoperative ject considered appropriate to be included in a textbook of
cognitive dysfunction, ischemic optic neuropathy) now complications. Perhaps future editions will address them,
have attracted widespread attention. and no doubt some subjects of apparent importance today
In a previous textbook (1996), we remarked on will be less so in years hence. Dr. Julius Comroe, author
the substantial investment of time and effort in the of the classic textbook, The Lung, once said he welcomed
American Society of Anesthesiologists reports of Closed suggestions for additional material so long as they were
Claims Studies. These reports continue to the present accompanied by recommendations for the deletion of a
and have done much to elucidate the mechanisms by similar amount. We concur with that approach.
which many adverse events occur and how (perhaps) they Once again, we thank Hope Olivo, the incredibly
may be avoided. Much of the information that has been skilled, best, and most patient faculty editor we ever have
forthcoming is summarized in several chapters contained known. We also are in debt to the editors at Lippincott
herein. Williams & Wilkins: Lisa McAllister, Executive Editor;
Originally, Drs. Frederick Orkin and Lee Cooperman Brian Brown, Senior Acquisitions Editor; and Brigitte
stated that complications generally are related to the sur- Wilke, Developmental Editor, all of whom tolerated the
gical procedure, the patients medical condition, and the slings and arrows emanating from our word processors
anesthetic. This categorization may be a bit oversimpli- when we became frustrated and/or irritated at some in-
fied, and some gray zones exist. As an example, does surmountable problem that reared its ugly head. Almost
hypothermia result primarily from an open abdominal without exception, they managed to soothe the untamed
cavity attendant to an exploratory laparotomy, to insuffi- spirits (aka Drs. Gravenstein, Lobato, and Kirby) and ac-
cient efforts at warming by the anesthesiologist or nurse tually brought the project to fruition. We are in their debt.
anesthetist, or (more probably) to a combination of the
two exacerbated by conductive, convective, and radiant Emilio B. Lobato, MD
heat loss to the operating room and ventilation system? Nikolaus Gravenstein, MD
Hence, the classification is less important than is the Robert R. Kirby, MD
xi
CONTRIBUTORS
Leonard Allmond, MD
Fellow, Obstetric Anesthesia
Clinical Instructor
Department of Anesthesiology
University of Colorado Health Sciences Center
Denver, Colorado
Giuditta Angelini, MD
Assistant Professor of Anesthesiology and Medicine
University of Wisconsin-Madison
Madison, Wisconsin
Miguel Bejar, MD
Assistant Professor of Anesthesiology
University of Florida
Shands Hospital at Jacksonville
Jacksonville, Florida
M. Tariq Bhatti, MD
Associate Professor
Departments of Ophthalmology and Medicine (Division of Neurology)
Duke University Eye Center
Duke University Medical Center
Durham, North Carolina
xiii
xiv CONTR I BUTOR S
Michael J. Bishop, MD
Professor of Anesthesiology and Medicine (Adjunct)
Veterans Affairs Puget Sound Health Care System
University of Washington School of Medicine
Seattle, Washington
Barbara W. Brandom, MD
Professor of Anesthesiology
Attending Anesthesiologist
Childrens Hospital of Pittsburgh
University of Pittsburgh Medical Center
Pittsburgh, Pennsylvania
Sorin J. Brull, MD
Professor of Anesthesiology
Mayo Clinic College of Medicine
Jacksonville, Florida
Virginia M. Buchanan, JD
Levin, Papantonio, Thomas, Mitchell, Echsner & Proctor, P.A.
Pensacola, Florida
Brenda A. Bucklin, MD
Professor
Department of Anesthesiology
University of Colorado Health Science Center
Denver, Colorado
CONTR I BUTOR S xv
Janet M. Caruso, MD
Courtesy Clinical Assistant Professor of Nephrology
University of Florida College of Medicine;
Malcom Randall VA Medical Center
Gainesville, Florida
Lawrence J. Caruso, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Linda Chi, MD
Associate Professor
Department of Diagnostic Imaging
The University of Texas MD Anderson Cancer Center
Houston, Texas
Jerry A. Cohen, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Douglas B. Coursin, MD
Professor of Anesthesiology and Medicine
University of Wisconsin School of Medicine and Public Health
Madison, Wisconsin
xvi CONTR I BUTOR S
Laurie K. Davies, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Hernando Desoto, MD
Associate Professor of Anesthesiology
Director Pediatric Anesthesia
University of Florida College of Medicine
Jacksonville, Florida
Christine A. Doyle, MD
Staff Anesthesiologist and Intensivist
OConnor Hospital
San Jose, California;
Former Fellow in Critical Care Medicine
Stanford University Medical Center
Stanford, California
Jan Ehrenwerth, MD
Professor of Anesthesiology
Yale University School of Medicine;
Attending Physician
Yale New Haven Hospital
New Haven, Connecticut
CONTR I BUTOR S xvii
John H. Eichhorn, MD
Professor
Department of Anesthesiology
University of Kentucky
College of Medicine/Medical Center
Lexington, Kentucky
John E. Ellis, MD
Professor
Department of Anesthesia and Critical Care
The University of Chicago Medical Center
Chicago, Illinois
David Ferson, MD
Professor
Department of Anesthesiology and Pain Medicine
The University of Texas MD Anderson Cancer Center
Houston, Texas
Rainer Georgi, MD
Kinikum Stuttgart Katharinenhospital
Klinik fur Anaesthesiologie und operative Intensivmedizin
Stuttgart, Germany
Gordon L. Gibby, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
xviii CONTR I BUTOR S
Salvatore R. Goodwin, MD
Chairman, Department of Anesthesia
Nemours Childrens Clinic
Jacksonville, Florida;
Associate Professor of Anesthesiology
Mayo Medical School
Rochester, Minnesota
Dietrich Gravenstein, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
J. S. Gravenstein, MD
Graduate Research Professor, Emeritus
Department of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Nikolaus Gravenstein, MD
The Jerome H. Modell, MD Professor and Chairman of Anesthesiology
Professor of Neurosurgery
University of Florida College of Medicine
Gainesville, Florida
Roy A. Greengrass, MD
Associate Professor of Anesthesiology
Mayo Clinic College of Medicine
Jacksonville, Florida
Jacob T. Gutsche, MD
Cardiac Anesthesia/Critical Care Fellow
Department of Anesthesiology and Critical Care
Hospital of the University of Pennsylvania
Philadelphia, Pennsylvania
Ala S. Haddadin, MD
Assistant Professor
Department of Anesthesiology
Yale University School of Medicine
New Haven, Connecticut
CONTR I BUTOR S xix
Carin A. Hagberg, MD
Professor, Department of Anesthesiology
Director of Neuroanesthesia and Advanced Airway Management
The University of Texas Medical School at Houston
Houston, Texas
Joy L. Hawkins, MD
Associate Chair for Academic Affairs
Department of Anesthesiology
University of Colorado Health Sciences Center
Denver, Colorado
Christopher F. James, MD
Consultant, Department of Anesthesiology and
Assistant Professor
Mayo Clinic College of Medicine;
Director, Obstetric Anesthesia
St Lukes Hospital
Jacksonville, Florida
xx CONTR I BUTOR S
Gregory M. Janelle, MD
Associate Professor of Anesthesiology and Surgery
Chief, Division of Cardiothoracic Anesthesia
University of Florida College of Medicine
Gainesville, Florida
Jennifer Janelle, MD
Staff Physician
Malcom Randall Veterans Administration Medical Center
Gainesville, Florida
W. Russ Jones
Uzick, Oncken, Scheuerman & Berger, P.C.
Houston, Texas
Robert R. Kirby, MD
Professor Emeritus of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Charles T. Klodell, MD
Assistant Professor of Thoracic and Cardiovascular Surgery
Director, Cardiac Mechanical Assist Device Program
University of Florida College of Medicine
Gainesville, Florida
Robert S. Lagasse, MD
Professor of Clinical Anesthesiology
Albert Einstein College of Medicine & Montefiore Medical Center
Bronx, New York
Melissa A. Laxton, MD
Assistant Professor of Anesthesiology
Wake Forest University School of Medicine
Winston-Salem, North Carolina
Jerrold H. Levy, MD
Professor
Department of Anesthesiology
Emory University School of Medicine
Atlanta, Georgia
Marjorie A. Lewis, MD
Pediatric Anesthesiologist
Nemours Childrens Clinic
Jacksonville, Florida
Edwin B. Liem, MD
Staff Anesthesiologist, ORA PLLC
Norton Hospitals;
Assistant Professor of Anesthesiology (Courtesy)
University of Louisville
Louisville, Kentucky;
Assistant Professor (Courtesy)
University of Florida College of Medicine
Gainesville, Florida
Emilio B. Lobato, MD
Professor of Anesthesiology
University of Florida College of Medicine;
Chief, Anesthesiology Service
North Florida/South Georgia Veterans Health System
Malcom B. Randall Veterans Affairs Medical Center
Gainesville, Florida
Stuart B. Mushlin, MD
Brigham Circle Medical Associates
Brigham and Womens Hospital;
Assistant Professor of Medicine
Harvard Medical School
Boston, Massachusetts
J. Mauricio Palacios, MD
Assistant Professor
Department of Anesthesiology
College of Medicine
University of Arkansas for Medical Sciences
Little Rock, Arkansas
xxiv CONTR I BUTOR S
Charise T. Petrovitch, MD
Clinical Professor
Department of Anesthesiology and Critical Care Medicine
The George Washington University Hospital
Washington, DC
Patricia H. Petrozza, MD
Professor of Anesthesiology
Associate Dean for Graduate Medical Education
Wake Forest University School of Medicine
Winston-Salem, North Carolina
Christian Popa, MD
Chief, Critical Care Medicine
Walter Reed Army Medical Center
Washington, DC;
Assistant Professor of Anesthesiology
Uniformed Services University of the Health Sciences
Bethesda, Maryland
Christine S. Rinder, MD
Department of Anesthesiology & Laboratory Medicine
Yale University School of Medicine
New Haven, Connecticut
Stanley H. Rosenbaum, MD
Professor of Anesthesiology, Internal Medicine, and Surgery
Department of Anesthesiology
Yale University School of Medicine
New Haven, Connecticut
Myer H. Rosenthal, MD
Professor of Anesthesia, Medicine and Surgery
Stanford University School of Medicine
Stanford, California
Avner Sidi, MD
Associate Professor of Anesthesiology
Tel Aviv University
Tel Aviv, Israel;
Vice Chairman and Head of PACU
Anesthesiology Department
Sheba Medical Center
Tel Hashomer, Israel;
Associate Professor of Anesthesiology
University of Florida College of Medicine
Gainesville, Florida
Joshua D. Stearns, MD
Clinical Fellow, Cardiothoracic Division
Department of Anesthesiology & Critical Care
University of Pennsylvania School of Medicine
Philadelphia, Pennsylvania
Cheri A. Sulek, MD
Associate Professor of Anesthesiology
University of Florida College of Medicine;
North Florida/South Georgia Veterans Health System
Malcom B. Randall Veterans Affairs Medical Center
Gainesville, Florida
Murat Sungur, MD
Assistant Professor of Anesthesiology and Critical Care Medicine
University of Florida College of Medicine
Division of Critical Care Medicine
Gainesville, Florida
Mark D. Tasch, MD
Associate Professor of Clinical Anesthesia
Department of Anesthesia
Indiana University School of Medicine
Indianapolis, Indiana
Angela T. Truong, MD
Assistant Professor
Department of Anesthesiology & Pain Medicine
The University of Texas MD Anderson Cancer Center
Houston, Texas
Felipe Urdaneta, MD
Clinical Associate Professor of Anesthesiology
University of Florida College of Medicine;
Assistant Chief
Anesthesiology Service
Malcom Randall Veterans Administration Medical Center
NF/SG Veterans Health System
Gainesville, Florida
Cengiz Utas, MD
Professor of Medicine and Nephrology
Erciyes Universitesi Rektorlugu
Kayseri, Turkey
Deborah M. Whelan, MD
Assistant Professor of Anesthesiology
Wake Forest University School of Medicine
Winston-Salem, North Carolina
Luis M. Zabala, MD
Assistant Professor of Anesthesiology
University of Arkansas for Medical Sciences;
Staff Anesthesiologist
Arkansas Childrens Hospital
Little Rock, Arkansas
Sonal Zambare, MD
Department of Anesthesiology and Pain Medicine
The University of Texas MD Anderson Cancer Center
Houston, Texas
SECTION
1
GENERAL
AND MEDICOLEGAL
C O N S I D E R AT I O N S
CHAPTER ANESTHESIA, PERIOPERATIVE
1
MORTALITY, AND PREDICTORS
OF ADVERSE OUTCOMES
A
mass index = 63) presented to a recently (QA) system, and the state department of health.
opened bariatric surgery center at a 200-bed
exurban community hospital for laparo-
scopic gastric bypass surgery. She had type
2 diabetes and dyspnea, with palpitations Has Anesthesia Outcome
from minor effort but no cardiac diagnosis. A nonsmoker,
she slept most nights sitting in a 60-degree position in a Improved?
lounge chair but never had a sleep study. Her internist
and surgeon cleared her for anesthesia. The surgeon had Anesthesia practitioners who have practiced more than
been recently recruited from fellowship training. 25 years widely accept that anesthesia care is safer now
PREOPERATIVE FINDINGS: Blood pressure 155/95, than, for example, in 1980at least regarding catastrophic
heart rate 90, respiratory rate 22, temperature 37 C, adverse outcomes and probably for complications in gen-
and oxygen saturation as measured by pulse oximetry eral. Traditionally, because surgical anesthesia care is
(SpO2 ) 91% while she breathed room air, rising to facilitative rather than therapeutic, a good outcome is
95% on O2 2 L per minute through nasal prongs. Her measured in terms of the absence of complications or
electrocardiogram showed normal sinus rhythm at 96 adverse outcomes. Avoidance of preventable complica-
bpm with left ventricular hypertrophy and nonspecific tions that lead to adverse patient outcomes has been the
ST-T wave changes. A right subclavian double-lumen focus of the organized patient safety movement that be-
catheter was inserted on the second pass. General gan in the United States in the early 1980s and spread
anesthesia was conducted with midazolam, fentanyl, worldwide.1
etomidate, rocuronium, and desflurane in O2 , with Many reasons can be cited as to why anesthesia care
additional intermittent boluses of fentanyl. The surgeon has become safer. Utilization of sensitive electronic mon-
requested limitation of crystalloid infusion. Peak airway itors, such as the capnograph and pulse oximeter, extends
pressures exceeded 45 cm H2 O. With positive end- the human senses of the physician. Application of mon-
expiratory pressure of 7 cm H2 O, SpO2 stayed in the itoring standards24 is a dramatic example of anesthesia
low 90s. At the end of the case, the surgeon urged that the patient safety improvement, particularly regarding intra-
patient be extubated immediately. When she was placed operative anesthesia catastrophes that lead to cardiac
in a semisitting position, she met the usual criteria and arrest, permanent brain damage, or death. Other im-
was extubated in the operating room at 5:15 PM.
portant components contributing to improved anesthesia
She received no pain medication in the postanesthe-
patient safety include:
sia care unit and was stable, with SpO2 91% while she
breathed O2 3 L per minute through nasal prongs. When Improved training of practitioners (better students
the postanesthesia care unit closed at 7:00 PM, she was entering training, better teachers, a longer training
transported in a semisitting position to a standard single period, improved/expanded textbooks and journals,
room on the floor, with orders for sliding scale insulin teaching with simulation)
and morphine through patient-controlled analgesia. At Better communication of safety information and rec-
10:00 PM, her vital signs (including spot-check SpO2 ) were ommendations through multiple professional organiza-
unchanged, and she was alone, dozing (with snoring) and tions and the Anesthesia Patient Safety Foundation
watching television. At next contact, almost 3 hours later More extensive and focused research into safety and
(12:45 AM), the patient was found dead and did not respond human factors
3
4 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
More and improved anesthetic and ancillary medica- anesthesia practice. As an example, in 2002, the inflation-
tions adjusted malpractice insurance premium for Harvard
Other equipment and technologies, such as fiberoptics Medical School-insured anesthesiologists was approxi-
for airway management; and the development of mately one fourth of the rate in the mid-1980s.5 For all
practice checklists, protocols, and algorithms from anesthesiologists, based on analysis of premium amounts
various sources for $1 million or $3 million policy limits, the average
However, some counterbalancing forces have tended premium in 2006 was $19,558 compared with an average
to increase the risk of complications and adverse patient inflation-adjusted premium of $32,502 in 1985.6 While the
outcomes. Surgical procedures are expanded in scope, overall trend toward safer anesthesia should continue, in
complexity, and length, and older patients and patients light of the competing forces outlined in the preceding
who are more ill are considered potential surgical text, it is appropriate to consider anesthesia perioperative
candidates than in previous decades. The combination mortality and predictors of adverse outcomes.
of these factors has increased the aggregate intensity of
the challenge involved in providing safe anesthesia care.
Further, the demand for anesthesia care in many locations
outside the traditional hospital or surgicenter operating How Is Anesthesia
room is perceived by many practitioners as a risk factor for Perioperative Mortality
increased complications and adverse patient outcomes.
Evidence that anesthesia care is safer now includes Defined?
an important alternate definition of truth that goes
beyond the traditional p < 0.05 statistical approach of Although death is clearly definable, historically, the con-
data analysis. The nonparallel construction of studies cept of anesthesia mortality has been confused and very
and databases use variable definitions and methods. In poorly articulated. Various definitions of perioperative
addition, no adjustments are made for differing settings mortality have been employed since the 1950s; accord-
and patient populations. Comparatively small samples are ingly, the evaluation and comparison of the various
assessed in an arena where it would take huge population reports and statistics are difficult.
groups to demonstrate statistically valid changes in the
rate of very rare events. Therefore, it is very difficult to
draw conclusions by comparing statistical studies and POSTOPERATIVE TIME
reports in the literature.5
One approach in the United States is to consider the FRAME
conclusions on this point by an integral component of the
A key factor is the issue of time frame in relation to
American health care systemthe medical malpractice
the surgery. Beyond the agreement that intraoperative
insurance industry. Malpractice insurance premiums
deaths qualify as perioperative mortality, there is little
directly reflect the losses of insurance companies from
consistency as to how long to extend the postoperative
adverse patient outcomes that result in insurance claims,
period in order for deaths to count as perioperative.
settlements, and court judgments. Since the end of the
Different sources have used widely differing postoperative
1980s, malpractice insurance premiums in the United
time boundaries:
States actually decreased for many anesthesiologists and
did not increase at the same rate as they did for virtually The American College of Surgeons30 days
all other physicians. This observation provides significant The American Hospital Associationno identifiable
functional evidence indicating that anesthesia care has limit
become safer. The Joint Commission on Accreditation of Healthcare
Insurance company actuaries are not charitable peo- Organizationswithin 48 hours
ple, but they will reduce premiums, actually and relatively, The U.S. Center for Medicare and Medicaid Ser-
only when the calculations show a decreased loss by their vicesno identifiable limit
companies. Given the nature of the medical-legal system The British National Health Service in their National
and associated public attitudes over the last 3 decades Confidential Enquiry into Perioperative Deathwithin
in the United States, the vast majority of unexpected 29 days
catastrophic patient outcomes in the perioperative period Study groups, such as the South Australian Periopera-
should have been brought to the attention of plaintiffs tive Mortality Committeewithin 24 hours, compared
attorneys. The decrease in insurance company losses in with the neighboring Anesthesia Mortality Commission
this time frame is a result of fewer and less serious insur- of Western Australiawithin a 48-hour time frame
ance cases involving anesthesia catastrophes. Insurance
premiums have increased in this decade, and the avail-
ability of malpractice insurance is problematic in certain CAUSE OF DEATH
locations. Nonetheless, the trend of proportionately less
increases in premiums persist for anesthesiologists com- Beyond the significant difficulty resulting from nonparal-
pared with those for physicians in the high-risk specialties lel definitions of the time frame that defines perioperative
of obstetrics and gynecology, neurosurgery, and orthope- mortality is the even greater problem of dissecting and
dic surgery, thereby supporting the improved safety in attributing the causation for the deaths. Some discussions
C H A P T E R 1/ A N E S T H E S I A , P E R I O P E R AT I V E M O R TA L I T Y , A N D P R E D I C T O R S O F A D V E R S E O U T C O M E S 5
consider death from any cause as perioperative mortal- PERIOPERATIVE MORTALITY: 13 results, range 1:53
ity, and record it as such, be it direct surgical error or (1 death per 53 cases) to 1:5417
complications; direct anesthesia error or complications; ANESTHESIA-RELATED MORTALITY: 22 results, range 1:388
the underlying surgical problem (such as sepsis following to 1:85,708 to 0 deaths
a perforated bowel); any preexisting concurrent medical ANESTHESIA SOLELY RESPONSIBLE MORTALITY: 7 results,
condition (such as coronary artery disease); any preexist- range 1:9,090 to 1:200,200 to 0 deaths
ing evolving medical condition (such as acute respiratory PREVENTABLE ANESTHETIC MORTALITY: 6 results, range
distress syndrome in a patient in the intensive care unit); 1:1,707 to 1:48,748 to 0 deaths (the 0 deaths study was
or any new problem (such as a fatal postoperative pul- of 27,184 inpatients at four Canadian hospitals9 )
monary embolus).
The wide variability among these results clearly
This expansive but nonspecific approach makes the
illustrates the difficulties in comparing disparate sources
meaningful analysis of death rates virtually impossible.
at different times using different definitions of mortality.
However, this method of statistical analysis must be ac-
This problem, of course, makes it difficult to identify
knowledged because it is the basis of information that
trends and epidemiologic patterns. Lagasse added data
various government agencies, regulatory bodies, insur-
from his own university hospital system for two time
ance providers, and media are trying to collect and
periods: 1992 to 1994 and 1995 to 1999. The perioperative
publish. By doing so, they may promote competition
mortality was 1:332 and 1:632, respectively. These figures
among health care providers and facilities, in addition
may seem high but they included a mortality rate of 1:4.6
to providing public information based on which choices
for American Society of Anesthesiologists (ASA) Physical
in health care can be made. The specific consideration of
Status (PS) class V patients. Using death within 48 hours
the role of anesthesia care in perioperative mortality is ex-
with anesthetist contribution (human error) as the
tremely difficult to derive, because there is no consistency
in the available information to be examined. definition of anesthesia-related mortality, the mortality
rates were 1:12,641 and 1:13,322, respectively, with 0
deaths in the anesthesia-solely-responsible category for
either group. These data include patients of all ASA PS
ANESTHETIC DEATHS classes.7 The vast majority of perioperative deaths was in
Some authors have accepted all operative and postopera- ASA PS class V, with a few in class IV, and very few in
tive deaths as anesthetic deaths. Others have attempted to class III. On the explanatory graph, the anesthesia-related
define anesthesia-related deaths or very limited examples mortality did not register for ASA PS class I patients and
of death or catastrophe caused solely by anesthesia care. was barely perceptible for PS class II patients. The rate
Sometimes they refer specifically to an anesthesia acci- appeared to be 1:6,883 for PS class III, 1:2005 for PS class
dent, such as an unrecognized esophageal intubation that IV, and 1:715 for PS class V patients.
causes an immediate hypoxemic cardiac arrest and death, In the discussion, Lagasse noted that, of the 14
as opposed to an intraoperative myocardial infarction reported anesthesia-related deaths, 4 were the result
during an anesthetic that seemed to go well but caused of major contribution from the anesthesia personnel,
death days later. yielding a rate of 1:46,118 anesthetics. Of those four, only
one was in ASA PS class I or II, yielding a mortality
rate of 1:126,711 for these healthy patients; that death
Definitions of Terms was not intraoperative and, therefore, would not have
Consideration of anesthesia perioperative mortality was met the inclusion criteria in the Eichhorn study3 of
undertaken in an extensive review by Lagasse that also intraoperative catastrophes in 1,001,000 healthy ASA PS
incorporated original data from one hospital system on classes I and II patients. Again, specifically noting that
perioperative mortality.7 One conclusion was a challenge there was no death considered due solely to anesthetic
to the claim that anesthesia care is much safer now management in the 184,472 anesthetics reported, Lagasse
than in previous decades. This challenge was opposed in concluded: Therefore, our results are consistent with the
an accompanying editorial5 that noted it is notoriously Eichhorn study, which may have been the basis of the IOM
difficult to assess the contribution of anesthesia care claim. That very widely publicized 1999 report from the
to perioperative mortality, and even more problematic to Institute of Medicine (IOM) heralded the improvement in
make comparisons across time periods using small sample anesthesia patient safety, citing a decrease in anesthesia
sizes from widely differing institutions involving deaths mortality rates from approximately 1:5,000 anesthetics
that anesthesia variably solely caused, or was related at the start of the 1980s to 1:200,000 to 300,000 (in the
to, contributory to, or associated with. late 1990s).10
Lagasse, nonetheless, attempted to assemble virtually
all the data in the literature through 1997. He did not Data from Other Countries
include the landmark 1954 Beecher and Todd study,8 in
which a death rate with anesthesia as a very important Other relatively recent publications, including some from
contributing factor of 1:1,560 was reported, because other countries, report values that are not too dissimilar,
it was published before the beginning of the Medline particularly because the rates of extremely rare events
database in 1966. The various mortality statistics from can sometimes be dramatically altered or even quite
the 23 studies reviewed were presented as: skewed by a single occurrence. A 2001 multicenter study of
6 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
869,483 patients of all ASA PS classes in the Netherlands rate totally attributable to anesthesia management was
identified 811 patients (1:1,072) who died or suffered 1:720,388 (two patients in that group suffered deaths for
brain damage within 24 hours of surgery.11 One hundred which anesthesia care was solely responsible).
and nineteen deaths were determined to be anesthesia-
related (1:7,306), and seven were solely attributable to
anesthesia, for a rate of 1:124,000. A comprehensive Numbers Do Not Lie (Do They?)
Norwegian study12 from a single hospital of 83,844
Overall, these representative reports illustrate the scope
patients of all ASA PS classes over a 5-year period
of the type of information that is available in the lit-
revealed 111 patients with serious problems for which
erature regarding anesthesia mortality. Although some
anesthesia was the major contributing factor. Forty-two
generalizations might be possible, they are flawed, as was
patients died intraoperatively (1:1,996), and all belonged
previously noted, because of the wide variability in clas-
to ASA PS class IV or V, except three with uncontrollable
sifications, definitions, different time periods, locations,
surgical bleeding. No intraoperative deaths solely related
institutions, and patient populations.
to anesthesia were reported. One patient had a cardiac
One approach to this question, among many possi-
arrest from an anaphylactic reaction to a muscle relaxant
bilities, is a rigorous attempt to isolate the anesthesia
and later died in the intensive care unit (making the
care component for consideration. This approach seeks
anesthesia mortality 1:83,844, if that case is counted as
an anesthetic death). to remove the influence of patients intercurrent medical
A report of a huge Australian database from the conditions, the surgical issue provoking the administra-
mid-1990s13 cited anesthesia-related deaths at a rate of tion of an anesthetic, and the surgical factors including
1:62,500. A French report from the early 1990s,14 covering misadventure during the surgery. This paradigm led to
101,769 anesthetics in patients of ASA PS classes I to IV the well known (but often misquoted) statistics from
(class V specifically excluded) began with an examination the Harvard-associated teaching hospitals regarding only
of intraoperative cardiac arrests. Of the 11 anesthesia- intraoperative catastrophes that were solely due to anes-
related cardiac arrests, 3 were considered totally related thesia care (cardiac arrest, permanent brain damage, and
to anesthesia, and 8 considered only partially related to death).3 Of the 1,329,000 patients anesthetized from 1976
anesthesia. Six of the 11 patients died (1 PS class II patient, through June, 1988, 1,001,000 ASA PS classes I and
3 PS class III patients, and 2 PS class IV patients), leading II healthy patients were the subject of the report. A
to a cited anesthesia-related mortality of 6:101,769, or central component was the comparison of the rate of
1:16,961. If the one ASA class II patient had been removed, anesthesia-caused catastrophes before and after the im-
the rate for classes I and II would have been 1:101,769. plementation of the Harvard monitoring standards in
A US study of the same type covered 72,959 anesthetics July, 1985. Ten intraoperative catastrophes occurred out
from 1989 to 1999 in a university hospital15 and focused on of 757,000 anesthetics (1:75,700), five of which resulted
anesthesia-related cardiac arrests (extending to 24 hours in death (1:151,400) before the implementation of the
postsurgery). The death rate from anesthesia-attributable standards. In contrast, one catastrophe occurred without
perioperative cardiac arrest in the total study population death in 244,000 anesthetics after implementation.
was 1:18,181. The aggregate number of five deaths solely related to
Nevertheless, one of the most quoted reports involving anesthesia care in 1,001,000 cases was mentioned in the
anesthesia mortality is the British National Confidential discussion of the report almost as an aside. Recall that the
Enquiry into Perioperative Deaths,16 because it involved a focus of the study was on the before-and-after comparison.
systematic epidemiologic study of a large population. This The aggregate mortality rate for this population calculated
database used a 30-day postoperative reporting period and over that period was 1:200,200. This citation in that
a fairly broad set of criteria. Ultimately, an anesthesia- context was offered originally as a complement to and
related mortality of 1:1,351 was reported. An anesthesia confirmation of the British National Confidential Enquiry
solely responsible mortality rate of 1:185,056 was also into Perioperative Deaths figure of 1:185,056 from 2 years
reported, which was a remarkably low figure in 1987. earlier16 that had been questioned so vigorously. Thus
The Japanese Society of Anesthesiologists (JSA) was born (however unintended) the very widely quoted,
Subcommittee on Surveillance of Anesthesia-Related improved anesthesia mortality rate of approximately
Critical Incidents generated thorough statistics. From 1:200,000 anesthetics.
1999 through 2002, among 3,855,384 anesthetics in JSA- Data on ASA PS classes I and II patients anesthetized
certified training hospitals, 2,638 intraoperative deaths at Harvard hospitals were collected for 2 more years and
occurred (1:1,461), of which anesthetic management was published4 as shown in Table 1.1. A principal focus of
responsible for 40, yielding an anesthesia intraoperative that citation was the continued, very busy, high-acuity
mortality rate of 1:96,384, which incorporated patients in anesthesia practices in nine diverse teaching hospitals
all ASA PS classes.17 In the subset of ASA PS classes without a single intraoperative anesthesia catastrophe
I and II patients, the anesthesia mortality rate was that met the inclusion criteria for the study. In fact,
reported as 1:250,000 (within the range of Eichhorns the only poststandards, anesthesia catastrophe reported
reports3,4 ). A separate analysis (with the data collection occurred in August 1985, meaning that by the end of
period extending anesthesia deaths to 7 postoperative June 1990, 467,000 consecutive anesthetics had been
days) of the same database considered only 1,440,776 ASA performed without an intraoperative catastrophe or death
PS class I patients having elective surgery.18 The death solely due to anesthesia care.
C H A P T E R 1/ A N E S T H E S I A , P E R I O P E R AT I V E M O R TA L I T Y , A N D P R E D I C T O R S O F A D V E R S E O U T C O M E S 7
TABLE 1.1 Anesthesia Catastrophes and Deaths in Harvard-Affiliated Teaching Hospitals Among
Healthy Patients Before and After Implementation of Monitoring Standards in July 1985
a By
Fishers exact test.
ASA PS, American Society of Anesthesiologists Physical Status (classification).
The combined total aggregate calculation yielded an but ultimate full recovery; and severe hypoxic brain
overall death rate (ASA PS classes I and II patients damage resulting in a permanent vegetative state.
suffering intraoperative catastrophes caused solely by The term, adverse outcome, is encompassing enough
anesthesia care) of 1:229,800, but this figure was not to allow an effective discussion of the straightforward con-
included in the publication. As noted by Lagasse, however, cept that an adverse anesthesia outcome is undesirable.
this omission may have contributed to the greatly Harm may be implied, but blame is not necessarily in-
simplified and unqualified statistic that has been cited cluded. The patient who awakens from a prolonged spine
repeatedly, including the statistic by the IOM.10 surgery in the prone position with impaired visionin
The subsequent publication by the JSA noted an over- spite of great efforts by the anesthesia personnel to
all intraoperative anesthesia mortality rate of 1:96,384 in maintain optimum hemodynamics, intravascular volume,
teaching hospitals. The rate was 1:250,000 in the sub- and head positionhas experienced an adverse outcome,
set of ASA PS classes I and II patients17 (1:720,388 in with no direct, demonstrable cause and effect or blame.
ASA PS class I patients having elective surgery).18 The Similarly, the patient who suffers an intraoperative or
figures supported the idea of very low magnitude anesthe- postoperative pulmonary embolus, even when all known
sia mortality rates. Review of the parallel numbers from preventive precautions have been taken, will be consid-
Lagasses institution for patients in all PS classes, includ- ered by some as suffering an adverse anesthesia outcome,
ing the postoperative period, showed that the aggregate although no specific evidence or causation is identified.
death rate was higher (as is logical), at approximately Adverse anesthesia outcomes, other than death,
1:13,000. However, the vast majority of these anesthesia occur more often than cardiac arrests. However, they
deaths were in extremely ill, ASA PS class V, high-acuity have widely varying frequencies that have been poorly
patients. The mortality rate in ASA PS classes I and II documented in medical records and within an anesthesia
patients was functionally imperceptible, and no anesthe- practice, group, or department. Formal studies resulting
sia deaths were judged to be solely due to anesthesia in publications or discussion of the frequency and
care. These excellent statistics are consistent with the oth- severity of adverse anesthesia outcomes are infrequent.
ers cited previously; they demonstrate that the anesthesia All anesthesia personnel are familiar with the concept and
mortality risk for healthy patients is vanishingly small. are interested in avoiding adverse outcomes, but precious
little organized, substantive information is available to
work on. Textbooks that detail the many and varied
complications associated with anesthesia care may be
What Are the Predictors of valuable aids to alerting practitioners about potential
adverse outcomes and, it is hoped, to offering suggestions
Adverse Anesthesia Outcomes? as to how to prevent them.
Many patients receive muscle relaxants, tracheal intu- one demographic component of each patient, along with
bation, and mechanical ventilation. They face the risk of age and gender, for use by anesthesia and surgical services
Y-connector separation from the endotracheal tube, with for epidemiologic research and outcomes.
resultant rapid hypercarbia and dangerous hypoxemia.
Such disconnects continue to occur currently. However, National Surgical Quality Improvement
the standard regimen of safety monitoring includes provid-
er vigilance and electronic monitors that are exquisitely Program
sensitive to this adverse event. Low pressure and apnea
The anesthesia mortality rate for ASA PS class V patients
alarms on the ventilator and the spirometer, as well as pa-
is dramatically higher than for ASA PS class I patients.7
rameter limit alarms on the capnograph and pulse oxime-
Extensive discussion of the correlation of the ASA PS
ter, will sound loudly, well before this patient is in real
classification with surgical risk factors in the database
imminent danger (assuming the alarm limits and activa-
of the National Surgical Quality Improvement Program
tion are appropriately set). Therefore, mechanical ventila-
(NSQIP) has taken place. The NSQIP database was shown
tion through an endotracheal tube with relaxation includes
to correlate with and, in some circumstances, predict
risk, but it is not a predictor of a potential adverse outcome.
surgical outcome and validate the ASA PS classification.21
On the other hand, a morbidly obese patient with
Although the ASA PS classification was a strong predictor
documented, severe obstructive and central sleep apnea,
of surgical outcome, other NSQIP risk factors were better
who is oxygen-dependent at home and scheduled for open
predictors as a group.
gastric bypass surgery, faces clear risk and also has defined
Previous epidemiologic research showed that the
predictors of potential adverse anesthesia outcome. These
NSQIP had the best risk-adjustment methodology of
predictors must be recognized and should stimulate extra
those applied to surgery databases, and that NSQIP was
efforts to deal with the risk and avoid the obvious potential
best able to predict mortality.22 In the original NSQIP,
adverse outcomes.
49 preoperative and 17 intraoperative risk factors were
examined for correlations with 33 outcome variables.23
American Society of Anesthesiologists Table 1.3 shows selected preoperative risk factors, most of
Physical Status Classification which were demonstrated as having positive correlations
with surgical mortality and/or morbidity. Therefore, these
The ASA PS classification (see Table 1.2) has already been correlating clinical factors can be considered, in one way,
discussed in some detail. It was not created or intended to be predictors of surgical outcome.
for risk classification or as a predictor of potential adverse Logic suggests that several of the more severe fac-
outcomes, although correlations do exist. Because of torsincluding acute preoperative conditions such as
repeated requests for further detail and clarification, the ventilator dependent, pneumonia, congestive heart fail-
ASA includes this statement with the classification system: ure (CHF), renal failure, and sepsis, as well as in-
There is no additional information that will help you traoperative factors such as transfusion of more than
further define these categories.19 four units of blood, emergency cases, and contaminated
The original intent of the 1941 classification was woundswould be considered predictors of increased
to facilitate studies of patients undergoing anesthesia adverse surgical outcome. Many abnormal laboratory
and surgery. The system was enhanced into its modern values also seem logical correlates; nevertheless, among
form in 1961.20 It has facilitated research and has the chronic preoperative conditions, correlations are not
provided a basis for communication among anesthesia always obvious. Interestingly, coronary artery disease
and nonanesthesia professionals. It describes the acuity manifesting as chronic angina did not appear on the
and the likely anesthetic challenge of a patient and list of positive correlates with adverse surgical outcomes
facilitates the most logical assignment of personnel and (Table 1.3). Diabetes mellitus, smoking, and dialysis ap-
resources for a given case. Because the ASA PS class is pear to predict morbidity but not increased mortality.
usually recorded on the anesthesia record and is often Age, chronic obstructive pulmonary disease, and stroke
captured in the database of surgical facilities, it provides with residual deficit predict both increased morbidity and
mortality. Male gender compared to the total population
is correlated with increased morbidity. In this initial anal-
TABLE 1.2 The American Society of Anesthesiologists ysis, ASA Classification is listed as having a positive
Physical Status Classification correlation and has been further refined as noted.21
TABLE 1.3 Positive Correlations Between National Surgical Quality Improvement Program
(NSQIP) Preoperative Risk Factors and 30-Day Surgical Mortality and Morbidity for Combined
NSQIP Databases for General and Vascular Surgery
X, significant bivariate relation for both the VA and combined non-VA databases.
ETOH, ethyl alcohol; DNR, do not resuscitate; COPD, chronic obstructive pulmonary disease; CHF, congestive
heart failure; TIA, transient ischemic attack; CNS, central nervous system; CA, carcinoma; Abn., abnormal; BUN,
blood urea nitrogen; SGOT, serum glutamate oxaloacetate transaminase; WBC, white blood cell; HCT, hematocrit;
ASA, American Society of Anesthesiologists; PT, prothrombin time; OPS, operation complexity score, which is a
predefined score reflective of operative complexity.
Source: Reprinted with permission, from Fink AS, Campbell DA Jr., Mentzer RM Jr., et al. The National Surgical
Quality Improvement Program in non-veterans administration hospitals: initial demonstration of feasibility. Ann
Surg. 2002;236:344.
10 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
absolute certainty (statistical analysis applies to popu- TABLE 1.4 Traditional Adverse Outcomes of Anesthesia
lations, not individuals). However, the correlations that Care
were demonstrated with the rigorous statistical analysis
of abundant data from large populations allow predic- Broken teeth from laryngoscopy
tions that similar populations in the future will likely Unplanned admission to the hospital following
show the same patterns. Therefore, application of a cor- outpatient surgery
relation demonstrated by the NSQIP may be reasonably Unplanned postoperative admission to an intensive care
used to predict that a group of ASA PS classes IV and V unit
patients will have more adverse surgical outcomes than a Pneumonitis from aspiration of acid stomach contents
group of ASA classes I and II patients. Emergency reintubation in the immediate postoperative
A comparable, national organized study of correla- period
tions between preanesthetic and intraanesthetic factors Peripheral nerve injury
and anesthesia outcomes has not been performed. Again, Neuraxial nerve injury, including stroke, with deficit
logic suggests that a group of ASA classes IV and V Unanticipated pulmonary edema
patients, almost by definition, will have more adverse Myocardial infarction
anesthesia outcomes than a group of ASA classes I and II Cardiac arrest
patients. Hence, assignment of classes IV and V ratings Death
should be carefully considered in any given patient pop-
ulation. Whether the extrapolated congruency between
surgical and anesthesia outcomes will also be true for
other preoperative and operative risk factors has not been
systematically investigated. For example, most anesthesia
What Is the American Society
professionals would probably predict that a preoperative, of Anesthesiologists Closed
ventilator-dependent patient with pneumonia, CHF, renal
failure, and sepsis, who is expected to receive intraopera-
Claims Project?
tive transfusion of more than four units of blood during
emergency surgery, and who has a contaminated wound is Insight about predictors of adverse outcomes unique to
more likely to have adverse anesthesia outcomes than an anesthesiology is represented by the ASA Closed Claims
ASA class I patient scheduled for minor elective surgery. Project. Initiated in 1984 by the ASA Committee on
Although this extreme example will most likely be correct Professional Liability, the team of investigators and staff
for a specific patient, we again emphasize that anesthe- developed and implemented a precedent-setting study
sia outcome predictions should be based on overall risk that continues today. Medical malpractice insurance
factors for populations. companies were persuaded to open their closed files
and allow confidential examination of the litigation. All
these cases, by definition, involved adverse anesthesia
Anesthesia Indicators
outcomes. They were examined for patterns of causes
Many preoperative and intraoperative risk factors overlap and effects regarding untoward clinical events, usually
between surgery and anesthesia, but traditional outcomes severe and involving at least perceived damage to the
are more dissimilar. Lists of anesthesia indicators or qual- patients that led to lawsuits against anesthesiologists.
ity parameters for anesthesia have been promulgated by A summary of the positive impact of the project, over
various accrediting and regulatory agencies. Functionally, its first 17 years,24 reveals the extensive breadth of the
these lists contain the main adverse outcomes of anesthe- findings.
sia care that practitioners strive to avoid and risk-factor One initial important finding was that respiratory
analysis attempts to predict. Traditional adverse outcomes system adverse events accounted for a significant fraction
caused solely or largely by the anesthesia care and man- of the claims, particularly lawsuits involving death or
ifested at the time of surgery or within 24 to 48 hours brain damage.25 The concept of preoperative predictors
are listed in Table 1.4. Some authors may add failed of adverse anesthesia outcome was not the main focus
regional anesthesia, pneumothorax from central venous as much as causation of the adverse events. It was
catheter placement, corneal abrasion, postspinal anesthe- determined that failure to monitor adequately was the
sia headache, severe bronchospasm, burns, or unintended main reason for the adverse outcomes and that the correct
awareness during general anesthesia to this list. application of capnography and pulse oximetry together
The concept of extended hospital stays, when caused would have prevented 93% of the preventable mishaps.26
by unanticipated anesthesia accidents or untoward events, The landmark initial publication analyzed sudden
is included in some manner on these lists and would be cardiac arrest during spinal anesthesia in healthy pa-
counted statistically as an anesthesia adverse outcome. tients undergoing elective surgery.27 Predisposing factors
Several specific instances of potential predictors of these included intravenous sedation sufficient to stop verbal-
and lesser anesthesia adverse outcomes are discussed ization and the failure to immediately treat the relative
subsequently. A logical inference would conclude that, hypovolemia caused by the spinal-induced sympathetic
in many cases, the NSQIP risk factors in Table 1.3 are blockade. The panoply of subsequent studies and inter-
also relevant as general predictors of potential anesthesia pretations into the mechanisms of adverse anesthesia
adverse outcomes. outcomes from the Closed Claims Project are reviewed
C H A P T E R 1/ A N E S T H E S I A , P E R I O P E R AT I V E M O R TA L I T Y , A N D P R E D I C T O R S O F A D V E R S E O U T C O M E S 11
annually in a Refresher Course Lecture at the ASA Na- been published on this issue.30 The greatest risk for
tional Meeting. This approach consistently provides new perioperative cardiac complications include the usual sus-
insights related to identified predictors of potential ad- pectsrecent myocardial infarction, unstable or severe
verse outcomes and recommends guidelines intended to angina, decompensated CHF, significant dysrhythmias,
help avoid them. heart block, and untreated valvular disease. Intraoperative
risk is highest for major emergency surgery (particularly
in the elderly), aortic surgery and other major and periph-
eral vascular surgery, and prolonged surgery with major
What Are the Major Risk fluid shifts. Many risk-scoring schemes for cardiac status
Categories? exist. One validated method is the Cardiac Anesthesia Risk
Evaluation score,31 which has been offered as a particu-
larly useful tool for cardiac anesthesiologists to accurately
GENERAL CONSIDERATIONS predict the outcome of cardiac surgery.
5. Adverse anesthesia outcome has widely varying defi- 12. Fasting S, Gisvold SE. Serious intraoperative problems
a five-year review of 83,844 anesthetics. Can J Anaesth. 2002;
nitions. All patients anesthetized face some degree of
49:545.
risk. Of these, some exhibit characteristics that have
13. Davis NJ, ed. Anaesthesia related mortality in Australia
been associated with an adverse anesthesia outcome; 19941996. Australian and New Zealand College of Anaes-
these are predictors in populations, not necessarily in thetists, Melbourne, Australia: Capitol Press; 1999.
individuals. 14. Biboulet P, Aubas P, Dubourdieu J, et al. Fatal and non fatal
6. The strongest predictors of adverse surgical outcomes cardiac arrests related to anesthesia. Can J Anaesth. 2001;
include: ASA PS Class; preexisting, significant car- 48:326.
diac, pulmonary (including smoking), renal, or CNS 15. Newland MC, Ellis SJ, Lydiatt CA, et al. Anesthetic-related
disease; emergency surgery; transfusion of four or cardiac arrest and its mortality: A report covering 72,959
more units of blood; contaminated wound; sepsis; anesthetics over 10 years from a US teaching hospital.
and a variety of lab value abnormalities. A logical Anesthesiology. 2002;97:108.
assumption is that the same factors could also be 16. Lunn JN, Devlin HB. Lessons from the confidential enquiry
into perioperative deaths in three NHS regions. Lancet. 1987;
predictors of an adverse anesthesia outcome.
2(8572):1384.
7. Cardiac failure and ischemia, hypertension (some-
17. Irita K, Kawashima Y, Iwao Y, et al. Annual mortality and
what), the nature of the surgery (i.e. cardiovascular morbidity in operating rooms during 2002 and summary of
and abdominal), ASA PS class, advancing age, obe- morbidity and mortality between 1999 and 2002 in Japan:
sity, smoking, and male gender predispose to some A brief review. Masui. 2004;53:320.
adverse anesthesia outcomes. 18. Irita K, Kawashima Y, Morita K, et al. Critical events in the
8. Extended periods of time at profoundly deep levels operating room among 1,440,776 patients with ASA PS 1 for
of unconsciousness are potential predictors of an elective surgery. Masui. 2005;54:939.
adverse anesthesia outcome. 19. American Society of Anesthesiologists. ASA physical status
9. A prospective awareness of the predictors of adverse classification system. Available at: http://www.asahq.org/
anesthesia outcome should help promote measures to clinical/physicalstatus.htm. Accessed June 2, 2006.
prevent precisely those identified adverse outcomes, 20. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia
in surgical mortality. JAMA. 1961;178:261.
thereby further lowering anesthesia mortality.
21. Davenport DL, Bowe EA, Henderson WG, et al. National
surgical quality improvement program (NSQIP) risk factors
REFERENCES can be used to validate American Society of Anesthesiologists
Physical Status Classification (ASA PS) levels. Ann Surg.
1. Pierce EC Jr. The 34th Rovenstine Lecture. 40 years
2006;243:636.
behind the mask: Safety revisited. Anesthesiology. 1996;84:
22. Atherly A, Fink AS, Campbell DC, et al. Evaluating alter-
965.
native risk-adjustment strategies for surgery. Am J Surg.
2. Eichhorn JH, Cooper JB, Cullen DJ, et al. Standards for
2004;188:566.
patient monitoring during anesthesia at Harvard Medical
23. Fink AS, Campbell DA Jr, Mentzer RM Jr, et al. The
School. JAMA. 1986;256:1017.
national surgical quality improvement program in non-
3. Eichhorn JH. Prevention of intraoperative anesthesia acci-
veterans administration hospitals: Initial demonstration of
dents and related severe injury through safety monitoring.
feasibility. Ann Surg. 2002;236:344.
Anesthesiology. 1989;70:572.
24. Domino K. Role of the closed claims project in improving
4. Eichhorn JH. Monitoring standards: Role of monitoring in
outcome. In: Fleisher LA, Prough DS, eds. Problems in
reducing risk of anesthesia. Prob Anesth. 2001;13:430.
5. Cooper JB, Gaba DM. No myth: Anesthesia is a model for anesthesia, Vol. 13. Philadelphia, PA: Lippincott Williams
addressing patient safety (Editorial). Anesthesiology. 2002;97: & Wilkins; 2001:491.
1335. 25. Caplan RA, Posner KL, Ward RJ, et al. Adverse respiratory
6. Domino KB. Malpractice insurance premiums: Greater events in anesthesia: A closed claims analysis. Anesthesiology.
stability for most anesthesiologists. Am Soc Anesthesiol 1990;72:828.
Newsl. 2006;70(6):6. 26. Tinker JH, Dull DL, Caplan RA, et al. Role of monitoring
7. Lagasse RS. Anesthesia safety: Model or myth? A review of devices in prevention of anesthetic mishaps: A closed claims
the published literature and analysis of current original data. analysis. Anesthesiology. 1989;71:541.
Anesthesiology. 2002;97:1609. 27. Caplan RA, Ward RJ, Posner K, et al. Unexpected cardiac
8. Beecher HK, Todd DP. A study of the deaths associated arrest during spinal anesthesia: A closed claims analysis of
with anesthesia and surgery. Based on a study of 599,518 predisposing factors. Anesthesiology. 1988;68:5.
anesthesias in ten institutions 19481952, inclusive. Ann 28. Forrest JB, Rehder K, Cahalan MK, et al. Multicenter study
Surg. 1954;140:2. of general anesthesia. III. Predictors of severe perioperative
9. Cohen MM, Duncan PG, Pope WD, et al. The Canadian four- adverse outcomes. Anesthesiology. 1992;76:3.
centre study of anaesthetic outcomes: II. Can outcomes be 29. Larach MG, Landis JR, Bunn JS, et al. The North American
used to assess the quality of anaesthesia care? Can J Anaesth. Malignant Hyperthermia Registry. Prediction of malignant
1992;399:430. hyperthermia susceptibility in low-risk subjects. An epi-
10. Committee on Quality of Health Care in America of the demiologic investigation of caffeine halothane contracture
Institute of Medicine. To err is human: Building a safer responses. Anesthesiology. 1992;76:16.
health care system. Washington, DC: National Academy Press; 30. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide-
1999:32. line update for perioperative cardiovascular evaluation for
11. Arbous MS, Grobbee DE, van Kleef JW, et al. Mortality noncardiac surgeryexecutive summary: A report of the
associated with anaesthesia: A qualitative analysis to identify American College of Cardiology/American Heart Association
risk factors. Anaesthesia. 2001;56:1141. Task Force on Practice Guidelines (Committee to Update the
14 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
1996 Guidelines on Perioperative Cardiovascular Evaluation 35. The American Society of Anesthesiologists Task Force on
for Noncardiac Surgery). J Am Coll Cardiol. 2002;39:542. Postoperative Management of Obstructive Sleep Apnea.
31. Dupuis JY, Wang F, Nathan H, et al. The cardiac anesthesia Practice guidelines for the perioperative management of pa-
risk evaluation score: A clinically useful predictor of mor- tients with obstructive sleep apnea. Anesthesiology. 2006;104:
tality and morbidity after cardiac surgery. Anesthesiology. 1081.
2001;94:194. 36. Monk TG, Saini V, Weldon BC, et al. Anesthetic management
32. Burkle CM, Walsh MT, Harrison BA, et al. Airway manage- and one-year mortality after noncardiac surgery. Anesth
ment after failure to intubate by direct laryngoscopy: Out- Analg. 2005;100:4.
comes in a large teaching hospital. Can J Anaesth. 2005;52: 37. The American Society of Anesthesiologists Task Force on In-
634. traoperative Awareness. Practice advisory for intraoperative
33. Shiga T, Wajima Z, Inoue T, et al. Predicting difficult intuba- awareness and brain function monitoring. Anesthesiology.
tion in apparently normal patients: A meta-analysis of bed- 2006;104:847.
side screening test performance. Anesthesiology. 2005;103: 38. Aitkenhead AR. Injuries associated with anaesthesia. A global
429. perspective. Br J Anaesth. 2005;95:95.
34. Peterson GN, Domino KB, Caplan RA, et al. Management of
the difficult airway: A closed claims analysis. Anesthesiology.
2005;103:33.
CHAPTER EVALUATION OF ANESTHETIC RISK
A
with renal insufficiency, a history of tran-
sient blindness in the right eye, and chronic
stable angina class II was scheduled to
undergo an elective lumbar vertebrectomy How Is Anesthesia Risk
(L1-4) and posterior spine stabilization. As Quantified and Predicted?
part of the preoperative workup, the patient had a stress
thallium study, which was positive for inducible ischemia
Risk is defined as the possibility of suffering harm or loss
in the area of the left anterior descending (LAD) artery. Af-
(also see Chapters 1 and 3). Why study anesthesia risk?
ter an angiogram showed 80% stenosis in the LAD artery,
First, because do no harmprimum non nocereis a
he underwent an uneventful coronary angioplasty with
fundamental doctrine in the practice of medicine. Second,
placement of a bare metal stent in the LAD. Thereafter,
as part of a medical specialty that has championed
he began to experience symptoms of spinal cord compres-
sion, and surgery was urgently performed 6 weeks after patient safety, our quest as anesthesiologists is to reduce
the angioplasty. Aspirin was continued perioperatively but anesthetic-related mortality and accurately predict the
clopidogrel was discontinued. The surgery lasted 8 hours, likelihood of complications in the perioperative period
and four units of packed red blood cells were transfused for a patient undergoing a particular surgical procedure.
intraoperatively. The patient experienced brief periods Finally, through active interventions, our intent is to
of hypotension during the procedure, which responded reduce the incidence of complications and improve
to fluid therapy and intermittent boluses of ephedrine. outcome by positively modifying the impact of the
He was transferred to an intensive care unit (ICU) for inherent risk factors.
postoperative care after the procedure. His trachea was Since anesthesia is rarely therapeutic and always
successfully extubated the next morning. He complained administered to facilitate a diagnostic or therapeutic
of reduced vision in his left eye along with weakness and intervention, the evaluation of anesthesia risk is, in
reduced sensation in his upper right arm on the first post- fact, the assessment of perioperative risk. Therefore, any
operative day, and was diagnosed with posterior ischemic study of risk during the perioperative period involves
optic neuropathy and brachial plexopathy. On the second understanding the interactions between the patients
postoperative day, he suffered an acute anterior wall my- comorbid conditions and functional status, the surgical
ocardial infarction (MI) caused by stent thrombosis, along technique, surgeons skill, complexity of the surgical
with transient blindness in the right eye, which quickly procedure, and, finally, the anesthetic and postoperative
resolved. The patients hospital course was further com- management.
plicated by a chest infection that necessitated antibiotic It is challenging to accurately quantify and reliably
treatment and noninvasive ventilation. The patient was predict anesthesia risk through a review of the available
discharged on postoperative day 45 with some residual literature for the reasons outlined in Table 2.1.
upper extremity weakness and improving visual acuity. Given these challenges, a systematic and standardized
Six months after the surgery, the visual acuity in the left approach to studying risk is much needed. Our goal
eye had improved to 20/25, with a persistent left temporal should be to accurately predict and modify the risk
field deficit. factors, if possible, and to manage patients appropriately
The questions we, as anesthesiologists, should ask with the state-of-the art technology and best evidence-
ourselves are as follows: Could any of these complications based practices to achieve optimal outcomes for our
have been predicted? Was the preoperative workup patients.
15
16 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
TABLE 2.1 Reasons for Difficulties in Reliable Prediction and Quantification of Anesthetic Risk
1. Differences in operational definitions (e.g., postoperative mortality)
2. Differences in the methodology of the studies conducted
3. Differences in the source of data for the studies (e.g., patient population, geographic location,
type of practice, etc.)
4. Difficulty in accurately predicting the contribution of anesthetic management to the perioperative
outcome
5. Validity of earlier studies becomes questionable as practice changes with the acquisition of
new knowledge. This is particularly applicable for studies conducted over a prolonged period
of time
to act appropriately with existing knowledge (rather than patients. Data collected were American Society of Anes-
lack of knowledge); equipment malfunction; and fatigue.17 thesiologists (ASA) physical status, emergency or elective
At the same time, a Canadian study reported that no surgery, presence of medical comorbid conditions, type of
deaths could be directly attributable to anesthesia.18 anesthesia, type of surgery, and the operating time. The
Four factor groups (patient, surgical, anesthesia, and surgeries were classified by the Hoehn system (commonly
other) were assessed by logistic regression analysis used in Germany) into minor, moderate, or major surg-
to ascertain which variables were predictive of death eries. Logistic regression analysis was used to study the
within a week of surgery. Advanced age, male gender, correlation between the preoperative risk factors and post-
compromised physical status, major surgery, emergency operative complications. Within the study group of 6,304
procedure, procedures performed from 1975 to 1979, patients, 140 died postoperatively, and 1,432 patients de-
intraoperative complications, narcotic techniques, and veloped complications that they survived. The variables
having one or two anesthetic drugs administered were that had the most influence on the risk for postoperative
associated with a higher mortality rate, whereas duration complications were ASA class IV (odd ratio [OR], 4.2), fol-
of anesthesia, experience of the anesthesiologist, and lowed by ASA class III (OR, 2.2), and severity of surgery
inhalation techniques did not affect the mortality rate. (OR, 1.86). The model described by Wolters et al.22 is 96%
The authors concluded that the patient and surgical risk specific (able to predict uncomplicated course correctly
factors were much more important in predicting 7-day predicted in 96% of cases) but has a very low sensitivity
mortality than the anesthesia factors studied.18 of only 16% (ability to correctly predict complications),
The same authors19 reported the results of a follow- giving a positive predictive value of 57% and a nega-
up study in 27,184 anesthetic procedures performed on tive predictive value of 80%. The authors concluded that
inpatients in four major teaching hospitals in Canada. despite studying a large number of variables, they were
Again, no deaths were directly attributed to anesthesia. unable to predict the risk of complications for individ-
Data on these patients were collected, and the outcomes ual patients with any degree of accuracy using statistical
determined for the intraoperative, immediate postanes- methods. They also implied that basing the surgical care
thetic, and postoperative time periods. Logistic regression of a patient on the probability of complications predicted
was used to control for differences in patient populations by a statistical model is not any more accurate or reliable
across all four hospitals. There were large variations (two- than qualified clinical judgment.
to five-fold after case-mix adjustment) in minor outcomes The results of a national confidential inquiry into
across the four hospitals. The rates of major events and perioperative deaths conducted in France were reported in
deaths were similar in three hospitals. One of the hospitals 2004.23 In this report, an expert committee anonymously
had a significantly lower mortality rate but a significantly analyzed the patients charts to determine the cause of
higher rate of all major events (cardiac arrest, MI, and death and its relation to anesthesia. The annual rates of
stroke). Possible reasons for these differences include lack deaths that were totally or partially related to anesthesia
of compliance in recording events, inadequate case-mix were 7 (95% confidence intervals [CI]; 95% CI, 212)
adjustment, differences in interpretation of the variables and 47 (95% CI, 3163) per million, respectively. These
(despite guidelines), and institutional differences in mon- mortality rates increased with patient comorbidity from 4
itoring, charting, and observation protocols. The authors per million in patients of ASA physical status class I to 554
concluded that measuring the quality of care in anesthesia per million for those in ASA class IV. Similarly, the rates
by comparing major outcomes is unsatisfactory because increased with age from 7 per million in patients younger
the contribution of anesthesia to perioperative outcomes than 45 years to 32 per million in older patients. Accidents
is uncertain, and variations may be explained by insti- were of respiratory (38%), cardiac (31%), ischemic (25%),
tutional differences that are beyond the control of the and vascular origin (30%). The main surgical procedures
anesthesiologist. that contributed to the mortality were orthopedic (50%:
Warner et al.20 studied major morbidity and mor- hip fracture, hemorrhagic surgery) and colorectal (24%:
tality occurring within 1 month of ambulatory surgery occlusion, peritonitis).
in 38,598 patients and reported no deaths directly at- Institut National de la Sante et de la Recherche
tributable to anesthesia. Fleisher et al.21 performed a Medicale reported data between 1978 and 1982 on annual
claims analysis of a nationally representative sample of complications associated with anesthesia and found that
Medicare beneficiaries for the years 1994 to 1999. Pa- 76 and 263 per million, respectively, were totally or
tients undergoing 16 different surgical procedures under partially related to anesthesia. The two aforementioned
anesthesia in an outpatient hospital, ambulatory surgery studies indicate that the anesthesia-related mortality
center, and a physician office were studied. The 7-day rate has decreased 10-fold during the subsequent study
mortality rate was 50 per 100,000 in the outpatient hospi- periods, whereas the number of anesthetic procedures
tal, 35 per 100,000 in the office setting, and 25 per 100,000 doubled. In addition, the number of procedures involving
in the ambulatory surgery setting. elderly patients and patients with poor physical status was
Wolters et al.22 prospectively studied 6,304 con- four times higher. The authors attributed these improved
secutive patients admitted to surgery for perioperative results to enhanced safety and adherence to practice
complications in Germany. The objective of their study guidelines published after 1982. They hoped that the rate
was to correlate variables recorded perioperatively with of 1 per 145,000 (at the time of publication) would serve
morbidity and mortality in an attempt to assess the pre- as a basis for systematic analysis of accidents and review
dictive value of the variables for the outcome of individual of future trends.
18 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
Mortality (%)
anesthesia were 7.12 (95% CI, 6.30, 7.94) and 1.00 (95%
CI, 0.88, 1.12) per 10,000 cases, respectively. The average 20
annual mortality within 7 postoperative days due to all
etiologies and that totally attributable to anesthesia were
7.18 (95% CI, 6.22, 8.13) and 0.21 (95% CI, 0.15, 0.27) per
10,000 cases, respectively. Two principal causes of cardiac
10
arrest during anesthesia and surgery due to all etiologies
were massive hemorrhage (31.9%) and surgery-related
factors (30.2%); those totally attributable to anesthesia
were drug overdose or selection error (15.3%) and
serious dysrhythmias (13.9%). This report indicates that 0
preventable human errors caused 53.2% of cardiac arrests,
Standard High risk ICU Ward to ICU
and 22.2% of those deaths in the operating room were
totally attributable to anesthesia. The authors concluded
FIGURE 2.1 Mortality rates for general surgical patients
that the rates of cardiac arrest and death during anesthesia
identified from the * Clinical Accountability, Service Planning
and surgery due to all etiologies, as well as those totally
and Evaluation Health Care Knowledge Systems (CHKS Ltd.)
attributable to anesthesia, were comparable to those of
and ICNARC databases. CHKS database: standard, all patients
other developed countries.
admitted to hospital for a general surgical procedure with an
The National Surgical Quality Improvement Program
overall mortality rate of <5%; high risk, subpopulation of
(NSQIP) is the first national, validated, outcome-based,
patients undergoing a procedure with an overall mortality rate
risk-adjusted, and peer-controlled program for measure-
of 5% or more. ICNARC database: intensive care unit (ICU),
ment and enhancement of the quality of surgical care in
general surgical patients admitted directly to the ICU following
the Veterans Administration (VA) System. Khuri et al.25
surgery; ward to ICU, patients admitted to the ICU following
reported the results of the first prospective multicenter
initial postoperative care on a standard ward. (From Pearse RM,
study that identified the predictors of 30-day and long-
Harrison DA, James P, et al. Identification and characterisation
term survival after major surgery, taking into account the
of the high-risk surgical population in the United Kingdom. Crit
compendium of preoperative, intraoperative, and postop-
Care. 2006;10:R81.)
erative variables that define an episode of surgical care
from the NSQIP database. This study showed that events
in the postoperative period are more important than pre-
operative patient risk factors in determining the survival surgical procedures, of which a high-risk population of
after major surgery in the VA. After correcting for the con- 513,924 patients was identified. These high-risk patients
founding variables collected in the NSQIP, the occurrence accounted for 83.8% of deaths but for only 12.5% of pro-
of a complication within the first 30 days postoperatively, cedures. Despite the high mortality rates, <15% of these
independent of the patients preoperative risk, reduced patients were admitted to the ICU (see Fig. 2.1). The au-
median patient survival by 69% in the total patient study thors concluded that the incidence of postoperative deaths
group. This study did not focus on the role of anesthetic in the United Kingdom has changed little in recent years,
management in the survival rate of patients after major and that most deaths occur in older patients with coex-
surgery. isting medical disease who undergo major surgery. They
The number of deaths identified each year by the recommended that higher ICU resources be provided for
NCEPOD changed little between 1989 and 2003. Suc- these high-risk patients.
cessive reports showed that most deaths occur in older
patients who undergo major surgery and who have severe Regional Anesthesia
coexisting disease.26 A recent analysis identified higher
mortality rates in a UK hospital when compared with a Auroy et al.29 reported the results of a 10-month prospec-
similar institution in the United States.27 Pearse et al.28 tive survey of 8,150 French anesthesiologists on major
extracted data from the Intensive Care National Audit and complications related to regional anesthesia based on
Research Center (ICNARC) database and the Clinical Ac- (a) voluntary reporting during the study period using a
countability, Service Planning and Evaluation Health Care telephone hotline service available 24 hours a day and
Knowledge Systems database on inpatient general surgi- managed by three experts; and, (b) voluntary reporting of
cal procedures and ICU admissions in 94 National Health the number and type of regional anesthesia procedures
Service hospitals in the United Kingdom between January performed using pocket booklets. A total of 487 partici-
1999 and October 2004. There were a total of 4,117,727 pants reported 56 major complications in 158,083 regional
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 19
anesthetized annually in the United States are at risk the prevalence of the outcome of interest. A model is
for or have clinically significant cardiovascular disease. considered perfect when the ROC area is 1.0, useless when
More than 50% of the deaths after surgery are related to it is <0.5 (i.e., under a line of no discrimination), has a
cardiac events.33 In patients who have ischemic heart low accuracy if it is between 0.5 and 0.7, and becomes
disease, the incidence of perioperative MI following useful with an area >0.7.
noncardiac surgery is 5.6%.34 The incidence of cardiac Dupuis et al.43 have developed a simple Cardiac Anes-
adverse events in high-risk patients undergoing major thesia Risk Evaluation (CARE) score based on clinical
vascular surgery is higher at 10% to 18%.35 Raby et al.36 judgment and three clinical variables: comorbid condi-
showed that the odds ratio (OR) for adverse cardiac events tions categorized as controlled or uncontrolled; surgical
increased significantly to 2.7 in patients with preoperative complexity; and urgency of the procedure. In a prospec-
ischemia and to 16 in those with postoperative ischemia. tive observational study at a major university hospital, this
The perioperative morbidity and mortality as a result clinical scoring system (CARE) was compared with the
of this complex disease process can be reduced by a Parsonnet et al., Tuman et al., and Tu et al., multifactorial
comprehensive clinical evaluation, use of reliable cardiac risk indexes,4042 the three existing indexes for prediction
risk indexes, and appropriate risk-modifying measures. of morbidity and mortality after cardiac surgery. The first
2,000 patients enrolled in this study served as a reference
group to determine discrimination of each classification
(CARE, Parsonnet index, Tuman index, and Tu index) us-
Are Risk Indexes Reliable ing ROC curves (see Fig. 2.3). The next 1,548 patients were
used to calibrate the scoring system. Areas under the ROC
Predictors of Outcome? curves were similar for all the indexes. The authors con-
cluded that the CARE score performs as well as the others
One of the earliest and most widely used clinical indexes for outcome prediction after cardiac surgery. They claim
for risk stratification is the ASA physical status classifica- that the CARE score may be more appealing to health care
tion.7 However, the ASA physical status was never intended providers because of its simplicity and focus on evaluation
to be an outcome predictor. Goldman et al.37 have devel- of the patients condition at a much more clinical level.
oped a multifactorial index of cardiac risk for noncardiac Both Gilbert et al.44 and Dupuis et al.43 have shown that
surgical procedures. This method generated much inter- scoring systems based on subjective assessment may have
est and has led to numerous other indexes evaluating a much wider acceptability in clinical use, because they
perioperative risk for both cardiac and noncardiac pro- are easier to use in everyday practice and are as accurate
cedures.3843 Most of these indexes are multifactorial and
developed from observational studies. They are not widely
Specificity
used in everyday clinical practice because of their complex-
1.0
ity and inaccuracy in predicting outcome for individual
patients. Anesthesiologists and other perioperative health
care providers commonly use ASA and New York Heart
Association classification for risk stratification. However, 0.8
these two indexes are physical and functional status classi-
fication tools, which, as noted, were not originally designed
to predict outcome after major surgery. 0.6
1-Sensitivity
ativeevaluationoftheASA,Goldman,Detsky,andCanadian
Cardiovascular Society (CCS) indexes for predicting the
0.4
overall rate for cardiac complications in patients undergo-
ing noncardiac surgery. Cardiac complications occurred
in 6.4% of all patients. The areas under their respective
receiver operating characteristic (ROC) curves for the dif- 0.2 CARE AUC = 0.801 0.027
ferent indexes were 0.625 (ASA), 0.642 (Goldman index), Parsonnet AUC = 0.808 0.024
0.601 (Detsky), and 0.654 (CCS). The authors concluded Tuman AUC = 0.782 0.030
Tu AUC = 0.770 0.028
that existing indexes for predicting cardiac complications 0.0
after noncardiac surgery perform better than chance, but 0.0 0.2 0.4 0.6 0.8 1.0
no index is significantly superior to any other. 1-Specificity
The ROC curve is a graphic representation of the
relation between sensitivity and specificity in a given FIGURE 2.3 Receiver operating characteristic curves obtained
model. An important advantage of ROC analysis over with each risk model for prediction of mortality in the reference
traditional sensitivity and specificity analysis is that the group (n = 2,000 patients). CARE, Cardiac Anesthesia Risk
area under the ROC curve is independent both of the cut- Evaluation; AUC, area under the curve. (From Dupuis JY,
point criteria chosen and the prevalence of the outcome Wang F, Nathan H, et al. The cardiac anesthesia risk evaluation
of interest.45 This independence allows comparisons of score: A clinically useful predictor of mortality and morbidity
ROC curves across study populations in which sensitivity after cardiac surgery. Anesthesiology. 2001;94:194.)
and specificity would be distorted by differences in
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 21
TABLE 2.2 Perioperative Evaluation of Patient with Coronary Artery Disease for Noncardiac
Surgery
Perioperative cardiac failure may only be clinically appar- interest in looking at physiologic capacity (as measured by
ent in the postoperative period when oxygen demand is cardiopulmonary exercise testing [CPET] using exercise
increased. It may also occur independently of both conges- gas exchange data) as an objective method for predict-
tive cardiac failure and MI, although all three conditions ing overall postoperative morbidity and mortality.52,5564
may coexist. Therefore, preoperative evaluation should be The underlying principle is that the perioperative period
focused on the detection of cardiac failure under stress. imposes a degree of metabolic stress on the body similar
Cardiac failure in the elderly is frequently occult, because to that of physical exercise, and similar demands are im-
they often adjust their level of activity to their physical posed on the heart, lungs, and the peripheral circulation
capacity. These patients may therefore be totally asymp- system to support the bodys metabolic needs in both sit-
tomatic at rest for angina and heart failure. However, in uations. This principle assumes that a patients capacity
the postoperative period when oxygen demand exceeds to increase oxygen delivery during exercise may correlate
supply, this adjustment may not be an option. Hence, with his or her capacity to meet and sustain the increased
these patients are prone to cardiovascular complications. metabolic demand on the body during the perioperative
period.
Echocardiography
Echocardiographic determination of resting left ventricu-
lar function may help predict adverse cardiac outcomes CARDIOPULMONARY
perioperatively. However, it does not appear to add impor- EXERCISE TESTING
tant information to the clinical evaluation.53 In patients
undergoing vascular surgery, there seems to be no rela- CPET is an objective method of evaluating cardiopul-
tion between resting preoperative ejection fraction and monary function in which the patient exercises on a
postoperative MI or death.54 Given the limited evidence of bicycle ergometer or, failing that, on a treadmill. During
demonstrable benefit of testing the resting left ventricular exercise, the inspired and expired gases from the patient
function by echocardiography, it is not currently recom- are continuously sampled through a mouthpiece, and both
mended for routine preoperative left ventricular function oxygen uptake and carbon dioxide elimination are com-
evaluation. puted. Cardiac function is evaluated in terms of aerobic
capacity (by gas exchange data), and the pulmonary func-
tion is evaluated by dynamic flow volume loops and V/Q
measurements performed during exercise52 on a contin-
What Are the Requirements for uous basis throughout the duration of the test. The most
important variables related to cardiopulmonary function
Preoperative Assessment? (measured by CPET) that determine or predict postoper-
ative morbidity and mortality are the anaerobic threshold
A comprehensive preoperative evaluation should objec- (AT) and the peak oxygen consumption, VO2 .
tively assess the risks of a patient for cardiac and/or The advantage of using the AT over maximum aerobic
respiratory complications in the perioperative period. The capacity is that the AT does not rely on the motivation of
testing methods used for the evaluation need to be non- the patient, and it occurs well before maximum aerobic ca-
invasive, reliable, reproducible, and valid at submaximal pacity. Therefore, the AT is readily obtained even in elderly
exercise capacity. To meet this end, there is a growing patients because it does not require great physical stress.
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 23
Older et al.65 previously have shown that cardiovascular plasma BNP and troponin may have greater prognostic
mortality was restricted to patients with an AT of utility than measurement of either marker alone.70
<11 mL/minute/kg. The combination of ECG evidence Postoperative plasma BNP and cardiac troponin I
of myocardial ischemia and an AT indicative of moderate levels have been shown to be independent predictors of
cardiac failure (i.e., <11 mL/minute/kg) was associated postoperative cardiac dysfunction after cardiac surgery.71
with a high mortality rate and the worst outcomes. Many This prospective observational study in 92 consecutive pa-
elderly patients have a low AT without demonstrable tients in a university hospital also showed that simultane-
ischemia. Aging results in reduced left ventricular com- ous measurement of BNP and cardiac troponin I improved
pliance, making patients in this age group candidates for the risk assessment of postoperative cardiac dysfunction.
heart failure based on diastolic dysfunction. A low AT in However, the association between BNP levels and 1-year
the elderly, even in the absence of ischemia, is associated outcome was no longer significant after adjusting for
with a high mortality risk. Therefore, solely relying on LVEF. The clinical utility of these newer biomarkers in
symptoms of angina or the results of an ECG may not the management of heart failure and other causes of car-
accurately predict cardiac risk in this set of patients. diac dysfunction merits further study through carefully
CPET is a totally noninvasive test that is quick, cheap, designed prospective investigations. It is likely that, in
easy to perform, and requires no special preparation. It is the future, the use of multiple biomarkers in combination
promoted as being able to objectively evaluate the extent may improve the diagnosis, treatment, and risk stratifi-
of any cardiac failure and/or myocardial ischemia, provide cation of patients with heart failure.72 However, to date
insight into stroke index and the presence of pulmonary there is no data on the use of biomarkers as reliable tools
artery hypertension, and define obstructive and restrictive for predicting postoperative morbidity and mortality or as
lung disease and V/Q inequality better than conventional a measure for preoperative risk stratification.
preoperative respiratory function tests.52 The challenge
still remains to incorporate risk evaluation to positively
change outcome through appropriate perioperative mea-
sures. What Should the Focus Be: Risk
The American Thoracic Society/American College of Stratification or Modification?
Chest Physicians joint statement on the use of CPET is as
follows, Indications for cardiopulmonary exercise testing
as a preoperative assessment tool, is encouraged for lung A risk stratification tool should be accurate, reliable,
cancer surgery, lung volume reduction surgery and evalua- and must add to the pretest knowledge; it should result
tions for both lung and heart transplantation. In a section in interventions that positively affect the outcome and
titled Preoperative Evaluation for Other Procedures it have a favorable riskbenefit ratio. Current conventional
states, Work has shown that cardiopulmonary exercise noninvasive cardiac testing methods tend to have fairly
testing is helpful in objectively assessing the adequacy of good sensitivity but lack in specificity (e.g., exercise ECG
cardiovascular reserve and in predicting cardiovascular has a sensitivity of 68% and a specificity of 77%).73
risk in elderly patients.66 It remains to be seen if this Sensitivity and specificity for SPECT MPI (single photon
preoperative testing will be widely accepted and adopted emission computed tomography myocardial perfusion
in the future. imaging) and exercise echocardiography are 85% and
87%, and 77% and 64%, respectively.74 Given this level
of accuracy, risk stratification may be valuable for
predicting postoperative complications in intermediate-
BRAIN NATRIURETIC risk and high-risk patients. However, the routine use of
PEPTIDE these tests to stratify all patients, or only patients at low
risk for perioperative complications, would not be cost
Strong evidence exists now for the use of brain natriuretic effective and could subject patients to unnecessary tests
peptide (BNP) in the diagnosis of acute heart failure and interventions that have both primary and secondary
and for improving clinical outcomes with a BNP-guided complications.75
approach to heart failure care. In the setting of acute
breathlessness, plasma BNP enabled the discrimination
of cardiac causes from noncardiac causes of dyspnea with NONINVASIVE TESTS
high accuracy.67 In a smaller study of patients referred
for evaluation for cardiac transplantation, plasma NT-pro- Current noninvasive tests, such as exercise ECG and
BNP was able to predict the combined endpoint of death DSE, were originally designed as tools to measure the
and the need for transplantation better than several key intermediate and long-term prognosis of patients with
prognostic indicators, including left ventricular ejection coronary artery disease in a nonsurgical setting. Use
fraction (LVEF), peak oxygen consumption (VO2 ), and the of these tests in the perioperative period, therefore,
heart failure survival score.68 In patients with heart failure may result in low specificity for predicting postoperative
having an LVEF of 45%, the combination of plasma complications in the surgical population, especially if the
BNP and percentage of maximal predicted VO2 attained patient is taking risk-modifying medications. Boersma
on cardiopulmonary testing predicted cardiovascular et al.76 concluded that the additional predictive value of
mortality similarly.69 The combined measurement of DSE is limited in clinically low-risk patients receiving
24 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
-blockers. In clinical practice, DSE may be avoided in a no role for prophylactic percutaneous coronary interven-
large number of patients (with low risk) who can proceed tions before noncardiac surgery if there is no other reason
safely to surgery without delay. In clinically intermediate for the intervention (i.e., the intervention would not be
or high-risk patients receiving -blockers, however, DSE done if the patient was in a nonsurgical setting).
may help identify those on whom surgery can still be
performed and on whom cardiac revascularization should
be considered.
Surprisingly, multiple studies are evaluating risk PERIOPERATIVE -BLOCKER
stratification with noninvasive stress testing when no THERAPY
randomized controlled trials (RCTs) support revascular-
ization before noncardiac surgery. On the contrary, the Given the ineffectiveness of prophylactic revascularization
strategy of catheterization before surgery in patients with in reducing cardiac adverse events before noncardiac
a noninvasive test result positive for inducible ischemia re- surgery, efforts at risk stratification in low-risk groups
sulted in higher morbidity and mortality rates and costs.77 have been questioned. In the last 20 to 30 years, clinical
Stable patients who have undergone coronary artery by- studies have shown that -blockers may decrease the risk
pass grafting (CABG) surgery within the last 5 years, of cardiac-related complications in patients who undergo
and patients undergoing low-risk noncardiac surgery are noncardiac surgery. Prys-Roberts et al.83 conducted one
unlikely to benefit from revascularization and further non- of the earliest studies with -blockers. They showed
invasive cardiac testing, because they have a very low that practolol-treated patients had a marked reduction
mortality rate (<1%).78 in dysrhythmias and myocardial ischemia (4% vs. 38%)
compared with historical controls. Subsequently Stone
et al.84 in a prospective, randomized, nondouble-blinded
study, reported a significant decrease in the incidence of
PERCUTANEOUS myocardial ischemia in mild hypertensive patients with a
REVASCULARIZATION single dose of -blockers compared with the control group.
Mangano et al.85 and Poldermans et al.86 performed
Percutaneous revascularization before noncardiac surgery randomized, double-blind, placebo-controlled trials on
has not been proved to be beneficial in RCTs. A higher rate perioperative -adrenergic blockade. Although Mangano
of adverse perioperative cardiac events has been demon- et al.85 reported a statistically significant reduction in
strated if the noncardiac surgery was performed within 90 mortality for patients given atenolol when compared
days of angioplasty. In a retrospective study of 686 patients with placebo in the perioperative setting, there was no
who underwent percutaneous transluminal coronary an- statistically significant difference between the two groups
gioplasty (PTCA) before noncardiac surgery, patients who when deaths that occurred during the hospital stay were
had PTCA had twice the rate of adverse cardiac outcomes included in the analysis (while the patients were receiving
as healthy subjects, seven times the rate of angina, almost the intervention).
four times the rate of MI, and twice the rate of conges- Poldermans et al.86 discussed 112 patients undergoing
tive heart failure (CHF). Twenty-six percent of the patients vascular surgery, selected from 173 who had positive DSE
who underwent PTCA <90 days before noncardiac surgery test results, out of 846 patients identified at screening to
had adverse cardiac outcomes. The OR of adverse cardiac have one or more cardiac risk factors. Although this study
outcome, angina, CHF, and MI in patients undergoing demonstrated that -blockade therapy with metoprolol
PTCA <90 days before noncardiac surgery compared with reduced mortality following noncardiac surgery in those
healthy subjects were 2.8, 26.0, 2.4, and 34.0, respectively. with inducible ischemia on DSE, the generalization of this
Patients who underwent PTCA within 90 days of non- result is limited because of the large difference between
cardiac surgery suffered twice the rate of perioperative the numbers of patients selected and those finally included
MI compared with patients with uncorrected coronary in the study. Lee87 reported that -blocker therapy may
artery disease.79 Mason et al.77 and Fleisher and Tuman80 reduce the need for additional tests and revascularization
found that patients with coronary artery disease under- procedures, thereby reducing the cost of care. However,
going vascular surgery without coronary intervention had wider use of this therapy will be better supported if
better outcomes than patients who were revascularized. findings from existing studies are replicated in large
For nonvascular surgery patients, the risk-to-benefit randomized trials.
ratio of coronary revascularization is probably even Given the usefulness of a perioperative -blockade to
poorer, owing to the lower baseline risks of adverse reduce cardiac risk, some authors wonder whether risk
cardiac outcomes. Kaluza et al.81 reported that among stratification is still necessary.8890 However, current data
40 patients who had noncardiac surgery within 6 weeks do not support the idea that -blockers alone will reduce
of successful coronary stent placement, 8 died, 7 had a the risk of postoperative cardiac events below thresholds
nonfatal MI, and 11 had major bleeding problems. Mc- suggested by the American College of Physicians (ACP)46
Fadden et al.82 suggested that the problems related to or the American Heart Association/American College of
late stent thrombosis in patients who undergo noncardiac Cardiology risk stratification guidelines.47
surgery within 6 weeks of stent placement may be due to Auerbach and Goldman91 from their review of
preoperative discontinuation of antiplatelet medications. five RCTs looking into the efficiency of perioperative
Therefore, on the basis of the current knowledge, there is -blockade in reducing myocardial ischemia, MI, and
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 25
cardiac or all-cause mortality, concluded that despite the 0.58 (95% CI, 0.50 to 0.67), respectively. The authors con-
heterogeneity of trials, -blockade therapy may be use- cluded that perioperative -blocker therapy is associated
ful in preventing perioperative cardiac morbidity. The with a reduced risk of in-hospital death only among high-
authors noted that because no randomized studies have risk patients undergoing major noncardiac surgery (see
been done to date assessing the impact of such therapy Fig. 2.4). Therefore, patient safety may be enhanced by
in the general population of patients undergoing surgery, increasing the use of -blockers only in high-risk patients.
little direct evidence describes the impact of -blockers
in average patients, such as those who have stable coro- Recommendation
nary disease and are undergoing elective surgery. The
investigators recommend further studies to determine the The following is the recommendation of Auerbach and
optimal duration of therapy, to identify populations of Goldman91 from their review on the perioperative use
patients in which -blocker use is cost effective, and to of -blockade to reduce cardiac risk. Currently, it is
develop new perioperative risk management algorithms considered the best, evidence-based, practical approach.
that reflect the impact of -blockers on patient outcomes. The initial consideration should include risk stratification
Devereaux et al.92 performed a systematic review and according to clinical criteria.39 The first step in risk
meta-analysis to determine the effect of perioperative stratification is to identify patients at lowest risk (those
-blocker treatment in patients undergoing noncardiac whose estimated risk for perioperative cardiac events is
surgery. The authors included 22 RCTs with 2,437 pa- <1% without -blockers [RCRI score of 0]) and patients
tients, which evaluated -blocker treatment in patients at highest risk (those whose estimated risk is >10% [RCRI
having noncardiac surgery in the analysis. Periopera- score of 3 or higher]). -Blockers in patients at low risk
tive outcomes within 30 days of surgery included total impart little absolute benefit, and surgery can proceed
mortality, cardiovascular mortality, nonfatal MI, nonfatal without this medication. In contrast, patients at highest
cardiac arrest, nonfatal stroke, CHF, hypotension need- risk require additional risk stratification using noninvasive
ing treatment, bradycardia needing treatment, and bron- or invasive testing. As described in the preceding text,
chospasm. The perioperative administration of -blockers preoperative revascularization is not useful, except in
did not show any statistically significant beneficial effects patients with an indication for these procedures in the
on any of the individual outcomes. The only nominally sta- absence of the planned surgical procedure. The authors
tistically significant beneficial relative risk was 0.44 (95% recommend noninvasive testing only in high-risk patients
CI, 0.20 to 0.97; 99% CI, 0.16 to 1.24) for the composite and in moderate-risk patients whose exercise capacity
outcome of cardiovascular mortality, nonfatal MI, and cannot be determined by their history.
nonfatal cardiac arrest. The individual safety outcomes Patients who are at high risk and have negative nonin-
in patients treated with perioperative -blockers showed vasive testing results and those at intermediate risk (RCRI
a relative risk for bradycardia needing treatment of 2.27 score of 12) should begin taking a -blocker if they are
(95% CI, 1.53 to 3.36; 99% CI, 1.36 to 3.80) and a nomi- not taking one long-term. Optimally, medications should
nally statistically significant relative risk for hypotension be started before hospitalization and, if possible, as long
needing treatment of 1.27 (95% CI, 1.04 to 1.56; 99% CI, as 30 days before surgery. This period will allow adequate
0.97 to 1.66). The authors concluded thus: The evidence titration of the medication to the target heart rate. Patients
that perioperative -blockers reduce major cardiovascular receiving long-term -blockers should have their dose eval-
events is encouraging but too unreliable to allow definitive uated and adjusted appropriately. Dose titration up to
conclusions to be drawn. the induction of anesthesia may be performed with intra-
More recently, Lindenauer et al.93 conducted a retro- venous atenolol. Postoperatively, oral -blocker use should
spective cohort study of patients 18 years of age or older be restarted as soon as possible, with intravenous atenolol
who underwent major noncardiac surgery at 329 hospi- used for stable patients who are unable to take medications
tals throughout the United States over a 2-year period. orally. Unstable patients should receive a short-acting, in-
The authors used propensity-score matching to adjust for travenous -blocker such as esmolol until they are able
differences between patients who received perioperative to tolerate longer-acting oral medications. The transition
-blockers and those who did not receive such therapy to oral medications should include an overlap with intra-
and compared in-hospital mortality using multivariable venous medications to maintain a target heart rate. Oral
logistic modeling. Of 782,969 patients, 663,635 (85%) had -blocker use should be continued at least through hos-
no recorded contraindications to -blockers. A total of pitalization and up to 1 month postoperatively, when a
122,338 patients (18%) received such treatment during gradual reduction in the dose can be initiated in patients
the first 2 hospital days, including 14% of patients with without an indication for long-term therapy.
a revised cardiac risk index (RCRI) score of 0, and 44% Still, there are a few unanswered questions: Which
of patients with a score of 4 or higher. The relation be- patients should receive -blockers perioperatively? Which
tween perioperative -blocker treatment and the risk of -blocker should be used, at what dose, and for how
death varied directly with cardiac risk; among the 580,665 long? Do patients on chronic -blocker therapy need extra
patients with an RCRI score of 0 or 1, treatment was dosing?
associated with no benefit and possible harm, whereas To answer some of these questions, the Perioperative
among the patients with an RCRI score of 2, 3, or 4 or Ischemic Evaluation (POISE) trial was initiated.94 The
higher, the adjusted OR for death in the hospital were 0.88 POISE trial is a blinded, randomized, and controlled
(95% CI, 0.80 to 0.98), 0.71 (95% CI, 0.63 to 0.80), and trial of controlled-release metoprolol versus placebo in
26 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
Propensity-matched cohort
FIGURE 2.4 Adjusted odds ratio (OR) for in-hospital death associated with perioperative -blocker
therapy among patients undergoing major noncardiac surgery, according to the revised cardiac risk
index (RCRI) score and the presence of other risk factors in the propensity-matched cohort and the
entire study cohort. Open boxes represent patient subgroups within the listed RCRI category. (From
Lindenauer PK, Pekow P, Wang K, et al. Perioperative beta-blocker therapy and mortality after
major noncardiac surgery. N Engl J Med. 2005;353:349.)
10,000 patients at risk for a perioperative cardiovascular causing a reduction in the norepinephrine levels in
event who are undergoing noncardiac surgery. Patients patients undergoing surgery.95,96 -Blockers inhibit the
in this trial will receive the study drug (metoprolol) 2 peripheral effects of catecholamines and thereby block
to 4 hours before surgery and subsequently for 30 days. the effects of the stress response, whereas 2 -agonists
The primary outcome is a composite of cardiovascular inhibit the release of catecholamines, thereby blocking
death, nonfatal MI, and nonfatal cardiac arrest at 30 days. the pathway for the stress response. Clonidine has some
Patients will also be followed up for events at 1 year. advantages over perioperative -blockers because it has a
To date, the POISE trial has recruited more than 6,300 lower risk of bronchospasm in asthmatics, and it comes in
patients from 182 centers in 21 countries. The patients a transcutaneous form that can be used in patients who are
mean age is 69 years; 63% are men; 43% have a history of unable to take oral medications. It is also a viable option
coronary artery disease; 43% have a history of peripheral for patients with conduction abnormalities or those who
arterial disease, and 30% have diabetes. Participants have cannot tolerate -blocker therapy. In one study of 297
undergone vascular (42%), intra-abdominal (23%), or patients undergoing vascular surgery, clonidine-treated
orthopedic (19%) surgery. patients had fewer episodes of ischemia.97 Mivazerol,
an 2 -agonist that reduces postganglionic norepinephrine
availability and spinal efferent sympathetic output, has
also been shown to reduce the incidence of perioperative
Do Other Medications Modify ischemia.98 A large randomized trial of 1,897 patients
undergoing noncardiac surgery produced mixed results.99
Perioperative Cardiac Risk? In this study, mivazerol had no statistically significant
effect on all-cause mortality or MI, but cardiac mortality
was reduced by half (a relative risk of events among
2 -AGONISTS treated patients of 0.50; 95% CI, 0.250.96). In a planned
subgroup analysis of the same data, a more marked
Clonidine lowers blood pressure and heart ratekey impact was observed among patients undergoing vascular
factors in preventing myocardial ischemiaprobably by surgery, in whom the relative risk of postoperative MI and
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 27
death among treated patients was 0.67 (95% CI, 0.450.98) cardiovascular events, defined as cardiac death, nonfatal
and the relative risk for cardiac death was 0.32 (95% CI, MI, stroke, or unstable angina pectoris. After 6 months,
0.120.76).99 These findings attest to the role of adrenergic the incidence of cardiac events was more than three
blockade in preventing cardiac events. times higher with placebo (26%) than with atorvastatin
Wallace et al.100 conducted a prospective, random- (8%; p = 0.031). ONeil-Callahan et al.110 reported a ret-
ized, placebo-controlled clinical trial to assess the re- rospective cohort study, in which they evaluated 1,163
duction of mortality with perioperative clonidine in hospitalizations of 997 patients who underwent carotid,
noncardiac surgery patients. The incidence of periopera- aortic, or lowerextremity vascular surgery in a 2-year pe-
tive myocardial ischemia (defined by 1 mm of ST-segment riod; in this cohort were 571 statin users. Perioperative
depression lasting at least 1 minute during Holter ECG outcomes studied in this trial were death, MI, ischemia,
monitoring) was reduced in patients treated with cloni- CHF, and ventricular tachyarrhythmias. After adjustment
dine compared with those receiving placebo preopera- for other significant predictors of perioperative compli-
tively on the day of surgery and on postoperative days 1 cations, statins significantly reduced the occurrence of
through 3. Patients assigned to the clonidine group had cardiac events (OR, 0.52; p = 0.001). Investigators noted
a lower 30-day mortality rate (1 in 125 [0.8%] compared that the highly protective effects of statins in this study
with 4 in 65 [6.2%] in the placebo group; p = 0.048). In were largely accounted for by reductions in myocar-
long-term followup for up to 2 years, the incidence of dial ischemia and CHF. However, these results must be
postoperative death was reduced in patients who received interpreted in light of the retrospective nature of this
clonidine compared with those who had placebo (rela- study.
tive risk = 0.43; 95% CI, 0.210.89; p = 0.035). Wallace Kennedy et al.111 and McGirt et al.112 reported two
et al.100 concluded that the perioperative administration retrospective cohort studies of statin use in patients un-
of clonidine for 4 days to patients at risk for coronary dergoing carotid endartectomies (CEA). In these studies,
artery disease significantly reduces the incidence of peri- patients significantly benefited from statin therapy with
operative myocardial ischemia and postoperative death. respect to mortality (adjusted ORs 0.25 and 0.20, p < 0.05)
Current evidence suggests that the rates of myocar- and stroke (adjusted ORs 0.55 and 0.35, p < 0.05), respec-
dial ischemia and postoperative mortality are identical tively. In both studies, a lower insignificant incidence of
in patients managed with atenolol85,101 and those man- MIs was observed in statin users compared with nonusers.
aged with clonidine. The current belief is that initiation of These insignificant results were probably due to the low
anti-ischemia therapy before surgery results in better out- incidence of events in low-risk patients. Biccard et al.113
comes from the cardiac standpoint.86,100 Presently, among reviewed 11 papers on perioperative statin use and con-
the available 2 -adrenoceptor agonist medications, a re- cluded that the use of statins improved postoperative
duction in myocardial ischemia and improved outcomes cardiac outcome.
for patients at risk of cardiac events has been documented
only for clonidine. The only available data for dexmedeto- Adverse Effects
midine showed that perioperative infusion appeared to
benefit the perioperative hemodynamic management of A major deterrent to perioperative statin therapy has been
surgical patients undergoing vascular surgery.102 Further the fear of statin-induced myopathy and rhabdomyolysis.
studies are needed to examine whether dexmedetomidine, Other factors that could influence the risk of statin-
like clonidine, can reduce the incidence of myocardial induced myopathy are impaired renal function after major
ischemia and postoperative mortality.103 surgery and multiple drug use during anesthesia.114,115
Failure to detect statin-induced myopathy early in the
perioperative course could lead to continuous statin use
and the subsequent development of rhabdomyolysis and
PERIOPERATIVE STATIN acute renal failure.
THERAPY Schouten et al.116 retrospectively studied 981 consec-
utive patients undergoing vascular surgery from 1998 until
Besides lowering cholesterol, statins have significant 2004. Forty-four patients with an elevated cholesterol level
pleiotropic effects. These include the increased expression at preoperative screening were started on statin therapy
of endothelial nitric oxide synthase; generation of reactive (acute statin users). The average duration of statin use
oxygen species; reduction of endothelin-1 production; im- before surgery was 40 days. A total of 182 patients already
proved thrombogenic profile; reduction in inflammation taking statin therapy (long-term statin users) continued
through reduced expression of cytokines, chemokines, their statin use. Patients were monitored for symptoms,
and adhesion molecules; lowering of C-reactive protein and blood samples were taken on days 1, 3, and 7 and
levels; and the general inhibition of other aspects of the at discharge. After correcting for cardiac and clinical risk
atherosclerotic process.104106 These effects are believed factors for myopathy, length of surgery was the only
to be the cause of atherosclerotic plaque stabilization independent predictor of myopathy. No case of rhab-
during surgical procedures.107,108 domyolysis was observed in the study period. Therefore,
Durazzo et al.109 randomized 100 patients to treat- despite the lack of large randomized clinical trials, it
ment with either 20 mg atorvastatin or placebo. Patients seems justifiable to prescribe statins to high-risk patients
received treatment for a total of 45 days, starting at least during the perioperative period considering the low in-
2 weeks before surgery. The endpoints in this trial were cidence of statin-induced myopathy and rhabdomyolysis.
28 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
The potential benefits of perioperative statin use seem to complications than those who smoked less. Patients who
outweigh the potential hazards.117 continue to smoke have a higher risk than those who have
stopped for >8 weeks. Furthermore, patients who stopped
smoking for >6 months effectively reduced their risks
THE FUTURE to rates that were comparable to those who had never
smoked. However, smoking cessation for <8 weeks before
To assess the impact of pharmaceutical strategies surgery has not been shown to decrease postoperative
(-blockers, statins, or both) in preventing perioperative morbidity. A case-controlled cohort study showed a higher
cardiac complications, a large prospective, randomized rate of postoperative pulmonary complications in smokers
trial (the Dutch Echocardiographic Cardiac Risk Evalu- who had stopped or reduced smoking within 8 weeks of
ation Applying Stress Echo-IV) was initiated.118 Initial surgery than in those who continued to smoke.124 This
recruitment of patients started in 2004, and the results are finding illustrates that recent smoking cessation may,
expected in 2008. paradoxically, increase short-term risk, owing to transient
Current evidence indicates that accurate risk stratifi- increased mucus production resulting from improvement
cation and subsequent aggressive therapy with -blockers, of mucociliary activity and decreased coughing from
2 -agonists, and statins in the appropriate patient pop- reduced bronchial irritation.
ulation could significantly minimize risk and reduce
morbidity and mortality from cardiovascular complica- Obesity
tions.
Pulmonary effects related to obesity are secondary to de-
creased chest wall compliance, increased work of breath-
ing, decreased inspiratory and expiratory reserve volumes,
What Factors Determine decreased functional residual capacity, increased oxygen
consumption and carbon dioxide production, secondary
Perioperative Pulmonary Risk? polycythemia, pulmonary hypertension, and right heart
failure. Furthermore, obesity is associated with a higher
Despite the emphasis on cardiac complications, pul- incidence of obstructive sleep apnea. One would expect
monary complications in the perioperative period are that these changes would result in an increased incidence
not only more frequent, but also require longer and of pulmonary complications. However, studies have not
costlier hospital stays for patients undergoing abdomi- shown obesity to be a consistent independent risk fac-
nal surgeries.119 Risk factors for perioperative pulmonary tor for increased postoperative pulmonary risk.125 Flier
complications may be divided into factors related to the and Knape126 reported the results of a literature search
patient, the surgical procedure, and anesthesia. (213 articles from 1966 to 2004) on postoperative pul-
monary complications of abdominal surgery in morbidly
obese patients. The authors showed that for morbidly
PATIENT-RELATED FACTORS obese patients who undergo abdominal surgery under
general anesthesia, the likelihood of developing atelecta-
Asthma sis is 10.4% (p < 0.001). The likelihood of developing both
atelectasis or pneumonia was 29.3%, with an adjusted OR
Preexisting chronic obstructive pulmonary disease of 2.82 (95% CI, 1.664.78; p = 0.0001).
(COPD) is one of the most significant patient-related Morbidly obese patients are typically considered at
risk factors for perioperative pulmonary complications. high risk for perioperative complications and often un-
Mohr and Jett120 reported a 26% rate of respiratory com- dergo extensive preoperative testing. Ramaswamy et al.127
plications in patients with COPD compared with 8% in analyzed prospectively collected data from 193 patients
controls. Kroenke et al.121 found that patients with severe undergoing weight loss surgery between November 2000
COPD were six times more likely to suffer a major post- and November 2002. Preoperative chest radiographic ex-
operative pulmonary complication following abdominal amination, pulmonary function tests (PFTs), noninvasive
or thoracic surgery than those without COPD. Although, cardiac testing, and blood work were performed routinely
symptomatic and uncontrolled asthma is a significant in all patients. Abnormalities were detected on 4% of chest
predictor of postoperative complications, well controlled radiograph examinations and 15% of ECGs, none of which
asthma, indicated by the absence of symptoms and an required preoperative intervention. Spirometry result was
FEV1 of >80% of predicted, is not associated with higher abnormal in 21% of patients. Logistic regression analysis
than expected pulmonary risk.122 identified preexisting asthma as predictive of obstructive
physiology (OR, 3.3; 95% CI, 1.28.9), and body mass
Smoking index as predictive of restrictive physiology (OR, 1.1; 95%
CI, 1.011.2). Arterial blood gas analysis identified only
Patients who smoke have an increased risk for pulmonary one case of severe hypoxemia requiring intervention. Mild
complications compared with nonsmokers, even if they hypoxemia was associated with increasing age (OR, 14.5;
have no clinical evidence of chronic lung disease. Warner 95% CI, 1.8114). Echocardiographic abnormalities were
et al.123 reported that cigarette smokers with a >20- demonstrated in four patients (2%); previous history of
pack-year history had a higher incidence of pulmonary cardiac disease was the only risk factor (OR, 14.5; 95%
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 29
CI, 1.8114) in the presence of echocardiographic ab- complications than transverse incisions.131 Laparoscopic
normalities. Age, body mass index, and history of asthma cholecystectomy has a lower incidence of postoperative
were associated with abnormal PFTs and previous cardiac pulmonary complications than open procedures. Laparo-
disease with abnormal cardiac testing. The authors con- scopic techniques may reduce pulmonary risks by causing
cluded that chest radiograph, arterial blood gases, PFTs, less pain and less disruption of diaphragmatic activity,
and routine, noninvasive cardiac testing are not manda- thereby leading to enhanced recovery in the postoperative
tory; each of these tests can be used selectively on the period.132 Procedures >3 to 4 hours in duration are as-
basis of medical history. sociated with a higher risk of pulmonary complications.
In one study involving 520 patients, 40% of individu-
als whose surgeries lasted >4 hours had postoperative
Obstructive Sleep Apnea
courses complicated by pneumonia, compared with 8% of
Obstructive sleep apnea is characterized by frequent patients with surgeries lasting <2 hours.133
episodes of apnea during sleep, heavy snoring, and day- Periera et al.134 performed a prospective risk as-
time sleepiness. Patients with sleep apnea are at increased sessment of postoperative pulmonary complications in
risk of airway problems, severe hypoxemia and hyper- 408 patients undergoing upper abdominal surgery. The
capnia, pulmonary hypertension, and right heart failure patients were prospectively analyzed during the pre-
during the perioperative period. Recent ASA guidelines operative period and followed up postoperatively for
on the management of patients with obstructive sleep ap- pulmonary complications. Patient characteristics (based
nea recommend a systematic approach to preoperative, on clinical and physical evaluations), related diseases,
intraoperative, and postoperative care of patients in this smoking habits, and duration of surgery on the incidence
high-risk group.128 of postoperative pulmonary complications were studied.
The authors reported an overall postoperative pulmonary
complication rate of 14%. Significant predictors of post-
Aging operative pulmonary complications in univariate analyses
were age >50 years, smoking, chronic pulmonary disease
Aging is associated with physiologic changes affecting or respiratory symptoms at the time of evaluation, dura-
multiple organ systems. Yet, significant individual varia- tion of surgery >210 minutes, and comorbid conditions.
tion occurs in the rate of decline, and biologic age does not In a logistic regression analysis, the statistically signifi-
always correlate with chronologic age. In the past, studies cant predictors were the presence of chronic pulmonary
suggested an increased risk of pulmonary complications disease, surgery lasting >210 minutes, and comorbidity.
with advanced age.129 However, more recent studies have Patients with all three risk factors were three times more
confirmed that age alone is not an independent predictor likely to develop a postoperative pulmonary complica-
of risk for perioperative pulmonary complications. Cerfo- tion than were patients without any of these risk factors
lio and Bryant130 performed a case-control study of 6,450 (p < 0.001).
patients with nonsmall cell lung cancer who underwent
complete resection over a 5-year period. Patients 70 years
or older, 75 years or older, and 80 years or older were
matched 1:1 to younger controls for stage, pulmonary
ANESTHESIA-RELATED
function, performance status, and type of pulmonary re- FACTORS
section. No significant differences in length of hospital
Past studies yielded conflicting results as to which type of
stay, major morbidity, or operative mortality were iden-
anesthesia is associated with a higher risk of pulmonary
tified between any of the elderly groups and the younger
complications. To help clarify this debate, a compre-
controls. Therefore, elderly patients should not be de-
hensive review was performed evaluating the results of
nied major surgery based solely on chronologic age, as
141 trials. According to this review, general anesthesia
their short-term risks and long-term survival are similar
leads to a higher risk of pulmonary complications than
to those of younger patients.
central neuraxial anesthesia such as epidural or spinal
anesthesia.135 During general anesthesia, neuromuscular
blocking drugs are often used to facilitate surgery. No dif-
SURGICAL FACTORS ference was seen in the rates of postoperative pulmonary
complications between intermediate-acting drugs such
The site of surgery has been found to be the most im- as atracurium and vecuronium. However, pancuronium
portant factor in predicting the risk of postoperative was associated with a higher incidence of residual post-
pulmonary complications. The closer the surgical inci- operative neuromuscular blockade compared with the
sion is to the diaphragm, the higher is the incidence of shorter-acting drugs,136 probably due to a higher inci-
complications. Therefore, thoracic and upper abdominal dence of postoperative residual curarization associated
surgeries showed the highest rates of pulmonary compli- with long-acting neuromuscular blockers. Patients with
cations when compared with all other surgical procedures. residual neuromuscular blockade who received pancuro-
These rates were in the range of 19% to 59% for thoracic nium were three times more likely to develop pulmonary
surgery, 17% to 76% for upper abdominal surgery, and complications postoperatively. This finding was largely
0% to 5% for lower abdominal surgeries.120 Vertical la- due to the lack of objective monitoring of neuromuscu-
parotomy incisions are associated with more pulmonary lar function; inability to recognize residual blockade with
30 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
simple monitors or by clinical testing; and inadequate carbon monoxide [DLCO]) have the greatest utility in
or lack of antagonism of residual neuromuscular block- documenting physiologic operability in low-risk patients.
ade. Further, long-acting neuromuscular blocking agents Other diagnostic modalities, such as CPET and/or split
are more difficult to reverse than those of intermediate function assessment by quantitative lung scintigraphy,
durations. are often necessary to more accurately assess moderate
to high-risk patients for lung resection, volume reduction,
and lung transplantation surgeries.57,59,141
Although proponents advocate both approaches,55,142
How Can Risks for Pulmonary current evidence suggests that CPET and the measure-
ment of V O2 peak (especially when expressed as a per-
Complications Be Assessed and centage of the predicted value) appear to be particularly
Managed? useful in predicting postoperative pulmonary complica-
tions. A VO2 peak <50% to 60% predicted is associated
with higher morbidity and mortality rates after lung re-
section.59,63,64 Preoperative CPET and split function study
ASSESSMENT results may be used in tandem to predict the risk of postop-
erative pulmonary complications and functional capacity;
The first step in preoperative assessment that helps to this may be most beneficial to borderline patients who
identify patients at risk for perioperative pulmonary com- might otherwise be precluded from surgery.57,62,63 In ad-
plications is a complete history and physical examination. dition, CPET permits the detection of clinically occult
Further laboratory tests (chest radiography, PFTs, arte- heart disease and provides a more reliable estimate of
rial blood gas analysis and exercise testing) should be functional capacity postoperatively than PFTs, which rou-
used in select patients on the basis of clinical evaluation. tinely overestimate functional loss after lung resection.142
In the large majority of cases, preoperative chest radio- Wyser et al.143 reported improved outcomes for pa-
graphs provide very little help in identifying risk in healthy tients undergoing lung resection surgeries when a func-
patients. A meta-analysis of studies evaluating the value tional preoperative algorithm for pulmonary workup was
of routine preoperative chest radiographs confirmed this utilized. This algorithm incorporated cardiac history, in-
fact.137 Preoperative PFTs alone are not a more sensi- cluding an ECG, and the three parametersFEV1 , DLCO,
tive predictor of occult pulmonary disease than a careful and maximal oxygen uptake (V O2 max)as well as their re-
history and physical examination.138 Their use as a preop- spective predicted postoperative values (FEV1 -ppo, DLco-
erative screen for pulmonary disease in patients without ppo, and V O2 max-ppo) calculated on the basis of radionu-
relevant symptoms or physical findings is inappropriate. clide perfusion scans. This algorithm resulted in a 50%
PFTs should be considered in patients with COPD or reduction in mortality compared with the authors pre-
asthma in whom clinical evaluation cannot determine viously published series. They concluded that adherence
the level of airflow obstruction. In these cases, preopera- to an algorithm resulted in a very low complication rate
tive spirometry may guide more aggressive preoperative (morbidity and mortality), and they excluded more rigor-
management. ous patient selection as a bias for the improved results.
Lung Resection
MANAGEMENT
Lung resection results in more severe impairment of pul-
monary function than other types of surgery. The severity Lawrence et al.144 systematically reviewed the English-
of postoperative pulmonary complications is related to language literature on interventions to prevent postoper-
the extent and functional status of the segment resected ative pulmonary complications after noncardiothoracic
and the amount and function of the remaining lung tissue. surgery from January 1980 through June 2005. The
In the immediate postoperative period, the severity of the authors concluded that there was good evidence (two
surgical trauma sustained to the remaining lung during systematic reviews and five additional RCTs) to indicate
resection also plays a very important role in determining that lung expansion interventions (e.g., incentive spirom-
risks of postoperative respiratory complications. One mul- etry, deep breathing exercises, and continuous positive
ticenter study reported a patient mortality rate of 3.8% airway pressure) reduce pulmonary risk. There was fair
following wedge resection, 4.2% following lobectomy, and evidence (two meta-analyses) to suggest that selective use
11.6% following pneumonectomy.139 of nasal gastric tubes after abdominal surgery and short-
In patients considered high risk (FEV1 <2 L) who acting intraoperative neuromuscular blocking agents (one
are undergoing pneumonectomy, split-perfusion lung RCT) reduce risk.
scanning should be performed to assess the relative The evidence was conflicting or insufficient for pre-
contribution of each lung to total ventilation and perfusion operative smoking cessation (one RCT), epidural anes-
in the postoperative period.140 A predicted postoperative thesia (two meta-analyses), epidural analgesia (six RCTs,
FEV1 of <800 mL, while associated with an operative one meta-analysis), and laparoscopic (vs. open) surgeries
mortality rate of 15%, is generally believed to represent (one systematic review, one meta-analysis, two additional
a reasonable level of postoperative lung function.138 RCTs), although laparoscopic surgeries reduce pain and
Routine PFTs (FEV1 , diffusing capacity of the lung for pulmonary compromise, as measured by spirometry.
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 31
While malnutrition is associated with increased pul- Endarterectomy Trial, European Carotid Surgery Trial,
monary risk, routine total enteral or parenteral nutrition and Asymptomatic Carotid Atherosclerosis Study, have
does not reduce risk (one meta-analysis, three additional reported the effectiveness of CEA in preventing stroke
RCTs). Enteral formulations designed to improve immune in patients with high-grade carotid artery stenosis.145148
status may prevent postoperative pneumonia (one meta- However, the occurrence of perioperative complications
analysis, one additional RCT). such as stroke or death is still a major concern, because
The overall conclusion was that few interventions have these complications negate the benefits of the procedure.
clearly been shown to reduce postoperative pulmonary Hence identification of risk factors for adverse outcomes
complications. Although there is a lack of definitive evi- after carotid surgery is very important in surgical patient
dence in many areas, aggressive management with the fol- selection and patient counseling.
lowing measures during the pre-, intra-, and postoperative Rothwell et al.147 conducted a systematic review
periods may reduce pulmonary risks. Recommendations of CEA studies published from 1981 to 1996 that
for care in each phase are summarized in Table 2.3. reported perioperative risk data by more than one clinical
or angiographic characteristic. The authors carefully
reviewed the literature to select only studies meeting strict
criteria; of 126 studies reviewed, only 35 met the criteria.
Why Do Cardiac and The selected studies included a variety of designs ranging
Cerebrovascular Diseases Pose from retrospective case series to prospective, randomized
clinical trials. The findings of this review were that a
a Significant Risk for history of cerebral or ocular transient ischemic attack,
Thromboembolic Events? age >75 years, systolic hypertension, female sex, and
peripheral vascular disease are significant, independent
Cardiac and cerebrovascular diseases are the leading predictors of perioperative stroke and death.
causes of death worldwide and pose significant risks Tu et al.42 reported on perioperative risk factors for
for perioperative thromboembolic complications (stroke, stroke and death within 30 days of CEA performed in
myocardial infarction, deep vein thrombosis, pulmonary Ontario, Canada, from 1994 through 1997 using a large
embolism).32 population-based multicenter registry. The overall 30-day
death or stroke rate after surgery was 6.0%. A history of
transient ischemic attack or stroke (OR, 1.75; 95% CI,
DEATH OR STROKE AFTER 1.39 to 2.20), atrial fibrillation (OR, 1.89; 95% CI, 1.29 to
CAROTID ENDARTERECTOMY 2.76), contralateral carotid occlusion (OR, 1.72; 95% C.I.,
1.25 to 2.38), CHF (OR, 1.80; 95% CI, 1.15 to 2.81), and
Several large randomized, controlled clinical trials, diabetes (OR, 1.28; 95% CI, 1.01 to 1.63) were significant,
including the North American Symptomatic Carotid independent predictors of 30-day death or stroke. Of
Intraoperative Period
1. Consider laparoscopic procedures instead of open procedures when possible
2. Limit duration of surgery to <4 hours, if possible
3. Avoid long-acting neuromuscular blocking agents and ensure complete recovery from neuromuscular blockade
Postoperative Period
1. Provide adequate pain control, preferably with regional nerve blocks or neuraxial analgesia instead of parenteral opioids
2. Use noninvasive lung expansion techniques to reduce the incidence of postoperative atelectasis and pneumonia
3. Continue measures for prevention of deep vein thrombosis and pulmonary embolism in high-risk patients
32 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
particular concern were patients with risk scores >2, Kikura et al.154 performed a prospective 12-year study
who have a high risk of perioperative complications. enrolling 21,903 consecutive patients who underwent
Notably, although the risk factors differed between the elective general surgery, orthopedic surgery, or thoracic
two studies, common factors were a history of transient and peripheral vascular surgery in a major university-
ischemic attack, stroke, and peripheral vascular disease. affiliated general hospital in Japan between January 1991
and December 2002. Patients undergoing head and neck
and thoracic aortic surgery were excluded. All patients
LATE STROKE AFTER were monitored for new onset of MI, pulmonary em-
CORONARY ARTERY bolism, deep vein thrombosis, ischemic stroke, and death
from cardiovascular causes for 30 days postoperatively.
BYPASS GRAFTING They underwent preoperative evaluation of preexisting
morbidities (hypertension, diabetes mellitus, hyperlipi-
Stroke as a perioperative complication of CABG ranges
demia, hyperuricemia, atrial fibrillation, stable angina,
between 1.1% and 3.8% according to the published lit-
unstable angina, MI, valvular disease, pulmonary hy-
erature. The hospital mortality rate in these patients is
pertension, ischemic stroke, and cancer). The authors
as high as 14% to 21%, which is a 10-fold increase
reported that a history of atrial fibrillation and coronary
compared with that seen in patients undergoing CABG
artery disease increased the risk of MI (OR [95% CI],
who do not have a stroke.149 Atherosclerosis of the as-
4.3 [2.86.7]). A history of stroke increased the risk of
cending aorta is associated with a patients history of
stroke (2.4 [1.44.1]) and death (4.7 [1.317.3]). Diabetes
neurologic events150 and perioperative stroke in patients
mellitus increased the risk of MI (2.1 [1.33.2]), and hy-
undergoing cardiac surgery.151 However, not much infor-
peruricemia increased the risk of stroke (3.5 [1.29.8]),
mation exists concerning the incidence of late stroke after
whereas both increased the risk of death (4.3 [1.314.1]
CABG. and 11.8 [2.263.5], respectively). A history of MI in-
Schachner et al.152 performed a prospective study creased the risk of deep vein thrombosis (7.7 [1.734.7]).
of 500 hospital survivors who had undergone epiaortic Cancer increased the risk of all thromboembolism (2.4
ultrasonography during CABG from 1998 through 2003. [1.93.2]). The authors also stated that trend analysis
Followup data on only 395 patients could be obtained. showed that the incidences of preexisting morbidities
In telephone interviews, information regarding the oc- will increase 1.5-fold, and thromboembolic events will
currence of postoperative strokes was collected from the increase 3-fold during the next decade in Japan. They
patient or the referring physician. The median followup concluded that by identifying the patients at risk for pe-
time was 52 months (9 to 74 months). Patients who had rioperative acute thrombotic complications, appropriate
a perioperative stroke were excluded from the results. All measures can be taken to reduce the risks and associated
patients had received aspirin in the postoperative period. costs, and thereby improve patient outcomes.
Stroke occurred in 26 (7%) of 387 patients in this study.
A significantly lower freedom from stroke was present
in patients with an age of 70 or more (p = 0.007), pre-
operative unstable angina (p = 0.031), COPD (p = 0.009), Why Are Perioperative Nerve
carotid artery disease (p < 0.001), preoperative history of
neurologic events (p < 0.001), and a maximum ascend- Injuries a Significant Source of
ing aortic wall thickness of 4 mm or more (p = 0.010). Morbidity?
Multivariate analysis revealed a preoperative history of
neurologic events (p = 0.021) to be an independent risk
Perioperative peripheral nerve injuries are a significant
factor. The authors concluded that a history of neurologic
source of morbidity and, unfortunately, will be encoun-
events is of special predictive importance for late stroke
tered by even the most conscientious anesthesia provider.
after CABG.152
These injuries are the second most common cause (after
death) of professional liability among anesthesiologists,
accounting for 16% of claims in the ASA closed claims
database (see Fig. 2.5).155 Nerve injuryrelated claims
What Is the Overall Risk for have steadily increased in the past decade. Unfortunately,
Perioperative Acute the relation between current conventional perioperative
care and positioning and development of postoperative
Thromboembolic Syndrome? nerve injury is poorly understood.
The ulnar nerve, brachial plexus, and the lumbosacral
Surgical patients are prone to acute, perioperative throm- roots account for most of the claims, with the ulnar nerve
boembolic events if they have obesity, hypertension, being the most frequently injured during the perioperative
hyperlipidemia, hyperuricemia, ischemic heart disease, period. Mechanisms for nerve injury include compression,
cerebrovascular disease, diabetes mellitus, or cancer.153 stretching, ischemia, direct trauma, and laceration. While
Furthermore, the perioperative period is associated with injuries to the brachial plexus and lumbosacral roots may
hemodynamic disturbances, prolonged bedrest, and hy- be secondary to stretching or compression associated with
percoagulability, all of which may contribute to a throm- malpositioning, injuries to the ulnar nerve are usually
boembolic episode. unexplained and often puzzling.
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 33
(2,904) 19801989
(783) 19701979
Total = 4,361
0 5 10 15 20 25 30 35 40 45
Total claims (%)
FIGURE 2.5 The incidence of death, brain damage, and nerve injury as a percentage of total claims
in a given time period. A significant reduction in the proportion of claims for death and brain
damage occurred between 1970 to 1979 and 1990 to 1994. (From Cheney FW. The American Society
of Anesthesiologists Closed Claims Project: What have we learned, how has it affected practice, and
how will it affect practice in the future? Anesthesiology. 1999;91:552.)
rotation of the thigh during leg positioning can produce Anesthesiologists should be familiar with and follow
excessive stretching of the sciatic nerve resulting in injury. the recommendations of the ASA practice advisory for the
The common peroneal nerve is particularly vulnerable to prevention of perioperative neuropathies (Table 2.4).161
compression injury, as it wraps around the head of the During particularly long procedures, consideration should
fibula. Femoral neuropathy is more commonly associated be given to minimizing the time spent in a position that
with surgical factors, although ischemic injury can re- amplifies physiologic perturbations or injury to the pa-
sult from extreme abduction and external rotation of the tient. It may be advisable to look for and document
thighs during lithotomy positioning. symptoms of nerve dysfunction preoperatively in high-risk
TABLE 2.4 Summary of American Society of Anesthesiologists (ASA) Task Force Consensus
Practice Advisory for the Prevention of Perioperative Neuropathies
Preoperative Assessment
When judged appropriate, it is helpful to ascertain that patients can comfortably tolerate the
anticipated operative position
Upper Extremity Positioning
1. Arm abduction should be limited to 90 degrees in supine patients; patients who are positioned
prone may comfortably tolerate arm abduction >90 degrees
2. Arms should be positioned to decrease pressure on the postcondylar groove of the humerus
(ulnar groove). When arms are tucked at the side, a neutral forearm position is recommended.
When arms are abducted on arm boards, either supination or a neutral forearm position is
acceptable
3. Prolonged pressure on the radial nerve in the spinal groove of the humerus should be avoided
4. Extension of the elbow beyond a comfortable range may stretch the median nerve
Lower Extremity Positioning
1. Lithotomy positions that stretch the hamstring muscle group beyond a comfortable range may
stretch the sciatic nerve
2. Prolonged pressure on the peroneal nerve at the fibular head should be avoided
3. Neither extension nor flexion of the hip increases the risk of femoral neuropathy
Protective Padding
1. Padded armboards may decrease the risk of upper extremity neuropathy
2. The use of chest rolls in laterally positioned patients may decrease the risk of upperextremity
neuropathies
3. Padding at the elbow and at the fibular head may decrease the risk of upper extremity
neuropathies, respectively
Equipment
1. Properly functioning automated blood pressure cuffs on the upper arms do not affect the risk of
upper extremity neuropathies
2. Shoulder braces in steep head-down positions may increase the risk of brachial plexus
neuropathies
Postoperative Assessment
A simple postoperative assessment of extremity nerve function may lead to early recognition of
peripheral neuropathies
Documentation
Charting specific positioning actions during the care of patients may result in improvements of care
by the following:
From American Society of Anesthesiologists. Practice advisory for the prevention of perioperative peripheral
neuropathies: A report by the American Society of Anesthesiologists Task Force on Prevention of Perioperative
Peripheral Neuropathies. Anesthesiology. 2000;92:1168.
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 35
patients (those with risk factors for perioperative neu- and consensus expert opinion about this devastating
ropathies or those undergoing high-risk surgery, including problem.166
long procedures and procedures involving surgical posi-
tions at risk for injury). A description of the intraoperative
positioning and measures taken to prevent injury should
be documented in the anesthetic record at the beginning What Is the Incidence of
of surgery and thereafter on a regular basis.
Maternal Mortality?
Worldwide, it is estimated that the overall maternal
How and Why Does Visual mortality rate is approximately 400 per 100,000 live births.
In the most recent Report on Confidential Enquiries
Injury Occur Perioperatively? into Maternal Deaths in the United Kingdom, the overall
incidence of maternal mortality in the United Kingdom
Postoperative visual complications range from temporary was 12 per 100,000 live births.167 In the new millennium,
loss of visual acuity to devastating permanent loss the likelihood of dying from anesthesia given for caesarean
of visual function. Corneal abrasions, periorbital and section is >30 times lower than it was in the mid-1960s
conjunctival edema, ocular hemorrhage, vitreous loss, (see Fig. 2.6). The most salient factor contributing to
retinal detachment, central retinal artery occlusion, and this reduction is the increase in regional anesthesia for
ischemic optic neuropathy can all be encountered in the caesarean section and analgesia in labor. In this recent
perioperative period. report, anesthesiologists were directly responsible for
There is a wide variation in the reported incidence seven deaths (compared with only four deaths in the
(<0.06% to 25.6%) of perioperative visual injury. The previous two reports). Three of these deaths were due to
American Association of Nurse Anesthetists (AANA) Foun- failure to recognize esophageal intubation. Each of these
dation Closed Claims Project162 shows an incidence of deaths involved a trainee in anesthesia and clearly point
3.3%, with the ASA Closed Claims Project163 reporting a to their lack of experience. Capnography was used in only
similar incidence of 3.47% for all types of eye injury. In one of these cases, and the gas sampling tube became
both projects, corneal abrasions are the most common blocked with gastric contents.
complications encountered. Patient movement, chemical
irritation from preparatory solutions, direct trauma from
the face mask, pressure from the laryngoscopic blade,
50
pressure effects on the globe from lateral and prone posi-
tioning, prolonged procedures on the spine in the prone Direct anesthetic deaths
position, and intraoperative hypotension and anemia have Direct anesthetic deaths (CS)
all been implicated. 40
The lateral, prone, and Trendelenburg positions in-
crease the risk for visual complications. In all these
Number of mothers
The same trend was noted in the United States. The 9. With an increase in atherosclerotic disease, can-
anesthesia-related maternal mortality rate decreased from cer, older patients, and obesity worldwide, acute
4.3 per million live births in the triennium of 1979 to 1981 thromboembolic and respiratory complications con-
to 1.7 per million in the triennium of 1988 to 1990.168 tribute widely to perioperative morbidity and
In contrast to the number of deaths related to general mortality.
anesthesia, the number of regional anesthesiarelated 10. Postoperative peripheral neuropathies are the sec-
deaths has decreased since 1984.168 It is disheartening to ond most common cause of professional liability
note, however, that the risk for general anesthesia death claims against anesthesiologists.
is estimated at 1 in 20,000, which has not altered since 11. Postoperative visual loss is a devastating complica-
1982.169 tion that can occur after long spine procedures in
Analysis of the ASA Closed Claims database showed the prone position.
that the frequency of claims for esophageal intubation was 12. Anesthesia-related maternal mortality continues to
greater in nonobstetric cases than in obstetric cases.170 decline, owing to an increase in the use of regional
One patient died of aspiration of gastric contents and one techniques for labor analgesia and operative de-
death was attributed to anaphylaxis from an anesthetic liveries. However, the risk associated with general
drug. Obesity was a contributory factor in anesthetic anesthesia is estimated to be 1 in 20,000 and has not
deaths; 35% of all the women who died of direct obstetric changed since 1982.
causes were obese. The ASA Closed Claims database
disclosed that the use of regional anesthesia has no
impact on malpractice litigation.171 However, the toxicity REFERENCES
of local anesthetics and high regional blocks remain a 1. Snow J. On chloroform and other anaesthetics; their action
problem. and administration. Br J Anaesth. 1957;29:524.
2. Beecher HK. Deaths during anaesthesia. Lancet.
1954;267:922.
3. Dornette WH, Orth OS. Death in the operating room. Curr
Res Anesth Analg. 1956;35:545.
4. Bodlander FM. Deaths associated with anaesthesia. Br J
KEY POINTS
Anaesth. 1975;47:36.
1. Assessing risk for perioperative morbidity and mor- 5. Hovi-Viander M. Death associated with anaesthesia in
tality is important in helping the patient make an Finland. Br J Anaesth. 1980;52:483.
informed decision. 6. Marx GF, Mateo CV, Orkin LR. Computer analysis of
2. Perioperative risk should be individualized on the postanesthetic deaths. Anesthesiology. 1973;39:54.
basis of the patients comorbid burden and func- 7. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia
in surgical mortality. JAMA. 1961;178:261.
tional status, the surgeons skill, the complexity
8. Farrow SC, Fowkes FG, Lunn JN, et al. Epidemiology in
of the procedure, local factors (availability of re-
anaesthesia. II: Factors affecting mortality in hospital. Br J
sources, expertise of support staff), and anesthetic Anaesth. 1982;54:811.
and postoperative management. Risk indexes are 9. Farrow SC, Fowkes FG, Lunn JN, et al. Epidemiology in
useful for defining risk. anaesthesia: A method for predicting hospital mortality.
3. An ideal risk index should be simple, reliable, Eur J Anaesthesiol. 1984;1:77.
accurate, and able to guide further testing and 10. Harrison GG. Death attributable to anaesthesia. A 10-year
management. Risk indexes that score the functional survey (19671976). Br J Anaesth. 1978;50:1041.
status of the patient may be more predictive of 11. Holland R. Anaesthetic mortality in New South Wales. Br J
postoperative complications. Anaesth. 1987;59:834.
4. Cardiovascular and carotid artery diseases are the 12. Tiret L, Desmonts JM, Hatton F, et al. Complications as-
leading cause of perioperative morbidity and mor- sociated with anaesthesiaa prospective survey in France.
Can Anaesth Soc J. 1986;33:336.
tality. ACC/AHA and the ACP guidelines are a good
13. Tikkanen J, Hovi-Viander M. Death associated with anaes-
framework for risk stratification and management thesia and surgery in Finland in 1986 compared to 1975.
strategies for patients with ischemic heart disease. Acta Anaesthesiol Scand. 1995;39:262.
5. Perioperative heart failure is the most common 14. Lunn JN. The study on anaesthetic-related mortality.
cardiac complication after noncardiac surgery. It is Anaesthesia. 1980;35:617.
very important to aggressively manage heart failure 15. Lunn JN, Mushin WW. Mortality associated with anesthe-
during the perioperative period. sia. Anaesthesia. 1982;37:856.
6. Anesthesiologists and perioperative physicians can 16. Lunn JN, Hunter AR, Scott DB. Anaesthesia-related surgical
improve the outcomes of patients by employing mortality. Anaesthesia. 1983;38:1090.
aggressive measures to modify risk factors. 17. Lunn JN, Devlin HB. Lessons from the confidential enquiry
7. Current literature suggests that perioperative - into perioperative deaths in three NHS regions. Lancet.
1987;2:1384.
blockade, 2 -agonists, and lipid-lowering drugs may
18. Cohen MM, Duncan PG, Tate RB. Does anesthe-
improve outcomes in high-risk patients.
sia contribute to operative mortality? JAMA. 1988;260:
8. There is evidence of increasing morbidity and 2859.
mortality secondary to coronary revascularization 19. Cohen MM, Duncan PG, Pope WD, et al. The Canadian four-
procedures before noncardiac surgery. centre study of anaesthetic outcomes: II. Can outcomes be
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 37
used to assess the quality of anaesthesia care? Can J Anaesth. 38. Detsky AS, Abrams HB, Forbath N, et al. Cardiac as-
1992;39:430. sessment for patients undergoing noncardiac surgery. A
20. Warner MA, Shields SE, Chute CG. Major morbidity and multifactorial clinical risk index. Arch Intern Med. 1986;146:
mortality within 1 month of ambulatory surgery and 2131.
anesthesia. JAMA. 1993;270:1437. 39. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation
21. Fleisher LA, Pasternak LR, Herbert R, et al. Inpatient and prospective validation of a simple index for prediction
hospital admission and death after outpatient surgery of cardiac risk of major noncardiac surgery. Circulation.
in elderly patients: Importance of patient and system 1999;100:1043.
characteristics and location of care. Arch Surg. 2004;139:67. 40. Parsonnet V, Dean D, Bernstein AD. A method of uni-
22. Wolters U, Wolf T, Stutzer H, et al. Risk factors, complica- form stratification of risk for evaluating the results of
tions, and outcome in surgery: A multivariate analysis. Eur surgery in acquired adult heart disease. Circulation. 1989;
J Surg. 1997;163:563. 79:I3.
23. Lienhart A, Auroy Y, Pequignot F, et al. Preliminary 41. Tuman KJ, McCarthy RJ, March RJ, et al. Morbidity and
results from the SFAR-iNSERM inquiry on anaesthesia- duration of ICU stay after cardiac surgery. A model for
related deaths in France: Mortality rates have fallen ten-fold preoperative risk assessment. Chest. 1992;102:36.
over the past two decades. Bull Acad Natl Med. 2004;188: 42. Tu JV, Wang H, Bowyer B, et al. Risk factors for
1429. death or stroke after carotid endarterectomy: Observations
24. Kawashima Y, Takahashi S, Suzuki M, et al. Anesthesia- from the Ontario Carotid Endarterectomy Registry. Stroke.
related mortality and morbidity over a 5-year period in 2003;34:2568.
2,363,038 patients in Japan. Acta Anaesthesiol Scand. 2003; 43. Dupuis JY, Wang F, Nathan H, et al. The cardiac anesthesia
47:809. risk evaluation score: A clinically useful predictor of mor-
25. Khuri SF, Henderson WG, DePalma RG, et al. Determinants tality and morbidity after cardiac surgery. Anesthesiology.
of long-term survival after major surgery and the adverse 2001;94:194.
effect of postoperative complications. Ann Surg. 2005; 44. Gilbert K, Larocque BJ, Patrick LT. Prospective evaluation
242:326. of cardiac risk indices for patients undergoing noncardiac
26. Cullinane M. The 2003 report of the national confiden- surgery. Ann Intern Med. 2000;133:356.
tial enquiry into perioperative deaths. London: NCEPOD; 45. Hanley JA, McNeil BJ. The meaning and use of the area
2003. under a receiver operating characteristic (ROC) curve.
27. Bennett-Guerrero E, Hyam JA, Shaefi S, et al. Comparison Radiology. 1982;143:29.
of P-POSSUM risk-adjusted mortality rates after surgery 46. Palda VA, Detsky AS. Perioperative assessment and man-
between patients in the USA and the UK. Br J Surg. 2003; agement of risk from coronary artery disease. Ann Intern
90:1593. Med. 1997;127:313.
28. Pearse RM, Harrison DA, James P, et al. Identification and 47. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide-
characterisation of the high-risk surgical population in the line update for perioperative cardiovascular evaluation for
United Kingdom. Crit Care. 2006;10:R81. noncardiac surgeryexecutive summary: A report of the
29. Auroy Y, Benhamou D, Bargues L, et al. Major compli- American College of Cardiology/American Heart Associ-
cations of regional anesthesia in France: The SOS re- ation Task Force on Practice Guidelines (Committee to
gional anesthesia hotline service. Anesthesiology. 2002;97: update the 1996 guidelines on perioperative cardiovascu-
1274. lar evaluation for noncardiac surgery). J Am Coll Cardiol.
30. Irita K, Kawashima Y, Morita K, et al. Critical inci- 2002;39:542.
dents during regional anesthesia in Japanese Society of 48. Gibbons RJ, Balady GJ, Bricker JT, et al. ACC/AHA 2002
Anesthesiologists-Certified Training Hospitals: An analysis guideline update for exercise testing: Summary article. A re-
of responses to the annual survey conducted between 1999 port of the American College of Cardiology/American Heart
and 2002 by the Japanese Society of Anesthesiologists. Association Task Force on Practice Guidelines (Committee
Masui. 2005;54:440. to Update the 1997 Exercise Testing Guidelines). J Am Coll
31. Lagasse RS. Anesthesia safety: Model or myth? A review Cardiol. 2002;40:1531.
of the published literature and analysis of current original 49. Mangano DT, Browner WS, Hollenberg M, et al. The Study
data. Anesthesiology. 2002;97:1609. of Perioperative Ischemia Research Group. Long-term
32. Murray CJ, Lopez AD. Mortality by cause for eight regions cardiac prognosis following noncardiac surgery. JAMA.
of the world: Global Burden of Disease Study. Lancet. 1992;268:233.
1997;349:1269. 50. Mangano DT, Browner WS, Hollenberg M, et al. The Study
33. Mangano DT, Goldman L. Preoperative assessment of of Perioperative Ischemia Research Group. Association of
patients with known or suspected coronary disease. N Engl perioperative myocardial ischemia with cardiac morbidity
J Med. 1995;333:1750. and mortality in men undergoing noncardiac surgery.
34. Badner NH, Knill RL, Brown JE, et al. Myocardial infarction N Engl J Med. 1990;323:1781.
after noncardiac surgery. Anesthesiology. 1998;88:572. 51. Charlson ME, MacKenzie CR, Gold JP, et al. Risk for
35. Raby KE, Goldman L, Creager MA, et al. Correlation postoperative congestive heart failure. Surg Gynecol Obstet.
between preoperative ischemia and major cardiac events 1991;172:95.
after peripheral vascular surgery. N Engl J Med. 1989;321: 52. Older P, Hall A, Hader R. Cardiopulmonary exercise testing
1296. as a screening test for perioperative management of major
36. Raby KE, Barry J, Creager MA, et al. Detection and surgery in the elderly. Chest. 1999;116:355.
significance of intraoperative and postoperative myocardial 53. Halm EA, Browner WS, Tubau JF, et al. Study of
ischemia in peripheral vascular surgery. JAMA. 1992;268: Perioperative Ischemia Research Group. Echocardiography
222. for assessing cardiac risk in patients having noncardiac
37. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial surgery. Ann Intern Med. 1996;125:433.
index of cardiac risk in noncardiac surgical procedures. 54. Franco CD, Goldsmith J, Veith FJ, et al. Resting gated
N Engl J Med. 1977;297:845. pool ejection fraction: A poor predictor of perioperative
38 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
myocardial infarction in patients undergoing vascular 73. Garber AM, Solomon NA. Cost-effectiveness of alternative
surgery for infrainguinal bypass grafting. J Vasc Surg. 1989; test strategies for the diagnosis of coronary artery disease.
10:656. Ann Intern Med. 1999;130:719.
55. Smith TP, Kinasewitz GT, Tucker WY, et al. Exercise 74. Fleischmann KE, Hunink MG, Kuntz KM, et al. Exercise
capacity as a predictor of post-thoracotomy morbidity. Am echocardiography or exercise SPECT imaging? A meta-
Rev Respir Dis. 1984;129:730. analysis of diagnostic test performance. JAMA. 1998;
56. Miyoshi S, Nakahara K, Ohno K, et al. Exercise tolerance 280:913.
test in lung cancer patients: The relationship between 75. Sisson JC, Schoomaker EB, Ross JC. Clinical decision
exercise capacity and postthoracotomy hospital mortality. analysis. The hazard of using additional data. JAMA. 1976;
Ann Thorac Surg. 1987;44:487. 236:1259.
57. Olsen GN. The evolving role of exercise testing prior to lung 76. Boersma E, Poldermans D, Bax JJ, et al. Predictors of
resection. Chest. 1989;95:218. cardiac events after major vascular surgery: Role of clinical
58. Sue DY, Wasserman K. Impact of integrative cardiopul- characteristics, dobutamine echocardiography, and beta-
monary exercise testing on clinical decision making. Chest. blocker therapy. JAMA. 2001;285:1865.
1991;99:981. 77. Mason JJ, Owens DK, Harris RA, et al. The role
59. Morice RC, Peters EJ, Ryan MB, et al. Exercise testing in of coronary angiography and coronary revasculariza-
the evaluation of patients at high risk for complications tion before noncardiac vascular surgery. JAMA. 1995;273:
from lung resection. Chest. 1992;101:356. 1919.
60. Epstein SK, Faling LJ, Daly BD, et al. Predicting compli- 78. Eagle KA, Rihal CS, Mickel MC, et al. Cardiac risk of
cations after pulmonary resection. Preoperative exercise noncardiac surgery: Influence of coronary disease and
testing vs a multifactorial cardiopulmonary risk index. type of surgery in 3368 operations. CASS Investigators
Chest. 1993;104:694. and University of Michigan Heart Care Program. Coronary
61. Wasserman K. Preoperative evaluation of cardiovascular Artery Surgery Study. Circulation. 1997;96:1882.
reserve in the elderly. Chest. 1993;104:663. 79. Posner KL, Van Norman GA, Chan V. Adverse cardiac
62. Gilbreth EM, Weisman IM. Role of exercise stress testing in outcomes after noncardiac surgery in patients with prior
preoperative evaluation of patients for lung resection. Clin percutaneous transluminal coronary angioplasty. Anesth
Chest Med. 1994;15:389. Analg. 1999;89:553.
63. Bolliger CT, Wyser C, Roser H, et al. Lung scanning 80. Fleisher LA, Tuman KJ. What can we learn from provoking
and exercise testing for the prediction of postoperative ischemia? Anesth Analg. 1997;84:1177.
performance in lung resection candidates at increased risk 81. Kaluza GL, Joseph J, Lee JR, et al. Catastrophic outcomes
for complications. Chest. 1995;108:341. of noncardiac surgery soon after coronary stenting. J Am
64. Bolliger CT, Jordan P, Soler M, et al. Exercise capacity Coll Cardiol. 2000;35:1288.
as a predictor of postoperative complications in lung 82. McFadden EP, Stabile E, Regar E, et al. Late thrombosis
resection candidates. Am J Respir Crit Care Med. 1995;151: in drug-eluting coronary stents after discontinuation of
1472. antiplatelet therapy. Lancet. 2004;364:1519.
65. Older P, Smith R, Courtney P, et al. Preoperative evaluation 83. Prys-Roberts C, Foex P, Biro GP, et al. Studies of anaesthe-
of cardiac failure and ischemia in elderly patients by sia in relation to hypertension. V. Adrenergic beta-receptor
cardiopulmonary exercise testing. Chest. 1993;104:701. blockade. Br J Anaesth. 1973;45:671.
66. Ross RM. ATS/ACCP statement on cardiopulmonary 84. Stone JG, Foex P, Sear JW, et al. Myocardial ischemia in
exercise testing. Am J Respir Crit Care Med. 2003;167:1451; untreated hypertensive patients: Effect of a single small oral
author reply 1451. dose of a beta-adrenergic blocking agent. Anesthesiology.
67. Cowie MR, Struthers AD, Wood DA, et al. Value of 1988;68:495.
natriuretic peptides in assessment of patients with possible 85. Mangano DT, Layug EL, Wallace A, et al. Multicenter
new heart failure in primary care. Lancet. 1997;350:1349. Study of Perioperative Ischemia Research Group. Ef-
68. Gardner RS, Ozalp F, Murday AJ, et al. N-terminal pro- fect of atenolol on mortality and cardiovascular morbid-
brain natriuretic peptide. A new gold standard in predicting ity after noncardiac surgery. N Engl J Med. 1996;335:
mortality in patients with advanced heart failure. Eur 1713.
Heart J. 2003;24:1735. 86. Poldermans D, Boersma E, Bax JJ, et al. Dutch Echocardio-
69. de Groote P, Dagorn J, Soudan B, et al. B-type natriuretic graphic Cardiac Risk Evaluation Applying Stress Echocar-
peptide and peak exercise oxygen consumption provide diography Study Group. The effect of bisoprolol on
independent information for risk stratification in patients perioperative mortality and myocardial infarction in
with stable congestive heart failure. J Am Coll Cardiol. high-risk patients undergoing vascular surgery. N Engl J
2004;43:1584. Med. 1999;341:1789.
70. Horwich TB, Patel J, MacLellan WR, et al. Cardiac troponin 87. Lee TH. Reducing cardiac risk in noncardiac surgery.
I is associated with impaired hemodynamics, progres- N Engl J Med. 1999;341:1838.
sive left ventricular dysfunction, and increased mortality 88. Litwack RS, Gilligan DM, DeGruttola V. Beta-blockade
rates in advanced heart failure. Circulation. 2003;108: for patients undergoing vascular surgery. N Engl J Med.
833. 2000;342:1052; author reply 1052.
71. Provenchere S, Berroeta C, Reynaud C, et al. Plasma brain 89. Karthikeyan G, Bhargava B. Managing patients undergoing
natriuretic peptide and cardiac troponin I concentrations non-cardiac surgery: Need to shift emphasis from risk
after adult cardiac surgery: Association with postoperative stratification to risk modification. Heart. 2006;92:17.
cardiac dysfunction and 1-year mortality. Crit Care Med. 90. Grayburn PA, Hillis LD. Cardiac events in patients un-
2006;34:995. dergoing noncardiac surgery: Shifting the paradigm from
72. Lee DS, Vasan RS. Novel markers for heart failure diagnosis noninvasive risk stratification to therapy. Ann Intern Med.
and prognosis. Curr Opin Cardiol. 2005;20:201. 2003;138:506.
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 39
91. Auerbach AD, Goldman L. Beta-blockers and reduction 110. ONeil-Callahan K, Katsimaglis G, Tepper MR, et al. Statins
of cardiac events in noncardiac surgery: Scientific review. decrease perioperative cardiac complications in patients
JAMA. 2002;287:1435. undergoing noncardiac vascular surgery: The statins for
92. Devereaux PJ, Beattie WS, Choi PT, et al. How strong is risk reduction in surgery (StaRRS) study. J Am Coll Cardiol.
the evidence for the use of perioperative beta blockers in 2005;45:336.
non-cardiac surgery? Systematic review and meta-analysis 111. Kennedy J, Quan H, Buchan AM, et al. Statins are associated
of randomised controlled trials. Br Med J. 2005;331:313. with better outcomes after carotid endarterectomy in
93. Lindenauer PK, Pekow P, Wang K, et al. Perioperative symptomatic patients. Stroke. 2005;36:2072.
beta-blocker therapy and mortality after major noncardiac 112. McGirt MJ, Perler BA, Brooke BS, et al. 3-hydroxy-3-
surgery. N Engl J Med. 2005;353:349. methylglutaryl coenzyme A reductase inhibitors reduce
94. Devereaux PJ, Yang H, Guyatt GH, et al. Rationale, design, the risk of perioperative stroke and mortality after carotid
and organization of the PeriOperative ISchemic Evaluation endarterectomy. J Vasc Surg. 2005;42:829.
(POISE) trial: A randomized controlled trial of metoprolol 113. Biccard BM, Sear JW, Foex P. Statin therapy: A poten-
versus placebo in patients undergoing noncardiac surgery. tially useful peri-operative intervention in patients with
Am Heart J. 2006;152:223. cardiovascular disease. Anaesthesia. 2005;60:1106.
95. Ellis JE, Drijvers G, Pedlow S, et al. Premedication with 114. Owczarek J, Jasinska M, Orszulak-Michalak D. Drug-
oral and transdermal clonidine provides safe and effica- induced myopathies. An overview of the possible mech-
cious postoperative sympatholysis. Anesth Analg. 1994;79: anisms. Pharmacol Rep. 2005;57:23.
1133. 115. Waters DD. Safety of high-dose atorvastatin therapy. Am J
96. Quintin L, Bouilloc X, Butin E, et al. Clonidine for major Cardiol. 2005;96:69F.
vascular surgery in hypertensive patients: A double-blind, 116. Schouten O, Kertai MD, Bax JJ, et al. Safety of perioperative
controlled, randomized study. Anesth Analg. 1996;83:687. statin use in high-risk patients undergoing major vascular
97. Stuhmeier KD, Mainzer B, Cierpka J, et al. Small, oral surgery. Am J Cardiol. 2005;95:658.
dose of clonidine reduces the incidence of intraoperative 117. Dunkelgrun M, Schouten O, Feringa HH, et al. Beneficial
myocardial ischemia in patients having vascular surgery. effects of statins on perioperative cardiovascular outcome.
Anesthesiology. 1996;85:706. Curr Opin Anaesthesiol. 2006;19:418.
98. Fox K, Dargie HJ, de Bono DP, et al. Effect of an alpha(2) 118. Schouten O, Poldermans D, Visser L, et al. Fluvastatin and
agonist (mivazerol) on limiting myocardial ischaemia in bisoprolol for the reduction of perioperative cardiac mor-
stable angina. Heart. 1999;82:383. tality and morbidity in high-risk patients undergoing non-
99. Oliver MF, Goldman L, Julian DG, et al. Effect of cardiac surgery: Rationale and design of the DECREASE-IV
mivazerol on perioperative cardiac complications during study. Am Heart J. 2004;148:1047.
non-cardiac surgery in patients with coronary heart disease: 119. Lawrence VA, Hilsenbeck SG, Mulrow CD, et al. Incidence
The European mivazerol trial (EMIT). Anesthesiology. and hospital stay for cardiac and pulmonary complications
1999;91:951. after abdominal surgery. J Gen Intern Med. 1995;10:671.
100. Wallace AW, Galindez D, Salahieh A, et al. Effect of 120. Mohr DN, Jett JR. Preoperative evaluation of pulmonary
clonidine on cardiovascular morbidity and mortality after risk factors. J Gen Intern Med. 1988;3:277.
noncardiac surgery. Anesthesiology. 2004;101:284. 121. Kroenke K, Lawrence VA, Theroux JF, et al. Postoperative
101. Wallace A, Layug B, Tateo I, et al. McSPI Research Group. complications after thoracic and major abdominal surgery
Prophylactic atenolol reduces postoperative myocardial in patients with and without obstructive lung disease. Chest.
ischemia. Anesthesiology. 1998;88:7. 1993;104:1445.
102. Talke P, Li J, Jain U, et al. The Study of Periopera- 122. Warner DO, Warner MA, Barnes RD, et al. Perioperative
tive Ischemia Research Group. Effects of perioperative respiratory complications in patients with asthma. Anesthe-
dexmedetomidine infusion in patients undergoing vascular siology. 1996;85:460.
surgery. Anesthesiology. 1995;82:620. 123. Warner MA, Offord KP, Warner ME, et al. Role of pre-
103. Paris A, Tonner PH. Dexmedetomidine in anaesthesia. Curr operative cessation of smoking and other factors in post-
Opin Anaesthesiol. 2005;18:412. operative pulmonary complications: A blinded prospective
104. Ray KK, Cannon CP. The potential relevance of the multiple study of coronary artery bypass patients. Mayo Clin Proc.
lipid-independent (pleiotropic) effects of statins in the 1989;64:609.
management of acute coronary syndromes. J Am Coll 124. Bluman LG, Mosca L, Newman N, et al. Preoperative smok-
Cardiol. 2005;46:1425. ing habits and postoperative pulmonary complications.
105. ODriscoll G, Green D, Taylor RR. Simvastatin, an HMG- Chest. 1998;113:883.
coenzyme A reductase inhibitor, improves endothelial 125. Pasulka PS, Bistrian BR, Benotti PN, et al. The risks of
function within 1 month. Circulation. 1997;95:1126. surgery in obese patients. Ann Intern Med. 1986;104:540.
106. Yamakuchi M, Greer JJ, Cameron SJ, et al. HMG-CoA 126. Flier S, Knape JT. How to inform a morbidly obese patient
reductase inhibitors inhibit endothelial exocytosis and on the specific risk to develop postoperative pulmonary
decrease myocardial infarct size. Circ Res. 2005;96:1185. complications using evidence-based methodology. Eur J
107. Liao JK. Clinical implications for statin pleiotropy. Curr Anaesthesiol. 2006;23:154.
Opin Lipidol. 2005;16:624. 127. Ramaswamy A, Gonzalez R, Smith CD. Extensive preoper-
108. Arnaud C, Veillard NR, Mach F. Cholesterol-independent ative testing is not necessary in morbidly obese patients
effects of statins in inflammation, immunomodulation and undergoing gastric bypass. J Gastrointest Surg. 2004;8:
atherosclerosis. Curr Drug Targets Cardiovasc Haematol 159.
Disord. 2005;5:127. 128. Gross JB, Bachenberg KL, Benumof JL, et al. Practice
109. Durazzo AE, Machado FS, Ikeoka DT, et al. Reduction in guidelines for the perioperative management of patients
cardiovascular events after vascular surgery with atorvas- with obstructive sleep apnea: A report by the American
tatin: A randomized trial. J Vasc Surg. 2004;39:967. Society of Anesthesiologists Task Force on Perioperative
40 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
Management of patients with obstructive sleep apnea. 147. Rothwell PM, Slattery J, Warlow CP. Clinical and angio-
Anesthesiology. 2006;104:1081. graphic predictors of stroke and death from carotid en-
129. Djokovic JL, Hedley-Whyte J. Prediction of outcome of darterectomy: Systematic review. Br Med J. 1997;315:1571.
surgery and anesthesia in patients over 80. JAMA. 1979;242: 148. Ferguson GG, Eliasziw M, Barr HW, et al. The North Amer-
2301. ican symptomatic carotid endarterectomy trial: Surgical
130. Cerfolio RJ, Bryant AS. Survival and outcomes of pul- results in 1415 patients. Stroke. 1999;30:1751.
monary resection for non-small cell lung cancer in the 149. Roach GW, Kanchuger M, Mangano CM, et al. Adverse cere-
elderly: A nested case-control study. Ann Thorac Surg. 2006; bral outcomes after coronary bypass surgery. Multicenter
82:424. Study of Perioperative Ischemia Research Group and the Is-
131. Halasz NA. Vertical Vs Horizontal Laparotomies. I. Early chemia Research and Education Foundation Investigators.
postoperative comparisons. Arch Surg. 1964;88:911. N Engl J Med. 1996;335:1857.
132. Couture JG, Chartrand D, Gagner M, et al. Diaphragmatic 150. Davila-Roman VG, Barzilai B, Wareing TH, et al.
and abdominal muscle activity after endoscopic cholecys- Atherosclerosis of the ascending aorta. Prevalence and role
tectomy. Anesth Analg. 1994;78:733. as an independent predictor of cerebrovascular events in
133. Moller AM, Maaloe R, Pedersen T. Postoperative intensive cardiac patients. Stroke. 1994;25:2010.
care admittance: The role of tobacco smoking. Acta 151. Hogue CW Jr, Murphy SF, Schechtman KB, et al. Risk
Anaesthesiol Scand. 2001;45:345. factors for early or delayed stroke after cardiac surgery.
134. Pereira ED, Fernandes AL, da Silva Ancao M, et al. Prospec- Circulation. 1999;100:642.
tive assessment of the risk of postoperative pulmonary 152. Schachner T, Zimmer A, Nagele G, et al. The influence of
complications in patients submitted to upper abdominal ascending aortic atherosclerosis on the long-term survival
surgery. Sao Paulo Med J. 1999;117:151. after CABG. Eur J Cardiothorac Surg. 2005;28:558.
135. Rodgers A, Walker N, Schug S, et al. Reduction of 153. King H, Aubert RE, Herman WH. Global burden of
postoperative mortality and morbidity with epidural or diabetes, 19952025: Prevalence, numerical estimates, and
spinal anaesthesia: Results from overview of randomised projections. Diabetes Care. 1998;21:1414.
trials. Br Med J. 2000;321:1493. 154. Kikura M, Takada T, Sato S. Preexisting morbidity as an
136. Berg H, Roed J, Viby-Mogensen J, et al. Residual neu- independent risk factor for perioperative acute thromboem-
romuscular block is a risk factor for postoperative pul- bolism syndrome. Arch Surg. 2005;140:1210.
monary complications. A prospective, randomised, and 155. Cheney FW, Domino KB, Caplan RA, et al. Nerve injury
blinded study of postoperative pulmonary complications af- associated with anesthesia: A closed claims analysis.
ter atracurium, vecuronium and pancuronium. Acta Anaes- Anesthesiology. 1999;90:1062.
thesiol Scand. 1997;41:1095. 156. Warner MA, Warner ME, Martin JT. Ulnar neuropathy.
137. Archer C, Levy AR, McGregor M. Value of routine preop- Incidence, outcome, and risk factors in sedated or anes-
erative chest x-rays: A meta-analysis. Can J Anaesth. 1993; thetized patients. Anesthesiology. 1994;81:1332.
40:1022. 157. Warner MA, Warner DO, Harper CM, et al. Ulnar neuropa-
138. Zibrak JD, ODonnell CR, Marton K. Indications for thy in medical patients. Anesthesiology. 2000;92:613.
pulmonary function testing. Ann Intern Med. 1990;112:763. 158. Cooper DE, Jenkins RS, Bready L, et al. The prevention
139. Romano PS, Mark DH. Patient and hospital characteristics of injuries of the brachial plexus secondary to malposition
related to in-hospital mortality after lung cancer resection. of the patient during surgery. Clin Orthop Relat Res. 1988;
Chest. 1992;101:1332. 228:33.
140. Boysen PG, Block AJ, Olsen GN, et al. Prospective evalua- 159. Prielipp RC, Morell RC, Walker FO, et al. Ulnar nerve
tion for pneumonectomy using the 99mtechnetium quanti- pressure: Influence of arm position and relationship to
tative perfusion lung scan. Chest. 1977;72:422. somatosensory evoked potentials. Anesthesiology. 1999;91:
141. American Thoracic Society. Standards for the diagnosis 345.
and care of patients with chronic obstructive pulmonary 160. Warner MA, Warner DO, Harper CM, et al. Lower ex-
disease. Am J Respir Crit Care Med. 1995;152:S77. tremity neuropathies associated with lithotomy positions.
142. Pierce RJ, Copland JM, Sharpe K, et al. Preoperative risk Anesthesiology. 2000;93:938.
evaluation for lung cancer resection: Predicted postopera- 161. American Society of Anesthesiologists. Practice advisory for
tive product as a predictor of surgical mortality. Am J Respir the prevention of perioperative peripheral neuropathies:
Crit Care Med. 1994;150:947. A report by the American Society of Anesthesiologists
143. Wyser C, Stulz P, Soler M, et al. Prospective evaluation Task Force on Prevention of Perioperative Peripheral
of an algorithm for the functional assessment of lung Neuropathies. Anesthesiology. 2000;92:1168.
resection candidates. Am J Respir Crit Care Med. 1999;159: 162. Jordan LM, Kremer M, Crawforth K, et al. Data-driven
1450. practice improvement: The AANA Foundation closed mal-
144. Lawrence VA, Cornell JE, Smetana GW. Strategies to practice claims study. AANA J. 2001;69:301.
reduce postoperative pulmonary complications after non- 163. Gild WM, Posner KL, Caplan RA, et al. Eye injuries
cardiothoracic surgery: Systematic review for the American associated with anesthesia. A closed claims analysis.
College of Physicians. Ann Intern Med. 2006;144:596. Anesthesiology. 1992;76:204.
145. Rothwell PM, Slattery J, Warlow CP. A systematic com- 164. Rupp-Montpetit K, Moody ML. Visual loss as a com-
parison of the risks of stroke and death due to carotid plication of nonophthalmologic surgery: A review of the
endarterectomy for symptomatic and asymptomatic steno- literature. AANA J. 2004;72:285.
sis. Stroke. 1996;27:266. 165. Shaw PJ, Bates D, Cartlidge NE, et al. Neuro-
146. Young B, Moore WS, Robertson JT, et al. An analy- ophthalmological complications of coronary artery bypass
sis of perioperative surgical mortality and morbidity in graft surgery. Acta Neurol Scand. 1987;76:1.
the asymptomatic carotid atherosclerosis study. ACAS In- 166. American Society of Anesthesiologists. A report by the
vestigators. Asymptomatic carotid artheriosclerosis study. American Society of Anesthesiologists Task Force on
Stroke. 1996;27:2216. Perioperative blindness. Practice advisory for perioperative
C H A P T E R 2 / E V A L U AT I O N O F A N E S T H E T I C R I S K 41
visual loss associated with spine surgery. Anesthesiology. 169. Drife J. Fifty years of the confidential enquiry into maternal
2006;104:1319. deaths. Br J Hosp Med (Lond). 2006;67:121.
167. Why Mothers Die 20022004. The sixth report of the 170. Ross BK. ASA closed claims in obstetrics: Lessons learned.
confidential enquiries into maternal deaths in the United Anesthesiol Clin North Am. 2003;21:183.
Kingdom. London: RCOG Press; 2004. 171. Chadwick HS. An analysis of obstetric anesthesia cases from
168. Hawkins JL, Koonin LM, Palmer SK, et al. Anesthesia- the American society of anesthesiologists closed claims
related deaths during obstetric delivery in the United States, project database. Int J Obstet Anesth. 1996;5:258.
19791990. Anesthesiology. 1997;86:277.
CHAPTER QUALITY ASSURANCE AND RISK
3
MANAGEMENT
R
and risk management (RM) present in a sin- mance, would expose a hospital to severe liability. This
gle case. However, in Walker vs. Anesthesia horrifying scenario is much less likely now, owing to the
Associates and others,1 many did. The plain- following factors:
tiff, an otherwise healthy, 4-year-old grand-
child of a physician, was to undergo repair of The evolution of practice guidelines and standards
an asymptomatic atrioseptal defect at the hospital where The role of peer review and quality management
his grandfather was on staff. A graduate registered nurse essential for accreditation
anesthetist (GRNA), supervised at intervals by an anesthe- The regulatory requirements and legal precedents
siologist, performed the case, although she was ineligible requiring competent credentialing and granting of
to practice under Alabama law, having failed in her last privileges and
attempt to pass the certification examination. Monitoring The pressures on individuals and institutions linking
was limited to a blood pressure cuff, which was placed pay to outcome
on the arm with the only intravenous infusion catheter,
which was tucked at the side. An arterial catheter was not
used for monitoring, as it was inaccessible for the sam-
pling of blood. In addition, coagulation parameters were Why Is Quality Management
not monitored, and some syringes were not labeled. Increasingly Important Now?
Anesthesia proceeded smoothly, until massive clot-
ting of the four cardiac chambers and primary and
secondary pulmonary artery (PA) segments was noted
almost immediately after cardiopulmonary bypass began.
RECENT PERSPECTIVES
The child was declared dead, and some of the heparin bot- What has changed most about quality management in
tles were removed from the scene. The medical record was recent years is not the methodology used, but rather
covertly modified (although this action was denied by the the recognition of its importance in controlling the cost
defendant despite the crudeness of the alteration) to indi- and efficacy of medical practice by a growing number
cate the dose of heparin that had been given before bypass of hospitals, regulators, and insurers.2 Anesthesiologists
began. The case was lost because of the following reasons: have a long history of concern for safety and an
(i) Patent failure to supervise, monitor, and exert due appreciation for data-driven decisions and the innovations
care, (ii) violation of the States Certified Registered Nurse derived from them. Most people believe that the practice
Anesthetist licensure law, (iii) failure to meet minimal of anesthesiology is now safer and less frightening than it
standards of practice, (iv) forgery of the medical record, was just a few decades ago.
and (v) numerous contradictions in depositions, some-
times within the same deposition, regarding who kept the
record, how heparin was determined to have reached the
circulation in the proper dosage, how corrections were What Important Historical
made to the record, the presence or absence of an arterial
catheter, and inability to establish the presence of the Events Affected Quality?
attending anesthesiologist for large portions of the case.
These practices were beyond the pale of reasonable The motives and future of contemporary quality improve-
care in 1981, as they would be now. Moreover, had the ment are founded on events of the last century. To a large
42
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 43
extent, it is still the product . . . of events set in motion the Joint Commission of Accreditation of Hospitals
by private foundations, organized medicine, and politics (JCAH) was established in 1951. It refined the survey
acting through government institutions. . .3 Our govern- standards of the ACS and issued its first edition of
ment, medical societies, and politicians, working together the Standards for Hospital Accreditation in 1953. When
(and sometimes at cross-purposes), have produced the Medicare was created by the Social Security Act of 1965,
current state of quality management. We need not look JCAH accreditation was deemed to qualify hospitals for
too far for those responsible; as the comic strip character, payment. By 1982, the major portion of the JCAH audit
Pogo, put it, We have met the enemy and he is us!.4 For- related to quality assurance, with 62% of challenges to
tunately, during our search for quality improvement, the accreditation of the hospitals falling in this category.10
standards of performance improvement used by indus-
trial engineeringwith their more objective, clear-headed
thinkinghave been of great value.5 Although medical
practice may be more akin to the repair business than
LEGISLATIVE AND JUDICIAL
the assembly line, there is a message to be gleaned from HISTORY
those who made a success of converting a series of ad hoc
solutions into an efficient production pathway. In 1965, the statutory standards for participation in the
Medicare Program were written. They required JCAH
accreditation, and, by 1967, the amendments to the
Social Security Act included requirements for utilization
RISE OF THE JOINT review.3 In 1972, the Professional Standards Review
COMMISSION ON Organizations (PSRO) was established by federal law
(Public Law 92603). The practice of medicine changed
ACCREDITATION OF over the ensuing 20 years in response to a plethora of
HEALTHCARE new regulations that required physicians to: (i) meet
national standards of practice, (ii) be credentialed on
ORGANIZATIONS the basis of their ability to meet these standards,
The current quality and RM procedures and the develop- (iii) contain costs, (iv) limit hospital confinement, and
ment of outcome measurement have a history reaching (v) establish quality assurance programs to realize these
back over 100 years. Although the Joint Committee on Ac- goals. The Social Security Act of 1983 established the
creditation of Healthcare Organizations (JCAHO) is not concept of reimbursement on the basis of diagnosis-
the only determinant of how institutions measure quality related groups. The Social Security Act of 1986 mandated
and improve performance, it is undoubtedly the largest, the establishment of a federal system for reporting quality
single driving force for hospital quality management. of care problems on a recurrent basis.
Early in the 20th century, events leading to the es-
tablishment of the JCAHO began to unfold. Following the Malpractice Legislation
Flexner report in 1910,6 a commentary on the shortcom-
ings of the apprenticeship system of medical education As a result of the virtually uncontrolled malpractice cli-
in the early 1900s, E.A. Codman, at the Clinical Congress mate at that time, Florida in 1985 adopted legislation man-
of the American College of Surgeons (ACS) held in 1912, dating an internal RM program11 for all licensed medical
encouraged the development of outcomes approach for care facilities. This act mandated disclosure to the state of:
evaluating the competence of surgeons and developing the
All adverse occurrences to an institutional risk manager
concept of hospital standardization.7 In part, he stated
Frequency and cause of problems, and the providers
that hospitals should look critically at their outcomes
responsible (reported annually)
and strong and weak points; contrast their results with
Malpractice claims brought against the institution
other hospitals; base physician credentials on demon-
or its practitioners, and measures taken to reduce
strated ability; and be forthcoming about bad outcomes
risk, including reduction and/or suspension of clinical
as a lever for increasing resources. His key concepts, al-
privileges and
though extremely well developed and the impetus for the
Incidents resulting in death or central nervous system
current hospital survey system, were left to twist idly in the
damage (report required within 3 days)
wind. They were to be rediscovered later by Donabedian8
and then systematically applied to healthcare by George When this legislation did not stop the exodus of
Labovitz through his company, Organizational Dynamics, insurance carriers from the state, the legislature extended
Inc. A few years after Codman presented his outcome- the law with the passage of the Medical Incident Recovery
oriented approach, the ACS adopted it as their official po- Act in 1988.12 In addition to a system for arbitration of
sition, leading ultimately to the Hospital Standardization and limits on malpractice claims, the Division of Medical
Program (1917) which was the grandparent of the JCAHO. Quality Assurance was created within the Division of
In 1918, the first audit of almost 700 hospitals Professional Regulation (DPR). The DPR was charged
was carried out, 87% of which failed to meet the with developing a list of adverse incidents that would be
minimum standards.9 Thirty-two years later, the failure deemed reportable under the law. The law held all persons
rate was less than 6%. Because of the enormous task immune from civil liability (including antitrust) who
of surveying approximately 2,500 participating hospitals, reported incidents of incompetence to DPR, except when
44 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
fraudulent or malicious. This step was significant because based on procedure and diagnosis-related audits. (In 1987,
the statute obligated all practitioners to report adverse JCAH changed its name to JCAHO to reflect its expanded
incidents that they observed. All documentation involved scope of activities.) The organization was required, on a
in this quality assurance process was mandated by state quarterly basis, to select a problem for evaluation and
law, and therefore immune to discovery outside the DPR. then eliminate it. The audits were episodic in nature, did
not encompass overall practice, and often concluded with
the recommendation that the problem should continue
Case Law and Corporate Negligence
to be studied. The audit era was a well meaning attempt
The concept of corporate negligence, as expressed in the to look at major problems in care and correct them, but
courts, has served to write into case law that which was it lacked the following: (i) The tools to conceptualize the
established by statute. The case of Darling vs. Charleston root causes of problems, (ii) the concept of the relation-
Community Memorial Hospital13 established that a hospi- ship of resources and processes to outcome, and, most
tal which extends credentials to a physician is ultimately importantly, (iii) a means of linking episodic improve-
held liable for controlling the quality of care by establish- ments to long term gain. It had no adequate theory of how
ing a functioning peer review process. In this particular to evaluate and improve quality. It relied on the point and
legal case, a leg was amputated after a cast was improp- shoot approachgetting rid of problem equipment and
erly applied by a physician who did not request orthopedic problem people to produce a good outcome, rather than
consultation. making a series of improvements. This JCAHO approach
The responsibility of a hospital in determining the mirrored a management style typical in America and in
capability of its staff was reviewed in Johnson vs. medicine at that timestrong managers eliminated prob-
Misericordia Community Hospital.14 The court found the lems once and for all, and then moved on to the next
hospital liable for not discovering that the physician in problem until they achieved perfection.
question had a long history of malpractice litigation and We still retain more than a hint of this belief system in
revoked privileges at other hospitals that appointed him our morbidity and mortality conferencesthat is, the idea
to their staff. that if we can just understand and correct each individual
mistake, we will improve. The chief problem with this
theory is that it does not work, and for a very good reason:
Liability of Peer Reviewers The number of possible mistakes we can make is infinite,
and problems tend to recur. What was needed was a
Because of the potential legal hazards of the peer review
concept of how to relate what we do in medical practice,
process (mandated by Title 19 of the Social Security Act),
and how we do it, to the type of outcomes that result; this
the process of review is protected from legal discovery
linkage of cause to effect is essential to achieving solutions
by statute (section 1160 of the Social Security Act).
that produce better outcomes.
Disclosure of information, except for legitimate peer
review purposes, is punishable by a $1,000 fine and up to 6
months incarceration in a federal institution. Nonetheless,
the protection extended by federal and state statutes was
eroded when a physician sued on the basis that his What Are the Scientific Tools
hospitals peer review committee violated the Sherman for Quality Management?
Antitrust Act when it stripped him of staff privileges. In the
malpractice suit of Patrick vs. Burget,15 this physician was
allowed by the federal district court to obtain confidential
peer review documents normally considered protected ANALYSIS OF STRUCTURE,
from legal discovery by Oregon law and was awarded
nearly $3 million. The case was remanded for retrial
PROCESS, AND OUTCOME
by the Federal Court of Appeals, which ruled that the Throughout the 1970s, as the JCAHO explored ways to
disclosure was prohibited. measure and affect quality, the tools for quality improve-
The concept of due process and the various state and ment slowly matured under the influence of W. Edwards
federal acts protecting confidential peer review creates a Deming in industry16 and Donabedian in medicine.8,17 In
potential conflict and exposes peer reviewers to substan- the 1960s, Donabedian established a model for the ob-
tial liability. Most of the suits impugning the peer review jective assessment of the quality of care. The concepts
process have been brought by physicians who have been were derived from, but different than, industrial quality
judged improperly by their peers. The only real solution control, in which the elimination of variation in pro-
for this problem is scrupulous fairness and use of review duction through standardization provides the basis for
procedures based on written criteria, accepted in writing quality improvement.18 Donabedians approach to quality
by the hospital staff, and applied in a uniform manner. evaluation and improvement introduced three enduring
interdependent elements that continue to form the core of
Rise of the Joint Commission quality assessment systems today: Structure, process, and
outcome.
Throughout the 1970s and early 1980s, JCAHO accredita- Each of these elements applies to the quality manage-
tion required the adoption of quality assurance programs ment activities of administrators, nurses, and physicians
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 45
in healthcare organizations. Individual elements specific reproducible standards for monitoring, techniques, or
to each group will tend to receive more emphasis by that procedures should also be included.22
group, although all require the collection of verifiable data
on the basis of predefined criteria. The goal is to define Process Review
the causes of adverse outcomes and provide a basis for
assessing improvements that translate into reduced risk. Elements that comprise process review include the proper
Historically, the elements have come into common use in use of techniques, management strategies and judgments,
the order listed. drugs, blood products, medical records, and surgical
procedures according to accepted practice guidelines and
standards to produce an acceptable outcome. Differences
Structural Review between process and structure may be reduced when the
scientific basis of an action is so well understood and
This review validates the presence of adequate structural
developed that it has defined indications and methods of
elements, that is, physical facilities, equipment, and per-
execution. If medication orders are written with errors
sonnel, management algorithms (clinical and logistical),
in dosage and spelling, a review of the process or
safety measures, and expected performance limits. The
structure might both show that those errors could be
definition of what constitutes adequate structure is es-
eliminated through use of a computer-based order writing
sential for structural review to be useful. For example,
system.
expectations need to be identified for staffing levels and
A more complex example is the selection of patients
expected capacity to move patients along the surgical care
for elective tracheal intubation. Assume that reliable
pathway, from the operating room (OR) to the nursing
evidence shows rapid sequence inductions fail in an
unit. If the suitability of equipment is to be validated, the
unacceptable number of patients with a body mass
expected purpose, performance standards, and limits of exceeding a certain index. A policy for performing awake
that equipment should be specified. intubations in the entire cohort of these patients might
Often, obvious structural deficiencies go unrecog- be adopted, thereby converting a problematic decision-
nized as such, even in extensive, well planned studies. making process, by virtue of standardization, into one
Half of the deaths and neurological injuries in a clas- of structure. As with all standards, such a policy does
sic study of 198,103 patients in 460 French institutions not restrict the right of the individual anesthesiologist to
were caused by hypoventilation during the postoperative decide a matter of process to reduce a known risk based
period.19 The study concluded that this was a result of on evidence in a cohort when one cannot determine in
the popularity of narcotic anesthesia in France. Similar advance which member of the cohort is at risk.
results were obtained in a retrospective study of over two For the process of assessment to be successful, the
million anesthetics in North Carolina.20 Although narcotic following issues should be defined: (i) Adverse events to
usage may have been the precipitating factor, the lack of be reduced, (ii) ideal outcome, and (iii) highly specific and
recognition of hypoventilation (an issue of process) may verifiable changes in management. When these changes
have been the more correctable root cause not addressed are adopted, they should lead to the desired outcome
in either study. through a reduction in adverse events. Ultimately, it is not
Even when the objective of the study is to relate struc- necessary to make an absolute distinction between errors
tural failure to the process of anesthetic administration, in management (process), technical errors (process or
the definition of structure may be drawn too narrowly. structure), and purely equipment problems (structure), as
Cooper, using critical incident techniques, described a 4% long as the quality assessment process detects the problem
occurrence of critical incidents attributable to equipment and can correct it with the appropriate improvement in
failure in 1089 patients.21 Drug administration errors, IV outcome.
apparatus problems, gas flow errors, anesthesia circuit
disconnects, and other factors were defined primarily as Outcome Review
errors in the process rather than the structure of provid-
ing anesthesia. However, these elements have significant These types of evaluations involve endpoints of care,
structural significance, the appreciation of which is re- including morbidity and mortality, length of hospital
flected in the subsequent improvements to anesthesia stay, escalation of care including unexpected outpatient
machines. The study did conclude that many of these admission, and overuse or underuse of blood products,
incidents could be reduced by changes in monitoring drugs or monitoring techniques. The purpose of the
techniques or the adoption of different management algo- outcome review is to determine when a problem exists
rithms, that is, structure-related changes. that requires corrective action. Because the multiple
The key to maximizing the use of structure analysis, antecedents of outcome often reinforce or cancel each
therefore, is to include within its sphere the identifica- other, good care and bad care do not always result
tion of errors in the decision-making process that can in proportionally good and bad outcome. Therefore,
be modified by structural change, for example, intelligent although outcome is the result of its antecedent causes,
alarm systems, redundant syringe labeling, or the auto- inferring these antecedent causes from outcome is not
matic detection of potentially hazardous combinations straightforward in medicine.
of drugs. Elements of process, which can be reduced to Episodic outcome assessment has a long tradition in
algorithms or policies, for example, generally accepted, anesthesiology in the form of mortality and morbidity
46 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
conferences; they often served before the 1980s as the pointing to specific processes that need improvement is
only form of quality review. This practice was an integral limited by uncontrollable variables and the difficulty of
part of surgery, from which anesthesia emerged as a distinguishing provider-caused events from those caused
discipline.23 Our use of outcome analysis to point to by process. The same outcome may result from the
specific problems of structure and process is more recent combination of a competent provider using a flawed
and still evolving. Not until 1999 was a structured peer process and an incompetent provider using a well
review (SPR) model in anesthesiology introduced that designed process. In addition, delayed outcome blurs its
looked at system errors as critically as human errors.24 relation to the behavior of the provider and the quality of
The measurement of defined indicators of outcome along the process.
the care pathway became a central piece in the JCAHOs
Agenda for Change paradigm in the early 1980s and had
a major influence on outcome measurement. However,
measurable outcome rarely points to the root cause of a What Are the Roles of Industry,
problem, only its existence.
Outcome may be positive or negative, although the Statistical Quality Control,
terminology most commonly refers to adverse outcomes. and Continuous Quality
Adverse patient-related occurrence (APO) is a relatively
old term, but is as useful as any number of other Improvement?
termscomplication, adverse event, untoward outcome,
or variancethat refer to negative outcomes related to The origins of continuous quality improvement (CQI) in
care. While the occurrence of negative outcomes tends to industry began with the work of statisticians like Walter
be most frequently measured, positive outcomes, when Shewhart, W. Edwards Deming, and Joseph Juran,25 all
expectations are met, are also important. An APO in of whom introduced the concept that quality can be
this chapter designates a negative outcome related to measured and analyzed to reduce the incidence of defects
patient care. and improve the product quality. It had little immediate
Sometimes outcome causes are obscure or multivari- impact in America, but it revolutionized manufacturing by
ate. The usefulness of outcome studies in altering clinical the Japanese who added to and refined the methodology
practice depends on our ability to ferret out sometimes from 1950 to 1980. CQI was reintroduced to America in
complex cause-and-effect relationships. Good outcome the early 1980s as industry became aware of its potential
does not prove the absence of management errors. Con- to simultaneously improve the quality and reduce the cost
versely, the absence of patient management errors does of production of goods and services.
not preclude a bad outcome because the mechanisms can The application of statistical process control to the
be subtle and previously unknown.23 For example, the in- measurement of quality in health care has been discussed
judicious use of muscle relaxants may be reflected by the extensively,26 including specific examples of its use in
number of patients who require unexpected postoperative defining the quality of perioperative care.24 The JCAHO
ventilation or reintubation in the acute recovery period. consciously borrowed from the ideas of George Labovitz
Many coexisting variables interact to cause the specific in formulating its philosophy of quality improvement and
incidence of this problem: (i) The frequency with which using quality teams to improve performance.
relaxants are used determines the population of patients Although Demings original writings were highly
at risk and is inflated by the frequency with which relax- technical and difficult to read, even when they were
ants are used in excess of that needed; (ii) the methods intended for the public, the methodology used and its
of reversal and testing of adequate reversal of relaxants impact on the rise of Japanese industry were captured in
may be the proximate cause of residual postoperative re- a very accessible, well-written book by Walton.16 Today,
laxation; and (iii) the monitoring capabilities and size of we use Demings principles of statistical quality control to
the staff available in the postanesthesia recovery area can improve practices that work and eliminate those that do
amplify or reduce the frequency of the problem and may not. At the same time, we attempt to improve the efficiency
determine its early detection and intervention. and economy of medical practice.
Multiple process and structure variables present the Eventually the JCAHO began to adopt the methods of
potential for changing the outcome if any one of the quality improvement that were very successful in industry
variables changes. If there is a high institutional use of as a central principle of performance improvement. First,
muscle relaxants, but the intraoperative and postoperative it introduced quality screening as a means for populating
monitoring and control of these agents is excellent, then a statistically valid quality measurement database.27
the excess use is unlikely to be detected. However, if a Subsequently, it incorporated the structure, process, and
shortage of twitch monitors, oximeters, capnographs, or outcome concepts of Donabedian; the performance limits
postanesthesia recovery room nurses were to develop, a introduced by Deming; and the methods used by Labowitz
major increase in respiratory arrests might occur in the to focus on CQI. Responding to criticism that it provided
postoperative period. hospitals with no strategic tools to help them survive
In summary, outcome measurement is useful in in difficult times, or even guarantee that they would
evaluating the gross confirmation of successes and failures be accredited by their state inspectors,28 the JCAHO
and results of changes in processes and structures. became more collaborative as a consultant for quality
Its usefulness in evaluating individual performance or management in conjunction with its role as regulator.
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 47
example, if controlling dental trauma is the objective, are made on the basis of initial conclusions or through
one should be able to demonstrate that the number of continuous monitoring of statistical process control.
damaged teeth resulting from intubation is at a specific If a program designed to improve outcome does not
verifiable level through the application of statistical produce some verifiable reduction in risk, the program is
process control. Intraoperative mortality is another clearly superficial or the wrong indicators of quality are being
verifiable indicator of outcome. As with chipped teeth, observed. That is to say, if the practice profile of the
however, it does not in itself define causation and, providers always meets the established standards, the
therefore, should be combined with a peer review process standards probably are set too low.
and continuously monitored.24
An objective may be so broad that it must be verified
by breaking it down into its component parts. When
John Snow reported intraoperative deaths in 6 of 80 What Defines a Quality
patients undergoing ether anesthesia, he concluded that
the cause of death was not due to the administration
Management Program for
of the anesthetic.34 Any contribution of the anesthetic Anesthesiology?
to death was masked by the severe risks of surgery
in the mid-19th century. If he had participated in a To evaluate the quality of care, the services and associated
modern quality improvement program, he may have inherent risks within a department must be clearly de-
drawn more heavily on his extensive prior experience fined. Quality measures to be monitored are then drawn
as an epidemiologist to break down the outcome of from the resulting service-risk profile (Table 3.1). All pa-
intraoperative death into a series of objective, contributing tients should be monitored for these events through a uni-
factors. These factors could include indicators such as form, systematic screening process. Predefined criteria are
monitoring, hypoventilation, hypoxemia, hypotension, used to evaluate the findings, which are regularly dissemi-
pulmonary aspiration, duration of the procedure, ASA nated to department members. Recommendations for cor-
physical status, surgeon, anesthetist, and completeness of rective action are made and implemented, and resulting
the preoperative evaluation. Furthermore, he might have improvements are documented. Computer software pro-
evaluated all 80 patients in this manner and compared the grams that support the process may be obtained by ASA
frequency of adverse outcomes in the survivors to those members from the ASA web site (http://www.asahq.org).
patients who died. What he did, in fact, was an extremely The ASA Manual for Anesthesia Department Organi-
simple cohort study with a rudimentary denominator. zation and Management (MADOM) and the ASA Quality
Management Template37 are among the current quality
management tools available. The Quality Management
ESTABLISH REASONABLE Template is intended to be used off the shelf by depart-
ments of anesthesiology to improve the quality of patient
PARAMETERS TO DRAW care and to meet various internal and external administra-
VALID CONCLUSIONS tive requirements (e.g., accreditation). The need for such a
template became apparent from the ASA member requests
To produce valid conclusions, any quality management submitted to the Committee on Quality Management and
program should observe basic epidemiologic methods. Departmental Administration and from the findings of
An objective of improving the quality of care may be the ASA Anesthesia Consultation Program. Similarly, the
narrow or broad, as long at it can be confirmed using section of the MADOM on quality improvement has nu-
the same definitions of outcome and methodology in merous web links to standards, regulatory issues, and
different institutions. Calculation of the incidence of developments in quality management that are updated
contributing factors should be included so that cause- frequently and can be accessed on the web.38 A partial
and-effect relationships can be established. However, even listing of some of these links can be found in Table 3.2.
the most careful and extensive prospective cohort studies
do not guarantee that the proper relationship between
adverse outcome and cause will be found.
Despite valid epidemiologic methods, cause-and- IMPORTANT REQUIREMENTS
effect relationships may be initially obscured. The OF THE PROGRAM
Beecher-Todd study,35 performed between 1948 and 1952
on 599,548 patients in 10 hospitals, demonstrated that Any quality management program has the common
during anesthesia, the use of muscle relaxants, and not elements described in the following pages.
cyclopropane, was associated with a higher incidence
of intraoperative mortality than anesthesia given with- Service-Risk Profile
out muscle relaxants. This unwarranted conclusion was
based on the alleged, but nonexistent toxicity of muscle The specific data to be collected are based on the service-
relaxants, and the results were largely a consequence of risk profile, which defines all activities and services
the newness of the drugs in question.36 A valid cause- provided by an anesthesia department, as well as some
and-effect relationship can only be determined by reex- general ones (Table 3.1). It requires an appreciation of
amination of the problem after modifications in practice the inherent risks; adverse outcomes; when, where, and
50 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
under what circumstances they occur; and how they are selected, their occurrences are monitored patient by
evaluated. The profile must be modified to conform to the patient. A partial retrospective sample should not serve
characteristics of the department using it. as the basis of ongoing quality assessment
Quality referrals are systematically included (from
Systematic, Ongoing, and Routine other departments or committees). When other de-
partmental quality management committees or special
Collection of Data committees (e.g., transfusion committee) refer prob-
All major aspects of care are evaluated, including the lems related to the anesthesiology department, these
pain clinic and preoperative, postoperative, and intra- problems are addressed by the anesthesiology quality
operative activities defined in the service-risk profile management committee. Ideally, the intradepartmen-
All clinical divisions are involved. For practical pur- tal reporting system should already have detected these
poses, if the chairman of the anesthesia department is problems internally
responsible for the activities of the service, it is included
in the quality management program of the anesthesia Criteria Used to Evaluate Findings
department. Intensive care units (ICUs), postanesthesia Criteria are written prospectively and must be acceptable
care units (PACU), resuscitation teams, and diagnostic to the members of the department for the staff to have
laboratories are examples
faith in the validity of the system.
Ongoing data collection and evaluation is done for
each patient. Once specific indicators of quality (APOs, 1. PROCEDURES: In particular, risk-prone or hazardous
completeness of preoperative evaluation, adequacy of procedures should have generally recognized indica-
postoperative follow-up, efficient turnover, etc.) are tions that can be validated. Presumably, as medicine
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 51
becomes more of a science and less of an art, well Changes are made in standards of practice and policies
defined criteria will be developed on the basis of an in- of the department
creasing body of scientific knowledge. A good starting Equipment is upgraded or replaced
point for developing criteria based on current scien- Requirements for additional personnel are met
tific evidence is the ASAs Standards, Guidelines, and Delineation and/or restriction of staff privileges occur.
Practice Parameters.22 Continuing education, restriction of practice to proven
2. ASSESSMENT OF ADVERSE EVENTS AND THE APPROPRI- areas of competence, or decredentialing because of
ATENESS OF THEIR CORRECTION: The criteria must allow lack of improvement in unacceptable practices may
the reviewer to distinguish avoidable from unavoidable be required. For this reason, the quality management
problems, and appropriate versus inappropriate treat- process must be fair both in concept and application
ment. The process of assessment can be enhanced by to avoid the liability of unfair restriction or unjustified
careful definition of an adverse event. The parameters granting of privileges
accompanying the APOs listed on the sample qual-
ity management report form provide an example (see
Fig. 3.1). Dividing broad categories (e.g., airway, neuro-
Documentation of Improvements
logic, discharge planning) into discreet verifiable events The extent of improvement and what constitutes improve-
helps to improve the validity of the initial screening ment must be defined in advance. Both the quality and
process. extent of improvement should be reasonable, verifiable by
3. ASSESSMENT OF OUTCOME: A reduction in the incidence preexisting criteria, achievable by a well defined protocol,
of an APO that can be affected by a change in and reproducible. Failure to improve a particular problem
anesthesia-related structure or process depends on the may signal a failure to adhere to the prescribed solution
factors contributing to its outcome. The severity of or is simply a failure of the proposed solution itself. The
the problem to be resolved may be reflected by the
criteria are as follows:3
severity of the outcome (e.g., death) or by its frequency.
Stratifying outcome by severity helps to prioritize the Decrease in the frequency of problems may be adequate
need for action and justify the allocation of recourses proof of improvement. As noted previously, the absence
to do so. At a minimum, an assessment should be made of adverse events does not always prove the absence of
of the duration and severity of the APO. problems. However, they can be detected by monitoring
4. DEFINED MAXIMUM FREQUENCY OF ADVERSE EVENTS: the adherence to standards
Presently, the acceptable rate for many APOs is Improved outcome, other than a reduction in the
not universally agreed upon, and the frequency of frequency of APOs, may include more rapid recovery
their occurrence is uncertain. Therefore, national times, fewer unexpected referrals to ICUs, reduction in
benchmarks are poorly defined. Often, departments unexpected postoperative hospital admissions, or better
have chosen to compare this years frequencies to that utilization of resources
of last year, which shows the trend and sets statistical Increased productivity in terms of volume and speed
limits but does not precisely define acceptable limits. If may include less wasted time waiting for admission to
reliable external benchmarks are not available, internal or discharge from the PACU, more efficient anesthesia
benchmarks should be based on the incidence of APOs startup or emergence times, or shorter turnover times.
that are assessed to be unpreventable by improvements
to current practices.
5. ACCEPTANCE OF THE MEDICAL STAFF: Because each prac- Periodic Evaluation and Revision
tice is unique, the exact criteria by which that practice No method is forever. Therefore, a program should be
is judged will vary. Quality management programs do evaluated and revised periodically. Indicators, standards,
not succeed when peers do not acknowledge standards and criteria must conform to the problems observed, and
to which they will be held. screening indicators, which are rarely manifest, should
be eliminated. Although the principles outlined in this
Evaluation and Recommendations of Findings chapter conform to JCAHO requirements27 that have been
Findings are evaluated at monthly intervals, and rec- in effect since 1984, the latest Accreditation Manual for
ommendations are made and disseminated. A monthly Hospitals should be considered authoritative. Consider-
quality management conference is useful to provide a fo- ation should be given to including indicators from the
rum for discussion. Generally, the confidentiality of the ORYX indicator system, Sentinel Event Alerts, and the
proceedings is protected from outside discovery by law National Patient Safety Goals. Changes in the manual oc-
because they constitute a part of a state or federally man- cur yearly, although many are not substantive. Over the
dated peer review process. Where the law is unclear or years, the details of quality management and improve-
may not be deemed adequate to protect confidentiality, the ment have been shifted from the accreditation manuals to
findings may need to be de-identified before presentation. other publications of the JCAHO.
FIGURE 3.1 Quality Reporting Form. (Courtesy of Department of Anesthesiology, Shands Hospital
at the University of Florida, College of Medicine, Gainesville, Florida.)
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 53
quality report. They represent ideals and should be modi- relaxants to reduce the frequency and duration of
fied on the basis of the needs of individual departments. postoperatively retained endotracheal tubes)
is unclear.44,48 In actuality, however, there may be little practice guideline for acute pain management in 1992.
difference between a lack of efficacy and an unrealistic Key elements included patient education, assessment and
sample size required to demonstrate such efficacy. If pulse frequent reassessment of pain, and aggressive use of both
oximetry were 100% effective at preventing cardiac arrest drug and nondrug therapies for pain control. Although ev-
or hypoxic injuries during previous outcome studies, the idence supporting the guidelines recommendations was
number of patients who needed treatment to save a life impressive, the guideline had little effect on postopera-
would have ranged from 8,000 to 77,519. In contrast with tive pain. Possible reasons for this discrepancy include
other technologies, only 6 patients need to be treated by an ineffective guideline, undetected barriers to complete
bypass grafting for left main coronary artery disease, 8 implementation, insensitive outcome measures, and in-
high-risk patients need to be immunized with hepatitis B, sufficient time allowed for the diffusion of innovation.50
and 14 patients with transient ischemic attacks need to be Compare this type of guideline development, dissem-
started on a daily dose of aspirin to save a life. ination, and implementation practices to that of the ASA.
Clearly the benefit of pulse oximetry is relatively A task force was convened in 1991 to develop guide-
low when compared to other medical interventions, lines for the appropriate use of PA catheters in settings
yet most anesthesiologists believe that the routine use encountered by anesthesiologists based on a systematic
of pulse oximetry has reduced the rate of anesthesia- review of the clinical benefits and harms of PA catheteri-
related mortality.48 The faith of anesthesiologists in pulse zation.51 They updated their recommendations in 2002.
oximetry is largely based on this monitors ability to The original task force found it impossible to draw mean-
provide early warning, or lead time, of hypoxemia. In ingful conclusions about PA catheter effectiveness from
essence, the earlier warning of hypoxemia has been used the available literature due to poor study designs and lack
as an outcome indicator. of statistical power. Within 5 years, there were calls for a
These intermediate indicators, however, have not moratorium on the use of PA catheters because a prospec-
been validated in their ability to reflect the effectiveness of tive cohort study showed an increase in hospital stay,
patient care. Improved lead time does not necessarily hospital costs, and 30-day mortality in patients treated
result in improved outcomes. Consider the analogous with PA catheters.52 This publication was followed by a
situations of screening chest radiographs for lung cancer consensus statement of the U.S. Food and Drug Adminis-
in smokers and screening mammography for breast tration and the National Heart, Lung, and Blood Institute
cancer in women. Both of these modalities have resulted Consensus, along with the American College of Cardiol-
in earlier diagnoses, but little change in mortality. ogy, the American College of Chest Physicians, and the
Hypoxemia is also a poor indicator of outcome. In American Thoracic Society, asking for more randomized
reviewing the Danish study of pulse oximetry, one finds a controlled trials.53
19-fold increase in the incidence of diagnosed hypoxemia The ASA reconvened its task force on PA Catheteriza-
in the oximetry group over the control group. This tion in 2000 with the hope of making recommendations
increased incidence of diagnosed hypoxemia resulted based on a scientific review of the current literature, sup-
in significantly more therapy (i.e., longer PACU stays, plemented by revised expert opinions. An additional 90
more ICU admissions, increased use of supplemental articles published between 1994 and 2000 were reviewed.
oxygen, and more frequent administration of naloxone),
Unfortunately, most of the studies, including the random-
but no significant difference in postoperative mortality
ized control trials, were flawed because of poor study
nor cardiovascular and neurologic morbidity.
design, lack of statistical power, or incomplete documen-
In 2003, the Cochrane Collaboration published an
tation. The task force deduced that it is difficult to draw
outcome review49 which concluded that pulse oximetry
meaningful conclusions about the effectiveness and safety
can detect hypoxemia and related events, but there is
of PA catheterization based on currently available data.
no evidence that pulse oximetry affects the outcome of
In summary, the literature review clearly did not support
anesthesia. In summary, the reviewers concluded that the
the routine use of PA catheters if there was a low risk
value of perioperative monitoring with pulse oximetry is
of hemodynamic compromise. It was equivocal for situa-
questionable in relation to improved reliable outcomes,
tions with moderate or high-risk patients, procedures or
effectiveness and efficiency.49
practice settings, and expert opinion was not unanimous
for most of the clinical scenarios. Therefore, the task force
recommended that PA catheterization be performed only
EFFECT OF PRACTICE if the risks associated with the individual patient, surgery,
and the practice setting were taken into consideration.54
GUIDELINES ON OUTCOME Perhaps the most significant contribution the task
Little evidence shows that development and widespread force made to the development of guidelines for mon-
adoption of practice guidelines has affected perioperative itoring devices was the recognition of the limitations
outcomes. Consider postoperative pain. Over 23 million of expert opinion and the plea for a research agenda
surgical procedures were performed in the United States with meaningful outcomes. Moving toward that goal, a
in 1989, and most of these patients received postoper- recent multicenter, prospective, randomized, controlled
ative pain management. The former Agency for Health trial showed no change in mortality or major morbidity
Care Policy and Research convened a 2-year panel of when 994 high-risk patients underwent high-risk surgery
experts that published a coherent and flexible clinical followed by an ICU stay with the guidance of a PA catheter.
56 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
The power analysis in this study was sufficient to show Most hospital leaders have some appreciation for
that an adequate sample of patients was recruited.55 quality improvement, if only from the standpoint of
requirements for accreditation, but they often lack the
infrastructure to implement it. Quality initiatives often
PROBLEMS IN RELATING remain quite fragmented. This chapter is written from
the perspective that anesthesiologists must support their
ANESTHETIC MANAGEMENT quality improvement efforts internally because many
TO OUTCOME AND institutions do not provide the infrastructure to assist
them. To be most effective, an institution must take the
MORTALITY lead in developing and supporting common methods of
One might argue that lack of unequivocal data supporting measurement and decision-making support to address
these mechanisms of improved patient safety is not problems in an integrated manner along the entire care
evidence of the lack of their effect. Thus, one needs to look pathway and across all departments.
directly at measures of anesthesia-related mortality over For now, institutional leadership and support for
time. A review of Medline and HealthStar can generally quality management is still just over the horizon. Ef-
be summarized in four major categories: (i) Overall fective quality management methods and the resulting
perioperative mortality ranged from 1 death in 53 improvements in patient safety have emerged earlier and
anesthetics to 1 in 5,417 anesthetics; (ii) anesthesia- more effectively in anesthesiology than in other areas of
related mortality ranged from 1 in 1,388 anesthetics medicine, in part because of the clearly defined beginning
to 1 in 85,708 anesthetics; (iii) anesthesia considered and endpoints of care, the tight relationship of action
solely responsible for perioperative death ranged from to outcome, and the ability to rapidly test the effects of
1 in 6,795 anesthetics to 1 in 200,200 anesthetics; and, improvement efforts. These are the reasons why anesthe-
(iv) preventable anesthetic mortality ranged from 1 in siology continues to lead in efforts in patient safety; it is
1,707 anesthetics to 1 in 48,748 anesthetics.24 the prototype of quality improvement successes that will
The overall perioperative mortality rate for patients follow in other areas of medicine.
with ASA Physical Status 1 to 5 may be as high as
1 per 500 anesthetics. The literature suggests a wide
range of perioperative mortality rates, which are probably
caused by differences in operational definitions and What Are Risk Management
reporting sources, as well as a lack of appropriate risk
stratification. Data further suggest that the anesthesia-
Goals in Anesthesia Practice?
related mortality rate, as determined by peer review,
has been stable over the last decade, at approximately The concept of risk in medical care usually means
1 death per 13,000 anesthetics. Wide variations based on avoidable risk. Risk is often dramatized by the well known
methodologic differences reported in the literature make and widely quoted statistics from the Institute of Medicine
it impossible to detect trends in anesthesia safety.24 that tens of thousands of Americans are killed annually
Perhaps global trends toward anesthetizing more crit- by medical errors, and countless others are injured in
ically ill, and older, patients are making improvements in some way.40 The goal of RM in anesthesia is, foremost,
anesthesia safety,56 but mounting evidence demonstrates to prevent adverse outcomes and, secondarily, to deal
a gap between the optimal delivery of health care and with adverse events that do occur, thereby attempting to
the actual delivery of health care in the United States. limit damage both to the patient and to the anesthesia
Physicians are not always practicing in accordance with professionals involved.
accepted methods for achieving the best possible out- RM efforts that help prevent adverse patient outcome
comes. In 2001, the Institute of Medicine called these will, therefore, reduce liability exposure and resulting
gaps between best practices and actual practices a quality costs. In the early years of the new millennium, there
chasm.57 has generally been an alarming increase in medical-legal
It is the responsibility of hospital leadership to activity, reaching a new crisis stage in several states where
provide a model of quality improvement that goes beyond medical malpractice insurance has become outrageously
measurements of dysquality (a semi-neologism meaning expensive or simply unavailable at any price. The so-called
the presence of dysfunctional qualities) and establishes tort reform efforts in several states are intended to reduce
quality management as a key management tool that has malpractice insurance premiums and medical costs in
a major impact on patient care and institutional survival, general. A few examples appear possibly encouraging,
not just a paean to regulators. Such attempts often look but the overall impact is uncertain. Thus, awareness of
like a mishmash of measurements acquired in ways that medical-legal implications in practice is clearly necessary,
have little statistical validity. Useful quality measurements and case precedents have revealed specific features and
evaluate the major activities that a hospital needs to issues that increase legal liability risk.
work well for it to succeed, such as organ transplant Traditional RM was initially associated with the
survival, vascular graft patency, improved survival after financial side of business or professional activity. The
chemotherapy, or the effectiveness of trauma care. Useful insurance industry recognized riskcertain activities that
quality initiatives support institutional improvement and predictably led to a degree of loss. This relationship then
the meaningful awarding of hospital privileges. stimulated efforts to: (i) Plan to pay for it and (ii) reduce
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 57
the likelihood and/or amount of loss, thereby managing TABLE 3.3 Risk Management Issues in Anesthesia Care
the known risk. Regarding medical liability, financial loss
is caused by settlements and judgments from claims and Wrong site surgery (which will result in anesthesia
suits. RM stresses, first, the prevention of loss-generating professionals also being named as defendants)
adverse incidents or outcomes and, second, the effort to Operating room fires involving the patient (such as the
limit loss once an adverse event has occurred. cautery igniting a paper drape or residual alcohol prep
The classic RM process involves four steps: (i) Iden- solution on the face during MAC when supplemental
tification of a problem (actual or potential injury or loss); oxygen is flowing)
(ii) assessment and evaluation of the problem (deter- Untoward interactions of volatile anesthetics and CO2
mining the cause of injury or loss); (iii) resolution of absorbents generating dangerous substances
the problem (modification or elimination of the cause, Postoperative visual loss (particularly after prolonged
by change [practice, procedures, equipment, or behav- prone spine surgery)
ior]), and enforcement, (with sanctions if necessary); and Airway obstruction and hypoventilation after airway
(iv) follow-up on the resolution (to verify the desired result surgery (particularly in morbidly obese patients with
and ensure continued effectiveness). obstructive sleep apnea)
A major example in health care involved the stan- Postoperative hypoventilation after bariatric surgery
dards for intraoperative monitoring. An unacceptably Central or peripheral nerve injury from regional blocks
high number of major anesthesia malpractice claims led placed in adult patients rendered unconscious by
to research that ultimately yielded the original Harvard general anesthesia
standards.58 Issues identified were unrecognized hypoven- Peripheral neuropathies in general (even when all
tilation (primarily), inadequate inspired oxygen, and so known precautions have been taken)
on, and resolution involved the implementation and en- Damage from placement of central venous catheters
forcement of the standards (changes). Follow up studies (including infections and sepsis)
suggested that the new standards had a positive impact.59 Alleged awareness during general anesthesia (with or
The related RM history of monitoring standards from without pain)
the ASA in the United States, in other countries, and for
the world by the International Task Force on Anesthesia MAC, monitored anesthesia care.
Safety (and then, in 1992, the World Federated Societies
of Anesthesiologists60 ) is well known.
rewards for those who follow these strategies will lead
to their widespread, voluntary adoption. In this way, it is
hoped, functional crowding out of substandard or danger-
CLINICAL ASPECTS ous practices will result. The potential value of this best
practices model in anesthesia care remains to be seen.
Valid programs in anesthesia RM61,62 should cover all
relevant aspects of practice. They must emphasize the
creation of optimum conditions of the who, what,
and how of anesthesia practice, optimum preparation, SYSTEMS ASPECTS
awareness, and skill of the anesthesia professionals. It
is impossible to mention, in much less detail, all the Managed Care Impact and Production
relevant risks in anesthetizing a patient. Rigorous applica- Pressure
tion of the standards for basic intraoperative monitoring
(particularly for ventilation but also oxygenation, circu- The enormous emphasis on cost cutting in medical care
lation, and temperature) should help avoid catastrophic has created an entire new set of risks. Patients (or
anesthesia accidents. By generating the earliest possible their survivors), who believe that they were wrongly
warning of events such as accidental disconnection of the denied by such policies, are filing malpractice claims
breathing circuit Y-connector from the endotracheal tube, against their health maintenance organization/managed
such monitoring serves as an early warning system. Air- care organization (MCO) and the physicians who accepted
way management, particularly tracheal intubation after the denials of care.
muscle relaxation, appears least improved over the last Anesthesia professionals are likely to face denial of
25 years. Other frequently discussed topics are listed in MCO coverage for workups (such as a cardiac echocar-
Table 3.3. diogram) done for worrisome preoperative findings; pre-
System failure versus human error as causes of ad- operative admissions to tune up chronically ill patients
verse anesthesia outcomes has been extensively analyzed. (e.g., severe asthma); invasive intraoperative monitoring;
Both are contributors, and the application of these find- and postoperative admission for monitoring and care of
ings will be an important component of future clinical patients scheduled for outpatient surgery. Nevertheless,
anesthesia RM efforts. anesthesia personnel should put the welfare of the patient
A new approach to RM in anesthesia practice involves first and strive to do what is obviously reasonable. Such
persistent attention to good outcomes, not just adverse efforts, which advocate increased involvement and care,
events. Identification of strategies, practices, and proto- must be scrupulously documented. They may necessitate
cols that are successful in preventing adverse outcomes is postponement of a scheduled case (explain the reasons
important. It is hoped that subsequent emphasis to and to the irate surgeon, because he would be a defendant
58 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
too) or absorbing the associated costs (e.g., postoperative long history of adverse outcomes and resultant medicole-
admission and monitoring). gal actions. Legal doctrines such as vicarious liability
Anesthesia professionals must not be pressured into and agency must be considered. Specific applicability
activities that they know are unwise and unsafe. Pervasive varies by jurisdiction.
production pressure on anesthesiologists, can degener- In general, if an individual, group, or institution hires
ate into a form of economic credentialing. Providers who or even approves a physician by securing or granting
are judged too slow between cases, or use too many expen- privileges, that entity may be held liable along with
sive monitors and drugs, may face loss of patients from an the practitioner for his or her actions. This problem,
MCO, or loss of privileges at a facility. Intense pressure to of course, would be especially likely to occur if it was later
use as few people and resources as possible will inevitably discovered that the credentialing process failed to reveal
lead to cutting corners with potential danger to patients.63 or to examine something questionable in that persons
Safe, reasonable care must prevail. past. The honest majority of practitioners must recognize
that such efforts are intended to protect patients and
Offsite Anesthesia the integrity of the profession and that an applicants
references must be checked. Written references that say
Anesthesia providers work in a dizzying array of locations very little or have implications between the lines can be
outside the traditional hospital or ambulatory surgery followed up with (unrecorded) telephone calls. Also, when
center ORs. RM issues are significant. Wherever it may an anesthesia professional assumes any new position,
be, especially in office-based anesthesia,64 the equipment there must be a thorough orientation and checkout to
(anesthetic, monitoring, and resuscitation); facilities (O2 , prevent dangerous errors caused by unfamiliarity with
suction, tilting table, etc.); and resources (emergency help, the new practice setting.
recovery personnel, etc.), must meet the standards of and Facilities have protocols for granting clinical privi-
be functionally equivalent to a regular OR. leges. Periodic renewal of privileges must be taken just
Anesthesiologists may be pressured to train and cre- as seriously. Personal reluctance to revoke or restrict a
dential registered nurses (RNs) (sedation nurses) to colleagues privileges may stem from fear of a retaliatory
work in endoscopy or procedure suites in which such lawsuit. However, legal precedents have held a facility
requirements often are impossible to meet. Prevention and/or its staff liable if the incompetence of a practitioner
of dangerous, flying metal objects during magnetic res- was known, or should have been known, but was not
onance imaging is a concern. Anyone who agrees to addressed. Another question to be addressed is whether
administer anesthetics in substandard/inadequate loca- all anesthesia professionals should have blanket privileges
tions or conditions increases the risk to patients and his to undertake any anesthetic challenge. The RM consider-
or her medical-legal liability exposure. ations are significant if practitioners who are not really
A valid RM strategy is to insist that the location qualified or lack the necessary experience are allowed or
be certified by one of the three office-based anesthesia expected to undertake major challenges for which they
accreditation agencies: Accreditation Association for Am- are not prepared. The likelihood of complications in this
bulatory Health Care, Inc; American Association for Ac- scenario is increased and the ability to defend the prac-
creditation of Ambulatory Surgery Facilities, Inc; JCAHO; titioner against a claim is decreased. There is no clear
or by an equivalent state or local regulatory authority. The answer on this question of procedure-specific privileges.
demand for sedation/analgesia services is another trend Each facility, department, and group needs to address
fraught with major potential risks. these complex issues.
Much more invasive and longer procedures are being
attempted with minimal sedation that can often evolve to
a room-air general with a propofol/narcotic/midazolam Policies and Procedures
infusion. These patients must be fully monitored, espe-
cially their ventilation (using new capnography technol- Developing written policies and procedures often is per-
ogy made specifically for sedation), and provided with ceived as merely more bureaucratic drudgery. Creating
supplemental oxygen when indicated (although some or updating such a manual (prototype examples exist65 )
providers claim that oxygen administration delays the forces practitioners to think about both mundane and rare
detection of hypoventilation, this potential problem is issues and may reveal an accident waiting to happen that
unlikely to occur if capnographic monitoring is used). can be prevented. This manual logically is divided into two
parts: Organizational and procedural. Under organization
Credentialing and Clinical Privileges is the delineation of responsibilities and expectations and
a communication plan for all personnel. This section is
Licensure, credentialing, and securing practice privileges especially important for a practice that covers multiple fa-
are often perceived as an annoyance by health profes- cilities from the physicians home on nights and weekends.
sionals. However, there are important RM implications, The procedural component outlines the proposed actions
and these processes must be taken seriously. The guiding for particular situations and includes: ASA statements,
assumption is that appropriately educated, trained, and guidelines, and standards; disaster drills; and protocols
experienced practitioners will provide higher quality care required by JCAHO standards. A thorough policy and pro-
with less risk of danger to patients than will fraudulent, cedure manual can help prevent adverse events and help
criminal, deviant, or poor quality practitioners with a in crisis management.
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 59
KEY POINTS
FOLLOWUP 1. In anesthesia RM, the themes of quality patient care
The primary anesthesia provider and others involved and concern about liability exposure for anesthesia
must document relevant information. Never change any professionals are intertwined.
CHAPTER 3/QUALITY ASSURANCE AND RISK MANAGEMENT 61
2. Concerns about errors in medical care harming 16. Walton M. The Deming management method. Chicago, IL:
Putnam; 1986.
patients, and about malpractice liability insurance
17. Donabedian A. Evaluating the quality of medical care.
being exorbitantly expensive or unavailable, are
Milbank Mem Fund Q. 1966;44:166.
exacerbated by the talk-show mentality. 18. Laffel G, Blumenthal D. The case for using industrial quality
3. The temptation to cut corners is very powerful be- management science in health care organizations. JAMA.
cause of production and cost pressures and because, 1989;262:3869.
almost always, there is no adverse consequence. 19. Tiret L, Desmonts JM, Hatton F, et al. Complications
4. Avoidable risks always exist; unfortunately, it is associated with anaesthesia - a prospective survey in France.
impossible to predict when and where the rare but Can Anaesth Soc J. 1986;33:336.
devastating case will occur. 20. Bechtoldt A. Committee on anesthesia study anesthetic
5. Avoidable risks must be avoided. related deaths: 19691976. N C Med J. 1981;42:253.
6. Anesthesia care has become a model for safe patient 21. Cooper JB, Newbower RS, Kitz RJ. An analysis of major
care.73 errors and equipment failures in anesthesia management:
Considerations for prevention and detection. Anesthesiology.
7. Although anesthesia is safer for patients now com-
1984;60:34.
pared to 25 years ago,74 a distinct number of adverse
22. American Society of Anesthesiologists. Standards, guide-
events will continue to occur. lines, and statements. Park Ridge, IL: American Society
8. In this era of managed care, application of RM of Anesthesiologists. Available at: http://www.asahq.org/
principles and quality of care should help minimize publicationsAndServices/sgstoc.htm. Accessed June 28,
the risks of adverse patient events and resulting 2006.
malpractice claims against anesthesia professionals. 23. Keats A. Role of anesthesia in surgical mortality. In:
Orkin FK, Cooper LH, eds. Complications in anesthesiology.
Philadelphia, PA: JB Lippincott Co; 1983:3.
REFERENCES 24. Lagasse R. Anesthesia safety: Model or myth?: A review of
1. Walker vs. Anesthesia Associates and others. CV-79-00675, the published literature and analysis of current original data.
13th Cir, Alabama; 1981. Anesthesiology. 2002;97:1609.
2. Cohen JA, Cohen JA. Quality assurance and risk manage- 25. Wheeler DJ, Chambers DS. Understanding statistical process
ment. In: Gravenstein N, ed. Manual of complications during control, 2nd ed. Knoxville, TN: Statistical Quality Control;
anesthesia. Philadelphia, PA: JB Lippincott Co; 1991:144. 1992:4.
3. Duberman S. Quality assurance in the practice of anesthesiol- 26. Carey RG, Lloyd RC. Measuring quality improvement in
ogy. Park Ridge, IL: American Society of Anesthesiologists; healthcareA guide to statistical process control applications,
1986. quality resources. New York: American Society for Quality.
4. Kelly W. Zeroing in on those polluters: We have met the Quality Resources, A Division of the Kraus Organization
enemy and he is us. In: Kelly S, Crouch Bill Jr., Crouch Bill, Limited; 1995.
eds. The Best of Pogo. A Fireside Book. New York: Simon & 27. Joint Commission on Accreditation of Healthcare Orga-
Schuster Inc; 1982:224. nizations. Accreditation manual for hospitals. Chicago, IL:
5. Deming W. Out of the crisis. Cambridge, MA: Massachusetts JCAHO; 1984.
Institute for Advanced Engineering Studies; 1986. 28. Bogdanich W. Small comfort: Prized by hospitals, accredita-
6. Flexner A. Medical education in the United States and tion hides perils patients face. Wall St J. 1988;212(12):A1,A12.
Canada. New York: The Carnegie Foundation for the Ad- 29. Joint Commission on Accreditation of Healthcare Organi-
vancement of Teaching; 1910, Available at: http://www zations. Comprehensive accreditation manual for hospitals.
.medicinenet.com/script/main/art.asp?articlekey=8795. Ac- Oakbrook Terrace, IL: Joint Commission Resources Inc;
cessed February 2, 2006. 2005.
7. Passaro E, Organ CH Jr. Ernest A. Codman: The improper 30. Katz RI, Lagasse RS. Factors influencing the reporting of
Bostonian. Bull Am Coll Surg. 1999;84:16. adverse perioperative outcomes to a quality management
8. Donabedian A. The quality of medical care: Methods for program. Anesth Analg. 2000;90:344.
assessing and monitoring the quality of care for research 31. Joint Commission on the Accreditation of Health Organiza-
and quality assurance programs. Science. 1978;200:856. tions. Task forces lay groundwork for new survey process. In:
9. Codman EA. A study in hospital efficiency: As demonstrated Agenda for change update, Vol. 1, No. 1. Oakbrook Terrace,
by the case report of the first five years at a private hospital. IL: JCAHO; 1987:1.
Oakbrook Terrace, IL: Joint Commission on Accreditation of 32. Joint Commission on Accreditation of Hospitals. Medical
Healthcare Organizations; 1966. staff issues in the changing hospital environment. Chicago,
10. Roberts JS, Walczak RM. Toward effective quality assurance: IL: JCAH; 1987.
The evolution and current status of the JCAH QA standard. 33. Joint Commission on Accreditation of Healthcare Organiza-
QRB Qual Rev Bull. 1982;10:1115. tions. Public policy initiatives: Principles for the construct of
11. The Comprehensive Medical Malpractice Reform Act of pay-for-performance programs. Chicago, IL: JCAHO; 2005:3.
1985. Internal risk management program, Florida Statutes 34. Snow J. On the inhalation of the vapour of ether in surgical
395.041, 1986; Title XXIX FS 395.0195, 2005. operations. London: John Churchill; 1847.
12. Medical Incident Recovery Act, (CS/SB 6) 1988: Florida 35. Beecher HK, Todd DP. A study of the deaths associated
Statutes. 766.207, 766.209, 1988. with anaesthesia and surgery based on a study of 599,548
13. Darling v. Charleston Community Memorial Hospital. 33 Ill. anaesthesias in 10 institutions 19481952, inclusive. Ann
2d 326, 211 N.E. 2d 253, 1965. Surg. 1954;140:2.
14. Johnson v. Misericordia Community Hospital. 99 Wis 2d, 708, 36. Bunker JP. Historical aspects. In: Nunn J, ed. Epidemiology
301 N.W. 2d 156, 1981. in anaesthesia: The techniques of epidemiology applied to
15. Patrick v. Burget. 800 F.2d 1498 [9th Cir.], 1986. anaesthetic practice. London: Edward Arnold; 1986:17.
62 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
37. Bierstein K. Quality management tools for the anesthesia 56. Clergue F, Auroy Y, Pequignot F, et al. French survey of
department. ASA Newsl. 2005;69:34. anesthesia in 1996. Anesthesiology. 1999;91:1509.
38. Cohen, JA, ed. Quality improvement and peer review in anes- 57. Institute of Medicine Committee on Quality Health Care in
thesia. Manual of department organization and management America. Crossing the quality chasm: A new health system for
(MADOM). Park Ridge, IL: American Society of Anesthesiol- the twenty-first century. Washington, DC: National Academy
ogists; 2005. Press; 2001.
39. Cohen JA. Quality assurance initiatives. Probl Anesth. 58. Eichhorn JH, Cooper JB, Cullen DJ, et al. Standards for
1991;5:277. patient monitoring during anesthesia at Harvard Medical
40. Kohn LT, Corrigan, JM, Donaldson MS, eds. Committee on School. JAMA. 1986;256:1017.
Quality of Health Care in America. To err is human: Building 59. Eichhorn JH. Prevention of intraoperative anesthesia acci-
a safer health system. Washington, DC: Institute of Medicine. dents and related severe injury through safety monitoring.
National Academy Press; 2000. Anesthesiology. 1989;70:572.
41. Gaba DM. Anesthesiology as a model for safety in health 60. Vickers MD, Robins DS. International Task Force on
care. Brit Med J. 2000;320:785. Anaesthesia Safety. Eur J Anaesthesiol. 1993;10(Suppl 7):1.
42. Stoelting R. APSF response to IOM medical error report. 61. Pierce EC, ed. Risk management in anesthesia. Int Anesthe-
Anesth Patient Safety Found Newsl. 2000;15:1. siol Clin. 1989;27(3):217.
43. Joseph V, Lagasse R. Monitoring and analysis of outcome 62. Eichhorn JH. Risk management. In: Benumof JL, Said-
data. Int Anesthesiol Clin. 2004;42:113. man LJ, eds. Anesthesia and perioperative complications, 2nd
44. Moller JT, Johannessen NW, Espersen K, et al. Randomized ed. St Louis, MO: MosbyYear Book; 1999:771.
evaluation of pulse oximetry in 20,802 patients: II. Periopera- 63. Morell TC, Prielipp RC. Does the pressure to do
tive events and postoperative complications. Anesthesiology. more, faster, with less, endanger patients? APSF Newsl.
1993;78:445. 2001;6:1. Available at: http://www.apsf.org/resource center/
45. Websters new collegiate dictionary. Springfield, MA: G. & newsletter/2001/spring/01prodpressureintro.htm. Accessed
C. Merriam Company; 1976:526. June 28, 2006.
46. Clark D. The Hawthorne effect. Available at: http://www 64. Twersky TS. Standards for office anesthesia vary widely or
.nwlink.com/donclark/hrd/history/hawthorne.html. Ac- do not exist. APSF Newsl. 2000;15:1. Available at: http://www
cessed: February 8, 2006. .apsf.org/resource center/newsletter/2000/spring/05-Twersky
47. Mayo E. The human problems of an industrial civilization. .htm. Accessed June 28, 2006.
New York: Macmillan; 1933. 65. American Society of Anesthesiologists. Manual for anesthesia
48. Duncan PG, Cohen MM. Pulse oximetry and capnography department organization and management. Park Ridge, IL:
in anaesthetic practice: An epidemiological appraisal. Can J American Society of Anesthesiologists; 2005.
Anaesth. 1991;38:619. 66. American Society of Anesthesiologists. American Soci-
49. Pedersen T, Dyrlund Pedersen B, Mller AM. Pulse oximetry ety of Anesthesiologists guidelines for determining anes-
for perioperative monitoring. Cochrane Database Syst Rev. thesia machine obsolescense. Posted June 22, 2004.
2003;(2):CD002013. Available at: http://www.asahq.org/publicationsAndServices/
50. Jiang HJ, Lagasse RS, Ciccone K, et al. Factors influencing machineobsolescense.pdf. Accessed June 28, 2006.
hospital implementation of acute pain management practice 67. U.S. Food and Drug Administration. Department of Health
guidelines. J Clin Anesth. 2001;13:268. and Human Services. MedWatch. The FDA safety infor-
51. Roizen M. Practice guidelines for pulmonary artery catheter- mation and adverse event reporting program. Available at:
ization. A report by the American Society of Anesthesiologists http://www.fda.gov/medwatch/. Accessed June 28, 2006.
Task Force on Pulmonary Artery Catheterization. Anesthesi- 68. Cooper JB, Cullen DJ, Eichhorn JH, et al. Admin-
ology. 1993;78:380. istrative guidelines for response to an adverse anes-
52. Connors AF Jr, Speroff T, Dawson NV, et al. The effec- thesia event. J Clin Anesth. 1993;5:79. Also available
tiveness of right heart catheterization in the initial care of at: http://www.apsf.org/resource center/clinical safety/. Ac-
critically ill patients. SUPPORT Investigators. JAMA. 1996; cessed Jun 28, 2006.
276:889. 69. Bacon AK. Death on the table. Anaesthesia. 1989;44:245.
53. Bernard GR, Sopko G, Cerra F, et al. Pulmonary artery 70. Eichhorn JH. Patient perspectives personalize patient safety.
catheterization and clinical outcomes: National Heart, APSF Newsl. 20052006;20(4). Available at: http://www.apsf
Lung, and Blood Institute and Food and Drug Administra- .org/resource center/newsletter/2006/winter/01perspective
tion Workshop Report. Consensus Statement. JAMA. 2000; .htm. Accessed June 28, 2006.
283:2568. 71. Trombly ST. Adverse events require communication
54. Roizen M, Berger DL, Gabel RA, et al. Practice guidelines and disclosure. APSF Newsl. 2006;21(1). Available at:
for pulmonary artery catheterization: An updated report http://www.apsf.org/resource center/newsletter/2006/spring/
by the American Society of Anesthesiologists Task Force 01adverse event.htm. Accessed June 28, 2006.
on Pulmonary Artery Catheterization. Anesthesiology. 2003; 72. Kraman SS, Hamm G. Extreme honesty may be the best
99:998. policy. Ann Intern Med. 1999;131:963.
55. Sandham JD, Hull RD, Brant RF, et al. A randomized, 73. Cooper JB, Gaba DM. No myth: Anesthesia is a model for
controlled trial of the use of pulmonary-artery catheters addressing patient safety. Anesthesiology. 2002;97:1335.
in high-risk surgical patients. N Engl J Med. 2003; 74. Eichhorn JH. Monitoring standards: Role of monitoring in
348:5. reducing risk of anesthesia. Probl Anesth. 2001;13:430.
CHAPTER MEDICINE AND LAW: WHEN TWO
4
WORLDS COLLIDE
And for the greater security of the weak commons, he gave without incident) and ordered immediate treatment with
general liberty of indicting for an act of injury; if any one antihistamines and steroids.
was beaten, maimed or suffered any violence, any man Gradually, the swelling subsided, and the patients
that would and was able might prosecute the wrongdoer; diet was advanced over the next few days as bowel function
intending by this to accustom the citizens, like members of was restored. Persistent postoperative complaints of
the same body, to resent and be sensible of one anothers severe throat pain and dysphagia were largely ignored
injuries. And there is a saying of his agreeable to his law, for, for 7 days. Finally, the attending surgeon ordered an ENT
being asked what city was best modelled, That, said he, consultation. During diagnostic laryngoscopy, the patient
where those that are not injured try and punish the unjust lost consciousness and was transferred to a major medical
as much as those that are. center in a comatose state. Six months later, she died from
(Solon, Plutarchs Lives, Vol. I, The Dryden complications of overwhelming mediastinal sepsis.
Translation, Ed. A. H. Clough, 2001 Modern
Library Paperback Edition)
A
went surgical repair of an incisional hernia Anesthetic complications can occur without negligence,
and small bowel obstruction under general or even when physicians follow the standard of care.
anesthesia. Surgery was uneventful. As the However, for the patient and family, the loss of a loved
patient was emerging from anesthesia, she one is significant, especially when it results in death.
encountered difficulty in breathing. Auscul- Many lay people do not accept that serious complications
tation of the chest revealed severe bilateral wheezing. and death can happen without someone being at fault.
The anesthesiologist informed the patient that she had to Litigation over a patients death is an emotional event
reinsert the endotracheal tube and proceeded with a semi- for both the patients family and the physician. The
emergent reintubation of the patients airway, followed litigation arena is foreign to most people, particularly
by positive-pressure ventilation and gradual reemergence physicians. Anesthesiologists who may be defendants in a
from anesthesia. The patient was transferred to the malpractice action clearly are concerned not only with the
postanesthesia care unit (PACU), and, following report, case, but also with the process. The same concerns apply
the anesthesiologist returned to the operating room to to certified registered nurse anesthetists (CRNAs) and any
begin another case. other medical personnel who may be involved.
Within minutes, the patient began to develop signif-
icant neck and facial swelling. The PACU nurses notified
the Chief of Anesthesiology and the attending surgeon,
who ordered a stat internal medicine consultation. Ra- When You Hear Hoof Beats,
diographs of the neck and chest showed massive subcu-
taneous emphysema and pneumomediastinum. Palpable Should You Look For Zebras?
crepitus was noted. The internist diagnosed an anaphy-
lactic reaction to cephazolin or meperidine (which had Defendant physicians or their expert witnesses have often
been administered during three prior abdominal surgeries stated, When you hear hoof beats, you dont look for
63
64 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
zebras. In reality, beginning in medical school, first year The anesthesiologists semi-emergent reintubation of
medical students are trained in the art and science of the patient at the end of the case was done without
differential diagnosisan approach designed to reveal all the benefit of additional sedatives or paralytics. A 4-cm
potential causes for a patients clinical symptomatology. linear perforation of the posterior cervical esophagus
Using their full armamentarium of knowledge, training resulted, which went undiagnosed. To make matters
and experience, reasonably prudent physicians consider worse, the treating anesthesiologist failed to document the
all possible explanations for a patients presenting signs reintubation in the anesthesia record or in the anesthesia
and symptoms, and then systematically rule out potential PACU discharge note. Also, she did not mention the
causes until the correct diagnosis is reachedeven if it emergent reintubation to the attending surgeon, the Chief
has stripes! Attempts are made to eliminate potentially of Anesthesiology (who covered for her while she was in
lethal conditions first, then progress to those less likely to surgery), the consulting internist, or the radiologist. Had
result in significant morbidity or mortality. she done so, more probably than not, someone would
We all have our tools of the trade. In the legal profes- have ordered a Gastrografin swallowing study to rule out a
sion, words are the primary tool, and cross-examination perforated esophagus, because it is a known complication
is analogous to the surgeons scalpel. The doctors tool of an emergency intubation and one which presents with
box contains various techniques and approaches used to these rather classic signs and symptoms.
meticulously narrow the list of suspects (i.e., compet- The Chief of Anesthesiology claimed to have asked
ing diagnoses) until the truth is revealed, much as cross- the treating anesthesiologist whether she knew of any
examination is designed to reveal the truth in a court of law. possible explanation for the problems the patient had
The tools in this search for the truth include the detailed experienced in the PACU. Rather than admit to the
medical history, the meticulous physical examination, per- emergent reintubation at the end of the case, she said
tinent laboratory work, and various relevant diagnostic she knew of no reason to explain the patients new
tests and/or imaging modalities. As data from these tools problems. The Chiefs own credibility was an issue,
are collected, the list of potential diagnoses is refined and however. He denied on deposition that he had ever served
narrowed until, ideally, the proper diagnosis is reached. as the Chief of Anesthesiology at the hospital in question,
In the true life example set forth, the defendant physi- notwithstanding the fact that the position was set forth on
cians heard a stampede of thundering hoof beats, and, his resume. Nothing turns an otherwise defensible case
rather than search for horses, began searching for zebras. into a sure loser quicker than dishonesty or a refusal to
Apparently nobody considered a perforated hollow viscus concede the obvious.
as the source of the massive subcutaneous emphysema Because of the missed diagnosis, the esophageal
or pneumomediastinum. Witnesses described massive perforation went unrepaired for more than 7 days, and, as
swelling of the patients face and neck; one even de- the patients diet was gradually advanced, food and fluids
scribed her as having a pumpkin head. Unquestionably, tracked down into the mediastinum, ultimately resulting
angioedema from anaphylactic shock can cause dramatic in suppurative mediastinal sepsis, multiorgan failure, and
swelling but not palpable crepitus, massive subcutaneous death.
emphysema, and pneumomediastinum. In short, the di-
agnosis of anaphylaxis did not adequately explain the
patients objective signs noted with clinical examination.
No one could reasonably fault the health care team for
administering countermeasures for anaphylaxis, because
Could This Outcome Have Been
their administration is, at worst, benign and ineffective, Avoided?
and, at best, life-saving. They were faulted for failing
to explain the unexplained (i.e., how a patient who In reasonable medical probability, the answer is yes.
underwent abdominal surgery developed the sudden onset Had the treating anesthesiologist been forthcoming about
of massive subcutaneous emphysema in her face and neck her experience with the patient, most credible expert
in the immediate postoperative period). witnesses would have said, There but for the grace of
The general surgeon knew that all his work was God go I (i.e., that it could just as easily have happened
performed below the diaphragm, and therefore trapped air to them.) The diagnosis would have been confirmed by
in the neck and face resulting from the surgical procedure Gastrografin swallow, and the patient taken promptly
was virtually impossible. Even if he had perforated the back to surgery for mediastinal drainage and primary
patients bowel, an upright chest film would only have repair of the esophagus, followed by aggressive parenteral
shown trapped air beneath the diaphragm (if at all), antibiotic therapy.
and not trapped air in the neck and mediastinum. As A skilled defense attorney could easily defend the
the air was gradually absorbed into the patients system,
following facts, had they been present:
her swelling subsided, and everyone assumed that the
internist had not only made the proper diagnosis, but After surgery ended, I began to reverse the effects of
also possibly even saved the patients life with timely the anesthesia. As the patient regained consciousness, she
orders for diphenhydramine and methylprednisolone. In developed severe difficulty breathing. I quickly examined
reality, his misdiagnosis gave everyone a false sense of her, listened to her chest, and noted that she was suffering
security, and efforts to find the true cause of her clinical from apparent acute bronchospasms. I knew that if her
symptomatology were quickly abandoned. ability to breathe on her own was not restored immediately,
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 65
her life was in jeopardy. I had no choice but to perform attempt to obtain adequate professional liability insurance
reintubation so that I could breathe for her until she could to cover potential malpractice losses.1 The most common
regain the ability to breathe on her own. Time was of type of professional liability insurance is claims-made
the essence. I quickly selected an appropriate laryngoscope coverage. It provides coverage for claims that occur and
and endotracheal tube and inserted the breathing tube into are reported while the policy is in effect. Unless both
her wind pipe. I then connected the breathing tube to the conditions are met during the policy period, there is no
mechanical ventilator, and, after reconfirming proper tube coverage for the physician for the claim.
position and stable vital signs, I accompanied her to the A less common, harder to secure, and more expensive
PACU for continued monitoring by the nursing staff. type of coverage is occurrence insurance. It provides
After the PACU nursing staff accepted her care, I re- coverage for any incident that occurs during a policy
turned to the OR for my next scheduled case. Shortly after period, regardless of when it is reported.
that case began, I received word from the nursing staff that Tail coverage is an additional type of policy. It is a
the patient had a sudden onset of significant facial and neck supplemental type of policy, which insures for incidents
swelling, with palpable crepitus. Given the fact that she that occur during the active period of a claims-made
had required an emergent reintubation combined with her policy but are not brought as claims or reported to the
difficult airway anatomy, I had to entertain the possibility insurer by the time the claims-made policy is terminated.
that she had suffered an inadvertent esophageal perforation Tail coverage is important for situations in which a
or other airway injury. I instructed the PACU nurse to have physician changes his claims-made carrier or when a
the Chief of Anesthesiology call me ASAP so that I could physician retires or dies. It is typically purchased from the
bring him up to speed and suggest orders for a stat ENT previous claims-made carrier at a cost of 125% to 250%
consultation and possible gastrografin swallow study to rule of the previous years premium. Many physicians now
out a perforated hollow viscous in the oropharyngeal region. opt for self-insurance programs with physician groups.
He agreed with my recommendations, the orders were given Others practice without insurance coverage. In some
and promptly implemented, and the provisional diagnosis communities, hospitals provide coverage. In addition,
of a perforated esophagus was confirmed. The patient was some physicians select higher deductibles and lower
promptly taken back to surgery for mediastinal drainage and coverage limits in exchange for lower premiums.
primary repair of the esophageal perforation, followed by IV In recent years, insurers have increased premiums,
antibiotics. After 7 days, the patient was discharged home imposed various limitations on coverage, and dropped
in good condition with the expectation of a full recovery. out of the insurance market. Physicians blame the civil
tort system and trial lawyers for increases in medical
Unfortunately, rather than admit to what had tran- claim payouts and the escalation of professional liability
spired in the OR, the anesthesiologist attempted to hide insurance premiums.
the truth by refusing to document the emergency reintu-
bation in the medical record, by failing to disclose it to
her professional colleagues (even after serious complica-
tions arose with her patient), and by never admitting to
an honest mistake. As a result, she tendered her million
Do We Have the Most Efficient
dollar policy limit at mediation and was thankful to do so. Legal System?
We, as a society, place great value on our health. We
demand the most technologically advanced care, with
What Are the Origins treatment by highly skilled physicians who utilize state-
of-the-art equipment and procedures and the finest phar-
of Tort Law? maceuticals in an attempt to relieve every chronic and
acute health problem. We demand that care be avail-
Tort law evolved to punish wrongdoers and to compensate able immediately. We want life-threatening conditions
those harmed by wrongful actions. The philosophy of reversed, limbs salvaged, and pain eliminated. We look
such a system is to achieve justice by shifting loss to the to our physicians and other health care providers to meet
wrongdoer rather than to the victim or society. As the these demands and, for the most part, they do. The Centers
philosophy is applied to real life situations, this simple for Disease Control recently indicated that life expectancy
system, with its basic tenets of right and wrong, quickly is greater than ever in this country. For white men, the
becomes complex. Numerous competing interests exist life expectancy was 75.4 years in 2003 and 75.1 in 2002.
between the injured, the alleged wrongdoer, and third For white women, the life expectancy was 80.5 in 2003
parties such as insurers, creditors, and society in general. and 80.3 in 2002.2
Along with the complexity and competing interests We recognize that physicians should be very well
inherent in all claims, extreme animosity between doctors compensated for the years of training that never really
and lawyers has also developed in medical litigation. cease, for the long hours of work that continue through
Perhaps at no other point in history have lawyers been so nights, weekends and holidays, and for the chronically
bitterly distrusted by physicians as now. disturbed sleep and missed family times that are a part of
Physicians are often targets in civil tort litigation in the profession. We know that the price paid by our health
medical malpractice claims. To protect themselves, they care providers is very high, and the suicide,3 divorce,
66 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
and substance abuse rates among physicians confirm this The phrase, malpractice crisis, recently reared its
premise. Shockingly, it is reported that this country loses ugly head after a fairly stable medical-legal world since
the equivalent of one medical student class per year due the problems of the 1970s and 1980s. Close attention has
to physician suicide,3 and substance abuse is estimated to been paid by professional medical and legal associations,
affect one in six medical residents.4 consumer groups, insurance specialists, actuarial compa-
Physicians believe that their present and future in- nies, state legislatures, and federal agencies to determine
come is at risk and that they must practice under whether a crisis exists, and, if so, why. Perhaps the most
conditions that compromise their professional life. They important question we should ask is, what can be done
attribute these problems, in large part, to our legal sys- about it, if such a crisis exists?
tem and the lawyers who drive it. A major disconnect
between physicians and patients exists. Patients and their
lawyers often do not realize the plight of physicians who
are caught in a constant battle to survive the pressures What Is a Malpractice Crisis?
of practice. On the other hand, patients and their lawyers
could well argue that physicians do not understand the
inherent unfairness of having to accept a negligently in- When one refers to malpractice crisis, the implication
flicted life-altering injury that could have been readily is that physicians are being sued with greater frequency
avoided. All have a common goal: To have the best pos- than the norm, with claims based on more questionable
sible patient outcomes; yet, so far, with rare exceptions, grounds than usual, resulting in both greater payouts of
we have not been able to work together to find the best insurance benefits and the unavailability of affordable
way to accomplish that goal. We have no Saint Anthony of medical liability insurance coverage. It is difficult to
Padua to facilitate understanding and mediate conflicts, evaluate all the components accurately, but the irrefutable
and therefore the competing interests continue to collide fact is that, in recent years, some physicians have been
unchecked. unable to obtain adequate professional liability coverage
Because health care is valued by all, we must at an affordable price. However, the American Society
continuously examine our legal system to assure that of Anesthesiologists Committee on Professional Liability
it is fair and just and not a detriment to good health care. announced in its annual survey of premiums that the
We must ask the following questions: 2005 premiums for anesthesiologists were essentially
unchanged from the 2004 premiums ($20,572 average
1. Is this the best system available? Should specialty in 2005; $20,611 in 2004). Between 2002 and 2004,
courts be created and the lay jury system abolished? however, the premiums increased an average of >30%
Does the current system work? for anesthesiologists.5
2. Should people be permitted to sue for damages; if so, That component alone may be sufficient to justify
which damages should be recoverable in litigation? some type of intervention on behalf of anesthesiologists;
What standards should be applied? Who should bear however, determining what that intervention might be
the burden of proving fault and damages? would be quite a difficult task.
3. How can a person at risk protect himself or herself Presently, no adequate long-term solution has been
from claims, when we all must concede that human proposed. Until a remedy is in place, physicians can
perfection does not exist and that human error is likely gain a great deal of benefit by understanding the legal
in all human endeavors? Should physicians pay a larger system. Lore and exaggerated horror stories are too
price for their human mistakes than those in other often repeated and gradually assume the status of truth.
professions? Should a wrongdoer be responsible for Misapprehension of the system can obviously result in a
mere mistakes? flawed analysis of any given legal situation. Knowledge of
4. What will happen to the injured party without a system the system is empowering and will help physicians to be
of compensation that is adequate for the loss sustained? effective self-advocates.
Is it sufficient to give partial compensation?
5. Should society pay for the negligent actions of physi-
cians?
Few special interest groups or industries have consis- What Is the Basis of a Medical
tently voiced opposition to the current civil justice system.
Chambers of commerce, associated industries, and other Negligence Claim?
probusiness entities have regularly lobbied for certain
legislation and opposed other measures that could affect Legal claims cannot be brought merely because an adverse
businesses. However, there has been no consistent agenda medical outcome occurred. Some states have statutes
of specific reforms/changes that have been advocated by prohibiting legal claims based solely on a bad outcome.
these groups. The American medical profession, however, For example, Florida Statute Sec.766.102(3) provides that
has been vocal in the last 35 years about its opposition the existence of a medical injury shall not create any
to the American medical negligence justice system. As inference or presumption of negligence against a health
an industry, health care professionals have declared on care provider, and the claimant must maintain the burden
multiple occasions that a crisis exists in American health of proving that an injury was proximately caused by a
care. breach of the prevailing professional standard of care by
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 67
the health care provider. In other words, the plaintiff or Some physicians may have practices that exceed or
claimant must demonstrate by some means, other than fall below the community standard of care, which changes
the bad outcome, that a legal duty was breached which over time. The defendant physician in a medical neg-
caused harm.6 ligence case should understand the significance of the
The sole exception occurs when the theory of Res standard of care as it applies to his or her case. For
Ipsa Loquitor (the thing speaks for itself) can be applied. example, if the standard of care is to order a computerized
Although rarely available in medical negligence cases, tomographic (CT) scan and plain film in the emergency
when it is applied, expert testimony is still generally department to evaluate an acute abdomen, the emergency
provided to show that the injury typically does not occur in room (ER) physician cannot be held liable for failing to
the absence of negligence. To use the theory, the plaintiff also order a MRI, even if the MRI would have revealed the
must also show that the instrumentality responsible for problem. A plaintiff may offer expert testimony to show
the harm was in the exclusive possession of the health care that MRI was available, that it could have been read-
provider, and the patient is without responsibility for the ily performed with minimal risk, and that it would have
injury. Leaving medical instruments in the patients body yielded a diagnosis and led to effective treatment. Such
following a surgical procedure is an example in which res testimony could be compelling and result in an adverse
ipsa loquitor may be appropriate. The jury can still find verdict for the physician; however, it should not be heard
that the standard of care was not breached even when in the courtroom.
the theory of res ipsa loquitor is used. Although it is true What an extraordinary physician does, or what the
that most client inquiries about medical malpractice occur standard of care used to be, or has evolved to be since the
because of a bad result, numerous procedural, evidentiary, incident, should not be a part of the evidence that the jury
and statutory protections are in place to protect against considers.
claims being brought solely on that basis. In Linn vs. Fossum,7 the defendant physician offered
A legal claim for medical negligence requires that the testimony of an extraordinary expert witness who
several independent elements be demonstrated: performed above the standard of care. Her method of
practice was not offered to show what the defendant
1. The existence of a legal duty (a legal duty is generally
physician should have done. Instead, she testified that
owed only to the patient and the patients spouse and
the standard of care was less than what she typically did
minor children and not to adult children, stepchildren,
in her practice. She opined that the defendant physician
or business associates)
complied with the standard of care that other physicians
2. The breach of a duty such that a reasonable and prudent
in the field follow.
health care provider would not have performed in the
The plaintiffs lawyer attempted unsuccessfully to
same manner as the allegedly negligent physician when
preclude the testimony. The jury apparently understood
faced with similar circumstances
the distinction between the standards practiced by an
3. A causal link between the duty and breach of the duty
extraordinary physician and the lesser standard that
4. Proof that the breach of the duty resulted in specific or
was regularly practiced by physicians in the same
quantifiable harm to the plaintiff
circumstances as the defendant physician. The jury ruled
Failure to prove each of the elements can result in that the physician met the applicable standard of care.
the dismissal of a medical negligence claim. Although Interestingly, review was granted by the Florida
the elements sound simple and straightforward, failure Supreme Court about this method of proof by the defen-
to analyze each element will permit some claims to go dant physician. The plaintiff claimed that the defendant
forward which technically should not. For example, some improperly acted as a conduit for the admission of opin-
medical experts opine about how a particular medical ions of nonwitnesses and that she was prejudiced by her
procedure should be done. However, such an opinion inability to cross-examine the physicians with which the
is totally irrelevant to medical negligence litigation and extraordinary physician expert conferred to determine
should not be presented to a jury. The only relevant proof the applicable standard of care. She also challenged the
for the jury is the standard of care in the community ability of the defendant physician to offer the testimony
during the subject time framethat is, what a reasonable of an expert who does not adhere to the standard of care
and prudent health care provider would do under similar in the experts own practice. No final opinion has been
circumstances. issued by the Court, but the general propositionthat a
Community for purposes of medical negligence claims physician should be judged by the applicable standard
most often refers to the national medical community of care followed by similar health care providers under
rather than a specific locale, although the circumstances similar circumstancesis certain to remain intact.
in which the care is provided can be taken into account. The physicians attorney must properly identify the
A small community hospital without magnetic resonance differences between mere opinions, matters of judgment
imaging (MRI) capability would not be expected to have that are discretionary from physician to physician, and
the same diagnostic standards as a sophisticated tertiary true standards of care. A defendant physician should be
care center with all the latest imaging modalities. Yet, held liable only for violations of the standard of care.
physicians practicing at the two facilities would each be Determining which, if any, of the opinions constitute
expected to diagnose the same illnesses. There is no excuse standard of care must be determined before the evidence
for one at a smaller rural hospital to have diagnostic skills is heard in open court. If a judge erroneously permits the
that are lesser than a physician at a large urban hospital. evidence, it is critical that the physicians attorney makes
68 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
the proper objections on the record and preserves the issue states is very close to the standard applied to criminal
for the appellate court. Untimely or unstated objections conduct; in addition, many states have limitations on
are generally viewed as waived by appellate courts. the amount that can be awarded for punitive damages.
The flip side of this issue arises when objective proof 6. Courts can exclude evidence that is irrelevant, in-
reveals a breach in the standard of care, which resulted cluding any that is truly not related to determining
in loss to the claimant. In some instances, the physicians negligence by the physician, as well as information
attorneys and insurance carrier either fail or refuse to ac- that is related but deemed more prejudicial than pro-
knowledge that the physician has any substantial liability, bative (i.e., the court could exclude evidence that the
perhaps because the possible award to the claimant may physician was involved in extramarital relations at the
exceed the physicians applicable insurance coverage. time of the patients cardiorespiratory arrest but could
Another essential component to a medical negligence allow evidence that the physician did not respond to
claim is the presence of documented damage. The burden a page by the nursing staff. Similarly, the court could
of proving the damage is on the plaintiff. Simply because exclude evidence of prior racist comments by the
a plaintiff may require future surgeries or may be unable physician or any similar inflammatory evidence).
to work is not enough for the jury to award damages 7. The court makes a case-by-case determination of
for those losses. In addition, in nearly every jurisdiction, when evidence of certain professional standing issues
even the loss of a chance of incurring those losses is not can be presented to the jury, including medical class
sufficient to support a claim for medical negligence. Such rank, failure to obtain primary residency choices,
a loss is deemed too speculative, as an undetermined loss number of residency positions sought, failure to pass
in the future is just as likely not to occur as to occur. If the board examinations, license suspensions, privilege re-
plaintiff cannot prove by the greater weight of the evidence strictions, firings from professional positions, and
that he would have had a better outcome with reasonable other similar matters. More liberal rulings may be
care, as opposed to an improved chance for a good given when questions regarding competence are di-
outcome, the case must be dismissed. However, dismissal rected to an expert witness.
is not self-executing. The physicians attorney must 8. Information about a physician defendants income is
present the issue to the court by a motion for summary rarely discoverable or admissible; similarly, a plain-
judgment or other appropriate dispositive motion. tiffs reliance upon public assistance is often inadmis-
Other aspects of medical malpractice litigation, which sible, as a partys wealth or standing is not supposed
are often seen, include: to be considered by the jury.
9. Certain matters that would otherwise not be admis-
1. The plaintiff can almost never use prior legal claims sible may become admissible if a party is dishonest
against a physician in pending litigation, unless a about them, because each partys credibility is always
fairly clear pattern has been demonstrated by the an issue. It is better to be fully forthright about an
other litigation (i.e., the physician over-radiated 13 issue and have the court exclude the irrelevant infor-
other patients in essentially the same manner and mation at trial than to risk having the information
time frame). Likewise, a defendant cannot generally become admissible because of not being forthright.
introduce evidence about other legal claims filed by Similarly, certain matters that might otherwise be in-
the plaintiff. admissible may become admissible if one party opens
2. The plaintiff cannot use evidence that a physician has the door by introducing the issue.
drug or alcohol impairment problems or has required 10. The court rules upon issues of relevance by its judicial
treatment in an inpatient facility unless a clear tie to discretion, and any rulings that are objectionable to
substance abuse at the time of the alleged negligence one party may be overturned on appeal by a show-
can be shown. This requirement generally prohibits ing that the court abused its discretion, a difficult
any evidence of arrests for driving under the influence, standard to meet.
open container violations, disorderly intoxication, and
so on, as well.
3. Generally, the physician cannot blame the plaintiff for
the plaintiffs underlying condition to reduce any ver-
dict by comparative negligence, as the physician must What Potential Conflicts Exist
take the patient as he finds him (i.e., a patient who in Medical Malpractice Cases
presents to the hospital with a subdural bleed that is
negligently missed by the physician cannot be blamed (Are You in Good Hands)?
by the physician for incurring the injury while driving
under the influence of alcohol and striking a tree). An inherent conflict exists between the physician, the
4. Malpractice damages are generally calculated by the insurance-retained attorney for the physician, and the
life expectancy of the plaintiff, so in most jurisdic- insurance company. That is not to say that the attorney
tions, an elderly or ill plaintiff will recover far less and carrier will always abdicate their responsibilities to
damages for the same negligently inflicted injury than the physician. This is certainly not the case. However, it
a young and healthy plaintiff. is in the physicians best interest to recognize that the
5. Punitive damages are virtually never awarded in med- physicians role in litigation is not necessarily consistent
ical negligence cases, as the standard applied in most with that of the defense attorney and carrier.
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 69
Physicians are free to obtain independent personal favor of the entity that is paying his or her fee bills each
counsel for the physicians protection during the pendency monththe insurance carrier. Physicians should be aware
of litigation at the physicians expense. When a physician that this financial allegiance can make it difficult for some
is facing what could potentially result in a judgment insurance defense attorneys to demand that the insurance
in excess of applicable insurance coverage, a personal carrier pays money to settle the case, thereby ensuring
attorney can help the physician avoid an excess judgment the physicians financial protection from an adverse jury
and assure that the insurance carrier acts in good faith verdict.
in the handling of the claim; that is, the personal counsel Neither the insurance defense counsel nor the in-
can demand that a proper defense is mounted and that surance carrier should roll the dice with the physicians
settlement efforts are made if the case should be settled financial future if the circumstances are such that reason-
before trial. In instances where the physician has learned, ably prudent persons in the conduct of their own business
after an excess judgment has been rendered, that his affairs would try to get the case settled for an amount up
carrier did not act in good faith, the physician may still be to and including the policy limits.
able to avoid paying personally for the judgment. Some Sometimes, the insurance carrier, or even defense
states allow a resolution for the physician and/or the counsel, will try to persuade the physician to withdraw
injured patient against the carrier based upon a theory of the consent to settle the case. In most medical malpractice
insurer bad faith. Bad faith may be founded on common insurance policies, the insurance carrier cannot settle a
law contract and tort principles or statutory rights, and it case without the consent of the insured. By withdrawing
binds an insurance carrier to act fairly and honestly in the the consent to settle a malpractice claim, the physician
handling of claims for its insureds or to bear the financial has effectively taken the insurance carrier off the hook for
losses caused by its failure to act in good faith. negligently failing to settle a claim within policy limits. The
Ultimately, only the physician faces the jury and must insurance carrier cannot be faulted for failing to settle a
abide by the jurys ruling. The insurance carrier is not claim when their hands are tied through a lack of consent to
a party to the legal case and has limited liability, and settle. A wise defendant physician never agrees to withdraw
the insurance-retained attorney has no financial stake his or her consent to settle a case without an agreement,
in the outcome of the trial. In the case summary at in writing, in which the insurance carrier agrees to pay
the introduction of this chapter, four of six defendants the entire judgment, plus interest, in the event of an adverse
(the surgeon, the Chief of Anesthesiology, the attending jury verdict. Stated simply, if the carrier talks the physician
anesthesiologist, and the hospital) all gave consent and out of giving the consent to settle, thereby making any
settled at mediation for $1.65 million. The consulting settlement impossible, the carrier should be willing to
internist and the consulting ENT surgeon each gave their indemnify the physician for the entire judgment in the
consent for settlement, but their liability insurance carrier event that an excess jury verdict is rendered against the
refused to offer anything to get the claims against them physician.
settled and released. The plaintiffs had three qualified Physicians should consider retaining an independent
experts who were highly critical of their respective care counsel to monitor the case and ensure that both the carrier
and treatment of the patient that proximately resulted and defense counsel are looking out for their best interests.
in her death. The defendant physicians wanted the case In a serious case that should settle, but for reluctance
settled, but the carrier was intent on rolling the dice on the insurance carriers part to tender its policy limits,
with their economic well-being. As a result, the case a board certified personal injury plaintiffs attorney with
proceeded to trial against the two nonsettling defendants. experience in handling medical malpractice litigation can
After an emotional two-week jury trial, the jury returned be your best friend. An experienced plaintiffs attorney will
a unanimous verdict against the consulting internist for not allow the carrier or their chosen attorneys to roll the
$2.92 million. No liability was found against the ENT dice with the physicians financial well-being when the only
surgeon. The plaintiffs had offered to settle with the prudent course of action is to tender the policy limits to
two remaining defendants for $150,000 at the original secure a release of all claims on the physicians behalf.
mediation and again at a second court-ordered mediation. Physicians and other providers such as CRNAs and
The carrier for medical malpractice insurance cover- Anesthesia Assistants (AA) should be wary of the insurance
age undertakes two duties: representative who urges to you to be strong by not
insisting that they settle the case on your behalf and/or by
1. To provide a defense to the lawsuit at the carriers ex-
withdrawing your consent to settle and proceeding to trial.
pense
Like the police detective investigating a murder case, you
2. To pay up to the policy limits of liability coverage to
have to ask yourself, Qui bono? (Who benefitsor who
secure a release of all claims
has a motive to commit the crime?). What is the insurance
In most instances, the insurance carrier retains the carriers motive by encouraging you to withdraw your
right to hire a defense counsel of their choice. The carrier consent or by urging you to proceed to verdict in a case,
will assign the defense of the case to a lawyer or law which you feel should settle? If the case results in a
firm on their approved list. The defense attorney and defense verdict, no harm is done. However, if the case
firm may be reliant on the insurance carrier as a major should result in an adverse jury verdict in excess of the
source of their annual billings. From an ethical standpoint, policy limits, you will be personally liable for that portion
the physician is the client. From a practical standpoint, of the verdict, which exceeds the limits of your liability
however, the defense counsels allegiance may be tilted in insurance coverage.
70 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
CRNA, certified registered nurse anesthetist; OR, operating room; AA, anesthesia assistant.
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 71
I have talked with an anesthesia provider, _______________________, and I have reviewed the risks
of the various anesthesia techniques which are available for my surgical procedure, along with the benefits
of the anesthesia techniques. Based upon our discussion, I have elected to proceed with the following
anesthesia technique:_____________________________________. I understand that events could occur
during the surgical case which could preclude the use of the method I have selected as a first choice of
anesthetic technique. I authorize the use of alternative methods by my anesthesia provider should the
circumstances require an alternative method or combination of methods, within the judgment of my
anesthesia provider.
I understand that there are risks of anesthesia which are known and unknown. I realize that
anesthesia carries a risk of death, cardiac and respiratory arrest, brain damage, vocal cord damage, spinal
cord or other neurologic injury, chipped teeth, nerve or muscle damage, paralysis, or allergic reaction. I also
understand that my health status affects the degree of risk to which I am exposed while receiving
anesthesia. Certain conditions such as obesity, advanced age, hypertension, coronary artery disease,
diabetes, respiratory insufficiency and other acute or chronic conditions and diseases increase my risk of
having an adverse event while undergoing anesthesia. I have been advised that certain patients have
idiosyncratic, unpredictable responses to anesthetic agents and other medications as well. My anesthesia
provider has discussed with me my known health conditions and how they may affect my anesthetic risk.
I have advised my anesthesia provider of all allergies known to me and all prior anesthetic or other
medical complications I have experienced in the past. I have had a full opportunity to ask all questions I
desire to have answered by my anesthesia provider, and my provider has offered to provide additional
reference materials should I desire. I also have advised my anesthesia provider that I have complied with all
preoperative instructions. I am consenting to proceed with the anesthetic and other medical services by my
anesthesia provider with no outstanding or unanswered questions.
__________________________________
Patients Full Name
___________________________________
Witness
___________________________________
Witness
standard of care through competent expert testimony is sufficient to safeguard against junk science, according
mandates that the court dismiss the case. A well qualified to its proponents. It requires that the opinions presented
and articulate expert is important to the success of a to the jury are the type that are generally accepted within
malpractice claim or defense. the field to which the expert belongs and that the subject
Most states have criteria for the admission of expert matter is beyond the general knowledge of jurors.
testimony, including the type of work in which the expert Daubert, some say, is far better at keeping junk
must be regularly engaged, the time frame in which the science from jurors. It requires that the conditions
expert must have been active, the education and training mentioned below are fulfilled:
required to qualify as a medical expert, and the types of 1. The scientific theory or evidence be tested
opinions that can be rendered. The courts also have the 2. The general opinion be published or reviewed by peers
authority and responsibility to apply certain standards to 3. The potential rate of error be known
expert testimony and to act as gatekeepers for opinions 4. The theory or technique be generally accepted in the
to be presented to the jury. In many states, the Frye scientific community (reiterating the Frye standard)
standard is used;8 in federal courts and in some state 5. Standards exist and are maintained, which control the
courts, the Daubert standard is used.9 The Frye standard application of the technique or theory
72 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
Regardless of the standard applied to the proposed ex- looking for a quick buck, and that lawyers are more
pert testimony, unqualified experts should be challenged than happy to bring frivolous suits in hopes of reaping
in court before the expert offers substantive medical tes- a big reward in the end. Perhaps some people are
timony or standard of care opinions. Many states have motivated by such considerations, but those explanations
presuit screening requirements that are specific, detailed, fall short in explaining the problem or helping to find
and mandatory. Failure to comply with a prerequisite is workable and fair solutions. Moreover, evidence shows
fatal to a malpractice claim that is subsequently asserted, that many injuries and deaths actually do occur as the
that is, failure to indicate that an experts opinion has result of avoidable medical mistakes and that a substantial
never been previously disallowed in a court. Failure to majority of claims filed involve serious injury or death.14
allege the specific bases of negligence against a prospec- Legislative hearings were held in Florida when tort
tive defendant in an affidavit could bar a later assertion of reform measures were proposed in 2003.15 Testimony
negligence against that defendant in a medical negligence was heard from a number of interested parties, including
case.10,11 Otherwise, the nonobjecting party is not likely representatives of the insurance industry and the Florida
to be heard after an adverse outcome. Medical Association. None of the witnesses was able
Once an expert has been declared qualified to of- to identify a specific example of a frivolous medical
fer expert testimony on a given topic, the most effective negligence case when placed under oath and asked
and common means of neutralizing an adverse expert about frivolous cases. The head of the Florida Medical
is through cross-examination. A well prepared cross- Association, Sandra Mortham, testified that she did not
examination will often reveal an individual who (i) spends have the information to say whether there are frivolous
most of his or her time as a paid expert; (ii) predomi- lawsuits in the State of Florida.
nantly or exclusively testifies for plaintiffs or defendants; Consultation with a malpractice attorney usually
(iii) offers contradictory testimony from case to case; and occurs because of an unexpected, adverse outcome, after
(iv) has inadequate clinical experience. Cross-examination which no one took the time to explain to the patient
may further reveal that exorbitant fees are charged which or their loved ones, in language they could understand,
constitute a large percentage of the experts overall income exactly what happened. This is a lost opportunity to
and that he or she is never consulted by professional col- avoid costly and protracted litigation. Most people are
leagues for medical advice about the topic in which exper- not lawsuit-crazy and are not seeking to win the lawsuit
tise is claimed. In addition, the expert may never have done lottery. However, most patients or their family members
any scholarly writing or research on the topic in question. do struggle to make sense of a sudden and unexpected
It usually becomes readily apparent when someone adverse outcome, which resulted in death or significant
is acting as a hired gun rather than as a physician expert disability. They begin by looking to the physician for
who has special expertise on the subject at issue. The answers. If they do not get them in terms they can
physician defendant can offer invaluable insight and cru- comprehend, they are much more likely to consult a
cial substantive medical information to aid the attorney in medical malpractice attorney who will.
preparing the cross-examination of the opposing expert. Physicians may have been trained not to allow
Although others may suggest to the contrary, bad themselves to get too close to their patients. No doubt this
or unqualified experts are hired by both plaintiff and practice originates as a necessary self-defense mechanism,
defense attorneys. The physician defendant should be because to do otherwise would take a savage emotional
aware of who is hired as an expert on his behalf and toll on the physician with each patients death. However,
should independently research the experts credentials this self-protective behavior is often misinterpreted by
and objectively assess any opinions the expert is expected patients or their family members as an aloof, or worse,
to offer before that individual is declared a trial expert. uncaring attitude toward the patient. When people have
Professional organizations have grown increasingly suddenly and unexpectedly lost a loved one, they need to
concerned about expert testimony in medical negligence know that the physician did all within his or her power
cases. The American Medical Associations position is that and ability to prevent that loss from occurring.
physicians should, as a matter of public interest, serve as Most people are rational beings, and many people
impartial expert witnesses.12 The American Society of will be initially in a state of emotional shock. Physicians
Anesthesiologists (ASA) has been proactive for a number should tell the patient or family members that they may
of years in the area of expert testimony. The ASA has eventually have questions, and that they will be more than
developed guidelines for participation in medicolegal happy to sit down and spend as much time as necessary to
cases by its members13 (see Fig 4.2). fully address any questions and concerns to the best of the
physicians ability. Having done so, instead of appearing
as aloof or uncaring, the physician will be perceived as
a caring and compassionate professional who empathizes
Why Are Malpractice Claims with their loss.
Initiated when Things Go If a lawsuit is filed against you, you will able to testify
truthfully that you did your best to explain matters in
Wrong? detail to the patient and/or family members, and even
offered to sit down with them, whenever they desired,
Some individuals or groups opine that people are litigious, to answer their questions and address any concerns they
they blame others for their misfortune, and they are might have. The jury will appreciate your professionalism
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 73
3. The physician should be familiar with the clinical practice of anesthesiology at the time of
the occurrence and should have been actively involved in clinical practice at the time of
the event.
3. The physician should make a clear distinction between medical malpractice and adverse
outcomes not necessarily related to negligent practice.
4. The physician should make every effort to assess the relationship of the alleged
substandard practice to the patients outcome. Deviation from a practice standard is not
always causally related to a poor outcome.
5. Fees for expert testimony should relate to the time spent and in no circumstances should
be contingent upon outcome of the claim.
6. The physician should be willing to submit such testimony for peer review.
FIGURE 4.2 Guidelines for Expert Witness Qualifications and Testimony. American Society of
Anesthesiologists, Park Ridge, Illinois.
and accessibility, and will remember those qualities when fraught with numerous opportunities for fatal missteps
the time comes for them to deliberate on the verdict. by the lawyer. As a result, most trial lawyers do not
Most importantly, many lawsuits will be completely undertake the challenge of handling a plaintiff medical
circumvented. People are seeking closure; they can either negligence claim.
get it from you or a jury of their peers. The decision The U.S. Department of Justice released a civil justice
is largely yours, and depends on how you handle the survey of State courts from an analysis of 2001 data.14
situation following an unexpected, adverse outcome. Dr. Thomas H. Cohen, a statistician for the Bureau of
Malpractice claims are neither quick nor easy; they Justice Statistics, concluded from his review of medical
take considerable time to prepare and large sums of malpractice trials and verdicts in large counties in 2001
money to finance. Medical negligence cases are more that there were 1,156 medical malpractice trials in the
often lost than won by the plaintiff.14 The cases are countrys 75 most populated counties during 2001. In
hard fought against highly skilled insurance lawyers, 90% of the cases, the patient alleged either a permanent
and the legal and medical issues are complex.10,11 The injury or death. The success rate for plaintiffs was 27%
complicated statutory scheme is a virtual minefield, overall in the medical negligence trials.
74 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
If a claimant wants to file a baseless claimand can nursing judgment, physicians may not get accurate or
find an attorney and expert who are willing to assume the timely patient information. A solid support staff can avoid
personal, professional, and financial risk of joining the ef- a number of mishaps and aid the physician and patient.
fort to pursue a frivolous casethe court, bar association, Unfortunately, consistent high quality nursing care is a
professional medical organizations, and jury can vindicate resource upon which physicians can no longer rely.
the innocent physician and hold the individuals account- The physician must be hypervigilant in all matters
able who are abusing the system.15,16 An expert may be of patient care. Not surprisingly, some physicians, when
challenged by the filing of a Motion in Limine or Motion forced to work under such difficult circumstances, fall
to Strike. The challenge should be supported by the sworn short of rendering the care they would render under more
affidavit or testimony of a qualified expert, learned trea- tolerable conditions.
tises and other peer reviewed medical literature about the
medical topic, and factual information about the specific
expert. The objectionable opinions offered by the expert
should be carefully analyzed and thoroughly critiqued and What Should Be Known About
impeached for the benefit of the Court. Another method of
challenging an expert, which is used in some courts, is to Tort Liability Laws,
voir dire the witness out of the presence of the jury before Professional Liability Coverage,
substantive opinions are offered. Some lawyers choose
this method of attacking an experts qualifications due to and Payouts?
a perceived tactical advantage.
Claims may be brought for reasons other than the Heated debate, limited and frequently conflicting data,
ones that are ascribed. Perhaps litigious plaintiffs, un- questionable research, and a lack of consensus on the
scrupulous lawyers, and uncaring physicians are not to causes of increases in professional liability insurance
blame. Physicians have to undertake much greater profes- premiums muddle the issues. Unfortunately, the media
sional responsibility, with diminished reimbursement, in spin has only complicated the issues and stirred the
a highly regulated field and with far less hospital support animosity between physicians and lawyers.
than was available in the past. Physicians may not have On one side of the issue, considerable information
time for productive interactions with patients. Too little suggests that changing the civil tort liability system will
time for patients places the quality of medical care at risk not achieve a positive effect. The Congressional Budget
and increases the likelihood that a patient will consider lit- Office released its Economic and Budget Issue Brief on
igation if an injury occurs at the hands of the physician.17 January 8, 2004. The Brief stated that evidence from the
The costs of running a practice have not decreased, states indicates that premiums for malpractice insurance
and physicians feel pressure to work more hours to are lower when tort liability is restricted than they would
maintain the same standard of practice and living, even be otherwise. But even large savings in premiums can have
in the face of decreasing reimbursement from private only a small direct impact on health care spendingprivate
and government insurers, Medicaid, and Medicare.18,19 or governmentalbecause malpractice costs account for
The U.S. Department of Labor, Bureau of Vital Statistics less than 2% of spending. The Congressional Budget Office
reported that almost one third of physicians worked 60 obtained the 2% figure based on its own calculations, using
or more hours per week in 2002. The total compensation data from Tillinghast-Towers Perrin (an actuarial and
of anesthesiologists in that same time frame was, on management consulting firm) and the Office of the Actuary
average, approximately $306,000. Therefore, it may not at the Centers for Medicare and Medicaid Services.20
be a realistic goal to state that physicians can simply plan The Briefs somewhat surprising conclusion was
better, make the appropriate cutbacks, and learn to live that the available evidence does not demonstrate that
with the decreases in reimbursement. restricting malpractice liability would have a significant
Physicians have other challenges as well. They have positive or negative effect on economic efficiency. It also
to practice in a highly legislated environment and must added that there is only weak or inconclusive evidence to
comply with regulations of federal and state government, suggest that tort liability limitations result in less defensive
in addition to private insurers and federal and state health medical practices, improved access to health care, and
insurers. They also have to practice in a manner that is, decreased medical injuries.
of necessity, dependent upon others (i.e., nurses, other From 1986 to 2002, the cost for claims did, in fact,
hospital employees, and support staff). increase, but the rate of claims remained relatively con-
In the past, highly trained, experienced registered stant. Most medical injuries caused by medical negligence
nurses commonly provided care to sick patients. The never resulted in claims. A study, conducted in 1984,
hospital environment has changed dramatically. Travel of New York medical negligence revealed 27,179 errors,
nurses, prn nurses who float from floor to floor, nursing only 1.5% of which resulted in legal claims. When minor
assistants, and nursing technicians are the mainstay of injuries were factored out, only 7.7% of the seriously in-
patient care in most hospitals. Often, there is little or jured brought claims.21 These data challenge the idea that
no relationship between the physician and nursing staff plaintiffs and attorneys are overly aggressive in pursuing
because of the turnover and transient nature of nursing malpractice claims.
assignments in a hospital setting. A nursing shortage exists Payments made on behalf of physicians to settle
in many areas of the country. Without the benefit of sound medical negligence claims must be reported to the
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 75
National Practitioner Data Bank. An evaluation of the are not getting full compensation. It also referenced
reports to the Data Bank was undertaken and the results the studies showing that most victims of negligence do
published by Health Affairs journal in an online article not sue. The committee also suggested that the system
in May 2005.22 The analysis showed that, from 1991 to impedes efforts to improve patient safety and may, in
2003, the average medical negligence payout increased by turn, actually increase medical errors.
approximately 4% each year. This increase is consistent As a result of increasing premiums, physicians have
with the increases in health care costs. demanded tort liability limitations in medical cases, and
A 2005 study by the American Enterprise Institute legislatures have responded favorably to their requests.
entitled, Two Cheers for Contingent Fees,23 summarizes the By 2002, more than 40 states had at least one restriction
conclusions of two economics professors who evaluated in effect (see Table 4.2).
the contingent fee system commonly used to retain The efforts to achieve tort liability limitations in
attorneys in civil tort cases. The researchers concluded medical cases have not been limited to the state level.
that contingent fees do not cause higher court awards. President George W. Bush repeatedly urged the passage
They warned that restrictions on contingent fees are likely of a federal law which would limit noneconomic damage
to produce unintended negative consequences. These awards to $250,000, limit punitive damage awards, reduce
observations raise questions about whether an actual the time frames in which claims could be filed, and limit
malpractice crisis really exists, and whether taking away fees paid to lawyers handling medical negligence cases.
or limiting the rights of the victims of medical negligence State legislatures have vigorously opposed the federal
will make a meaningful difference for physicians or the governments attempt to regulate medical negligence laws.
health care system. States accuse the federal government of attempting to
However, all data is subject to attack, and a wealth invade the purview of the states to regulate this area of
of information exists contrary to these observations. the law. An Action Alert was issued in July 2005 by the
A subcommittee of the Joint Economic Committee, vice- National Conference of State Legislatures25 in which they
chaired by Republican Jim Saxton from New Jersey, urged member states to contact their representatives in the
concluded in its May 2003 report to the US Congress24 that US House to vote against H.R. 5, the HEALTH Act (Help
the present liability system fails in its two goals: It neither Efficient, Accessible, Low-cost, Timely Health care) of
compensates the negligently injured nor penalizes and 2005. The bill would preempt states in the area of medical
deters negligent acts. In support, the committee pointed malpractice reform and establish uniform procedures and
to the large attorney fee payments as proof that victims substantive requirements for state medical malpractice
States with special SOL requirements: Most states have 2 years or longer SOL with some discovery provision
triggering the commencement of the SOL. The following states have more stringent requirements:
AL; AZ; FL; GA; ID; IL; IN; KY; LA; MI; MO; NE; OH; OR; SD; TN; UT
States with limitations on damage awards (primarily limits on noneconomic damages and punitive damages):
AL; AK; CA; CO; FL; GA; HI; ID; IL; IN; KS; LA; ME; MD; MA; MI; MS; MO; MT; NE; NV; NJ; NM; NC; ND; OH; OK; SD;
TX; UT; VA; WA; WV; WI
States with presuit screening requirements, arbitration proceedings with limited damages or other special
requirements before suit can proceed:
AL; AZ; DE; FL; GA; HI; ID; IL; IN; ME; MD; MA; MI; MN; MS; MT; NE; NJ; NM; NY; OK; TX; VA; WA; WV; WY
Joint and several liability between defendants:
CA; DE; FL; HI; IN; IA; KS; ME; MD; MA; MI; MN; MT; NE; NJ; NY; NC; OH; OK; PA; RI; SC; SD; TX; VT; VA.
States without Expert Witness Regulations (includes training, education, certification in field, recent clinical experience,
etc.)
AZ; HI; KY; ME; MA; MT; NE; NM; ND; OR; SC; SD; UT; VT; WA; WI; WY
States with limitations in attorneys fees
CA; CT; DE; FL; HI; IL; IN; IA; KS; ME; MA; MI; NV; NH; NJ; NY; OK; OR; TN; UT; WA; WI; WY
Patient compensation/stabilization/no-fault fund (often with limitations or unavailability of private civil claims for
medical negligence)
DE; FL; IN; KS; LA; MD; NE; ND; OR; PA; SC; VA; WI; WY
lawsuits. Pharmaceutical companies would also get some Dr. Girish Joshi included in his article,27 10 Things
special protection. The proposed law has been referred to that Changed Anesthesiology, for the commemorative
as the HEALTH Act. The House of Representatives has centennial issue of the ASA Newsletter. The 10 things
approved the bill multiple times, whereas the Senate has have been either developed or implemented by anesthesi-
defeated it each time. ologists, including:
According to a statistical analysis of 2001 data
compiled by A.M. Best, doctors spent $7.2 billion to obtain 1. Pharmacological advances
malpractice coverage. The alternative coverage market is 2. Fluid therapy
believed to be twice as large.26 Physicians in nearly every 3. Airway management
area of the country have complained about the increases 4. Ventilators and pumps
in premiums. 5. Monitoring devices
The American Academy of Actuaries Work Group 6. Closed Claims Project
study evaluated various factors that could help stabi- 7. Standards and guidelines
lize premiums. The efforts yielded recommendations that 8. Perioperative care
have been supported by others, including a cap on noneco- 9. Board certification
nomic damages and an offset for collateral payments from 10. Anesthesiology research
sources other than the defendant physician.1 Increases in The ASA Closed Claims Project began in 1985. It
premium rates were lower for each reported physician spe- consisted of a detailed analysis of claims resulting from
cialty in the states with noneconomic damage caps. The anesthesia mishaps. The goals of the Committee on Pro-
opponents of noneconomic caps point out the disparate fessional Liability of the American Society of Anesthesi-
impact on the elderly, children, and women. A brain in- ologists, under Dr. Frederick W. Cheney, Jr.s leadership,
jury, loss of sight, loss of limb, or paralysis for a nonwage were to identify losses and improve patient safety where
earning woman could only be awarded $250,000 for the possible, and thereby reduce insurance burdens on physi-
disability suffered. The award would be the same for a cians. The innovative approach to loss analysis and patient
young child, with 70 or more years of anticipated future safety is unique to anesthesiology. Anesthesia providers
disability from the injury. Given the increases in costs have a body of literature, standards, and guidelines for
for transportation, housing, and daily living expenses, a ready reference in daily practice, thereby facilitating the
single lifetime award of $250,000 could not be expected widespread dissemination of information for anesthesia
to provide lifetime compensation for an injury expected care. Patients benefit from an ongoing commitment to
to cause life-long loss. scholarly writing, focused medical research, and open dis-
The debate undoubtedly will continue, and the cussion and debate about patient careall of which are
interests of physicians and patients will continue in common in the specialty of anesthesiology. Patient anes-
conflict until a workable solution is found. thesia outcomes have improved dramatically over time,
and the measures outlined above must be given credit for
a substantial portion of the improvements.28
advice or be burdened and ill-advised by anyone with prudent physician would have done under the same or
ulterior motives. Because it is so difficult to remain objec- similar circumstances, or doing that which a reasonable
tive when evaluating ones own conduct, reliance on the and prudent physician would not have done under
advice and counsel of a trained, experienced professional the same or similar circumstances. Physicians are
who has only your best interest at heart is essential. defined by their respective specialties (anesthesiologists,
Following these simple but effective strategies will re- otolaryngologists, cardiologists, internists, etc.).
sult in a significant decrease in the number of unexpected In most cases, the plaintiff(s) must retain a testifying
adverse outcomes that end up in litigation, and will assist expert to elicit evidence through their testimony about the
the practicing anesthesiologist or anesthetist in ensuring standard of care, the breaches found in the case, and the
that his or her rights are adequately protected should proximate cause of injury. Theoretically, the defendant
litigation ensue. does not require a retained expert, and oftentimes the
defendant is his own expert; however, in most cases,
the defendants retain experts to rebut the opposing
experts opinions.
What Is the Process and How In the case under discussion, the first issue was
whether the ENT consultant should have responded to the
Does It Work? order on postoperative day 4 rather than late on postoper-
ative day 6. The defense position was that it was a weekend
and the consultation was routine, not STAT. Therefore,
CASE ANALYSISDEFENSE showing up on Monday was reasonable under the circum-
stances. Once the consultant examined the patient, there
PERSPECTIVE was no time to do anything other than look in the patients
On May 5, 2004, the United States House of Represen- throat with a laryngoscope, at which time she sustained
tatives introduced H.R. 4280, known as the HEALTH Act cardiopulmonary arrest and required resuscitation.
of 2004, in an attempt to nationalize medical negligence Overall, the consultant had two defenses to liability.
litigation. The act was tabled on May 13, 2004 and has not The first approach was to shift the blame to the anes-
been reconsidered. The failure of this national tort reform thesiologist who caused the tear, and the second was
measure for medical negligence lawsuits leaves the admin- to emphasize the failure of all the physicians to diag-
istration of these cases to the states. Currently, medical nose the tear for 6 days. Failure of the anesthesiologist
malpractice cases in each of our 50 states are governed by to document the second intubationand any possible
that states law. Generally, three considerations apply to complicationcould provide a defense to all of the physi-
any medical-legal case: Negligence, proximate cause, and cians who participated in the patients care after that
damages. Negligence and proximate cause considerations incident. This strategy necessarily pits physicians against
are similar in law and medicine. Negligence is generally each other. However, that defending attorney usually can
defined as not doing that which a reasonable and prudent avoid directly criticizing other physicians, at least initially,
physician under the same or similar circumstances would because the plaintiffs attorney has to retain and identify
do. Proximate cause means that the negligent conduct, to experts to make the case against them to prevail.
a reasonable degree of medical probability and in a natu- The most challenging aspect of defending a physician
ral and continuous sequence of events, caused an injury is to help the jury look at the case prospectively. Until
or death. the CT scan was obtained, none of the attending physi-
Damages are considered under each states common ciansanesthesiologist, surgeon, or internistsuspected
lawor case lawor legislation and usually include pain a tear with air and fluid leak. The injury was obvious
and suffering, mental anguish, disfigurement, physical after the CT scan and by direct visualization by the ENT
impairment, lost wages or loss of wage-earning capacity, consultant. Of course, the end result is known to the jury
and medical expenses. Spouses of injured patients can from the beginning of the case. Therefore, it is difficult to
recover for loss of consortium, which is the loss of those get the jurors to analyze the case as the events unfolded,
things intrinsic to a marital relationship such as compan- without the benefit of hindsight.
ionship, society, and sexual relations. Other survivors of
a patient who dies (i.e., children) can claim loss of con- Proximate Cause
sortium, mental anguish, and pecuniary loss. Tort reform
measures on the state and national levels are aimed at The next inquiry for a jury is proximate cause, whether
limiting the amount of damages recoverable, primarily any act or omission resulted in an injury to or death of
noneconomic damages such as pain and suffering, men- the patient. Because the order for consultation was made
tal anguish, disfigurement and physical impairmentas 4 days after surgery and the examination was 6 days
opposed to economic damages such as medical ex- after surgery, the question is whether anything the ENT
penses, lost wages, and loss of wage-earning capacity. consultant did, or allegedly failed to do, caused injury
to the patient. After 6 days of air and fluid leaking into
Negligence the mediastinum, serious injuries had already occurred,
and a diagnosis and rapid treatment at that time would
Negligence is generally defined as failure to use ordinary necessarily result in extensive treatment with its attendant
care, that is, failure to do what a reasonable and risks and morbidity.
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 79
Damages parties are deposed, the expert witnesses for the plaintiff
are deposed, then the experts for the defendants. Once
Medical negligence lawsuits are filed by patients and their a deposition is taken, that witness factual knowledge,
families against physicians to seek monetary damages. position, and opinions are known and locked in, because
Although the analysis of negligence and proximate cause the testimony is both under oath and preserved in writing
are medical issues, damage analysis is strictly a legal for all time.
one. For a patient who lives, elements of damages may
include pain and suffering, mental anguish, disfigurement,
physical impairment, lost wages, loss of wage-earning
Mediation
capacity, and medical expenses. For patients who do Most states courts and their legislatures have mandated
not live, their estate may have damages for mental alternative dispute resolution in all civil cases. Gener-
anguish, pain and suffering, and medical expenses. Their ally, this action entails a structured settlement conference
survivors (spouse, children, and parents) could have between all parties, their lawyers, and other decision mak-
claims for pecuniary loss, mental anguish, and loss of ers such as the representatives of the physicians medical
consortiumthat is, the loss of the personal relationship malpractice carriers. A trained facilitator manages the
between and husband and wife or parent and child. process, which usually takes a day. Under most schemes,
This case points out a typically catastrophic injury the process is nonbinding, meaning some settle and
from an anesthetic complication. It illustrates the serious some do not. Some cases partially settle, and some settle
nature of anesthetic risks that may result in ventilator all parties and all claims.
dependency, mediastinal infection, and eventually death; Partial settlements always change the dynamics of
6 months in a coma with hundreds of thousands of dollars the case. Mediation is usually scheduled many months
of attendant care costs; and the mental anguish, physical after the suit is filed, written discovery is exchanged,
impairment and, for those who are the sole support of motions are heard, and the depositions of the parties
their families, loss of income. and the experts have been taken. After so much time
and effort spent, the case is in a certain posture that
is difficult to change. The challenge for both sides is to
decide whether theories can be changed and how best to
PRETRIAL PROCEDURES take advantage of this change, if possible. If there is a
Discovery and Written Motions partial settlement, the remaining defendants can submit
the negligence of the settling defendants if they plead
In addition to the written discovery (interrogatories and and can prove their negligence. Oftentimes, the proof
requests for production) and depositions, pretrial motions comes from the plaintiffs expert who was critical of the
are a large part of legal practice. In every case, the defense physician when he or she was a defendant, and who
considers a Motion for Summary Judgment, either to was deposed before mediation. If the patients experts
have the case dismissed outright, or to narrow the issues. critical testimony is not available, then the defendant has
A summary judgment involves gathering evidence through to obtain critical testimony elsewhere.
medical records or other documents, depositions, and Here, the anesthesiologist who caused the esophageal
affidavits to present the case to the judge for a decision on tear settled, as did an anesthesiologist who evaluated the
written motion. If any of the three elementsnegligence, patient a day after surgery. The hospital also settled. The
proximate cause, or damagescan be presented without plaintiffs experts had already been deposed before medi-
controversion by the plaintiffs attorneys, the case can be ation, so their criticisms of the settling anesthesiologists
decided by the court. Under the facts of this case, if the were preserved and available for presentation to the jury
esophageal tear occurred at the time of surgery and was at the time of trial. The active defendants at trial want the
undiagnosed for 6 days, the ENT consultants actions did jury to consider the negligence of the settling defendants.
not change anything about the patients treatment. If the jury finds the settling physicians negligent, they will
Summary judgments are only granted if no evidence award a percentage of negligence against them. Any per-
is presented contrary to the movants position. If the centage of negligence awarded to the settling physicians is
nonmovant responds, the motion is usually denied. In subtracted from the total award. Generally speaking, set-
this case, the plaintiffs attorney responded with an tling defendants are more likely to have liability awarded
affidavit from their retained testifying expert to show the against them, as they are not present at the trial to defend
consultant violated the standard of care by failing to show themselves, or if they testify, they usually do not have an
up on postoperative day 4 instead of postoperative day 6. attorney defending their position.
Their position was that, had diagnosis and treatment
started earlier, the patient would not have arrested
Monday evening. The summary judgment was denied.
THE TRIAL
Depositions Impact of Partial Settlements
The depositions of all parties on both sides normally are Because only the internist and ENT consultant were de-
taken within the first few months of filing suit. After the fendants at the time of the jury trial, a number of strategies
80 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
had to be resolved. Trial can be analogous to surgery and The percentages you find must total 100%. The per-
anesthesiologymuch of it can be anticipated, but some- centages must be expressed in whole numbers. Negligence
times things that were not anticipated happen. With the attributable to any one named below is not necessarily
settlement of several parties, much of the story was miss- measured by the number of acts or omissions found. The
ing. The defense strategy was to show that the complica- percentage attributable to any one need not be the same
tion of esophageal tear occurred at the time of surgery, was percentage attributed to that one in answering another
not documented as a complication, and the patients con- question.
dition in the day or two after surgery indicated the diagno-
sis should have been made days earlier. This is a strategy Question No. 2
that dovetails with the plaintiffs best theory of the case.
The ENT consultant could refer to the issue of What percentage of the negligence that caused the
proximate causeif the tear occurred on the day of occurrence do you find to be attributable to each of those
surgery and was not treated for 6 days, the damage was found by you, in your answer to Question No. 1, to have
not going to be reversed on the day the patient was seen by been negligent?
the consultant. Indeed, treatment was started immediately a. ENT Consultant 0
after the consultant saw the patient the first time. If the b. Hospital 0
jury believed the consultant should have seen the patient c. Internist 35%
on postoperative day 4, they had to be convinced that d. Anesthesiologist 1 35%
the mediastinal infection was active, and a delay of a e. Anesthesiologist 2 30%
day or two would not have made any difference in the Total 100%
patients recovery.
After answering these questions 1 and 2, the questions
regarding damages were answered by the jury:
Impact of Pretrial Motions
Because a request for Summary Judgment on behalf of Question No. 3
the ENT consultant had been filed earlier in the case, the What sum of money, if paid now in cash, would fairly and
plaintiffs lawyer responded with an affidavit of his expert, reasonably compensate the son for his damages, if any,
criticizing the consultant for failure to show up on the resulting from the death of his mother?
day the order requesting consultation was written. This Consider each element of damages listed below and
document could be used by the plaintiffs attorneyor none other. Consider each element separately. Do not
the codefendants attorneyin front of the jury to create include damages for one element in any other element. Do
an issue against the consultant; it represents one of the not include interest on any amount of damages you find.
dangers of filing Motions for Summary Judgment, and it
invites criticism from the plaintiffs experts that can be a. Pecuniary loss
used against the defendant at trial. In this case, however, Pecuniary loss means the loss of the care, mainte-
the attorney for the patient was focusing on the events of nance, support, services, advice, counsel, and reason-
surgery and the day or two after, and did not bring out able contributions of a pecuniary value, excluding loss
this specific criticism of the ENT consultant. of inheritance, that the son, in reasonable probability,
would have received from the mother had she lived
b. Loss of companionship and society
The Result Loss of companionship and society means the loss
After 10 days of trial, the jury returned a verdict, answering of the positive benefits flowing from the love, comfort,
companionship, and society that the son, in reasonable
the following questions:
probability, would have received from the mother had
she lived.
Question No. 1 c. Mental Anguish
Did the negligence, if any, of the persons named below Mental Anguish means the emotional pain, torment,
proximately cause the occurrence in question? and suffering experienced by the son because of the
Answer Yes or No for each of the following: death of mother/patient.
In determining damages for elements a. and b., you
a. ENT Consultant No
may consider the relationship of the son and his mother,
b. Hospital No
their living arrangements, any extended absences from
c. Internist Yes
one another, the harmony of their family relations, and
d. Anesthesiologist 1 Yes
their common interests and activities. You are reminded
e. Anesthesiologist 2 Yes
that elements a. and b., such as the other elements of
damages, are separate, and, in awarding damages for one
After answering this question, the jury turned its
element, you shall not include damages for the other.
attention to Question No. 2. If you have answered Yes
Answer, with respect to the elements listed above, in
to Question No. 1, for more than one of those named
dollars and cents for damages, if any, that:
above, then answer the following question. Otherwise, do
not answer the following question. a. Were sustained in the past: $925,000.00
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 81
b. In reasonable medical probability, did not find the ENT consultant did anything negligently
Will be sustained in the future: $925,000.00 and probably agreed that the infection was present and
extensive before day four.
Question No. 4
What sum of money would have fairly and reasonably
compensated the mother?
Does the Future Present an
a. Pain and mental anguish
Pain and mental anguish means the conscious phys- Opportunity for Physicians?
ical pain and emotional pain, torment, and suffering
experienced by the mother before her death as a result The Institute of Medicine in 1999 prepared an often cited
of the occurrence in question. report: To Err Is Human: Building a Safer Healthcare
b. Medical expenses System.30 The report was compiled by a group of medical
Medical expenses means the reasonable expense of experts who tried to develop a method of identifying errors
the necessary medical and hospital care received by his and practices that may lead to medical losses. Data was
mother for treatment of injuries sustained by her as a collected from a varied range of hospitals, physicians,
result of the occurrence in question. and other health care providers. The report concluded
c. Funeral and burial expenses that errors and complex failures can be identified and
Funeral and burial expenses means the reasonable corrected, particularly errors that were inherent in the
amount of expenses for funeral and burial for the system itself.
mother reasonably suitable to her station in life. Similarly, in 2000, Dr. Lucian L. Leape wrote an ar-
Do not include any amount for any condition existing ticle titled Can we make health care safe?31 as part of a
before the occurrence in question, except to the extent, collaborative initiative between the National Coalition on
if any, that such other condition was aggravated by any Health Care and the Institute for Health Care Improve-
injuries that resulted from the occurrence in question. ment named Accelerating Change Today (A.C.T.). He
summarized the shocking findings released in the Insti-
Answer in dollars and cents for damages, if any.
tute of Medicine report about the frequency of medical
Answer: $1,029,721.44
errors. However, he also analyzed how dramatic improve-
The active defendant who was found negligent, the
ments in patient safety and substantial savings can result
internist, would pay 35% of this verdict of $2,879,721.44.
by implementing systemwide changes in health care.
The ENT consultant would pay nothing because he was
These promising statements encourage physicians to
found not negligent. The other parties had settled, so
invest time and effort to determine how best to reduce
they paid nothing more. However, the submission of these
individual practice and system errors, with the potential
parties was an important part of the strategy, because
for better care for patients and reduced costs, including
the total verdict was reduced by 65%the amount of
reduced medical liability premiums.
negligence and proximate cause attributed to the anesthe-
We hope that the worlds of medicine and law will
siologists. Because the hospital was not found negligent,
someday comfortably coexist, with physicians no longer
the decision to submit their negligence did not pay off
viewing lawyers as a threat to their way of life. Physicians
with a finding of negligence and award of a percentage.
face many difficult challenges in their day-to-day prac-
tice, but they have a substantial investment in medicine,
and will, no doubt, overcome each of the various obsta-
FINAL ANALYSIS cles and continue to provide meaningful contributions to
humankind.
The complication of esophageal intubation and tear dur-
ing tracheal intubation is a known risk and, so long as
proper technique is employed, should not be considered
negligent. However, most experts would agree that the KEY POINTS
possibility should be documented in the chart. Failure
to diagnose mediastinal infection in the day or two af- 1. Documentation of anesthesia complications is im-
ter intubation (in an otherwise healthy patient) is harder portant, not only for the acting physician, but also
to defend with the signs the patient was exhibiting after for other caregivers to let them know what happened
surgery. The jury deciding this case obviously felt the anes- and perhaps know what to expect.
thesiologists and internist shared blame for this patients 2. Even if a complication, such as an esophageal tear, is
death. The first issue was the failure of the anesthesiolo- a remote possibility, signs of the complication need
gist to document the possibility of perforation. Not only to be investigated quickly after surgery.
does this fact make a jury think this was an attempt to 3. Credibility always is an issue. Do not misrepresent
cover up a mistake, it also fails to raise the suspicions of facts. Nothing turns an otherwise defensible case
the physicians caring for this patient in the days follow- into a sure loser quicker than dishonesty or a refusal
ing. The jury also felt the second anesthesiologist and the to concede the obvious.
internist had sufficient signs of esophageal perforation to 4. Patients are often unwilling to accept known com-
diagnose the problem on day 2, or 3, or even 4. The jury plications that result in death or morbidity.
82 G E N E R A L A N D M E D I C O L E G A L C O N S I D E R AT I O N S
5. Patients and their lawyers often do not recognize the Available at: http://www.acponline.org/journals/news/feb00.
Accessed January 9, 2006.
plight of physicians caught in a constant battle to
5. Domino KB. Yearly survey of premiums: Professional liabil-
survive the pressures of practice, whereas physicians ity insurance for anesthesiologists. ASA Newsl. 2005;69(6):
often do not understand the inherent unfairness of Available at: http://www.asahq.org/Newsletters/2005/06 05/
a patient having to accept a negligently inflicted, domino0605 .html. Accessed January 3, 2006.
life-altering injury that could have been avoided. 6. Florida Statutes. Title XLV Torts Chapter 766 Medical
6. A legal claim for medical negligence requires that: Malpractice and Related Matters 766.102(3),2006. Medical
(a) a legal duty was owed to the patient and the pa- negligence; standards of recovery; expert witness.
tients spouse and minor children; (b) a reasonable 7. Linn v. Fossum, 894 So.2d 974 (Fla. 1st DCA 2004).
and prudent health care provider would not have 8. Frye v. United States, 293 F 1030 (DC Cir. 1923).
performed in the same manner under similar cir- 9. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 US 579
(1993).
cumstances; (c) a causal link between the duty and
10. Florida Statutes. Title XLV Torts Chapter 766 Medical Mal-
breach of the duty; and (d) the breach of the duty re- practice And Related Matters 766.203,2006. Presuit investiga-
sulted in specific or quantifiable harm to the plaintiff. tion of medical negligence claims and defenses by prospec-
7. To avoid informed consent claims, a physician tive parties.
should advise the patient of the risks, benefits, and al- 11. Georgia Code. Title 51, Torts Chapter 1. General Provisions
ternatives to a proposed plan of care. The physician O.C.G.A. 51-1-27, 2006. Recovery for medical malpractice
need not advise of every potential risk but should authorized.
identify the most common and the most serious. 12. AMA News. Opinion: expert witnesses need scrutiny: The
8. An inherent conflict exists between the defendant Tennessee approach. Editorial, December 19, 2005. Avail-
physician, the insurance-retained attorney for the able at: http://www.ama-assn.org/amednews/2005/12/19/
edsa1219.htm. Accessed January 7, 2006.
physician, and the insurance company.
13. Scott M. ASA adopts review procedure for expert
9. If the insurance carrier urges you to withdraw your witness testimony. ASA Newsl. 2006;67(12):1. Available
consent or to proceed to verdict in a case you feel at: http://www.asahq.org/Newsletters/2003/12-03/scott.html.
should settle, and a defense verdict results, no harm Accessed February 2, 2006.
is done. However, if the case results in an adverse 14. Cohen TH. Civil Justice Survey of State Courts, 2001. Medical
jury verdict in excess of the policy limits, you will malpractice trials and verdict in large counties, 2001. Bureau
be personally liable for that portion of the verdict of Justice Statistics, Civil Justice Data Brief, April 2004, NCJ
that exceeds the limits of your liability insurance 203098. U.S. Washington, DC: Department of Justice, Office
coverage. of Justice Programs.
10. Once a deposition is taken, that witness factual 15. American Bar Association Tort Trial and Insurance Practice
Section, Task Force on Contingent Fees: Report on Con-
knowledge, position, and opinions are known and
tingent Fees in Medical Malpractice Litigation, September
locked in, because the testimony is both under oath
20, 2004, Draft Report. Available at: http://www.abanet.org/
and preserved in writing for all time. tips/contingent/MedMalReport092004DCW2.pdf. Accessed
11. In most cases, the plaintiff(s) must retain one or January 30, 2006.
more testifying experts to elicit evidence through 16. Florida Capital News. Senate hearing clarifies malpractice
their testimony about the standard of care, the facts; more doctors coming in, no sharp rise in settlements.
breaches found in the case, and the proximate cause July 15, 2003. Published by Gannett News. Available at:
of injury. In most cases, the defendants retain ex- http://www.floridacapitalnews.com/legislature2003/stories/
perts to rebut the opposing experts opinions. 071503insurance.htm. Accessed January 7, 2006.
12. What an extraordinary physician does, or what the 17. Kahn O. Letter to the editor: The biggest issue in the complex
health care equation: Diminishing time spent with patients.
standard of care used to be, or has evolved to be since
American Medical Association News; December 19, 2005.
the incident, should not be a part of the evidence that
Available at: http://www.ama-assn.org/amednews/2005/12/
the jury considers. 19/ed1t1219.htm. Accessed January 7, 2006.
13. Litigation strategies are an evolving process through- 18. Bureau of Labor Statistics, U.S. Department of Labor. Occu-
out the life of a medical negligence suit. pational outlook handbook, 2006-07 edition. Physicians and
Surgeons; Bulletin 2600. Available at: http://stats.bls.gov/oco/
print/ocos074.htm. Accessed January 3, 2006.
REFERENCES
19. Florida Pediatric Society v. Levin. Case No. 0523037 (So.
1. U.S. General Accounting Office. Implications of rising pre- Dist. Fla. 2005).
miums on access to health care. GAO-03-836 (Washington, 20. Congressional Budget Office. Limiting tort liability for medical
DC, August 2003). Available at: http://www.gao.gov/ malpractice. Economic and Budget Issue Brief, January 8,
new.items/d03836.pdf. Accessed January 31, 2006. 2004. Available at: http://www.cbo.gov/showdoc.cfm?index=
2. Hoyert DL, Kung HC, Smith BL. Centers for Disease Control, 4968&sequence=0. Accessed November 22, 2005.
U.S. Department of Health and Human Services, National 21. Localio AR, Lawthers AG, Brennan TA, et al. Relation
Center for Health Statistics. Deaths: Preliminary data for between malpractice claims and adverse events due to neg-
2003. Natl Vital Stat Rep. 2005;53(15):25. ligence. Results of the Harvard Medical Practice Study III.
3. Soreff S. Suicide. eMedicine Journal [serial online]. 2006. N Engl J Med. 1991;325:245.
Available at: http://www.emedicine.com/med/topic3004.htm. 22. Chandra A, Nundy S, Seabury, SA. The growth of physi-
Accessed Date (February 8, 2006). cian medical malpractice payments: Evidence from the
4. Kelly CK. Facing up to an occupational hazard: Substance national practitioner data bank. Health Aff Policy J Health
abuse. American College of Physicians; 2000. February Sphere. Posted on internet May 31, 2005. Available at:
CHAPTER 4/MEDICINE AND LAW: WHEN TWO WORLDS COLLIDE 83
2
P E R I O P E R AT I V E
CLINICAL
C O N S I D E R AT I O N S
A . R E S P I R ATO R Y
CHAPTER RESPIRATORY COMPLICATIONS
5
OVERVIEW
Robert R. Kirby
CASE SUMMARY the PACU nurse observed that he was unable to maintain
his SpO2 in a satisfactory range, and the attending
14-year-old boy was admitted to an anesthesiologist was called. Bilateral fine crackles were
A
ambulatory surgical center for elective ton- noted. A chest radiograph was obtained and showed
sillectomy and adenoidectomy. He had un- diffuse interstitial infiltrates consistent with pulmonary
dergone general anesthesia for uneventful edema (see Fig 5.1). The patient was transferred to the
knee surgery 8 weeks previously. Past med- hospital and was admitted to the PICU. He was treated
ical history included remote asthma. He with nasal continuous positive airway pressure, and then
took no medications and had no history of allergies. Phys- weaned to nasal oxygen which was discontinued the next
ical examination showed a healthy male of average build morning. Subsequently, he was discharged home with no
who was 5 ft 10 in. (177 cm) tall and weighed 152 pounds sequelae 48 hours following PICU admission.
(69 kg). His blood pressure (BP) was 121/70 mm Hg, and
his heart rate (HR) was 44 beats per minute. Airway ex-
amination showed a mental-hyoid distance of three finger
breadths, a normal appearing mandible, and full range Why Do Pulmonary Problems
of motion of his head and neck. Chest auscultation was
normal, and his SpO2 was 100% while he breathed room
Occur?
air. A Mallampati III classification was noted because of a
deep airway and posterior pharynx. He was assigned an Pulmonary complications are the most frequently re-
American Society of Anesthesiologists physical status II. ported causes of postoperative morbidity and mortality,1,2
Before the induction of general anesthesia, 2 mg of in particular, after procedures involving the abdomen and
midazolam and 100 g of fentanyl were administered in- thorax.3,4 The incidence reportedly is as high as 75% in
some patient populations.5,6 Yet, as the case summary
travenously, followed by a rapid sequence induction with
shows, they may occur in the absence of major surgi-
300 mg of propofol and 50 mg of rocuronium. A grade 1
cal procedures when one might least expect them. When
laryngoscopic view was obtained, and easy intubation was
possible, the identification of patients at riskto allow
achieved with a Macintosh 3 blade and a 6.5 RAE endotra-
appropriate interventions that limit that riskis impera-
cheal tube. Maintenance of anesthesia utilized isoflurane,
tive before surgery. However, in this case, no amount of
nitrous oxide, and oxygen. The case proceeded unevent-
preanesthetic assessment would have indicated that this
fully and, at its conclusion, neostigmine and glycopyrro- perfectly normal and healthy young man would develop
late were administered for reversal of muscle relaxant. problems that would convert his anticipated outpatient
He was extubated in the operating room. Immediately surgical procedure into a 3-day hospital stay.
following extubation, his SpO2 decreased precipitously,
and he developed mild laryngospasm that responded
to the insertion of oral and nasal airways and mild
pressure. He then coughed, sat up, and was transferred to
RISK FACTORS
the PACU. Upon arrival, his BP was 142/76 mm Hg, HR 96 Prolonged mechanical ventilation, pneumonia, atelecta-
beats per minute, respiratory rate 18 breaths per minute, sis, changes in chest radiographs, respiratory failure,
and SpO2 85% while he inspired 100% oxygen through a prolonged ICU/hospital stays, and nosocomial infections
nonrebreathing face mask. Three hours postoperatively, are major factors. Clearly, the causes are multifactorial,7
87
88 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Data from Lang SA, Duncan PG, Shephard DAE, et al. Pul-
monary edema associated with airway obstruction. Can J
Anaesth. 1990;37:210.
ETIOLOGY
This condition will be described in some detail, because
FIGURE 5.1 - Diffuse pulmonary infiltrates consistent with it is most common to anesthesiology. Other forms of
interstitial pulmonary edema following tracheal extubation and pulmonary edema will be described in Chapter 12. The
transient airway obstruction. etiology involves a series of events characterized most
commonly by sudden relief of a partial or total obstruction
of the airway as, for example, with tracheal intubation (see
and their risk is increased by extremes of age, smok- Table 5.3)a syndrome most common in anesthesiology.
ing, preexisting lung disease, type of surgery, duration Luke et al.17 reported . . . Four patients (ages 3 to
of anesthesia, and poor general health.815 Patients with 6 years) with severe nasophayrngeal obstruction (had) . . .
preexisting lung disease are at greater risk for complica- cardiorespiratory complications ranging from moderate
tions.16 Identification of patients who may be at risk for cardiac enlargement and right ventricular hypertrophy to
postoperative pulmonary complications is relatively sim- cor pulmonale and pulmonary edema. . . . Wide swings in
ple compared to the complex task of modifying that risk. intrathoracic pressure probably played an important role
in the etiology of pulmonary edema.
Capitanio and Kirkpatrick18 noted, Obstructing le-
sions of the upper airway should be suspected when the
What Is Postobstructive transverse diameter of the heart appears larger during
the expiratory phase of respiration as opposed to its
Pulmonary Edema? size during inspiration. . . . Acute pulmonary edema with-
out cardiac enlargement may occur in patients with an
acute upper airway obstruction. Oswalt et al.19 described,
HISTORICAL Acute fulminating pulmonary edema (that) developed in
CONSIDERATIONS three patients after acute airway obstruction . . . (minutes
to hours). . . . The common etiologic factor was vigorous
In the case under discussion, several questions might
be askedmost, if not all of them, after the fact. The
questions of utmost importance are, what happened, and TABLE 5.2 Pediatric Epidemiology in Postobstructive
was it preventable? Almost certainly this young man sus- Pulmonary Edema
tained postobstructive pulmonary edema, often referred
to (incorrectly) as negative-pressure pulmonary edema. Age (years) 3 2.4 Range
The clinical course of this problem has been reported in 1/1210
the published literature for more than 40 years in pedi- Male:Female 2.4:1
atric patients17,18 and adults,19 and the epidemiology has Common obstructive Supraglottitis Croup
been reasonably well documented (see Tables 5.1 and 5.2). events (15/45) (18/45)
Could anything have been gleaned from this young mans Time to onset following 33 66 Range
history and physical examination? The only abnormal obstruction (minutes) 5240
finding was a class III Mallampati score, and the value Resolution (hours) 42 31 Range 296
of that routine examination has been subjected to recent
criticism.20 Regardless, in this case, the potentially diffi- Data from Lang SA, Duncan PG, Shephard DAE, et al. Pul-
cult airway or intubation that may have been indicated as monary edema associated with airway obstruction. Can J
a risk did not materialize. Anaesth. 1990;37:210.
C H A P T E R 5 / R E S P I R AT O R Y C O M P L I C AT I O N S O V E R V I E W 89
TABLE 5.3 Etiology of Postobstructive Pulmonary Edema TABLE 5.4 Clinical Presentation of Postobstructive
Pulmonary Edema
Relief of obstruction
Airway pressure Respiratory distress
Pleural pressure Tachypnea
Venous return and pulmonary artery blood flow Dyspnea
Microvascular pressure Paradoxical Breathing
Result: interstitial and alveolar edema SpO2
Cyanosis (late)
Modified from Kamal RS, Agha S. Acute pulmonary edema. A Wheezing, stridor
complication of upper airway obstruction. Anaesthesia. 1984;39: Pink, frothy secretions (sometimes frank blood)
464.
inspiratory effort against a totally obstructed upper airway mechanical ventilation seldom is required, unlike more
(and) ventilatory assistance (was required) to maintain severe forms of pulmonary edema that are discussed in
oxygenation . . . Chapter 12. Table 5.5 shows the elements that should be
Although one might briefly consider a differential considered.
diagnosisincluding anaphylactic or anaphylactoid reac- In summary, postobstructive pulmonary edema prob-
tions, pulmonary aspiration of gastric contents, some type ably occurs much more commonly than is generally
of latent cardiomyopathy (way down the list)a history recognized in the operative and perioperative periods.
of a brief period of laryngospasm during emergence Its features can easily be mistaken for other conditions
from anesthesia, the benign nature of the clinical course, such as the pulmonary aspiration of gastric contents, but
and the rapid response to minimal therapyargue for it differs in that resolution is usually rapid and complete
airway obstruction as the causative factor and against within hours of the inciting episode. Subtle episodes may
anything else. be detected by pulse oximetry when clinical manifesta-
tions are minimal to absent. Prevention clearly is superior
to treatment, but the onset is so rapid in some cases
CLINICAL COURSE that the anesthesia provider may be unaware that any ad-
verse outcome related to airway obstruction has occurred.
Fortuitously, most cases of this syndrome proceed more Therefore, the answer to the second question asked at the
or less uneventfully, with rapid recovery being the norm. beginning of this chapter, Is the condition preventable?
McConkey21 reported a small series of patients with is often no.
postextubation pulmonary edema. Of the six individuals Although rare, severe cases leading to death22 have
studied, all cases were preceded by an episode of been reported, as has pulmonary hemorrhage.23,24 The
laryngospasm; frank hemoptysis occurred in five; one incidence of postobstructive pulmonary edema has been
patient was reintubated and ventilated; two patients were suggested to be as high as 0.5% to 1% of all intubated,
admitted to the ICU for face mask CPAP; one patient anesthetized patients. Some cases have resulted from
was managed with CPAP in the PACU; two patients biting and occluding endotracheal tubes and laryngeal
received only oxygen; and all cases resolved fully within mask airways.25,26 They have also occurred in patients
24 hours. for whom there was no evidence of obvious airway
obstruction, although relative compromise was likely to
be present.
EPIDEMIOLOGY
Published literature suggests that postobstructive pul- TABLE 5.5 Supportive Therapy of Postobstructive
monary edema occurs in 20,000 to 30,000 cases annually Pulmonary Edema
in the United States and is often unrecognized or misdi-
agnosed.21 However, the benign nature of this problem is General supportive measures
not always seen. Adolph, et al., stated, Although the vast Maintenance of patent airway
majority of cases resolve quickly and with minimal prob- Oxygen
lems, death from ARDS and multisystem organ failure has CPAP
been reported despite relief of the airway obstruction.21 Mechanical ventilation with PEEP (seldom necessary)
Postobstructive pulmonary edema is only one of many Careful monitoring (invasive seldom required)
entities that manifest similar signs and symptoms (see No furosemide or other diuretics initially
Table 5.4). This similarity makes the historical aspects of Aggressive fluid administration if necessary to restore
a given case extremely important, because the prognosis is depleted intravascular volume in cases of fulminant
considerably better in postobstructive pulmonary edema pulmonary edema
than, for example, in aspiration of gastric content
or other forms of acute respiratory distress syndrome CPAP, continuous positive airway pressure; PEEP, positive-end
(ARDS). Therapy, for the most part, is simplified, because expiratory pressure.
90 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
HISTORICAL
What Were the Outcomes?
CONSIDERATIONS
Concepts involving treatment of atelectasis by the ap- Unfortunately, despite the apparent logic to this approach,
plication of mechanical ventilation trace their origins survival in ARDS remained disturbingly low, and com-
to Bendixen et al.2830 and their investigations of anes- plications that included pulmonary barotrauma (later
thetized patients in the 1960s. They described gradu- termed volutrauma) seemed unduly high. Alternative
ally increasing degrees of hypoxemia in spontaneously means of support that included high frequency ventila-
breathing patients undergoing surgical procedures and tion and extracorporeal membrane oxygenation (ECMO)
attributed the change to miliary atelectasis that oc- were introduced, but the outcome in adults was no better
curred as a result of monotonous low tidal volume than the conventional approach.34,35 By 1986, with the
(VT) breathing. Mechanical ventilators were rare in most publication of a large scale study that was an offshoot
operating theaters at that time, and patients were often of the original ECMO study,34 Bartlett et al.36 painted
allowed to breathe spontaneously during the adminis- a dismal view of mortality outcome (40%, if only the
tration of general anesthesia. Because of the respiratory lungs were involved, and up to 100% in patients older
depressant effects of many anesthetic agents or adju- than 65 years with four or more organ systems involved).
vants, the placement of retractors and packing in the PEEP and CPAP, although effective in improving oxygena-
abdominal cavity that impaired diaphragmatic excursion, tion, were thought by many clinicians to be ineffective in
and the lack of airway humidification that contributed decreasing mortality (although their efficacy in hyaline
to inspissation of secretions, predisposition to atelectasis membrane disease was unquestioned). Whatever the rea-
was common. Although later work showed that other sons, therapy designed to reverse atelectasis in ARDS, as
factors, including alterations in position and changes it was applied by the end of the 1980s, was not getting the
in diaphragmatic mechanics,30 were as important, the job done.
C H A P T E R 5 / R E S P I R AT O R Y C O M P L I C AT I O N S O V E R V I E W 91
LOW TIDAL VOLUME After enrollment of 861 patients, the trial was
stopped because mortality was significantly lower in
MECHANICAL VENTILATION the group treated with lower VT than in the group
Investigators began to look for other approaches,3742 and treated with traditional VT. Furthermore, the number
as so often is the case, by the mid to late 1990s, the pen- of days without ventilator use during the first 28 days
dulum began to swing the other way toward lower VT following randomization was greater in this group. The
and a renewed interest in the rationale for PEEP/CPAP. investigators concluded that, in patients with acute lung
The culmination of this new look was the publication injury and ARDS, mechanical ventilation with a lower
of a prospective, randomized, multi-institutional study VT than had been used traditionally since the mid-1960s
of mechanical ventilation in ARDS.43 This study com- resulted in decreased mortality and increased the number
pared traditional mechanical ventilation (initial VT 12 mL of days without ventilator use. Of particular interest, these
per kg and airway plateau pressure 50 cm H2 O) with benefits occurred despite higher requirements for PEEP
ventilation with a lower VT (initial VT 6 mL per kg and and FIo2 and the lower ratio of PaO2 to FIo2 in the group
plateau pressure <30 cm H2 O). treated with lower VT on days 1 and 3. Additionally, it was
ARDS, acute respiratory distress syndrome; DIC, disseminated intravascular coagulation; CHF, congestive heart
failure.
Source: Data from: Langevin PB. Pulmonary consultation. In: Kirby RR, Gravenstein N, Lobato EB, et al. eds. Clinical
anesthesia practice, 2nd ed. Philadelphia, PA: WB Saunders; 2002:148.
92 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
also noted that greater reductions of plasma interleukin-6 present with a mixed disorder (e.g., chronic obstructive
occurred in the low VT group, suggesting that this group lung disease [COLD] and restrictive lung disease). How-
had less lung inflammation. The investigators suggested ever, if the anesthesia providers know ahead of time that,
that greater reductions in plasma interleukin-6 may also whatever the values may be, they will not alter their plans;
reflect a reduced systemic inflammatory response to lung such consultations and tests are a waste of time and
injury, which could contribute to the higher number money.
of days without organ or system failure and the lower Spirometry, a component of PFTs, is a noninvasive
mortality in the group treated with lower VT. test that can be used to examine static lung volumes
On the basis of several studies,3743 lung injury and dynamic gas flows (respiratory mechanics). Static
appears related to repeated opening and closing of small volumes can be summed in various ways to give defined
airways or from excessive stress at the margins between capacities. The volume of air moving past the lips can be
aerated and atelectatic lung regions. Low VT with PEEP measured over time, and flow can be evaluated against
strategies attempt to keep the lungs open.42 Investigators volume to give rise to a flow volume loop. Clearly these
feel that lung injury appears preventable by the judicious measurements are dependent upon the patients effort
application of higher PEEP. This approach is thought to and cooperation in performing the tests being assessed. A
keep the alveoli open throughout the entire respiratory low value that suggests reduced functional capacity may
cycle, without the need for large VT that is associated with only reflect suboptimal effort. In an attempt to overcome
stretch injury and inflammation. If so, one might suppose this bias, effort-independent measurements (e.g., forced
that improved clinical outcomes could be seen in patients expiratory flow between 25% and 75% of vital capacity
with acute lung injury and ARDS. [FEF25 75 ]) can also be made.
In summary, anesthesia providers have come full
circle. Initially, low VT spontaneous breathing was felt Lung Disease
to predispose to atelectasis in the 1960s. Replacement
with large VT mechanical ventilation in the 1970s and On the basis of these measurements, lung diseases may
1980s was designed to prevent or treat atelectasis, be classified as obstructive or restrictive (see Table 5.6).47
but in the process damaged the lungs and led to Although most patients suffer from obstructive disease,
systemic complications. Most recently low VT mechanical restrictive diseases encompass a greater diversity of disor-
ventilation with PEEP/CPAP has been advocated to ders.48 Conditions that increase the risk of perioperative
ameliorate or prevent lung damage, on the one hand, pulmonary complications are listed in Table 5.7.
and prevent atelectasis on the other.
The debate, however, goes on in the related, but not
identical, area of one-lung ventilation.44,45 In thoracic Obstructive
surgery, this approach minimizes hypoxemia and has
Obstructive diseases are characterized by abnormally slow
reduced morbidity and mortality. More recently, the
alveolar emptying. Air trapping (incomplete expiration
low VT approach has been advocated for many of
before the next breath) results, and increased static
the same reasons as it was for conventional two-lung
volumes (e.g., residual volume [RV], total lung volume)
ventilation. Advocates for this approach44 suggest that
and capacities (e.g., functional residual capacity) develop.
lung-protective, low VT with the selective use of PEEP
Obstructive diseases also reduce the forced expiratory
is a logical choice for one-lung ventilation in an attempt
volume in 1 second (FEV1 ). The forced vital capacity
to prevent parenchymal lung injury. Those not favoring
(FVC) tends to be preserved, but it takes longer to exhale
the technique45 state that lung-protective strategies are
the volume. Hence, the FEV1 /FVC ratio is reduced in
unproven in the operating room, and that PEEP may result
obstructive disorders, and this depression is considered
in VT close to those associated with conventional two-lung
the hallmark of obstructive lung disease.
ventilation. If this supposition is true, the advantages of
One legitimate reason to perform preoperative PFTs is
one-lung ventilation may be theoretical, not real.
to see whether obstructive ventilatory disorders may have
a reversible component. Such knowledge can help the
anesthesia provider to anticipate and deal with possible
Which Patients Can Benefit
from Preoperative Pulmonary TABLE 5.7 Conditions that Increase the Risk of
Perioperative Pulmonary Complications
Consultation?
Thoracic and upper abdominal surgery
Patients who do not have symptoms that limit their activ- Age >70 years
ity are unlikely to benefit from a pulmonary consultation Morbid obesity
or pulmonary function tests (PFTs). Tests may confirm a Smoking
presumptive diagnosis, document the progress of a dis- Preexisting pulmonary disease
ease, and evaluate the results of therapeutic intervention.
Together with arterial blood gas analysis, PFTs may be From: Kirby RR. Respiratory system. In: Gravenstein N, ed. Manual
the least expensive method of evaluating suspected lung of complications during anesthesia. Philadelphia, PA: JB Lippincott
disease46 and may be particularly useful when patients Co; 1991:304.
C H A P T E R 5 / R E S P I R AT O R Y C O M P L I C AT I O N S O V E R V I E W 93
TABLE 5.8 Preoperative Treatment Regimen for Patients TABLE 5.10 Staging of Chronic Obstructive Lung Disease
with Chronic Obstructive Lung Disease
Stage I: FEV1 50% of predicted flow, and the disease
Smoking cessation 34 weeks before surgery minimally affects the patients life. Most patients with
Bronchodilator drugs with emphasis on -selective COLD have stage I disease
agents Stage II: FEV1 = 35%49% of predicted flow. A minority
Antibiotic therapy if infection is present of patients are involved at this stage, but it has a significant
Chest physiotherapy and bronchial drainage. Inter- impact on the quality of life
mittent positive-pressure breathing treatment is not Stage III: FEV1 35% of predicted flow and is associated
indicated with severely limited reserve and severe symptomsa
Delay of elective surgery in unusual cases if optimal
a
status is not achieved Bendixen HH, Hedley-Whyte J, Laver MB. Impaired oxygenation
in surgical patients during general anesthesia with controlled
From: Kirby RR. Respiratory system. In: Gravenstein N, ed. Manual ventilation: A concept of atelectasis. N Engl J Med. 1963;269:991.
of complications during anesthesia. Philadelphia, PA: JB Lippincott COLD, chronic obstructive lung disease.
Co; 1991:304.
FEV1 , forced expiratory volume in 1 s; FEF25 75 , forced expiratory flow between 25% and 75% of vital capacity;
FVC, forced vital capacity.
From: Langevin PB. Pulmonary consultation. In: Kirby RR, Gravenstein N, Lobato EB, et al. eds. Clinical anesthesia
practice, 2nd ed. Philadelphia, PA: WB Saunders; 2002:148.
94 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 5.11 Risk Factors for Postoperative Complications 4. Although most cases resolve quickly, death from
in Anesthetized Patients ARDS and multisystem organ failure has been
reported despite relief of the airway obstruction.
Position 5. Atelectasis occurs when unopposed elastic recoil of
Anesthetic agents the lungs leads to a decrease in lung volume, small
Muscle relaxants (inadequate reversal) airways collapse follows, and gas trapped in the
Positive-pressure ventilation alveoli is resorbed into the pulmonary blood.
Pulmonary aspiration of gastric and other contents 6. Large VT ventilation alone does not resolve the
Bronchospasm and airway edema problem of atelectasis; addition of PEEP/CPAP often
Embolic events (thromboembolism, fat/bone marrow, does.
air, carbon dioxide) 7. Low VT mechanical ventilation with PEEP/CPAP
Pulmonary barotrauma has been advocated to ameliorate or prevent lung
Fluid volume overload damage, on the one hand, and prevent atelectasis on
Atelectasis the other.
8. Lung-protective strategies, although popular in the
From: Kirby RR. Respiratory system. In: Gravenstein N, ed. Manual ICU for the treatment of ARDS, are unproven in the
of complications during anesthesia. Philadelphia, PA: JB Lippincott operating room.
Co; 1991:306. 9. PFTs may be the least expensive method of evaluat-
ing lung disease particularly when patients present
with mixed COLD and restrictive lung disease.
All patients are at risk for postoperative respiratory 10. All patients are at risk for postoperative respira-
complications following upper abdominal and thoracic tory complications following upper abdominal and
surgery.7,5052 No standardized staging system for COLD thoracic surgery.
exists.53,54 However, the FEV1 correlates directly with 11. Anesthesia providers should always be cognizant
morbidity and mortality. On the basis of this relationship, of their possible role in postoperative and post-
the American Thoracic Society established criteria for anesthetic pulmonary complications.
three stages of disease55 (see Table 5.10).
In patients with more than Stage I disease, mortal-
ity may reach 5%, and morbidity may exceed 80%,55 REFERENCES
primarily because ventilation becomes dependent on
1. Bartlett R, Bremman ML, Gazzaniga AB, et al. Studies on
the accessory muscles of respiration instead of the di-
the pathogenesis and prevention of postoperative pulmonary
aphragm. This observation is not attributable to pain
complications. Surg Gynecol Obstet. 1973;137:925.
but may explain why deep breathing, incentive spirom- 2. Garibaldi RA, Britt MR, Coleman ML, et al. Risk factors for
etry, or intermittent positive-pressure breathing fails to postoperative pneumonia. Am J Med. 1981;70:677.
completely eliminate postoperative pulmonary complica- 3. Pasteur W. Active lobar collapse of the lung after abdominal
tions. operations: A contribution to the study of post-operative lung
The physiological consequences of surgery are at- complications. Lancet. 1910;2:1080.
tributable to both surgery and anesthesia.50,51 The com- 4. Busch E, Verazin G, Antkowiak JG, et al. Pulmonary
plication rate for upper abdominal procedures is so high complications in patients undergoing thoracotomy for lung
in patients with Stage II and Stage III disease that surgery carcinoma. Chest. 1994;105:760.
that is not imperative should be avoided.55 When surgery 5. Kroehke K, Lawrence VA, Theroux JF, et al. Operative risk
must be performed, laparoscopic alternatives should be in patients with severe obstructive pulmonary disease. Arch
Intern Med. 1993;152:967.
considered56 and anesthesia tailored to the patient.50,55
6. Stein M, Cassara EL. Preoperative pulmonary evaluation and
Anesthesia providers should always be cognizant of their
therapy for surgery patients. JAMA. 1970;211:787.
possible role in postoperative and postanesthetic pul-
7. Zibrak JD, ODonnell CR. Indication for preoperative pul-
monary complications in patients with or without COLD monary function testing. Clin Chest Med. 1993;14:227.
(see Table 5.11). 8. Mohr DN. Estimation of surgical risk in the elderly: A
correlative review. J Am Geriatr Soc. 1983;31:99.
9. Warner MA, Divertie MB, Tinker JH. Preoperative cessation
of smoking and pulmonary complications in coronary artery
bypass patients. Anesthesiology. 1984;60:380.
KEY POINTS 10. Gracey DR, Divertie MB, Didier EP. Preoperative pulmonary
1. Whenever possible, the identification of patients at preparation of patients with chronic obstructive pulmonary
disease. Chest. 1979;76:123.
risk for pulmonary problems is imperative before
11. Tarhan S, Mottitt E, Sessler AD, et al. The risk of anesthesia
surgery.
and surgery in patients with chronic bronchitis and chronic
2. Postobstructive pulmonary edema has been reported
obstructive pulmonary disease. Surgery. 1973;74:720.
in the pediatric and adult published literature for 12. Wong DH, Weber EC, Schell MJ, et al. Factors asso-
more than 40 years. ciated with postoperative complications in patients with
3. Partial or total airway obstruction is the common severe chronic obstructive pulmonary disease. Anesth Analg.
factor in postobstructive pulmonary edema. 1995;80:276.
C H A P T E R 5 / R E S P I R AT O R Y C O M P L I C AT I O N S O V E R V I E W 95
13. Carr HD, Stevens PM, Adamiya R. Preoperative pulmonary 35. Carlon GC, Howland WS, Ray C, et al. High frequency jet
function and complications after cardiovascular surgery. ventilation. Chest. 1983;84:551.
Chest. 1979;80:276. 36. Bartlett RH, Morris AH, Fairley HB, et al. A prospective study
14. Celli BR, Rodriguez K, Snider GL. A controlled trial of inter- of acute hypoxic respiratory failure. Chest. 1986;89:684.
mittent positive pressure breathing, incentive spirometry and 37. Albert RK. Least PEEP: Primum non nocere. Chest. 1985;
deep breathing exercises in preventing pulmonary complica- 87:2.
tions after abdominal surgery. Am J Respir Dis. 1984;130:12. 38. Hickling KG, Walsh J, Henderson S, et al. Low mortality rate
15. Yeager MP, Glass DD, Neff RK, et al. Epidural anesthesia in adult respiratory distress syndrome using low-volume,
and analgesia in high-risk surgical patients. Anesthesiology. pressure-limited ventilation with permissive hypercapnia.
1987;66:729. Crit Care Med. 1994;22:1568.
16. Mitchell CK, Smoger SH, Pfeifer MP, et al. Multivariate 39. Amato MBP, Barbas CSV, Medeiros DM, et al. Effect of
analysis of factors associated with postoperative pulmonary a protective ventilation strategy on mortality in the acute
complications following general elective surgery. Arch Surg. respiratory distress syndrome. N Engl J Med. 1998;338:
1998;133:194. 347.
17. Luke MJ, Mehrizi A, Folger GM Jr, et al. Chronic naso- 40. Weg JG, Anzueto A, Balk R, et al. The relation of pneumo-
phyaryngeal obstruction as a cause of cardiomegaly, cor thorax and other air leaks to mortality in the acute
pulmonale, and pulmonary edema. Pediatrics. 1966;37:762. respiratory distress syndrome. N Engl J Med. 1998;338:341.
18. Capitanio MA, Kirkpatrick JA. Obstructions of the upper 41. Stewart TE, Meade MO, Cook DJ, et al. Evaluation of a
airway in children as reflected on the chest radiograph. ventilation strategy to prevent barotrauma in patients at
Radiology. 1973;107:159. high risk for acute respiratory distress syndrome. N Engl J
19. Oswalt CE, Gates GA, Holmstrom MG. Pulmonary edema Med. 1998;338:355.
as a complication of acute airway obstruction. JAMA. 42. Lachmann B. Open up the lung and keep the lung open.
1977;238:1833. Intensive Care Med. 1992;18:319.
20. Lee A, Fan LTY, Gin T, et al. A systematic review (meta- 43. The Acute Respiratory Distress Syndrome Network. Ventila-
analysis) of the accuracy of the Mallampati tests to predict tion with lower tidal volumes as compared with traditional
the difficult airway. Anesth Analg. 2006;102:1867. tidal volumes for acute lung injury and the acute respiratory
21. McConkey PP. Postobstructive pulmonary oedema: A case distress syndrome. N Engl J Med. 2000;342:1301.
series and review. Anaesth Intensive Care. 2000;28:72. 44. Slinger P. Pro: Low tidal volume is indicated during one-lung
22. Adolph MD, Oliver AM, Dejak T. Death from adult respiratory ventilation. Anesth Analg. 2006;103:268.
distress syndrome and multiorgan failure following acute 45. Gal TJ. Con: Low tidal volumes are indicated during one-lung
upper airway obstruction. Ear Nose Throat J. 1994;73:324. ventilation. Anesth Analg. 2006;103:271.
23. McConkey PP. Airway bleeding in negative-pressure pul- 46. Tisi GM. Preoperative evaluation of pulmonary function,
monary edema. Anesthesiology. 2001;95:272. validity, indications, and benefits. Am Rev Respir Dis.
24. Sow Nam Y, Garewal D. Pulmonary hemorrhage in associ- 1979;119:293.
ation with negative pressure edema in an intubated patient. 47. Langevin PB. Pulmonary consultation. In: Kirby RR, Graven-
Acta Anaesthesiol Scand. 2001;45:911. stein N, Lobato EB, et al. eds. Clinical anesthesia practice,
25. Devys JM, Balleau C, Jayr C, et al. Biting the laryngeal mask: 2nd ed. Philadelphia: WB Saunders; 2002:145.
An unusual cause of negative pressure pulmonary edema. 48. Kipp VJ, Arora SK, Boysen PG. Pulmonary consultation.
Can J Anaesth. 2000;47:176. In: Murray MJ, Coursin DB, Pearl RC, et al. eds. Criti-
26. Liu EH, Yih PS. Negative pressure pulmonary oedema cal care medicine perioperative management. Philadelphia:
caused by biting and endotracheal tube occlusiona case Lippincott-Raven; 1997:400.
for oropharyngeal airways. Singapore Med J. 1999;40:174. 49. Kroenke K, Lawrence VA, Theroux JF, et al. Postoperative
27. Nunn JF. Pulmonary collapse and atelectasis. Applied respi- complications after thoracic and major abdominal surgery
ratory physiology, 3rd ed. London: Butterworths; 1987:440. in patients with and without obstructive lung disease. Chest.
28. Bendixen HH, Hedley-Whyte J, Laver MB. Impaired oxy- 1993;104:1445.
genation in surgical patients during general anesthesia with 50. Ford GT, Rosenal TW, Clergue F, et al. Respiratory physiol-
controlled ventilation: A concept of atelectasis. N Engl J Med. ogy in upper abdominal surgery. Clin Chest Med. 1993;14:237.
1963;269:991. 51. Celli BR. Perioperative respiratory care of the patient
29. Bendixen HH, Bullwinkel B, Hedley-Whyte J. Atelectasis undergoing upper abdominal Surgery. Clin Chest Med.
and shunting during spontaneous ventilation in anesthesia 1993;14:253.
patients. Anesthesiology. 1964;25:297. 52. Merli GJ, Weitz HH. Approaching the surgical patient. Role
30. Froese AB, Bryan AC. Effects of anesthesia and paraly- of the medical consultant. Clin Chest Med. 1993;14:205.
sis on diaphragmatic mechanics in man. Anesthesiology. 53. Sweer L, Zwillich CW. Dyspnea in the patient with chronic
1974;41:242. pulmonary disease. Etiology and management. Clin Chest
31. Pontoppidan H, Geffin B, Lowenstein E. Acute respiratory Med. 1990;11:417.
failure in the adult. N Engl J Med. 1972;287:743. 54. Mahler DA, Weinberg DH, Wills CK, et al. The measure-
32. Gregory GA, Kitterman JA, Phibbs RH, et al. Treatment of ment of dyspnea. Contents, interobserver agreement, and
idiopathic respiratory distress syndrome with continuous physiologic correlates of two new clinical indexes. Chest.
positive airway pressure. N Engl J Med. 1971;284:1333. 1984;85:751.
33. Ashbaugh DG, Petty TL, Bigelow DB, et al. Continuous 55. American Thoracic Society. Standards for the diagnosis and
positive pressure breathing (CPPB) in adult respiratory care of patients with chronic obstructive pulmonary disease.
distress syndrome. J Thorac Cardiovasc Surg. 1969;57:31. Am J Respir Crit Care Med. 1995;152(5 Pt 2):S77.
34. Zapol WM, Snider MT, Hill JD, et al. Extracorporeal 56. MacFayden BV Jr, Wolfe BM, McKernan JB. Laparoscopic
membrane oxygenation in severe acute respiratory failure: A management of the acute abdomen, appendix, and small and
randomized prospective study. JAMA. 1979;242:2193. large bowel. Surg Clin North Am. 1992;72:1169.
CHAPTER THE DIFFICULT AIRWAY
CASE SUMMARY these claims, the standard of care was not performed. The
incidence of affected anatomical structures is shown in
38-year-old, 96 kg woman presents to the Table 6.1. Factors such as female gender, elective surgery,
A
emergency room complaining of a severe and outpatient procedures showed a higher proportion
sore throat, especially when swallowing. of injury, whereas there was no difference regarding ASA
Earlier that day, the patient had been ad- status or obesity.
mitted to the hospital for a pelvic exam-
ination under anesthesia and a diagnostic
laparoscopy. During routine intravenous induction, the
esophagus was inadvertently intubated; however, this was
detected immediately, and the endotracheal tube was re-
What Complications Can Be
moved. A second attempt at tracheal intubation was Seen when Supraglottic Devices
performed, but no end-tidal CO2 was detected by capnog- Are Used?
raphy. Bronchospasm was suspected and the endotracheal
tube was removed. After increasing the depth of anesthe-
sia, the trachea was then successfully intubated and the
correct position of the tube was confirmed. The remain- MASK VENTILATION
der of the anesthetic course and the operation itself were
uneventful. The maximum risk during airway management presents
During the patients visit to the emergency room, during the cannot intubate, cannot ventilate situation.2,3
an upper gastrointestinal GI series with gastrografin Difficult mask ventilation is an underestimated aspect of
is performed, which reveals an esophageal tear with the difficult airway. The ability to ventilate and oxygenate
leakage of contrast from the lower cervical esophagus, the patient sufficiently using a mask is essential. Face
posterolaterally, into the mediastinum. The patient is masks should be completely free of residual cleansing
admitted to the intensive care unit for observation agents, because these can cause serious mucosal, skin or
and conservative therapy. Subsequently, she develops eye injury (conjunctivitis, burning, irritation), and tongue
mediastinitis and undergoes a flexible bronchoscopy swelling (allergic glossitis).
with neck exploration and right hemithyroidectomy. While applying a mask to a patients face, soft tissue
The patient also requires a right-sided thoracotomy for damage may result if the tissue is subjected to excessive
drainage of the fluid collection from the mediastinum. pressure. Care must be taken to avoid contact with the eyes
Finally, a gastric tube and jejunostomy tube are placed to prevent corneal abrasions, retinal artery occlusions,
for chronic enteral feeding. Later, the patient required or blindness. Excessive pressure on the mandible may
placement of a drainage catheter for interventional damage the mandibular branch of the facial nerve,
radiology for treatment of a loculated pleural effusion. resulting in transient facial nerve paralysis. Pressure
Ultimately, the patient is discharged home with a gastric on the mental nerves has been implicated in causing
tube and jejunostomy tube and followup appointments. lower lip numbness. Oropharyngeal airways must be
Difficulty in managing the airway is the most im- gently inserted into the mouth to avoid injury (broken
portant cause of major anesthesia-related morbidity and teeth or mucosal tears). Improper placement can worsen
mortality. In the closed claims analysis of the American airway obstruction by forcing the tongue backward. Equal
Society of Anesthesiologists (ASA), 6% of all claims in- care should be given to the placement of nasopharyngeal
volved airway injury.1 Difficult intubation was a factor in airways to avoid epistaxis.
only 39% of airway injury claims; 87% of the airway in- During the course of induction, the lifting pres-
juries were temporary; and 8% resulted in death. In 21% of sure applied to the angle of the mandible during mask
96
C H A P T E R 6 / T H E D I F F I C U LT A I R W AY 97
Site of Injury Nondeath n (%) Death n (%) Standard n (%) Substandard n (%)
Larynx (n = 87) 86 (99) 1 (1) 74 (96) 3 (4)
Pharynx (n = 51) 46 (90) 5 (10) 29 (71) 12 (29)
Esophagus (n = 48) 39 (81) 9 (19) 25 (60) 17 (40)
Trachea (n = 39) 33 (85) 6 (15) 20 (63) 12 (38)
TMJ (n = 27) 27 (100) 0 21 (100) 0
Nose (n = 13) 13 (100) 0 11 (85) 2 (15)
ventilation is sometimes sufficient to subluxate the known disadvantage of this device is its inability to protect
temporomandibular joint. Patients may experience per- against pulmonary aspiration and regurgitation of gastric
sistent pain or bruising at these points, and can even contents. The incidence of regurgitation of small amounts
have chronic dislocation of the jaw, leading to severe of gastric contents was reported to be as high as 25%.5
discomfort. However, the overall risk of aspiration and regurgitation
Positive airway pressure can force air into the using the LMA is in the same low range as for endotracheal
stomach instead of the trachea, and therefore gastric intubation when the indications and contraindications of
distention may result, causing more difficult ventilation LMA usage are respected.6
and an increased propensity for regurgitation. For these Laryngospasm and coughing may result from inad-
reasons, mask ventilation should not be performed in the equate anesthesia, tip impaction against the glottis, or
following patients, unless necessary: aspiration. The incidence of sore throat is reported to
be 7% to 12%, an incidence similar to that seen with
Nonfasted patients
oral airways.7 The incidence of failed placement is 1% to
Morbidly obese patients
5%, although this tends to decrease with increasing op-
Patients with intestinal obstruction
erator experience. The LMA cuff is permeable to nitrous
Patients in the Trendelenberg position
oxide and carbon dioxide, which results in substantial
Patients with a tracheoesophageal fistula
increases in cuff pressure and volume during prolonged
Patients with massive oropharyngeal bleeding
procedures. Elevated intracuff pressures may increase the
Cricoid pressure can help reduce the amount of air incidence of postoperative sore throat or cause transient
being forced into the stomach and limit the likelihood of dysarthria. In addition, edema of the epiglottis, uvula,
vomiting. Nonetheless, gastric rupture has been reported and posterior pharyngeal wall can lead to airway ob-
with face mask ventilation. struction. Hypoglossal nerve paralysis, post obstructive
Recently, it was shown that independent risk factors pulmonary edema, tongue cyanosis, transient dysarthria,
for difficulties with mask ventilation include the presence tension pneumoperitoneum, and gastric rupture have also
of a beard, body mass index >26 kg per m2 , lack of been reported.
teeth, age >55 years, and history of snoring.4 Patients To minimize the risk of aspiration and regurgitation,
with trauma to the pharyngeal mucosa may be at the LMA-Proseala laryngeal mask with an esophageal
risk for subcutaneous emphysema. Pneumocephalus is a ventwas developed.8 Cases of gastric insufflation and
possibility whenever continuous positive airway pressure aspiration have been reported when this device was
is applied to patients with basilar skull fractures. malpositioned.9 Branthwaite reported a case of laryngeal
perforation leading to mediastinitis and patient death
following the blind insertion of an endotracheal tube
through the intubating LMA.10
LARYNGEAL MASK AIRWAY
The laryngeal mask airway (LMA) has been used in mil- Contraindications
lions of patients and is accepted as a relatively safe tech-
nique. Muscle relaxation is unnecessary, laryngoscopy is Contraindications for using an LMA include nonfasted
circumvented, and hemodynamic changes are minimized patients, morbid obesity, need for high inspiratory pres-
during insertion. Nevertheless, numerous complications sures (>20 to 25 cm H2 O) in the presence of low pul-
are associated with the LMA. The tip of the epiglottis can monary compliance or chronic obstructive pulmonary
be folded into the vocal cords during placement, which disease (COPD), acute abdomen, hiatal hernia, severe
can induce labored breathing, coughing, laryngospasm, gastroesophageal reflux, Zenkers diverticulum, trauma,
and sometimes complete airway obstruction. Excess lu- intoxication, airway problems at the glottic or infraglottic
bricant can promote coughing or laryngospasm. A well level, and thoracic trauma.
98 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
emergence, and 5% associated with regional anesthesia. incidence in this study was 41%.19 Fortunately, pain on
Dental trauma is also associated with LMA devices and swallowing usually lasts no more than 24 to 48 hours.
oropharyngeal airways. These injuries are most common Topical anesthesia, such as lidocaine jelly, applied to the
in small children, patients with periodontal disease or endotracheal tube does not lessen the incidence of this
fixed dental work, and patients in whom intubation is problem and may actually worsen it.
difficult. Preexisting dental pathology (protrusion of the
upper incisors, carious teeth, paradentosis or periodon- Trauma to the Larynx and Vocal Cords
titis) should be thoroughly explored before induction of
anesthesia, and the patient must be advised of the risk Trauma to the larynx and vocal cords is not uncommon
of dental damage. Although tooth guards may obstruct following endotracheal intubation. Whether it occurs
vision, their use is indicated in certain situations. depends on the experience and skill of the intubator,
In the event that an entire tooth is avulsed, it as well as the degree of difficulty. In one large study,
should be retrieved and saved in a moist gauze or in 6.2% of patients sustained severe lesions, 4.5% developed
normal saline. Aspiration of the tooth can induce serious a hematoma of the vocal cords, 1% developed a hematoma
complications requiring bronchoscopy for removal. With of the supraglottic region, and 1% sustained lacerations
a rapid response from an oral surgeon or dentist, an intact and scars of the vocal cord mucosa.20 Recovery is generally
tooth can often be reimplanted and saved, but only when prompt with conservative therapy, although hoarseness
performed within 1 hour. may appear even after a 2-week interval.21 Granulations
usually occur as a complication of long-term intubation
Macroglossia but can also be a result of short-term intubation. Injuries
of the laryngeal muscles and suspensory ligaments are also
Massive tongue swelling, or macroglossia, has been re- possible. Patients with hoarseness should be examined
ported in numerous instances in both adult and pediatric preoperatively by an ENT specialist.
patients. Although macroglossia (occasionally of life-
threatening proportions) is associated with angiotensin- Arytenoid Dislocation and Subluxation
converting enzyme inhibitors, some cases have occurred
Arytenoid dislocation and subluxation have been reported
while a bite block was in place and when there was sub-
as rare complications.22 Mitigating factors include trau-
stantial neck flexion during endotracheal intubation. Loss
matic and difficult intubations, repeated attempts at intu-
of tongue sensation is possible after a compression in-
bation, and attempted intubation using blind techniques
jury to the lingual nerve during forceful laryngoscopy or
such as light-guided intubation, retrograde intubation,
after LMA placement with an overinflated or malposi-
and the use of the McCoy laryngoscope (Penlon Lim-
tioned cuff. A reduced sense of taste and cyanosis of the
ited, Abingdon, UK). However, these complications are
tongue caused by lingual artery compression are addi-
also found after easy intubations. Early diagnosis and
tional injuries that can be incurred by using an oversized,
operative repositioning of arytenoid dislocation is nec-
malpositioned, or overinflated LMA.
essary, because fibrosation with consecutive malposition
and ankylosis can occur after 48 hours.
Damage to the Uvula
Damage to the uvula (edema and necrosis) is usually
Vocal Cord Paralysis
associated with an endotracheal tube, oropharyngeal Numerous investigators have reported vocal cord paralysis
and nasopharyngeal airways, an LMA, an alternative after intubation with no obvious source of injury. Paraly-
supraglottic airway device, or by overzealous use of sis may be unilateral (hoarseness) or bilateral (respiratory
a suction catheter. Sore throat, odynophagia, painful obstruction). The most likely source of injury is a mal-
swallowing, coughing, foreign body sensation, and serious positioned endotracheal tube cuff in the subglottic larynx
life-threatening airway obstruction have been reported. which exerts pressure on the recurrent laryngeal nerve.
Permanent voice change due to external laryngeal nerve
Sore Throat trauma following intubation results in up to 3% of pa-
tients undergoing surgery in sites other than the head or
The incidence of sore throat after intubation is approxi- neck. However, vocal cord paralysis after intubation is usu-
mately 40% to >65% when blood is noted on the airway ally temporary. Its incidence can be decreased by avoiding
instruments.17 The incidence of sore throat following LMA overinflation of the endotracheal tube cuff and by plac-
placement is 20% to 42%, depending on cuff inflation, ing the endotracheal tube at least 15 mm below the vocal
and 8% with face mask ventilation.18 Additionally, when cords. Vocal cord paralysis may also have a central origin.
comparing a double-lumen tube with an endobronchial Eroded vocal cords can adhere together, eventually form-
blocker, Knoll et al., determined that significant post- ing synechiae. Surgical correction is usually necessary.
operative hoarseness occurred more frequently in the
double-lumen group: 44% versus 17%, respectively. The Tracheobronchial Trauma
cumulative number of days with hoarseness and sore
throat were significantly increased in the double-lumen Tracheobronchial trauma has various causes. Injury can
group compared with the blocker group. Sore throat did result from an overinflated endotracheal tube cuff, inad-
not differ significantly between groups, but the overall equate tube size, malpositioned tube tip, laryngoscope,
100 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
stylet, tube exchanger, or related equipment. Predispos- difficult intubation may cause anesthesia of the upper
ing factors include anatomic difficulties, blind or hurried surface of the larynx. Transient palsies may result when
intubation, inadequate positioning, poor visualization, or, an LMA device is used because it affects the hypoglossal
most commonly, an inexperienced intubator. Edema af- and lingual nerves. The authors personally observed five
ter extubation decreases the lumen diameter and increases cases of hyposmia following uncomplicated nasotracheal
airway resistance. Small children are most susceptible to intubation for head and neck surgery, and one case
this problem; almost 4% of children within the age group of anosmia despite the use of preformed, warmed, and
of 1 to 3 years develop croup following endotracheal lubricated nasotracheal tubes. The hyposmias completely
intubation. Tracheal rupture, especially after emergency recovered in 3 to 6 months, whereas the anosmia became
intubation, has been reported, as well as a bronchial permanent.
rupture secondary to the use of an endotracheal tube
exchanger.23
Endotracheal tube cuffs inflated to a pressure greater
Cervical Spine Injury
than that of the capillary perfusion may devitalize the Airway management techniques such as chin lift, jaw
tracheal mucosa and lead to ulceration, necrosis, and loss thrust, and direct laryngoscopy transmit movement to
of structural integrity. Ulceration can result at even lower the cervical spine may injure the cervical spine. Attempts
pressures in hypotensive patients. The need for increasing to hyperextend the neck of a patient with ankylosing
cuff volumes to maintain a seal is an ominous sign of de- spondylitis can produce cervical fractures and quadriple-
veloping tracheomalacia. The various nerves in this region gia. Special attention should be given to patients with C1
of the neck are also at risk. Erosion of the endotracheal
or C2 fractures because any degree of extension might
tube into the paratracheal nerves can produce dysphonia,
compromise spinal cord function.
hoarseness, and laryngeal incompetence. Tracheomalacia
Several conditions, such as Down syndrome, Arnold-
results from erosion confined to the tracheal cartilages.
Chiari malformation, and rheumatoid arthritis, are asso-
It is imperative that the anesthesiologist inflates the cuff
ciated with atlantoaxial instability. Also, elderly patients
of the endotracheal tube only as much as is necessary
and those with pathological fragility, such as connec-
to ensure an adequate airway seal. If using nitrous oxide
tive tissue disorders, lytic bone tumors and osteoporosis,
during a lengthy surgical procedure, the pressure in the
should be intubated with caution. Awake fiberoptic intu-
endotracheal tube cuff should be checked by a cuff pres-
bation should be considered in all cases where time is not
sure control device. The cuff pressure should not exceed
crucial.
25 cm H2 O.
The incidence of granulomas has been reported to
range from 1:80024 to 1:20,000.25 Endotracheal intubation Eye Injuries
prolonged for several months can produce tracheal
stenosis and fibrosis, typically at the site of an inflated Corneal abrasions are the most common eye compli-
cuff and sometimes at the location of the endotracheal cations that occur during general anesthesia. They are
tube tip. Dilation of the stenosis is curative if it is caught primarily caused by a facemask being placed on an open
in its early stages. However, surgical correction may be eye or by the eyelids not being completely closed dur-
necessary once the tracheal lumen has been reduced to 4 ing anesthesia. Prevention consists of vigilance on the
to 5 mm. part of the anesthesiologist and application of adhesive
Supraglottic complications induced by long-term tape over the closed eyelids, especially during head and
intubation may be prevented by early tracheostomy. There neck surgery. Although these injuries typically heal within
is no evidence concerning the ideal time for tracheostomy 24 hours, they are usually painful and can lead to corneal
in long-term ventilated patients. ulceration. An immediate ophthalmologic consultation is
recommended. In the presence of a penetrating eye injury,
an increase in intraocular pressure should be avoided by
Barotrauma
adequate anesthesia.
Barotrauma, which can lead to pneumomediastinum or
tension pneumothorax, results from high-pressure disten- Temporomandibular Joint Injury
tion of intrapulmonary structures. High-flow insufflation
techniques are most often associated with barotrauma. Temporomandibular joint injury (TMJ) is a rare but
Such problems are common in microlaryngeal surgery in serious complication. Rupture of the lateral ligament is
which jet ventilation is used. possible. These injuries are caused during laryngoscopy
when increasing force is used to optimize the view of
Nerve Injury the glottis. As a result, limited mouth opening, pain
in the TMJ, lateral deviation of the mandible (in case
Laryngoscopy and cuffed supraglottic airway devices can of unilateral luxation), protrusion of the mandible, and
produce periodical or permanent nerve injury. Transient lockjaw can occur. Most of the cases of TMJ injury have
weakness, numbness, or paralysis of the tongue can result not been associated with difficult airway management.26
after laryngoscopy, presumably because of pressure on In the ASA Closed Claims Database, only 17% of the
the laryngeal and hypoglossal nerves. Damage of the claims had documented, preexisting TMJ disorders, such
internal branch of the superior laryngeal nerve during as pain.27
C H A P T E R 6 / T H E D I F F I C U LT A I R W AY 101
large can produce bronchial trauma. The overall incidence Laser Fires
of bronchial injuries in a recent comparison between
double-lumen tubes and endobronchial blockers was 25%, Lasers are frequently used in the operating room to
although the incidence did not differ significantly between ablate benign and neoplastic tissues in the airway.
groups. Additionally, vocal cord injuries occurred more Laser fire is a very serious complication. The use of
frequently in the double-lumen group compared with the special laser-guarded or metal tubes is recommended,
blocker group: 44% versus 17%, respectively. Most vocal and all inflammatory materials such as dentures and
cord injuries demonstrated redness and edema, and one nasogastric tubes should be removed. One of the most
hematoma was noted in the double-lumen group.19 catastrophic events associated with their use is an
airway fire, which can occur when the laser ignites the
endotracheal tube. The heat and fumes of the burning
plastic can cause severe damage to the airway. In this
instance, the circuit must immediately be disconnected
MAINTENANCE OF THE from the endotracheal tube and the burning tube removed
ENDOTRACHEAL TUBE from the airway. The fire should be extinguished with
saline, and the patient should be supported by facemask
Airway Obstruction ventilation. The airway should be evaluated for damage
with bronchoscopy.
Airway obstruction is possible at any time during general Numerous precautions can reduce the risk of an air-
anesthesia, particularly in prolonged surgery or in pa- way fire. If possible, avoid the use of an endotracheal
tients with predisposing anatomic abnormalities. Airway tube by employing other measures (ventilating laryngo-
obstruction can result from diverse factors, including a scope, jet ventilation system, intermittent apneic ventila-
sharp bend or kink in the endotracheal tube or a tube that tion).33 Endotracheal tubes can be protected by wrapping
is obstructed with mucus, blood, foreign bodies, or lubri- them with noncombustible tapes; alternatively, red rub-
cant. Reinforced wire tubes avoid kinking, and therefore, ber or noncombustible, metal endotracheal tubes may
their use is recommended in prolonged procedures, oral be used. Cuff ignition can be minimized by filling the
surgery, or during surgery associated with special posi- cuff with saline solution instead of air. Nitrous oxide
tioning of the patient. Nitrous oxide can cause expansion should not be used in laser surgery because it supports
of gas bubbles trapped in the walls of an endotracheal combustion. It is recommended that inert gases, such
tube, leading to airway obstruction. as helium or nitrogen, be used instead of nitrous ox-
The cuff of an endotracheal tube can also cause airway ide, and that concentrations of oxygen do not exceed
obstruction. An overinflated cuff can compress the bevel 40%.
of the endotracheal tube against the tracheal wall, thereby
occluding its tip. The cuff can also herniate over the
tip of the endotracheal tube. When faced with any of
these problems, the best solution is to pass a suction SPECIAL TECHNIQUES
catheter or a fiberoptic bronchoscope down the lumen
of the endotracheal tube and attempt to clear it. If the Fiberoptic intubation is one of the most common meth-
endotracheal tube is totally obstructed, passage of a stylet ods utilized in cases of anticipated difficult intubation.
should be attempted. Total obstruction that cannot be Intubation with a fiberoptic bronchoscope should not be
remedied quickly requires the removal of the endotracheal attempted when the pharynx is filled with blood or saliva,
tube followed by reintubation. when inadequate space exists within the oral cavity, or
when time is critical and creating a surgical airway is a
priority. Relative contraindications include marked tissue
Disconnection of Endotracheal Tube from edema, distortion of the oropharyngeal anatomy, blood in
the airway, soft tissue traction, or a severe cervical flexion
Anesthesia Circuit deformity.
A common and serious complication of endotracheal Potential complications associated with fiberoptic
intubation is disconnection of the endotracheal tube bronchoscopy include bleeding, epistaxis (especially if
from the anesthesia circuit. This was identified as the a nasal airway is attempted), laryngotracheal trauma,
laryngospasm, bronchospasm, and aspiration of blood,
most common critical incident in a study of anesthesia-
saliva, or gastric contents. Another possible hazard is
related human errors and equipment failures.32 Alarms to
associated with the practice of insufflating oxygen through
signal airway disconnection are included on all modern
the suction channel. Subcutaneous emphysema of the
anesthesia machines.
pharynx, face, and periorbital regions may occur if the
pharyngeal mucosa is injured.
Leaks in Airway Delivery Circuit The lighted stylet can facilitate intubation under both
local and general anesthesia. Sore throat, hoarseness,
Leaks in an air delivery circuit can cause hypoventilation arytenoid subluxation, and mucosal damage are possible.
and dilution of the inspired gases by entry of room air With inappropriate handling of the stylet, heat damage to
into the system. the tracheal mucosa can also occur.
C H A P T E R 6 / T H E D I F F I C U LT A I R W AY 103
TRANSTRACHEAL AIRWAY
What Types of Complications
Are Associated with Infraglottic Transtracheal Jet Ventilation
Procedures? Transtracheal jet ventilation (TTJV) is accomplished by
introducing a small percutaneous catheter into the trachea
Infraglottic airway access is the last step in the ASA airway and insufflating the respiratory tract with high pressure
management algorithm.15 In cases in which endotracheal oxygen over a jet ventilator or a hand jet device. Although
intubation is impossible and the patients condition deteri- this technique may be helpful in critical situations, life-
orates into a cannot ventilate, cannot intubate situation, threatening problems are associated with its use.
lifesaving steps must be immediately undertaken. Despite If the TTJV catheter is displaced from the trachea,
possible (and severe) complications, there are no con- subcutaneous emphysema, hypoventilation, pneumome-
traindications for infraglottic procedures in these critical diastinum, pneumothorax, severe abdominal distention,
situations. The most severe complication is failure to or death may result. Oxygen delivered through a transtra-
establish an airway before brain damage or death results. cheal catheter must be able to escape the lungs freely;
otherwise, overdistention and pulmonary rupture will oc-
cur. In cases of total airway obstruction, the risk for
TRANSLARYNGEAL AIRWAY pneumothorax is greatly increased because gas cannot
escape from the lungs. Strong consideration should be
Retrograde Wire Intubation given to placing a second transtracheal egress catheter
in these circumstances, or simply avoiding this technique
Retrograde wire intubation is an excellent technique for altogether. Laryngospasm can also impede the outward
securing a difficult airway. The procedure takes some time flow of oxygen from the trachea. Inadvertent placement
to perform and should not be considered under emer- of a gas delivery line into the GI tract can also result in
gency circumstances unless the practitioner is extremely complications (gastric rupture, esophageal perforations,
experienced in the technique. Bleeding may occur at the bleeding, hematoma, and hemoptysis). Damage to the tra-
site of the tracheal puncture. Cases of severe hemoptysis
cheal mucosa is possible in patients who are managed with
with resultant hypoxia, cardiopulmonary arrest, dysrhyth-
long-term TTJV, especially if the gas is not humidified.
mias, and death following retrograde wire intubation have
been reported. Subcutaneous emphysema localized to the
area of the transtracheal needle puncture is common.
In severe cases, pneumomediastinum and pneumothorax
Percutaneous Dilatational Tracheostomy
can be produced.23 Laryngospasm can result from irri-
Although percutaneous dilatational tracheostomy is not
tation by the retrograde wire unless the vocal cords are
usually recommended for emergency use, it appears to
anesthetized or relaxed. Other, less common complica-
be suitable for emergency situations in skilled hands.
tions include esophageal perforation, tracheal hematoma,
Many different sets are available. Bleeding, subcutaneous,
laryngeal edema, infection, tracheitis, tracheal fistula,
and mediastinal emphysema, pneumothorax, airway ob-
trigeminal nerve injury, and vocal cord damage.34
struction, aspiration, infection, accidental extubation, and
death are early complications. Delayed complications
Surgical Cricothyroidotomy are tracheal stenosis, scars, hoarseness, and tracheoe-
In both surgical cricothyroidotomy (using a scalpel) and sophageal and tracheocutaneous fistulae.
needle cricothyroidotomy (using a needle set) procedures, Minitracheostomy occasionally results in excess
the cricothyroid membrane requires penetration. Acute bleeding into the airway, necessitating progression to a
complications include bleeding (especially during surgical full surgical tracheostomy. Air embolism, subcutaneous
cricothyroidotomy), misplacement of the tube (especially emphysema, pneumomediastinum, and tension pneu-
after needle cricothyroidotomy) and barotrauma. During mothorax are also possible with this procedure.
this technique, subcutaneous emphysema, pneumotho-
rax, pneumomediastinum and pneumopericardium tube
malposition, failure of airway access, wound infection, Subglottic Stenosis
displaced cartilage fractures, and laryngotracheal separa-
tion can occur. Subglottic stenosis is a complication of long-term intuba-
Long-term complications sustained directly from tion. This is much more difficult to repair and frequently
these procedures are granulation tissue around the results in permanent speech impairment or laryngeal
tracheostomy site, subglottic stenosis, massive laryngeal damage. A tracheostomy tube can cause tracheal ero-
mucosa trauma, endolaryngeal hematoma and laceration, sion, particularly into the esophagus (tracheoesophageal
vocal cord paralysis, hoarseness, and thyroid cartilage fistula) or the brachiocephalic artery. Accidental extuba-
fracture with dysphasia. All emergency translaryngeal tion and dislodgement of the cannula happen occasion-
airways should be eventually changed to a formal tra- ally, most commonly in the early postoperative period.
cheostomy. Subglottic stenosis is a delayed complication, Infection, mediastinal sepsis, tracheal stenosis, and tra-
especially in children. cheomalacia are rare, late complications.
104 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
The Sellick maneuver, or cricoid pressure, has re- that a high percentage of children mechanically ventilated
moved much of the fear of aspiration during emergency at home have undiagnosed latex allergy.48
intubation. Cricoid pressure is effective in raising the pres-
sure in the upper esophageal sphincter, thereby preventing
aspiration.
What Problems Are
INTRAOCULAR AND Encountered with Extubation?
INTRACRANIAL PRESSURE Primary and secondary responses to extubation are pos-
With thiopental, etomidate, and halothane anesthesia, sible. The primary effects include local and systemic
an increase in intraocular pressure was observed dur- responses. The same responses that follow intubation may
ing laryngoscopy, as well as LMA insertion, but not with be observed at extubation. During intubation, the patient
total intravenous anesthesia or remifentanil and sevoflu- is more protected by anesthesia induction than during
extubation; therefore, the cardiovascular responses can
rane. Decreases in intraocular pressures were observed
be even more exaggerated. The most serious complication
under endotracheal intubation during general anesthe-
after extubation is acute airway obstruction. Decreased
sia with propofol and sevoflurane, both combined with
consciousness, central respiratory depression, decreased
remifentanil. Intraocular pressure may also increase dur-
muscle tone, and tongue obstruction may lead to inspira-
ing extubation.
tory or expiratory stridor, dyspnea, cyanosis, tachycardia,
Insertion of an LMA does not increase intraocular
hypertension, agitation, and sweating.
pressure in children after sevoflurane induction.41 Sufen-
tanil is also effective in preventing an intraocular pressure
increase caused by rapid-sequence induction with suc-
cinylcholine.42 It is extremely important that an increase
HEMODYNAMIC CHANGES
in intraocluar pressure be avoided in patients with pene- Hemodynamic changes, including a 20% increase in heart
trating eye injury. rate and blood pressure, occur in most patients at the time
Intracranial pressure markedly and transiently rises of extubation. Patients with cardiac disease, pregnancy-
during laryngoscopy and endotracheal intubation. Pa- induced hypertension,32 and increased intracranial pres-
tients with head injury are at higher risk from this sure may be at particular risk for life-threatening ischemic
increase, because it reduces cerebral perfusion, and there- myocardial episodes. Management of these changes con-
fore may increase secondary brain damage. Deep anesthe- sists of deep extubation or pharmacologic therapy.
sia during induction can prevent these adverse effects.
LARYNGOSPASM
LATEX ALLERGY
Laryngospasm, a protective reflex mediated by the vagus
Almost 17% of overall anaphylaxis in surgical procedures nerve, is the most frequent cause of postextubation airway
is related to latex anaphylaxis.43 To prevent anaphylaxis obstruction. It can be provoked by movement of the
in patients during anesthesia and surgery, the patients cervical spine, pain, vocal cord irritation by secretions,
history has to be carefully evaluated preoperatively. There or sudden stimulation while the patient is still in a light
is currently no therapy for latex allergy, and therefore, plane of anesthesia. In a large study in 136,929 patients,
avoidance of latex-containing products is mandatory for the incidence of laryngospasm was 50/1,000 in children
predisposed individuals.44 Latex allergy is present in 8% with bronchial asthma and airway infection and 25/1,000
of the general population in the United States, with a in children in the age group of 1 to 3 months when
prevalence of 30% in health care workers.45 There is an endotracheal intubation had been performed.49
increased incidence of type I and type IV latex sensitivity in The optimal course for dealing with laryngospasm
the general population. The prevalence of latex sensitivity is to avoid it. It is imperative that no saliva, blood, or
among anesthesiologists is approximately 12.5%, with gastric contents touch the glottis while the patient is
2.4% having a latex allergy.46 lightly anesthetized. In cases in which laryngospasm is
Patients with spina bifida, rubber industry workers, anticipated, the patient may undergo a deep extubation,
atopic patients, and patients with a multiple-surgery his- while placed in the lateral position, with the head down to
tory and allergies to certain exotic foods are most at keep the vocal cords clear of secretions during emergence.
risk. Contamination with latex in anesthesia is possible Because suctioning of the oropharynx does not adequately
through direct contact by face mask, endotracheal and remove secretions around the vocal cords, it is best to
gastric tubes, gloves, syringes, electrodes; through in- extubate patients during a positive-pressure breath; this is
halation from contaminated circuits and room air; and also the procedure of choice in children. In a recent study,
through the parenteral path with latex-containing intra- children were safely extubated in deep anesthesia from
venous administration sets. 1.5 minimum effective alveolar anesthetic concentration
Considerations for anesthesiologists who handle pa- of either sevoflurane or desflurane.50
tients with latex allergy are available at the ASAs In a survey of anesthesiologists in the United States,
website.47 In a pediatric study, Nakamura et al., found deep extubation was performed by 64% of the interviewed
106 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
practitioners.51 The study of Koga et al.,52 showed that can result in the spontaneously breathing patient. As a
the rate of airway obstruction in patients extubated result of the inspiratory effort against the closed glottis,
during deep anesthesia (17/20) was not higher than in these patients generate negative intrapleural pressure
patients extubated after regaining consciousness (18/20). greater than 100 cm H2 O. The following conditions can
Larson describes a maneuver that treats laryngospasm, a result in a marked increase in transmural pressure, fluid
maneuver in which intense medial pressure is applied filtration into the lung, and development of pulmonary
in the area between the angle of the mandible and edema:55
the mastoid process. This technique has also been
Increases in left ventricular preload and afterload
described as the Larson manoeuvre and has been
Altered pulmonary vascular resistance
used to greatly help all but the most severe cases of
Increased adrenergic state
laryngospasm. This maneuver is thought to stimulate
Right ventricle dilatation
the patient to take deep breaths and facilitate a rapid
Intraventricular septum shift to the left
transition to sustained minute volumes and maintenance
Left ventricular diastolic dysfunction
of oxygenation.53
Increased left heart loading conditions
Enhanced microvascular intramural hydrostatic pres-
sure
LARYNGEAL EDEMA Negative pleural pressure
Transmission to the lung interstitium
Laryngeal edema is a significant cause of postextu-
bation obstruction, especially in neonates and infants. Any of these conditions are seen within minutes after
Supraglottic edema most commonly results from surgical extubation. Management involves removing the obstruc-
manipulation, positioning, hematoma formation, overag- tion, supporting the patient with oxygen, monitoring the
gressive fluid management, impaired venous drainage, or patient closely, and reducing the afterload. Reintubation
coexisting conditions (such as preeclampsia or angioneu- is rarely necessary; most cases resolve spontaneously with-
rotic edema). Retroarytenoidal edema typically results out further complications.
from local trauma or irritation. Subglottic edema oc-
curs most often in children, particularly neonates and
infants.
Factors associated with the development of subglottic ASPIRATION
edema include traumatic intubation, intubation lasting
Pulmonary aspiration of gastric contents is a constant
longer than 1 hour, bucking on the endotracheal tube,
threat for any patient who has a full stomach or is at
changes in head position, or tight-fitting endotracheal
risk for postoperative vomiting. Laryngeal function is
tubes. Laryngeal edema usually presents as stridor within
altered for at least 4 hours after tracheal extubation. The
30 to 60 minutes after extubation, although it may start
depression of coughing reflexes, along with the presence
as late as 6 hours post extubation.54 Regardless of the
of residual anesthetic agents, places almost all recently
cause of laryngeal edema, management depends on the
extubated patients at risk. Aspiration is probably more
severity of the condition. Therapy consists of humidified
prevalent than is currently known. Most cases are so
oxygen, racemic epinephrine, head-up positioning, and,
minor, that they do not affect the patients postoperative
occasionally, reintubation with a smaller endotracheal
course. Reducing the gastric contents by suctioning
tube. The practice of administering parenteral steroids
through a gastric tube and extubating the patient in
with the goal of preventing or reducing edema is
the lateral position, with a head-down tilt, is the safest
controversial.54
protection against aspiration.
DIFFICULT EXTUBATION AND and long-lasting (nerve injuries) or mild and short-
lived (sore throat).
ACCIDENTAL EXTUBATION 3. To minimize injury to the patient, the anesthesiologist
The difficult removal of the endotracheal tube can be should examine the patients airway carefully, identify
caused by failure to deflate the cuff, use of an oversized any potential problems, devise a plan that involves
tube, adhesion of the tube to the tracheal wall, or the least risk for injury, and have a back-up plan
transfixation of the tube by an inadvertent suture to a immediately available.
nearby organ or a screw in the oromaxillofacial surgical 4. Each anesthesiology department should establish
site. Possible sequelae of these complications include guidelines/algorithms specific to their institution for
airway leak, aspiration, tube obstruction, and trauma difficult airway management.
from attempts at forceful extubation. In most cases, the
problem arises from the inability to deflate the cuff, REFERENCES
commonly as a result of failure in the cuff-deflating
1. Domino KB. Closed malpractice claims for airway trauma
mechanism. If this problem is encountered, the cuff during anesthesia. ASA Newsl. 1968;32.
should be punctured with a transtracheal needle. The 2. Adnet F. Difficult mask ventilation. Anesthesiology.
forceful removal of an endotracheal tube with an inflated 2000;92:1217.
cuff can result in damage to the vocal cords and arytenoid 3. Kroesen G. Guidelines for the advanced management of
dislocation. the airway and ventilation during resuscitation. A statement
Accidental extubation during anesthesia may occur by the Airway and Ventilation Management of the Working
with disposable tonsillectomy instruments and change Group of the European Resuscitation Council. Resuscitation.
in the patients head position. Most accidental extuba- 1996;31:201.
4. Langeron O, Masso E, Huraux C, et al. Prediction of difficult
tions that were reported took place in intensive care
mask ventilation. Anesthesiology. 2000;92:1229.
unit patients.56 Complications after accidental extubation 5. Joshi GP, Smith I, White PF. Laryngeal mask airway. In:
include hypoxia, hypercarbic respiratory failure, aspira- Benumof JL, ed. Airway management: Principles and practice.
tion, retention of pulmonary secretions, arrhythmias, and St. Louis: Mosby; 1996:353.
tachycardia. Reintubation may be very difficult, especially 6. Krier C, Georgi R. Airway management. Thieme Stuttgart.
if the first intubation was difficult. The use of the Com- 2001;36:193.
bitube or the LMA can be very useful in this critical 7. Alexander CA, Leach AB. Incidence of sore throats with the
situation. laryngeal mask. Anaesthesia. 1989;44:791.
8. Brain AIJ, Verghese C, Strube PJ. The LMA ProSeal
Airway exchange catheters can also be used as a
a laryngeal mask with an oesophageal vent. Br J Anaesth.
bridge to extubation in patients with a known difficult
2000;84:650.
airway, or if a successful extubation is questionable. These 9. Brimacombe J, Keller CH, Berry A. Gastric insufflation with
catheters have a dual function: oxygenation/ventilation the ProSeal laryngeal mask. Anesth Analg. 2001;92:1614.
and use as an intubation stylet. The literature suggests 10. Branthwaite MA. An unexpected complication of the intu-
four potential risks associated with the use of these bating laryngeal mask. Anaesthesia. 1999;54:166.
devices: (i) Catheter misplacement, (ii) bronchial or lung 11. Green KS, Beger TH. Proper use of the combitube. Anesthe-
trauma, (iii) laryngeal trauma, and (iv) barotraumas siology. 1994;81:513.
related to either oxygenation or jet ventilation through 12. Kraft P, Nikolic A, Fass M. Esophageal rupture associated
with the use of the Combitube. Letter to the editor. Anesth
these catheters.5759 These complications can be avoided
Analg. 1998;87:1453.
by proper placement of these catheters and the following 13. Vezina D, Lessard MR, Bussieres J, et al. Complications as-
measures: (i) Ensure appropriate length of placement sociated with the use of the esophageal-tracheal combitube.
(22 to 25 cm for oral placement and 27 to 30 cm Can J Anaesth. 1998;45:76.
for nasal placement in adults); (ii) confirm correct 14. Oczenski W, Krenn H, Dahaba A, et al. Complications
endotracheal placement (through direct laryngoscopy, following the use of the combitube, tracheal tube and
fiberoptic laryngoscopy, end-tidal CO2 monitoring); and laryngeal mask airway. Anaesthesia. 1999;54:1161.
(iii) proper use of oxygenation and ventilation techniques. 15. American Society of Anaesthesiologists Task Force on
Management of the Difficult Airway. Practice guidelines
for management of the difficult airway. Anesthesiology.
2003;98:1269.
16. Warner ME, Benefeld SM, Warner MA, et al. Perianesthetic
KEY POINTS dental injuries. Anesthesiology. 1999;90:1302.
17. Monroe MC, Gravenstein N, Saga-Rumley S. Postoperative
1. The inability to secure the airway, with consequent sore throat: Effect of oropharyngeal airway in orotracheally
failure of oxygenation and ventilation, is a life- intubated patients. Anesth Analg. 1990;70:512.
threatening complication. Failure of oxygenation 18. Brimacombe J, Holyoake L, Keller C, et al. Pharyngolaryn-
can lead to hypoxia followed by cerebral ischemia, geal, neck, and jaw discomfort after anesthesia with the
cardiovascular dysfunction, and finally death. Time is face mask and laryngeal mask airway at high and low cuff
volumes in males and females. Anesthesiology. 2000;93:26.
a very crucial factor in this context.
19. Knoll H, Ziegeler S, Schreiber J, et al. Airway injuries
2. Complications vary widely in severity, although some after one-lung ventilation: A comparison between double-
are dramatic and immediately life-threatening (unrec- lumen tube and endobronchial blocker. Anesthesiology.
ognized esophageal intubation). Others can be severe 2006;105:471.
108 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
20. Kambic V, Radsel Z. Intubation lesions of the larynx. Br J American Society of Anesthesiologists Task Force on man-
Anaesth. 1978;50:587. agement of the difficult airway, Anesthesiology. 1993;78:597.
21. Rieger A. Intubationsscha den: Inzidenz, komplikationen, 41. Duman A, Ogun CO, Okesli S. The effect on intraocular
konsequenzen. [Damage following intubation: Incidence, pressure of tracheal intubation or laryngeal mask insertion
complications, consequences]. In: Krier C, Georgi R, eds. during sevoflurane anaesthesia in children without the use
Airway management. Stuttgart, Germany: Thieme Medical of muscle relaxants. Paediatr Anaesth. 2001;11:421.
Publishers, Springer-Verlag; 2001. 42. Georgiou M, Parlapani A, Argiriadou H, et al. Sufentanil
22. Hiong YT, Fung CF, Sudhaman DA. Arytenoid subluxation: or clonidine for blunting the increase in intraocular pres-
Implications for the anaesthetist. Anaesth Intensive Care. sure during rapid-sequence induction. Eur J Anaesthesiol.
1996;24:609. 2002;19:819.
23. Marty-Ane Ch, Picard E, Jonquet O, et al. Membranous 43. Vervloet D, Magnan A, Birnbaum J, et al. Allergic emer-
tracheal rupture after endotracheal intubation. Ann Thorac gencies seen in surgical suites. Clin Rev Allergy Immunol.
Surg. 1995;60:1367. 1999;17:459.
24. Howland WS, Lewis JS. Postintubation granulomas of the 44. Hepner DL, Castells MC. Latex allergy: An update. Anesth
larynx. Cancer. 1956;9:1244. Analg. 2003;96:1219.
25. Snow JC, Harano M, Balough K. Postintubation granuloma 45. Garabrant DH, Schweitzer S. Epidemiology of latex sensi-
of the larynx. Anesth Analg. 1966;45:425. tization and allergies in health care workers. J Allergy Clin
26. Cheney FW, Posner KL, Caplan RA. Adverse respiratory Immunol. 2002;110:S82.
events infrequently leading to malpractice suits. Anesthesiol- 46. Brown RH, Schauble JF, Hamilton RG. Prevalence of latex
ogy. 1991;75:932939. allergy among anesthesiologists: Identification of sensitized
27. Domino KB, Posner KL, Caplan RA, et al. Airway injury but asymptomatic individuals. Anesthesiology. 1998;89:287.
during anesthesia: A closed claims analysis. Anesthesiology. 47. American Society of Anesthesiologists. Natural rubber latex
1999;91:1703. allergy: Considerations for anesthesiologists. Available at:
28. Green JG, Wood JM, Davis LF. Asystole after inad- http://www.asahq.org/publicationsAndServices/latexallergy
vertent intubation of the orbit. J Oral Maxillofac Surg. .pdf. Accessed July 17, 2006.
1997;55:856859. 48. Nakamura CT, Ferdman RM, Keens TG, et al. Latex allergy
29. Halac E, Indiveri DR, Obregon RJ. Complication of nasal on home mechanical ventilation. Chest. 2000;118:1000.
endotracheal intubation. J Pediatr. 1983;103:166. 49. Olsson GL, Hallen B. Laryngospasm during anaesthesia.
30. Caplan RA, Posner KL, Ward RJ, et al. Adverse respiratory A computer-aided incidence study in 136,929 patients. Acta
events in anesthesia: A closed claims analysis. Anesthesiology. Anaesthesiol Scand. 1984;28:567.
1990;72:828. 50. Valley RD, Freid EB, Bailey AG, et al. Tracheal extubation
31. Gamlin F, Caldicott LD, Shah MV. Mediastinitis and sepsis of deeply anesthetized pediatric patients: A comparison of
syndrome following intubation. Anaesthesia. 1994;49:883. desflurane and sevoflurane. Anesth Analg. 2003;96:1320.
32. Cooper JB, Newbower RS, Kitz RJ. An analysis of major 51. Daley MD, Norman PH, Coveler LA. Tracheal extubation of
errors and equipment failures in anesthesia management: adult surgical patients while deeply anesthetized: A survey of
Considerations for prevention and detection. Anesthesiology. United States anesthesiologists. J Clin Anesth. 1999;11:445.
1984;60:3442. 52. Koga K, Asai T, Vaughan RS, et al. Respiratory compli-
33. Henn-Beilharz A, Hagen R. Besonderheiten der cations associated with tracheal extubation. Anaesthesia.
atemwegssicherung in der hals-nasen-ohren-heilkunde. 1998;53:540.
[Special aspects of airway management in ear, nose, and 53. Larson CP. Laryngospasm: The best treatment. Anesthesio-
throat surgery]. In: Krier C, Georgi R, eds. Airway man- logy. 1998;89:1293.
agement. Stuttgart, Germany: Thieme Medical Publishers, 54. Darmon JY, Rauss A, Dreyfuss D, et al. Evaluation of risk
Springer-Verlag; 2001. factors for laryngeal edema after tracheal extubation in
34. Sanchez A. Retrograde intubation technique. In: Benu- adults and its prevention by dexamethasone: A placebo-
mof JL, ed. Airway management: Principles and practice. St controlled, double-blind, multicenter study. Anesthesiology.
Louis: Mosby; 1996:439. 1993;78:609610.
35. Oczenski W, Krenn H, Dahaba AA, et al. Hemodynamic 55. Broccard AF, Liaudet L, Aubert JD, et al. Negative pressure
and catecholamine stress responses to insertion of the post-tracheal extubation alveolar hemorrhage. Anesth Analg.
combitube, laryngeal mask airway or tracheal intubation. 2001;92:273275.
Anesth Analg. 1999;88:1389. 56. Kapadia FN, Bajan KB, Raje KV. Airway accidents in
36. Edwards ND, Alford AM, Dobson PM, et al. Myocardial intubated intensive care unit patients: An epidemiological
ischaemia during tracheal intubation and extubation. Br J study. Crit Care Med. 2000;28:659664.
Anaesth. 1994;73:537. 57. Loudermilk EP, Hartmannsgruber M, Stoltzfus DP, et al.
37. Brimacombe JR. Analysis of 1500 laryngeal mask uses by A prospective study of the safety of tracheal extubation using
one anaesthetist in adults undergoing routine anaesthesia. a pediatric airway exchange catheter for patients with a
Anaesthesia. 1996;51:76. known difficult airway. Chest. 1997;111:1660.
38. Engelhardt T, Webster NR. Pulmonary aspiration of gastric 58. Eisenach JH, Barnes RD. Potential disaster in airway
contents in anaesthesia. Br J Anaesth. 1999;83:453. management: A misguided airway exchange catheter via a
39. Brimacombe J, Berry A. The incidence of aspiration asso- hole bitten into a univent endotracheal tube. Anesthesiology.
ciated with the laryngeal mask airwaya metaanalysis of 2002;96:1266.
published literature. J Clin Anesth. 1993;7:297. 59. Fetterman D, Dubovoy A, Reay M. Unforeseen esophageal
40. American Society of Anesthesiologists. Practice guidelines misplacement of airway exchange catheter leading to gastric
for management of the difficult airway: A report by the perforation. Anesthesiology. 2006;102:1111.
CHAPTER SAFETY AND HAZARDS ASSOCIATED
7
WITH TRACHEAL INTUBATION AND
USE OF SUPRALARYNGEAL AIRWAYS
David Ferson, Linda Chi, Sonal Zambare, and
Wendy Botnick
CASE SUMMARY whereas tracheal tubes are always positioned below the
vocal cords, within the lumen of the trachea. Additionally,
42-year-old male patient was scheduled for currently available SLAs can only be inserted orally, but
A
an arthroscopic knee surgery under general tracheal tubes can be inserted through oral or nasal routes.
anesthesia. The patients medical history Complications can occur with the insertion of both
was unremarkable except for mild seasonal types of airway devices, and short- and long-term adverse
asthma. Height and weight were 180 cm effects have been described in the literature (see Fig. 7.1).
and 80 kg, respectively. In the operating This chapter will focus on the major hazards associated
room, general anesthesia was induced with midazolam, with tracheal intubation and SLA use, how to diminish
fentanyl, and propofol. A supraglottic airway device was the risk of complications, and, most important, how to
inserted easily on the first attempt. The cuff was inflated manage complications if they occur.
with air to the maximum volume recommended by the
manufacturer. Intracuff pressure was neither measured
nor monitored during the course of surgery. Anesthesia
was maintained with desflurane. At the conclusion of HAZARDS ASSOCIATED WITH
surgery, which lasted for 130 minutes, the patient was
awakened, and the supraglottic airway was removed
TRACHEAL INTUBATION
without incident. The patient was then transferred to MacEwen first described tracheal intubation for mandibu-
the recovery room. Shortly after arrival, he complained lar surgery in 1878; however, the current techniques of
of dysphagia. A specialist in head and neck surgery tracheal intubation were not developed, refined, and pop-
was consulted and, after performing a flexible nasal ularized until the 1920 and 1930s.1 4 Subsequently, the
fiberoptic laryngoscopy, bilateral hypoglossal nerve palsy development of different laryngoscopic blades has cor-
was diagnosed. A swallowing study demonstrated an responded with major improvements in tracheal tube
uncoordinated swallowing reflex. Because of the risk technology. Currently, tracheal intubation is the most
for aspiration, a nasogastric tube was placed for enteric widely used method of airway management; therefore,
feeding. The patients symptoms improved gradually and it stands to reason that its complications have been de-
on the 16th postoperative day, another swallowing study scribed numerous times in the literature. These can be
demonstrated near normal function. The nasogastric tube grouped into the following categories following a proce-
was removed. Despite enteric feeding, the patient lost 15 lb dural timeline: (i) Complications that result from direct
over 2 weeks. After 21 days, the symptoms of hypoglossal laryngoscopy and intubation, (ii) complications that result
nerve palsy resolved completely. from the pressure of the tracheal tube or cuff on airway
structures, and (iii) complications following extubation.
INTRODUCTION
One of the most critically important perioperative
functions of the anesthesiologist is establishing and How Can Complications
maintaining a patent airway. To achieve this goal, the Be Induced through Direct
anesthesiologist usually inserts an airway device to fa-
cilitate spontaneous, assisted, or controlled ventilation. Laryngoscopy and Intubation?
There are two distinctly different categories of airway
devicessupralaryngeal and tracheal. Supralaryngeal air- Tracheal intubation under direct observation requires vi-
ways (SLAs) do not descend below the vocal cords, sualization of the glottic opening and is usually achieved
109
110 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Thyrohyoid m.
Cricothyroid m. Genioglossus m.
Hyoglossus m.
Trachea
Esophagus
Inferior
pharyngeal
constrictor
Superior
laryngeal
n., a., v.
External
carotid a. Lingual v.
and a.
Hypoglossal n.
FIGURE 7.1 Structures that are potentially at risk for injury from tracheal intubation and
supraglottic airways include mucous membranes, soft tissues of the pharynx and larynx, nerves and
blood vessels of the neck, laryngeal cartilages, and bones of the neck.
n, nerve; a, artery; v, vein; m, muscle.
OA OA OA
PA
LA
PA
PA
LA
LA
A B C
FIGURE 7.2 Schematic diagram shows that to create a line of sight from the mouth to the larynx,
the laryngoscopist must minimize the angles between the oral axis (OA), pharyngeal axis (PA), and
laryngeal axis (LA). A: When the head is in a neutral position, the OA, PA, and LA are not aligned.
B: By flexing the neck, the angle between the PA and LA can be significantly reduced. This position is
known as the sniffing position. C: Extending the head at the atlanto-occipital joint brings the OA
close to the PA and LA and optimizes conditions for intubation by creating a line of sight from the
mouth to the larynx.
manner that is more applicable to their level of train- However, when intubation is difficult, excessive force is
ing (see Fig. 7.3). Furthermore, because of the higher usually required. Hemodynamic response to even rou-
speed of data acquisition possible with helical CT, mis- tine laryngoscopy and intubation can vary significantly
registration and image degradation caused by patient among patients.15 Most patients are reported to have tran-
motion is less of an issue. This is especially important sient tachycardia and hypertension during laryngoscopy.
when scanning uncooperative patients and trauma vic- These conditions are usually short-lived in nature and are
tims. In emergency cases, however, when there is no not associated with any significant morbidity. However,
time to perform radiological studies, it is important to in a patient with known cardiovascular disease, even a
minimize head and neck movement by having a skilled short duration of tachycardia and hypertension can lead
assistant apply in-line stabilization to the head and neck to myocardial ischemia, arrhythmias, or even myocar-
throughout the intubation. This maneuver, when prop- dial infarction.16 Various drugs have been used to abate
erly performed, minimizes the flexion-extension move- this detrimental cardiovascular response.17 Short-acting
ment and also decreases the rotation of the cervical -blockers such as esmolol and calcium channel blockers
spine.14 such as intravenous nicardipine have been shown to be
Laryngoscopy can also be particularly challenging in very effective in reducing tachycardia and hypertension
patients with degenerative disease of the cervical spine. caused by laryngoscopy and intubation.18,19 However,
This is because, in these patients, the movement of the cer- these agents, while attenuating the hemodynamic re-
vical spine is restricted or impossible, and therefore, the sponses, may result in excessive negative ionotropic and
optimal intubating position cannot be obtained. The most chronotropic action that precipitates cardiac failure in
common diseases in this group of patients are ankylosing susceptible patients. Various other techniques, such as in-
spondylitis, diffuse idiopathic skeletal hyperostosis, and jecting lidocaine through an intratracheal or intravenous
osteoarthritis. Also, the presence of cervical osteophytes route immediately before intubation, have been described,
can make visualization of the cords and passage of the but none has been proved to be completely reliable and
tube very difficult. In patients with restricted mobility of effective.20,21 Therefore, it is essential to use a combi-
the cervical spine, intubation is perhaps best achieved with nation of anesthetic and adjuvant agents to ensure that
the use of specialized equipment that is, a fiberoptic bron- the patient is adequately anesthetized before instrument-
choscope, video laryngoscope, or fiberoptically equipped ing the airway. This is particularly important in patients
intubating laryngeal mask airway (C-Trach LMA; LMA with increased intracranial or intraocular pressure, in
NA, San Diego, CA). whom the exaggerated hemodynamic responses can lead
to catastrophic consequences.2225
A B
C D
FIGURE 7.3 Helical computed tomography (CT) images of the airway anatomy. Because the
information acquired from a helical CT is volumetric, the entire anatomical region of interest can be
scanned in a matter of seconds. The images are displayed in the coronal (A), saggital (B), and axial
(C) planes, thereby allowing for careful analysis by a radiologist. Also, three-dimensional image
reconstructions give clinicians a view that is similar to gross anatomy (D).
extensively in a separate chapter of this textbook. There- approach to patients with difficult-to-manage airways that
fore, in this chapter we will not discuss the evaluation can be easily integrated into clinical practice during elec-
and management of these patients. However, it is im- tive and emergency airway-related situations.27 In 2003,
portant to remember that excessive force during direct the guidelines and the algorithm were revised and up-
laryngoscopy in patients with difficult-to-manage airways dated. Specifically, the new algorithm has expanded the
frequently results in tooth damage and trauma to the uses of the laryngeal mask airway (LMA Classic; LMA
soft tissues and cartilaginous and bony structures of the North America, San Diego, CA) in patients with difficult-
larynx. Because it is difficult to predict, in any patient, to-manage airways.28
how much force applied during laryngoscopy will result
in trauma to the pharyngeal and laryngeal structures, it
seems logical that if visualization of the larynx is difficult, TOOTH DAMAGE
one should switch to a different laryngoscopic blade or
consider an alternative technique for intubation instead of If tooth damage occurs during laryngoscopy, one should
increasing the amount of force while continuing to use a obtain a dental consultation in the postoperative period.
technique that has already failed.26 In 1993, the American More important, if a tooth is dislodged into the patients
Society of Anesthesiologists (ASA) established guidelines airway during laryngoscopy, it is imperative to retrieve it
for taking care of patients with difficult-to-manage air- from the pharynx before inserting an endotracheal tube to
ways and published a difficult airway algorithm. The avoid pushing the tooth into the trachea or bronchi. If the
ASA guidelines and the difficult airway algorithm pro- tooth does become wedged into the bronchus, one needs
vide an excellent framework for developing an organized to perform fiberoptic bronchoscopy immediately, using
C H A P T E R 7 / S A F E T Y A N D H A Z A R D S A S S O C I AT E D W I T H T R A C H E A L I N T U B AT I O N 113
will hopefully solve the problems associated with mucosal by several investigators, the incidence of residual mus-
pressure and also reduce the incidence of pulmonary cle relaxation in the recovery room can be as high
aspiration.39 as 50%.43
What is most concerning is that despite the avail-
ability of modern neuromuscular blocking agents with
short or intermediate durations of action, the incidence of
NERVE INJURIES residual neuromuscular blockade remains very high.44,45
There are also reports of nerve injuries caused by the Therefore, it is imperative that all patients satisfy extu-
inflated cuff.40 The nerves commonly affected are the bation criteria (i.e., sustained head lift for more than
recurrent laryngeal branch of the vagus nerve, the lingual 5 seconds or a strong handgrip) before removal of the
nerve, and the hypoglossal nerve. These nerve injuries are tracheal tube. It has been found, however, that even af-
usually temporary, and a complete recovery is possible, ter fulfilling the criteria for extubation, some patients
in most cases, within a couple of days. However, if may still have residual neuromuscular paralysis.46,47 This
symptoms persist, an otolaryngology consultation may be can have serious consequences, because the respiratory
required to assess the function of the vocal cords and the muscles and pharyngeal protective reflexes may not be
coordination of the structures associated with swallowing. functioning at optimal levels. Therefore, in patients who
The clinical presentation of recurrent laryngeal nerve are at high risk for developing hypoxemia or upper air-
palsy can be identical to that of arytenoid subluxation way obstruction and who have diminished respiratory
or arytenoid dislocation caused by forceful laryngoscopy. reserve, residual muscle paralysis is especially danger-
Only special studies can distinguish between these two ous and can lead to respiratory failure shortly after
clinical complications. Whereas hypoglossal or recurrent extubation.
nerve palsy can be diagnosed by nerve conduction studies, Also, muscle weakness, hypoxemia, upper airway
arytenoid displacement requires specialized CT scanning obstruction, and diminished pharyngeal reflexes increase
to establish the correct diagnosis. Cuff herniation used the chances for aspiration in the immediate postoperative
to be a common problem before the current standards period.48 In addition to patient-related factors, prolonged
for manufacturing endotracheal tubes were established. neuromuscular blockade can be caused or aggravated
Currently, the cuff is firmly attached to the tube, and this by an acid-base imbalance, electrolyte abnormalities,
complication is no longer seen. the use of magnesium, renal or hepatic disease, and
the administration of aminoglycosides, -blockers, and
calcium channel blockers.
HAZARDS ASSOCIATED WITH LMA Classic, whereas others differ significantly in their
functional design. These design differences include the
THE USE OF location and type of cuff, which serves as a sealing mecha-
SUPRALARYNGEAL AIRWAYS nism, the location of the ventilatory portion of the device,
and whether there is an attempt to seal off the entrance
to the esophagus. On the basis of an anatomical com-
parative analysis of currently available SLAs, they can be
grouped into the categories as shown in Table 7.1.53 (Also
How Do Supralaryngeal see Figs. 7.4 to 7.10.)
Airways Differ from Other The LMA Classic is the most widely used of all the
SLAs (over 200 million used worldwide, per the LMA
Airway Devices? companys personal communication in January 2006).
Consequently, several problems associated with the use
SLAs represent a class of medical devices designed to of the LMA Classic have been reported in the past
provide spontaneous, assisted, or controlled ventilation. twenty years. These complications include injuries to
SLAs differ from other airway devices, such as oropha- pharyngeal and laryngeal structures and problems related
ryngeal airways and tracheal tubes, in that they do not to the integrity of the device. It is important to note
require a facial seal or tracheal insertion for ventilation. that, owing to the recent introduction and infrequent
The laryngeal mask airway (LMA Classic; LMA North use of other SLA devices, their safety has not been
America, San Diego, CA) was introduced into clinical formally assessed. However, by combining information
practice in 1988 and became the first commercially suc- from the existent LMA Classic literature with crossover
cessful SLA. Indeed, the LMA Classic has demonstrated comparative anatomical studies of different SLA devices,
that, for many patients in select clinical situations, the one can predict with a certain degree of accuracy the
supraglottic airway approach is not only feasible, but also potential hazards associated with the new SLA devices.53
less invasive and more beneficial than endotracheal in- Accordingly, we will first review major complications
tubation.52 As a result of the clinical and commercial associated with the use of the LMA Classic and then
success of the LMA Classic, the following SLAs have address the clinically relevant potential hazards that might
been introduced into clinical practice: the cuffed oropha- be associated with the use of the new SLAs.
ryngeal airway (COPA) (Mallinckrodt Medical; St Louis,
MO), the pharyngeal airway express (PAEXPRESS ) (Vital
Signs, Inc., Totowa, NJ), the laryngeal tube (LT) (King
System Corporation, Noblesville, IN), the Portex soft seal What Risks Are Associated
laryngeal mask (Portex-Soft Seal) (Portex Inc., Keene,
NH), the perilaryngeal airway (COBRA) (Engineered Med- with the Use of the LMA?
ical Systems, Indianapolis, IN), the Ambu laryngeal mask
(AMBU-LM) (Ambu Inc., Linthicum, MD), and the stream- Although several problems associated with the use of
lined liner of the pharynx airway (SLIPA) (SLIPAmed SA the LMA Classic have been reported in the literature,
Pty Ltd, Cape Town, South Africa). Predictably, some of the occurrence of these problems appears to be inversely
the new supraglottic airways are close replicas of the related to the experience and skill level of the operator.
1. The laryngeal mask airway, which includes a ventilatory opening surrounded by a cuff that forms a seal with the
periglottic tissues (Fig. 7.5). The ventilatory opening and the cuff seal represent the most distal portion of the device.
The LMA Classic, AMBU-LM, and Portex-Soft Seal are examples of this airway type
2. The pharyngeal or pharyngeal-esophageal tube, which includes a ventilatory tube surrounded by a cuff in a
circumferential fashion located proximal to the ventilatory opening. The machine end of these devices is defined as
the proximal end. This design compartmentalizes the pharynx, with the cuff serving as a sealing divider between the
proximal and distal pharyngeal compartments (Figs. 7.4 and 7.6). The ventilatory openings are located in the distal
pharyngeal compartment. The PAEXPRESS , LT, and COBRA are examples of this airway type
3. The pharyngeal airway liner, which is represented by the SLIPA, is a shell-like, cuffless structure that, upon insertion,
expands the soft tissues of the neck (Figs. 7.7 and 7.8). The tension of the elastic soft tissues surrounding the device
provides the sealing mechanism. The ventilatory opening is located within the shell in the periglottic area
4. The cuffed oropharyngeal airway, where the ventilatory opening is located at the base of the tongue and a sealing
surface is located in the oropharynx, is represented by the COPA (Figs. 7.9 and 7.10)
PAEXPRESS , pharyngeal airway express; LT, laryngeal tube; COBRA, Consolidated Omnibus Budget Reconciliation Act; SLIPA, streamlined liner
of the pharynx airway; COPA, cuffed oropharyngeal airway.
Data from: Ferson DZ, Tamm E, Chi L, et al. Comparative anatomical study of supraglottic airways. Scientific Paper A-602 presented at: Annual
Meeting of the American Society of Anesthesiologists. Las Vegas, Nevada, October 227, 2004.
116 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
H
H
T
T
C
C
LMA Classic
Laryngeal tube
FIGURE 7.5 Three-dimensional reconstruction from a
computed tomography study of the human anatomy with the
FIGURE 7.4 Three-dimensional reconstruction from a laryngeal mask airway (LMA Classic; LMA North America, San
computed tomography study of the human anatomy with a
Diego, CA) in situ. The LMA Classic consists of a ventilatory
laryngeal tube in situ. The cuff of the laryngeal tube surrounds opening surrounded by a cuff that forms a seal with the
the ventilatory tube and is located proximal to the ventilatory periglottic tissues. The hyoid bone (H), thyroid cartilage (T), and
openings. This design compartmentalizes the pharynx, with the
cricoid ring (C) represent anatomical structures of the pharynx
cuff serving as a sealing divider between the proximal and distal that define the interaction between the natural and artificial
pharyngeal compartments. The ventilatory openings are located airway.
in the distal pharyngeal compartment as defined by the hyoid
bone (H), thyroid cartilage (T), and cricoid ring (C).
H
T
C
ASPIRATION
Aspiration during general anesthesia has an overall
incidence that ranges between 1.4 and 6.5 per 10,000 cases
and a mortality rate of 5%.6871 In a more recent study
H H of 215,488 anesthetic cases, Warner et al.48 reported that
T T the incidence of aspiration was 2.6 per 10,000 in patients
C C undergoing elective surgery and 11 per 10,000 in patients
Ambu LM LMA Classic undergoing emergency procedures. A mortality rate of
0.14 per 10,000 in patients with an ASA classification
FIGURE 7.12 Three-dimensional reconstructions obtained from of III-V was also reported. Warner et als study did not
computed tomography studies of the AMBU laryngeal mask include patients whose airways were managed using the
(AMBU-LM; Ambu Inc., Linthicum, MD) and the corresponding LMA Classic.
size laryngeal mask airway (LMA Classic; LMA North America, The LMA Classic does not reliably protect against as-
San Diego, CA) inserted sequentially into the same patient. The piration, and therefore, should not be used electively in
AMBU-LM has a curved, semirigid shaft, which determines the patients who are at high risk for this complication. How-
final position of the cuff inside the patients pharynx. As a result, ever, as stated earlier, most clinicians do not adhere to
the cuff of the correctly placed AMBU-LM is positioned much the recommended insertion or fixation techniques when
higher in relation to the hyoid bone (H), the thyroid cartilage using the LMA Classic, which is designed so that the
(T), and the cricoid cartilage (C) than the cuff of the same size seal of the distal end of the device is against the upper
LMA Classic. The high position of the AMBU-LM cuff can cause esophageal sphincter. Even so, a meta-analysis of pub-
the hypoglossal nerve to be compressed between the cuff and the lished literature on the LMA Classic by Brimacombe and
hyoid bone, leading to hypoglossal nerve palsy. This Berry72 revealed that the overall incidence of pulmonary
complication is likely to occur mainly in patients in whom the aspiration with this LMA is around 2 per 10,000. In fact,
hypoglossal nerve passes slightly lower and crosses the medial 18 cases of suspected pulmonary aspiration occurring
aspect of the greater cornu of the hyoid bone. Also, the higher during LMA Classic use were found by meta-analysis, and
position of the tip of the AMBU-LM in relation to the cricoid only 10 were radiographically confirmed. Moreover, only
cartilage, as compared to that of the same size LMA Classic, will 3 patients required between 1 and 7 days of ventilatory
cause the AMBU-LM to form a less effective seal against the support, and none of the patients who aspirated through
upper esophageal sphincter than the LMA Classic. the LMA Classic suffered any long-term effects. Of par-
ticular interest is that most of these patients had one or
more predisposing factors for aspiration, including emer-
gency surgery in a patient who had not fasted, a difficult
airway, obesity, steep Trendelenburgs position with in-
medial aspect of the greater cornu of the hyoid bone.
traabdominal insufflation, and previous gastric surgery.
Although with the LMA Classic, this high position of the
The meta-analysis study only included reports published
cuff represents malpositioning of the device (too shallow through September 1993, but a subsequent analysis of
insertion), with the AMBU-LM, the high position of the 11,910 LMA Classic anesthetic cases by Verghese and
cuff occurs even when the device is placed correctly. This Brimacombe73 yielded an even smaller incidence of aspi-
is because the curved shaft of the AMBU-LM determines ration (0.84 per 10,000). Even allowing for the fact that
the final position of the cuff inside the patients pharynx. critical events in anesthesia are frequently underreported,
Therefore, the AMBU-LM can, in patients in whom the the absence of intensive care unit admissions for ven-
hypoglossal nerve passes slightly lower in relation to the tilatory support indicates that aspiration with the LMA
greater cornu of the hyoid bone, cause nerve injury, even Classic is rare. Nevertheless, one must always be meticu-
when the device is placed correctly. To minimize the risk lous about appropriate patient selection, correct insertion
of hypoglossal nerve palsy when using the AMBU-LM, one technique to obtain the optimal LMA Classic position,
needs to monitor carefully the pressure inside the cuff and strict attention to the appropriate depth of anesthesia,
keep it at or below 60 cm H2 O. Also, perhaps the use of the and constant vigilance during surgery. If, despite all of
AMBU-LM for prolonged procedures should be avoided. these measures, regurgitation or aspiration occurs, a plan
Other SLAs, such as the PAEXPRESS , LT, COBRA, and of action should be implemented as shown in Table 7.2.
COPA that provide a seal at the base of the tongue carry a The risk of aspiration with the LMA Classic will
potentially higher risk for lingual nerve injury, in addition probably be significantly reduced with the LMA-ProSeal
to hypoglossal nerve palsy, and compression of blood (LMA North America, San Diego, CA), which incorporates
vessels that supply the tongue, particularly the lingual a drainage tube into the LMA Classic design, warns
veins (Fig. 7.4). When using these devices, one needs to of an inadequate upper esophageal sphincter seal, and
pay careful attention to cuff inflation pressures. Currently, permits more reliable positive-pressure ventilation than
only the LTs manufacturer supplies a pressure gauge the LMA Classic. However, if the LMA-ProSeal is inserted
120 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
1. Do not attempt to remove the LMA Classic, because a significant amount of regurgitant fluid may be trapped behind
the LMA Classics cuff. One should consider that the cuff shields and protects the larynx from the trapped fluid, and
removing the LMA Classic may worsen the situation
2. Temporarily disconnect the circuit to allow drainage of the fluid while tilting the patients head down and to the side
3. Suction the LMA Classic and administer 100% oxygen
4. Ventilate the patient manually using low gas flows and small tidal volumes to minimize the risk of forcing any fluid
from the trachea into the small bronchi
5. Use a large fiberoptic bronchoscope to evaluate the tracheobronchial tree, and suction any remaining fluid
6. If aspiration below the vocal cords is confirmed, consider intubating the patient with an endotracheal tube and
institute appropriate treatment protocols
incorrectly, aspiration can still occur.74 This is particularly the distal pharyngeal compartment, thereby significantly
worrisome in situations where, during insertion, the increasing the risk of aspiration. Of the currently available
tip of the LMA-ProSeal folds backwards. With this pharyngeal or pharyngeal-esophageal tube SLAs, only the
malposition of the LMA-ProSeal, one can obtain a very LT has a separate cuff at the esophageal portion of the
good seal and deliver positive-pressure ventilation, but device that attempts to seal off the esophagus and mini-
because the entry to the esophagus is exposed, it is mize the risk of aspiration (Fig. 7.4). Also, in the newer LT
easy to distend the stomach and provoke regurgitation there is a drainage tube that connects to the esophagus.
and aspiration.75 Therefore, before attempting positive- Other devices, such as the PAEXPRESS and COBRA do not
pressure ventilation with the LMA-ProSeal, it is important have features to seal off the entrance to the esophagus.
to verify the correct position of the device by following The SLIPA has no features to prevent regurgitation.
the manufacturers recommendations or by attempting to However, as reported by Miller et al., the volume inside the
pass a gastric tube through the drainage tube of the LMA- SLIPA is quite large, so should regurgitation occur, a large
ProSeal. If the gastric tube cannot be advanced easily, the amount of the fluid could theoretically be stored within
clinician should be highly suspicious that the device is the device before aspiration takes place.76 The downside
malpositioned and correct the situation before applying
positive-pressure ventilation.
Other SLA devices differ significantly in their ability
to minimize the risk of aspiration. This is particularly
worrisome with devices like the Portex-Soft Seal and
AMBU-LM, which seem, to an untrained eye, to closely
resemble the LMA Classic. However, there are significant
differences in how the LMA Classic, the Portex-Soft H
Seal, and AMBU-LM interact with pharyngeal anatomy. T
Because the cuff of the Portex-Soft Seal is shaped slightly
C
differently than the cuff of the LMA Classic, the cuff
CPM
of the Portex-Soft Seal herniates at the level of the Tip
cricopharyngeal muscle (CPM), forming a channel that
significantly increases the risk of aspiration (see Fig. 7.13).
On the other hand, the cuff of the AMBU-LM closely
resembles that of the LMA Classic. However, because the LMA Classic Portex-soft seal
shaft of the AMBU-LM is curved, it causes the AMBU-LM
to be positioned higher in the pharynx. The higher position FIGURE 7.13 Three-dimensional reconstructions obtained
of the AMBU-LM causes its cuff tip to be positioned from computed tomography studies of the laryngeal mask
farther away from the upper esophageal sphincter, thereby airway (LMA Classic; LMA North America, San Diego, CA) and
potentially increasing the chances of regurgitation and the corresponding size Portex-Soft Seal (Portex Inc., Keene, NH)
aspiration (Fig. 7.12). laryngeal mask inserted sequentially into the same patient.
The pharyngeal or pharyngealesophageal tube SLAs Because the cuff of the Portex-Soft Seal has a slightly different
also differ in their ability to protect against regurgi- shape than the cuff of the LMA Classic, the cuff of the
tation and aspiration. From the design point of view, Portex-Soft Seal herniates at the level of the cricopharyngeal
these devices compartmentalize the pharynx, with the cuff muscle (CPM), forming a channel that significantly increases the
serving as a sealing divider between the proximal and dis- risk of aspiration. Because the LMA Classic design was based on
tal pharyngeal compartments. Because their ventilatory careful anatomical studies of the human pharynx, the device
openings are located in the distal pharyngeal compart- conforms well to the soft tissues of the pharynx and is positioned
ment, these SLAs allow for effective positive-pressure optimally in relation to the hyoid bone (H), thyroid cartilage (T),
ventilation. However, the circumferential cuff prevents and criocoid ring (C).
any regurgitated fluid from draining from the proximal to
C H A P T E R 7 / S A F E T Y A N D H A Z A R D S A S S O C I AT E D W I T H T R A C H E A L I N T U B AT I O N 121
of this concept is that should aspiration still occur with leaks occurred in only 2.7% of patients. Subsequently,
the SLIPA in place, the large volume of aspirate could lead Verghese and Brimacombe reported using the LMA
to more severe pulmonary injury. Classic successfully in 5,236 patients undergoing a variety
of surgical procedures under general anesthesia with
positive-pressure ventilation.84
There are basic principles that should be considered
When Can Positive-Pressure when using the LMA with positive-pressure ventilation
(see Table 7.3). If leaking does occur when using the
Ventilation Be Used Safely with LMA Classic for positive-pressure ventilation, one should
a Laryngeal Mask Airway? immediately investigate the cause and try to correct the
situation. If the device has been correctly secured and
At the time of the first independent trial of the LMA inflated, the problem will usually be attributable to the
Classic, the only size of the device available was #3 need for more relaxant or deeper anesthesia.
and the correct fixation technique, with an adequate
seal against the upper esophageal sphincter, had not yet
been developed.77 Therefore, in large patients, the device
was not very reliable when applying positive-pressure What Is the Maximum
ventilation. Additionally, because the LMA Classic concept
was very new, it was prudent to limit the use of the LMA
Duration that LMAs Can Be
Classic to patients who were breathing spontaneously. Used Safely for Prolonged
Even now, positive-pressure ventilation should be
considered an advanced use of the LMA Classic. To be
Procedures?
sure, clinicians should gain considerable experience using
this device in patients who are breathing spontaneously Although the maximum duration for which the LMA
before attempting controlled ventilation. However, after Classic can be safely used is not well established, a limit
the LMA Classic was introduced commercially in a range of two hours was suggested soon after it became available
of sizesfirst for adults and then for childrenmany LMA for wide clinical use.85 This recommendation was based
Classic users found that, with experience, this device could on the possible increased risk of aspiration or pharyngeal
be very effective during positive-pressure ventilation.78 morbidity. Notably, subsequent reports demonstrated that
Devitt et al.79 compared the effectiveness of the LMA the LMA Classic can be safely used during procedures
Classic with that of the endotracheal tube in 48 patients lasting up to eight hours in the hands of experienced
and showed that positive-pressure ventilation (range, 15 to users.86,87 In fact, in rare circumstances, the LMA Classic
30 cm H2 O) through the LMA Classic was adequate and has been used in the intensive care arena to provide
comparable to that achieved through the endotracheal effective respiratory support for 10 to 24 hours, with
tube. Epstein et al. studied the effectiveness of the LMA no evidence of any adverse effects.88 The LMA-ProSeal,
Classic with controlled ventilation in children 3 months to which is designed for positive-pressure ventilation and has
17 years old and concluded that the device was effective a gastric drainage tube to minimize the risk of aspiration,
and reliable.80 In another study, Epstein et al. established may be best suited for prolonged procedures.89 Then
that the LMA Classic airway seal pressures (25.9 to 31.2 cm again, regardless of the model used, clinicians must always
H2 O) were well maintained in children.81 remain vigilant and provide adequate anesthesia, thereby
Two separate large clinical studies of more than minimizing the risk of complications. Conjointly, keep in
7,000 patients led by Verghese et al. and Van Damme mind that if nitrous oxide is administered, it diffuses into
established that the LMA Classic is as effective as the the LMA Classic cuff and increases intracuff pressure. For
endotracheal tube for controlled ventilation.82,83 Van this reason, intracuff pressure must be closely monitored
Dammes study also assessed the airway seal pressures during prolonged use of this device. Additionally, hourly
of this device and demonstrated that at 15 cm H2 O or less, removal of a few milliliters of air from the cuff is a
TABLE 7.3 Basic Principles when Using a Laryngeal Mask Airway and Positive-Pressure Ventilation
1. Patient Selection: Most patients with normal lung compliance who have fasted can be successfully mechanically
ventilated with the LMA Classic
2. Size Selection: Select the largest LMA Classic size that is appropriate for the patient to prevent overinflating the cuff
in an attempt to compensate for inadequate seal pressure
3. Insertion Technique: Carefully follow the correct insertion technique to ensure optimal positioning of the LMA
Classic in the airway
4. Fixation Technique: Use the correct method of taping the LMA Classic for proper contact between the LMA Classic
tip and the esophagus, thereby preventing gastric insufflation
5. Auscultation: Always auscultate over the stomach to confirm that gastric insufflation is not taking place
6. Ventilatory Parameters: Limit tidal volumes to 68 mL/kg and control the ETCO2 by adjusting the respiratory rate
122 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
prudent practice to minimize the risk of mucosal ischemia association between different design characteristics and
associated with high intracuff pressure.90 increased risk of aspiration, nerve palsies, and trauma to
Currently, there is no information on the use of other the delicate pharyngeal and laryngeal structures.
SLAs for prolonged procedures. Carefully designed studies
are needed to evaluate the efficacy and safety of the new
SLAs in patients undergoing prolonged surgeries with
either spontaneous or controlled ventilation. KEY POINTS
1. Before the surgery, always obtain patients history
regarding airway management and perform a phys-
DEVICE PROBLEMS ical examination. During physical examination, look
specifically for signs indicating possible intubation
As expected, the LMA Classic was subjected to rigorous difficulties.
testing before clinical release to ensure compliance with 2. Never use excessive force during laryngoscopy. If dur-
medical industry standards. Still, a number of device mal- ing laryngoscopy, the larynx is difficult to visualize,
functions have been reported with this device, including change to a different intubation technique, as recom-
separation of the cuff from the shaft,91,92 fragmentation mended by the ASA Difficult Airway Algorithm.
of the device inside of the patient,93 kinking of the shaft,94 3. Extubation carries a separate set of hazards and
and failure of the cuff to inflate or deflate.95,96 Most of requires the same level of vigilance and clinical
these complications are related to the use of the device attention as does intubation. Patients at increased
beyond the manufacturers recommendations, the use of risk include those with significant cardiopulmonary
the wrong chemicals for sterilization, and coating the de- disease, those who have undergone major abdominal
vice with silicone lubricants, which is not recommended or thoracic surgery, and the morbidly obese.
for the LMA Classic. Strict adherence to the manufac- 4. In the postoperative period, one should always
turers instructions for cleaning, sterilization, and use is evaluate the patient for any signs of complications
of paramount importance, especially the recommended related to intubation. Early detection and timely
limit of 40 uses per device. If the device malfunctions, the intervention greatly improves the chances for a good
clinician should consult the manufacturers manual for outcome.
instructions. Conceivably, the device problems linked to 5. SLAs represent a diverse group of airway devices,
overuse and incorrect sterilization techniques will be elim- and it appears that different complications can result
inated by the use of the disposable LMA Classic models. In from different design characteristics.
any case, all LMAs should always be inspected and tested 6. Of all the SLAs, the LMA Classic has currently the
before use to ensure optimal performance and safety. best record of safety, with no deaths linked directly to
its design.
7. Complications associated with use of the SLA include
CONCLUSIONS oropharyngeal morbidity, vascular compression and
nerve damage, and aspiration. There appears to
Complications associated with airway devices can occur be a direct relationship between the skill level
during insertion of the device as well as during surgery and experience of the clinician and incidence of
while the device is in place. Problems associated with complications reported with the use SLA devices.
tracheal intubation most commonly result from direct
laryngoscopy and intubation, the pressure of the tracheal
REFERENCES
tube or its cuff on airway structures, and complications
following extubation. Complications associated with the 1. MacEwen W. Clinical observations on the introduction
use of SLAs are not as well defined because SLAs represent of tracheal tubes by the mouth instead of performing
a diverse group of airway devices, and it appears that tracheotomy or laryngotomy. Br Med J. 1880;2:122.
2. Waters RM, Guedel AE. Endotracheal anesthesia: A new
different complications can result from different design
technique. Ann Otol Rhinol Laryngol. 1931;40:1139.
characteristics. Of all the SLAs, the LMA Classic has the
3. Rowbotham ES, Magill IW. Anaesthetics in plastic surgery
best record of safety, having been used safely in over of the face and jaws. Proc R Soc Med. 1921;14:17.
200 million patients worldwide, with no deaths linked 4. Waters RM, Rovenstine EA, Guedel AE. Endotracheal
directly to its design. This extremely impressive record of anesthesia and its historical development. Anesth Analg.
safety is a result of the anatomically and physiologically 1933;12:196.
based approach of Archie Brain, the inventor of the LMA 5. Goldberg W, Mueller C, Panacek E, et al. Distribution of
Classic. However, what is most important is that before patterns of blunt traumatic cervical spine injury. Ann Emerg
the LMA Classics design was finalized, the device was Med. 2001;38:17.
evaluated in more than seven thousand patients. No other 6. Ollerton JE, Parr MJA, Harrison K, et al. Potential cer-
vical spine injury and difficult airway management for
SLA has been designed with the same degree of preclinical
emergency intubation of trauma adults in the emergency
testing. Therefore, clinicians should always be vigilant for departmenta systematic review. Emerg Med J. 2006;23:3.
any signs of possible airway compromise that may point 7. Lennarson PJ, Smith DW, Sawin PD, et al. Cervical spine mo-
to an impending complication. Also, well designed studies tion during orotracheal intubation: Efficacy of stabilization
are clearly needed to evaluate the safety and effectiveness maneuvers in the setting of complete segmental instability.
of the new SLAs, with special attention paid to the J Neurosurg. 2001;94:265.
C H A P T E R 7 / S A F E T Y A N D H A Z A R D S A S S O C I AT E D W I T H T R A C H E A L I N T U B AT I O N 123
8. Russell WJ, Morris RG, Frewin DB, et al. Changes in 28. American Society of Anesthesiologists. Practice guidelines
plasma catecholamine concentrations during endotracheal for management of the difficult airway. An updated report
intubation. Br J Anaesth. 1981;53:837. by the American Society of Anesthesiologists Task Force
9. Derbyshire DR, Chmielwewski A, Fell D, et al. Plasma on management of the difficult airway. Anesthesiology.
catecholamine responses to tracheal intubation. Br J Anaesth. 2003;98:1269.
1983;55:855. 29. Filip L, Dinca M. Pulmonary suppurations: Etiologic profile.
10. Lowery DW, Wald MM, Browne BJ, et al. Epidemiology of Pneumologia. 2005;54:5.
cervical spine injury victims. Ann Emerg Med. 2001;38:12. 30. Takiguchi Y, Terano T, Hirai A. Lung abscess caused by
11. Chiu WC, Haan JM, Cushing BM, et al. Ligamentous injuries Actinomyces odontolyticus. Intern Med. 2003;42:723.
of the cervical spine in unreliable blunt trauma patients: 31. Mansharamani N, Balachandran D, Delaney D, et al. Lung
Incidence, evaluation, and outcome. J Trauma. 2001;50:457. abscess in adults; clinical comparison of immunocompro-
12. Crosby ET. Airway management in adults after cervical spine mised to non-immunocompromised patients. Respir Med.
trauma. Anesthesiology. 2006;104:1293. 2002;96:178.
13. Berne JD, Velmahos GC, El-Tawil Q, et al. Value of 32. McHardy FE, Chung F. Postoperative sore throat: Cause,
complete cervical helical computed tomographic scanning prevention, and treatment. Anaesthesia. 1999;54:444.
in identifying cervical spine injury in the unevaluable blunt 33. Ooi GC, Irwin MG, Lam LK, et al. An unusual complication
trauma patient with multiple injuries: A prospective study. of emergency tracheal intubation. Anaesthesia. 1997;52:154.
J Trauma. 1999;47:896. 34. Pan PH. Perioperative subcutaneous emphysema: Review
14. Ghafoor AU, Martin TW, Gopalakrishnan S, et al. Caring of differential diagnosis, complications, management, and
for the patients with cervical spine injuries: What have we anesthetic implications. J Clin Anesth. 1989;1:457.
learned? J Clin Anesth. 2005;17:640. 35. Rieger A, Hass I, Gross M, et al. Intubation trauma of the
15. Chung F, Evans D. Low-dose fentanyl: Haemodynamic larynxa literature review with special reference to arytenoid
response during induction and intubation in geriatric cartilage dislocation. Anasthesiol Intensivmed Notfallmed
patients. Can Anaesth Soc J. 1985;32:622. Schmerzther. 1996;31:281.
16. Chraemmer-Jorgensen B, Hoilund-Carlsen PF, Marving J, 36. Mikuni I, Suzuki A, Takahata O, et al. Arytenoid dislocation
et al. Left ventricular ejection fraction during anaesthetic caused by the double-lumen endobronchial tube. Br J
induction: Comparison of rapid-sequence and elective in- Anaesth. 2006;96:136.
duction. Can J Anaesth. 1986;33:754. 37. Young PJ, Ridley SA. Ventilator-associated pneumo-
17. Carabine UA, Allen RW, Moore J. Partial attenuation of the nia. Diagnosis, pathogenesis, and prevention. Anaesthesia.
pressor response to endotracheal intubation. A comparison 1999;54:1183.
of the effects of intravenous clonidine and fentanyl. Eur J 38. Young PJ, Rollinson M, Downward G, et al. Leakage of fluid
Anaesthesiol. 1992;9:325. past the tracheal tube cuff in a benchtop model. Br J Anaesth.
18. Ebert TJ, Bernstein JS, Stove DF, et al. Attenuation of 1997;78:557.
hemodynamic responses to rapid sequence induction and 39. Young PJ, Pakeerathan S, Blunt MC, et al. A low-volume, low-
intubation in healthy patients with a single bolus of esmolol. pressure tracheal tube cuff reduces pulmonary aspiration.
J Clin Anesth. 1990;2:243. Crit Care Med. 2006;34:632.
19. Mikawa K, Nishina K, Maekawa N, et al. Comparison of 40. Shafei H, El-Kholy A, Azmy S, et al. Vocal cord dysfunction
nicardipine, diltiazem and verapamil for controlling the after cardiac surgery: An overlooked complication. Eur J
cardiovascular responses to tracheal intubation. Br J Anaesth. Cardiothorac Surg. 1997;11:564.
1996;76:221. 41. Austin RL. Respiratory problems in emergence from anes-
20. Laurito CE, Baughman VL, Becker GL, et al. Effects of thesia. Int Anesthesiol Clin. 1991;29:25.
aerosolized and/or intravenous lidocaine on hemodynamic 42. Stoelting RK, Eger EI. II The effects of ventilation and anes-
responses to laryngoscopy and intubation in outpatients. thetic solubility on recovery from anesthesia. Anesthesiology.
Anesth Analg. 1998;67:389. 1969;30:290.
21. Helfman SM, Gold MI, DeLisser EA, et al. Which drug 43. Baillard C, Gehan G, Rebou-Marty J, et al. Residual
prevents tachycardia and hypertension associated with curarization in the recovery room after vecuronium. Br J
tracheal intubation: Lidocaine, fentanyl, or esmolol? Anesth Anaesth. 2000;84:394.
Analg. 1991;72:482. 44. Beaussier M, Boughaba MA. Residual neuromuscular block-
22. Martin DE, Rosenberg H, Aukburg SJ, et al. Low-dose ade. Ann Fr Anesth Reanim. 2005;24:1245.
fentanyl blunts circulatory responses to tracheal intubation. 45. Milliet N. Recurarization in the recovery room. Eur J
Anesth Analg. 1982;61:680. Anaesthsiol. 1999;16:493.
23. Cucchiara RF, Mahla ME. Anesthesia for patients with 46. Torda TA. Monitoring neuromuscular transmisson. Anaesth
increased intracranial pressure. ASA Refresher Courses. Intensive Care. 2002;30:123.
1993;21:177. 47. Osmer C, Vogele C, Zickmann B, et al. Comparative use
24. Antal M. Ketamine anaesthesia and intraocular pressure. of muscle relaxants and their reversal in three European
Ann Ophthalmol. 1978;10:1281. countries: A survey in France, Germany and Great Britain.
25. Ko SH, Kim DC, Han YJ, et al. Small-dose fentanyl: Optimal Eur J Anaesthsiol. 1996;13:389.
time of injection for blunting the circulatory responses to 48. Warner MA, Warner ME, Weber JG. Clinical significance
tracheal intubation. Anesth Analg. 1998;86:658. of pulmonary aspiration during the perioperative period.
26. Mort T. Emergency tracheal intubation: Complications Anesthesiology. 1993;78:56.
associated with repeated laryngoscopic attempts. Anesth 49. Capatanio MK, Kirkpatrick JA. Obstruction of the upper
Analg. 2004;99:607. airway in children as reflected on the chest radiograph.
27. American Society of Anesthesiologists. Practice guidelines Pediatr Radiol. 1997;107:159.
for management of the difficult airway. A report by the 50. Oswalt CE, Gates GA, Holmstrom FMG. Pulmonary edema
American Society of Anesthesiologists Task Force on man- as a complication of acute upper airway obstruction. JAMA.
agement of the difficult airway. Anesthesiology. 78:597, 1993. 1977;238:1833.
124 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
51. Lang SA, Duncan PG, Shephard DAE. Pulmonary oedema 73. Verghese C, Brimacombe J. Survey of laryngeal mask usage
associated with airway obstruction. Can J Anaesth. in 11910 patientssafety and efficacy for conventional and
1990;37:210. nonconventional usage. Anaesth Analg. 1996;82:129.
52. Williams PJ, Bailey PM. Comparison of reinforced laryngeal 74. Koay CK. A case of aspiration using the ProSeal LMA.
mask airway and tracheal intubation for adenotonsillectomy. Anaesth Intensive Care. 2003;31:123.
Br J Anaesth. 1993;70:30. 75. Brimacombe J, Keller C. Aspiration of gastric contents
53. Ferson DZ, Tamm E, Chi L, et al. Comparative anatom- during use of a ProSeal laryngeal mask airway sec-
ical study of supraglottic airways. Scientific Paper A-602, ondary to unidentified foldover malposition. Anesth Analg.
presented at: Annual Meeting of the American Society of Anes- 2004;99:312.
thesiologists. Las Vegas, Nevada, October 227, 2004. 76. Miller DM, Light D. Laboratory and clinical comparisons of
54. Rieger A, Brunne B, Hass I, et al. Laryngo-pharyngeal com- the Streamlined liner of the pharynx airway (SLIPA) with
plaints following laryngeal mask airway and endotracheal the laryngeal mask airway. Anaesthesia. 2003;58:136.
intubation. J Clin Anesth. 1997;9:42. 77. Brain AIJ, Brimacombe JR. Alternative insertion techniques.
55. Asai T, Murao K, Yukawa H, et al. Re-evaluation of In: Brimacombe JR, Brain AIJ, Berry AM, eds. The laryngeal
appropriate size of the laryngeal mask airway. Br J Anaesth. mask airwayA review and practical guide. London: WB
1999;83:478. Saunders; 1997:83.
56. Morikawa M. Vocal paralysis after use of the LM. J Clin 78. Wainwright AC. Positive pressure ventilation and the laryn-
Anesth. 1992;16:1194. geal mask airway in ophthalmic anaesthesia. Br J Anaesth.
57. Inomata S, Nishikiwa T, Suga A, et al. Transient biliteral 1995;75:249.
vocal cord paralysis after insertion of a laryngeal mask 79. Devitt JH, Wenstone R, Noel AG, et al. The laryngeal mask
airway. Anesthesiology. 1995;82:787. airway and positive-pressure ventilation. Anesthesiology.
58. Wynn JM, Jones KL. Tongue cyanosis after laryngeal mask 1994;80:550.
insertion. Anesthesiology. 1994;80:1403. 80. Epstein RH, Ferouz F, Jenkins MA. Airway sealing pressures
59. Lloyd Jones FR, Hegab A. Recurrent laryngeal nerve palsy of the laryngeal mask airway in children. Anesthesiology.
after laryngeal mask insertion. Anaesthesia. 1996;51:171. 1994;81:A1322.
60. Figueredo E, Vivar-Diago M, Munoz-Blanco F. Laryngo- 81. Epstein RH, Ferouz F, Jenkins MA. Airway sealing pressures
pharyngeal complaints after use of the laryngeal mask of the laryngeal mask in pediatric patients. J Clin Anesth.
airway. Can J Anaesth. 1999;46:220. 1996;8:93.
61. Keller C, Sparr HJ, Brimacombe JR. Laryngeal mask 82. Verghese C, Smith TCG, Young E. Prospective survey of
lubrication. A comparative study of saline versus 2% the use of the laryngeal mask airway in 2359 patients.
lignocaine gel with cuff pressure control. Anaesthesia. Anaesthesia. 1993;48:58.
1997;52:592. 83. Van Damme E. Die Kehlkopfmaske in der ambu-
62. Brimacombe J, Berry A. Laryngeal mask airway cuff pressure lanten anasthesiaeine antwertung vom 5000 ambulanten
and position during anaesthesia lasting one to two hours. Can narkosen. Anaesthesiol Intensivmed Notfalmed Schmerzther.
J Anaesth. 1994;41:589. 1994;29:284.
63. Grady DM, McHardy F, Wong J, et al. Pharyngolaryn- 84. Verghese C, Brimacombe J. Survey of laryngeal mask usage
geal morbidity with the laryngeal mask airway in sponta- in 11910 patientssafety and efficacy for conventional and
neously breathing patients: Does size matter? Anesthesiology. nonconventional usage. Anesth Analg. 1996;82:129.
2001;94:760. 85. Asai T, Morris S. The laryngeal mask airway: Its features,
64. Brimacombe J, Holyoake L, Keller C, et al. Pharyngolaryn- effects, and role. Can J Anaesth. 1994;41:930.
geal, neck, and jaw discomfort after anesthesia with the 86. Brimacombe J, Shorney N. The laryngeal mask airway and
face mask and laryngeal mask airway at high and low cuff prolonged balanced anesthesia. Can J Anaesth. 1993;40:
volumes in males and females. Anesthesiology. 2000;93:26. 360.
65. Brimacombe J, Holyoake L, Keller C, et al. Emergence char- 87. Brimacombe J, Archeacon J. The laryngeal mask air-
acteristics and postoperative laryngopharyngeal morbidity way for unplanned prolonged procedures. Can J Anaesth.
with the laryngeal mask airway: A comparison of high versus 1995;42:1176.
low initial cuff volume. Anaesthesia. 2000;55:338. 88. Arosio EM, Conci F. Use of the laryngeal mask airway for
66. Ouellette RG. The effect of nitrous oxide on laryngeal mask respiratory distress in the intensive care unit. Anaesthesia.
cuff pressure. Am Assoc Nurse Anesth J. 2000;68:411. 1995;50:635.
67. Brain AIJ. Course of the hypoglossal nerve in relation to 89. Braun U, Zerbst M, Fullekrug B, et al. A comparison of the
the position of the laryngeal mask airway. Anaesthesia. Proseal laryngeal mask to the standard laryngeal mask on
1995;50:82. anesthesized, non-relaxed patients. Anasthesiol Intensivmed
68. Cohen MM, Duncan PG, Pope WDB, et al. A survey of Notfallmed Schmerzther. 2002;37:727.
112000 anaesthetics at one teaching hospital. Can Anaesth 90. Hamakawa T. Sore throat after the use of the LM. J Clin
Soc J. 1986;33:22. Anesth. 1993;17:245.
69. Olsson GL, Hallen B, Hambraeus Jonzon K. Aspiration 91. Vickers R, Springer A, Hindmarsh J. Problem with the
during anesthesia: A computer-aided study of 185,358 laryngeal mask airway. Anaesthesia. 1994;45:892.
anaesthetics. Acta Anaesthesiol Scand. 1986;30:84. 92. Khoo ST. The laryngeal mask airwayan unusual complica-
70. Leigh JM, Tytler JA. Admissions to the intensive care unit tion. Anaesth Intensive Care. 1993;21:249.
after complications of anaesthetic techniques over 10 years. 93. Squires SJ. Fragmented laryngeal mask airway. Anaesthesia.
Anaesthesia. 1990;45:814. 1992;47:274.
71. Tiret L, Desmonts JM, Hatton F, et al. Complications 94. Martin DW. Kinking of the laryngeal mask airway in two
associated with anaesthesia: A prospective survey in France. children. Anaesthesia. 1990;45:488.
Can Anaesth Soc J. 1986;33:336. 95. George A. Failed cuff inflation of a laryngeal mask. Anaes-
72. Brimacombe J, Berry A. The incidence of aspiration asso- thesia. 1994;49:80.
ciated with the laryngeal mask airwaya meta-analysis of 96. Richards JT. Pilot tube of the laryngeal mask airway.
published literature. J Clin Anesth. 1995;7:297. Anaesthesia. 1994;49:450.
CHAPTER HYPOXEMIA AND HYPERCAPNIA
8 Roger S. Mecca
O
asks you to evaluate his first elective anes-
thetic in the ambulatory surgical unit that
day. A 42-year-old, obese woman is nearing How, When, Where, and Why
the completion of a laparoscopic cholecys-
tectomy with general, tracheal anesthesia Does Hypoxemia Occur in
utilizing sevoflurane, nitrous oxide, and oxygen (O2 ), with Anesthesia?
intravenous fentanyl and neuromuscular paralysis. Dur-
ing the last 60 minutes, she has exhibited a progressive
increase in the end-expired carbon dioxide (CO2 ) concen-
tration, from 35 to 48 mm Hg, coincident with a gradual PHYSIOLOGY OF OXYGEN
decrease in arterial O2 saturation (SpO2 ) from 99% to
In the perioperative interval, impairment of a patients
92% while she is breathing an inspired O2 concentration
ventilatory function by mechanical, hemodynamic, and
of 60%. Her systemic blood pressure and heart rate also
pharmacologic factors often manifests as a reduction in
have increased by approximately 20% during this period.
Your colleague assumed that these changes would resolve the partial pressure of arterial O2 (PaO2 ) or an increase
with termination of CO2 peritoneal insufflation; he tempo- in the partial pressure of arterial CO2 (PACO2 ).14 Clinical
rized by increasing the fraction of inspired O2 (FiO2 ) and etiologies for these problems are often multifactorial and
minute ventilation. However, deflation of the peritoneal sometimes are difficult to diagnose.
cavity had no effect on the end-expired CO2 or arterial O2
saturation. Restoration of spontaneous ventilation after Alveolar Partial Pressure
reversal of neuromuscular blockade only worsened the
picture. Several key physiologic variables influence a patients abil-
The patients chart reveals a 25-year history of smok- ity to maintain the PaO2 in systemic blood (i.e., arterial
ing, chronic asthma, and recent upper respiratory in- oxygenation). The alveolar partial pressure of O2 (PAO2 )
fection. She works in an industrial environment with provides the driving pressure for the diffusion gradient
potential exposure to chemicals, and has a brother who that moves O2 passively from exchanging airspaces into
suffered a serious problem during general anesthesia pulmonary capillary blood. The PAO2 varies with the FiO2 ,
for tonsillectomy. Physical examination reveals a spon- with the minute ventilation, and with the rate of O2 ex-
taneously ventilating patient with a respiratory rate of traction from the lungs by the pulmonary capillary blood.
30 breaths per minute, who is actively assisted with posi- Minute ventilation, in turn, is affected by the ventilatory
tive pressure in the anesthesia circuit. Auscultation of the drive, airway resistance, lung and chest wall compliance,
chest reveals somewhat distant breath sounds that seem neuromuscular status, and a host of other factors.
diminished in the left hemithorax. Mild end-expiratory
wheezing is detected. Ventilation/Perfusion Matching
Your concerned colleague asks your opinion about
the most likely causes of the patients respiratory status Arterial oxygenation is also affected by ventilation/
and requests your guidance on possible interventions to perfusion (V/Q) matching in the lungs, which determines
proceed with extubation. This chapter provides some the degree of contact between fresh gas in the airspaces
information that may answer the posed questions and and pulmonary capillary blood. Normally, hypoxic pul-
elucidates the surgical and anesthetic causes of hypoxemia monary vasoconstriction regulates the distribution of
125
126 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
pulmonary blood flow to match the distribution of venti- unreliable.9 Measuring the oxygenation of specific tissues
lation.5 Ventilation/perfusion mismatch generates hypox- or organs, although promising, is still of limited clinical
emia when airspaces do not receive sufficient O2 in fresh usefulness.10
gas to fully saturate the hemoglobin. Perfused regions that The appropriate lower limit for PaO2 varies with
receive no ventilation at all send blood with the systemic individual patient characteristics and clinical circum-
mixed venous partial pressure of O2 (PvO2 ) to the left stances.11 Although the life-threatening range for hypox-
heart, generating a true shunt across the lung. Regions emia is recognized,12 the lowest acceptable values during
with relatively low levels of ventilation compared with routine care are a matter of ongoing discussion,1315
perfusion (i.e., low V/Q) allow hemoglobin that is only especially because providers appreciate the degree of de-
partially saturated to reach the left heart. Low V/Q areas saturation that many individuals exhibit during normal
are caused by clinical conditions that decrease ventilation sleep.16 A PaO2 below 65 to 70 mm Hg causes significant
to an area with high perfusion or by factors that increase hemoglobin desaturation, although O2 delivery can be
perfusion to an area with low ventilation. In any of these maintained at lower levels. Generally, maintaining PaO2
circumstances, admixture of blood containing unsatu- above 80 mm Hg (saturation 93%) ensures adequate ar-
rated hemoglobin in the pulmonary veins and left atrium terial O2 content, assuming a reasonable hemoglobin
results in equilibration of O2 among red cells, decreasing concentration. Maintaining a high SpO2 also offers a
PaO2 and hemoglobin saturation. Peripheral O2 consump- time buffer against life-threatening hypoxemia should an
tion exerts an indirect effect on arterial oxygenation by acute event interrupt ventilation. Elevating PaO2 above
lowering the PvO2 . Usually, a low venous O2 content only 110 mm Hg offers negligible improvement in O2 -carrying
impacts PaO2 if PAO2 is marginal or if significant low V/Q capacity because hemoglobin is almost fully saturated,
units or shunting exist. and the incremental amount of O2 dissolved in plasma is
miniscule. Dissolved O2 has greater significance in min-
Central Nervous System Regulation imally perfused tissues during hyperbaric O2 therapy.
A high PaO2 may also improve wound healing, infec-
The body utilizes PaO2 to regulate spontaneous venti- tion rates, or postoperative nausea.17 During weaning
latory drive.6 Reduction of PaO2 increases the afferent from mechanical ventilation, a sustained PaO2 above
output from chemoreceptors in the carotid bodies and 80 mm Hg, with FiO2 of 0.4 or less and positive end-
central nervous system (CNS) to the ventilatory center in expiratory pressure (PEEP) or continuous positive airway
the medulla, generating an increase in spontaneous venti- pressure (CPAP) of up to 5 cm H2 O, should predict ade-
latory rate and depth. Also, insufficient delivery of tissue quate oxygenation after extubation.
O2 leads to anaerobic metabolism and lactic acidemia. Adequate PaO2 does not guarantee that cardiac
A spontaneously breathing patient will hyperventilate to output, arterial perfusion pressure, circulating red cell
generate a respiratory alkalemia to compensate for the mass, or peripheral blood flow distribution will support
metabolic acidemia. Hyperventilation, therefore, is a ma- tissue oxygenation. Hypotension, anemia, hemoglobin
jor compensatory response to systemic hypoxemia that abnormalities, or sepsis can generate profound tissue
increases O2 delivery to airspaces and perhaps expands ischemia in spite of a high PaO2 .
the functional residual capacity (FRC).
TABLE 8.1 Causes of Hypoxemia Related to a Global can fall dramatically, unless ventilation is intermittently
Reduction of Alveolar Partial Pressure of Oxygen supplied. Preintubation hyperventilation with 100% O2
and attention to the duration of apnea help avoid this
Severe airway obstruction problem. Esophageal intubation stops effective positive-
Inadequate positive-pressure ventilation pressure ventilation and causes a precipitous decline of
Suppression of ventilatory drive PAO2 . In a patient who is dependent on mechanical ven-
Disordered mechanics of ventilation tilation, other ominous problems that reduce PAO2 are
Neuromuscular paralysis failure to activate the mechanical ventilator, failure to set
Administration of a hypoxic mixture a sufficient rate or tidal volume, failure to supply hand-
Excessive alveolar concentration of a second gas bag ventilation, ventilator or bellows malfunction, and
unrecognized disconnection.
decreases the value of pulse oximetry for monitoring Inadequate Spontaneous Ventilation
ventilation.18 If hypoventilation is profound, or if ven-
tilation ceases, PaO2 rapidly declines at a rate that varies Spontaneously breathing patients frequently suffer hy-
with age, body habitus, underlying illness, and the initial poventilation and decreased PAO2 when medications de-
PAO2 .12 press the ventilatory drive. Opioids are potent ventilatory
suppressants, as are volatile anesthetics, sedative agents,
induction agents, and some antiemetics.3036 The depres-
Obstructive Sleep Apnea sant effects of these different agents are synergistic, and
During anesthesia, hypoxemia related to a global reduc- CNS depression often generates some upper airway ob-
tion of PAO2 frequently reflects suboptimal technique.1921 struction as well. In patients with OSA, the depressant
Loss of airway patency in spontaneously ventilating pa- effects of opioids and sedatives likely accentuate the fre-
tients is a common cause. Partial airway obstruction quency and depth of baseline desaturations that occur in
increases airway resistance, leading to hypoventilation, these patients during sleep.23,37,38 Ventilatory suppression
and sometimes hypercapnia. However, partial obstruc- and hypoxemia is a leading cause of morbidity during
tion alone does not usually reduce PAO2 to dangerously monitored anesthetic care and deep sedation.39 Local
low levels, especially with supplemental O2 administra- anesthetic toxicity secondary to inadvertent intravascu-
tion.22 Greater obstruction will cause a rapid decline lar injection or uptake can cause profound ventilatory
in PAO2 ; patients with obstructive sleep apnea (OSA) depression and airway obstruction after the administra-
are particularly vulnerable.23 Significant OSA occurs not tion of regional anesthetics. If local anesthetics reach the
only in obese patients but also in adults with average intracranial cerebrospinal fluid, ventilatory depression is
body habitus and in children.24 In patients with OSA, often immediate and complete. Depression from the cen-
airway dynamics are complex and unpredictable,25 man- tral spread of neuraxial opioids is often more insidious in
dating a high level of vigilance for adverse ventilatory onset.40
events.26 Impaired ventilatory mechanics can also lead to
hypoventilation sufficient to cause hypoxemia. An obvious
example is neuromuscular paralysis. However, the loss of
Airway Management intercostal and diaphragmatic muscle strength caused by
the spread of spinal or epidural anesthetics to higher levels
Inadequate positive-pressure ventilation is another com-
can seriously impair ventilation, as can reduced lung
mon cause of decreased PAO2 secondary to hypoventila-
or chest wall compliance secondary to pneumothorax,
tion. Loss of upper airway patency impairs the ability to
increased lung water, or extremes of position during
ventilate with positive pressure. Obstruction of this degree
surgery or recovery.41 Disruption of skeletal or muscular
can be caused by soft tissue, space-occupying lesions, for-
eign bodies, gastric contents, or airway edema secondary integrity in traumatic conditions, such as flail chest or
to trauma.27,28 Laryngospasm or bronchospasm can also diaphragmatic rupture, also cause hypoxemia. Children
impede ventilation.29 The presence of an airway device, with active or recent upper respiratory infection are
such as an endotracheal tube, does not automatically prone to breath-holding, severe straining, and arterial
eliminate airway obstruction as a cause of hypoxemia. desaturations below 90% during induction and recovery.
Kinking, luminal occlusion by clots or inspissated se- This problem is increasingly likely after intubation and/or
cretions, inadvertent extubation, or misplacement of a airway surgery, or if they have reactive airway disease or
laryngeal mask or oral airway can all cause upper airway secondhand smoke exposure.42
obstruction.
Coughing, straining, or chest wall rigidity will impede Gas Delivery
effective positive-pressure ventilation, as will a poor face
mask seal or a loss of pressure in the anesthesia circuit Hypoxemia from global reduction of PAO2 may result
due to leakage. Excessive leakage past an endotracheal from delivery of a hypoxic mixture of inspired gases
cuff, inadequate sealing of a laryngeal mask, or inad- (e.g., an inordinately high concentration of nitrous oxide
vertent tracheal placement of a gastric tube can reduce without sufficient O2 ) during anesthesia. With the excep-
PAO2 . During difficult intubation or tracheostomy, PAO2 tion of a gas pipe misconnection during maintenance or
128 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
renovation, safeguards on contemporary anesthesia ma- TABLE 8.2 Hypoxemia Related to V/Q Mismatching
chines, such as interlink systems, pin indexed mounts,
low pressure shutoffs, and O2 analyzers in the anesthesia Suboptimal Distribution of Ventilation
circuit, make delivery of a hypoxic mixture almost im-
possible. Redundant O2 sources such as transport and Obesity/increased intra-abdominal pressure
machine backup cylinders provide additional safety. Peritoneal insufflation
Rarely, an excessive alveolar partial pressure of an- Thoracic restriction/splinting
other gas may cause hypoxemia. Rapid outpouring of Hypoexpansion from low tidal volumes
nitrous oxide into alveoli can theoretically displace alveo- Extreme surgical positioning
lar O2 and lower PAO2 . The risk of this diffusion hypoxia Abdominal or thoracic retraction
is greatest immediately after discontinuation of nitrous Leaning on the patient by surgical assistants
oxide, especially if a patient is breathing ambient air Upper airway obstruction
or hypoventilating. The problem is easily countered by Retained secretions/small airway obstruction
ventilation for a short period with 100% O2 . Volume Mainstem intubation/one-lung ventilation
displacement of O2 can also occur with severe hyper- Increased lung water/pulmonary edema
capnia, again, if a patient is breathing ambient air. Aspiration pneumonitis
Serious respiratory acidemia usually precedes hypox- Pneumothorax/hemothorax
emia. Acute lung injuries (ARDS, TRALI, SIRS)
Increasing the O2 content in the FRC with supple- Absence of expiratory resistance
mental O2 helps to safeguard against hypoxemia caused Tracheal suctioning
by airway obstruction or hypoventilation. However, sup- Suboptimal Distribution of Perfusion
plemental O2 does not address the underlying cause of
hypoxemia and may limit the use of peripheral pulse Dependent positioning of diseased lung
oximetry as an early predictor of insidious hypoven- Vasodilators, anesthetic agents
tilation.18,22,43 Whether this possibility outweighs the Hepatic cirrhosis
benefits of supplemental O2 is a matter of individual Sepsis/circulating endotoxin
judgment. Changes in pulmonary arterial pressure
with volume and inversely with the pH of the acidic placing atelectatic or diseased lung tissue in a depen-
aspirate. The degree of dissemination of aspirate into dent position. Placing poorly ventilated parenchyma in a
small airways also impacts morbidity, as does the pres- nondependent, up position may improve V/Q matching,
ence of partially digested food or vegetable matter. In but could promote drainage of purulent material from
serious cases, epithelial degeneration, alveolar edema, diseased lung segments into unaffected areas. Avoiding -
and hemorrhage into air spaces progresses to ARDS with mimetic or vasodilatory medications may improve PaO2 ,
high-permeability pulmonary edema. but the benefits from using the medication almost always
The incidence of hypoxemia secondary to aspiration outweigh the drawback of impaired hypoxic pulmonary
during anesthesia is relatively low, but the possibility is vasoconstriction. Maintaining pulmonary artery and air-
always present. The greatest risk occurs between loss of way pressures within acceptable ranges likely optimizes
consciousness and intubation, but significant aspiration any yield from hemodynamic interventions aimed at im-
can occur anytime that airway reflexes are compromised. proving V/Q matching.
A well-seated laryngeal mask airway or an endotracheal
tube with an inflated tracheal cuff does not guarantee that
liquid in the pharynx will not gain access to the trachea.
During recovery, the incidence of aspiration is lower Why Does Reduced Mixed
but still significant. Vomiting after anesthesia remains
prevalent,70 especially if gas has accumulated in the
Venous Oxygen Content Cause
stomach. Protective airway reflexes are often depressed Hypoxemia?
by residual anesthetics, persisting laryngeal nerve blocks,
or residual neuromuscular paralysis.74,75 A patient who Mixed venous O2 is affected by arterial O2 content, cardiac
passes a head lift test and exhibits a train-of-four T4/T1 output, distribution of peripheral blood flow, and tissue
ratio >0.7 can still exhibit impaired airway reflexes O2 extraction. If arterial O2 content decreases or tissue
secondary to residual paralysis. The T4/T1 ratio may need extraction increases, PvO2 falls. The lower the PvO2 in
to exceed 0.9 to assure that the reflexes are restored.76 The blood that is shunted or flows through low V/Q units,
risk of aspiration also increases if reversal is omitted.77 the greater will be the reduction of PaO2 . Blood with a
low PvO2 also extracts larger volumes of O2 from alveolar
gas, amplifying the effect of hypoventilation or airway
MALDISTRIBUTION obstruction on PAO2 and increasing the risk of resorption
OF PERFUSION atelectasis in poorly ventilated alveoli when patients are
breathing high concentrations of O2 . Intraoperatively,
Causes low PvO2 seldom impacts PaO2 , given the reduction in
peripheral O2 utilization and the use of supplemental
Poor distribution of pulmonary blood flow also causes O2 , positive-pressure ventilation, and close monitoring.
V/Q mismatching and hypoxemia. Flow distribution However, a hypermetabolic state such as thyroid storm
is primarily determined by hydrodynamic factors (pul- or malignant hyperthermia could significantly increase
monary arterial and venous pressures, pulmonary vascu- peripheral O2 extraction, leading to a significant decease
lar resistance), which are, in turn, affected by gravity, in PvO2 . In postoperative patients, shivering, infection,
cardiac dynamics, vascular competence, airway pressure, and increased metabolism lower PvO2 by increasing
and lung volume. Patient positioning affects oxygenation peripheral O2 extraction. Supplemental O2 will reduce the
if gravity forces blood flow to areas with reduced ven- impact of low PvO2 on alveolar O2 extraction and PaO2 ,
tilation; for example, placing a poorly ventilated lung assuming that no true shunt is occurring across the lungs.
in a dependent position can reduce PaO2 . Intraoperative
or postoperative changes in pulmonary artery pressure,
airway pressure, and lung volume have complex effects
on blood flow distribution that can adversely affect V/Q
TISSUE HYOXIA IN SPITE
matching. Inhalational anesthetics, vasodilators, and sym- OF ADEQUATE Pa O 2
pathomimetic agents directly affect vascular tone and
hypoxic pulmonary vasoconstriction, partially explaining The ultimate endpoint for arterial oxygenation is to avoid
larger alveolar/arterial O2 gradients during and after gen- end-organ damage from inadequate availability of O2 . In
eral anesthesia. Patients with cirrhosis of the liver exhibit perioperative patients, conditions sometimes arise that
disordered blood flow distribution and V/Q mismatching deprive vital organs of sufficient O2 to meet metabolic
caused by circulating humoral substances resulting from demand in spite of an adequate PaO2 .
abnormal hepatic metabolism. Circulating endotoxin also
impairs hypoxic pulmonary vasoconstriction, contribut- Oxygen-Carrying Capacity
ing to hypoxemia in septic patients.
A patients preoperative hemoglobin level, intraoperative
Management blood loss, and the amount and freshness of blood
replacement determine the viable red cell mass and
Few interventions are practical to improve V/Q matching O2 -carrying capacity during and after surgery. Reduc-
by managing pulmonary blood flow. When possible, avoid tion of hematocrit caused by crystalloid infusion has less
CHAPTER 8/HYPOXEMIA AND HYPERC APNIA 131
impact on O2 -carrying capacity because the same red elimination. In severe cases, hyperbaric O2 therapy is
cell mass remains in the circulating blood volume and indicated.
expansion of intravascular volume often leads to better
tissue perfusion. The hematocrit at which O2 delivery
Methemoglobinemia
becomes insufficient to match tissue needs varies with
cardiac reserve, O2 consumption, PaO2 , and blood flow Other toxic hemoglobin alterations decrease the
distribution. Regardless, at any given time, every pa- O2 -carrying capacity. Methemoglobinemia occurs when
tient has a minimum level of tolerable hemoglobin; if hemoglobin is exposed to O-toluidine, a metabolic
hemoglobin falls below this level, the peripheral delivery by-product of the local anesthetic, prilocaine. Methe-
of O2 becomes suboptimal. Patients with vascular occlu- moglobinemia interferes with O2 binding and affects the
sive diseases are at increased risk of organ ischemia in this accuracy of standard pulse oximeters by trending readings
circumstance. toward 87%. Cyanmethemoglobinemia, which is formed
when methemoglobin is intentionally generated to chelate
cyanide, also precludes the binding and release of O2 , as
Hemoglobin Alterations does sulfhemoglobinemia.
Shifts in the O2 hemoglobin dissociation curve related to
changes in temperature, pH, and 2,3-diphosphoglycerate Sickle Cell Disease
level affect the amount of O2 that associates with Unusual hemoglobin moieties affect O2 -carrying capacity
hemoglobin at a given PaO2 and subsequently dissoci- and O2 delivery. In patients suffering from homozygous
ates to tissues at a given tissue PO2 . Although usually sickle cell disease, low partial pressures of O2 cause
insignificant, changes in oxyhemoglobin dissociation can conformational changes in red cells that impede passage
impair O2 delivery in marginal patients. through the capillaries.79 Additionally, the hemoglobin
dissociation curve of hemoglobin S is shifted to the right
Carbon Monoxide compared to normal hemoglobin.
Exogenous toxins can alter hemoglobin molecules and
impair O2 delivery. Carbon monoxide (CO) reversibly Hypoperfusion
binds to hemoglobin with 200 times the affinity of O2
and decreases O2 -carrying capacity by impeding both the Peripheral hypoperfusion is often caused by low cardiac
binding and the dissociation of O2 . Heavy smokers are output secondary to hypovolemia, cardiac failure, my-
chronically exposed to CO, but carboxyhemoglobinemia ocardial ischemia, or dysrhythmia. Decreased systemic
from smoking seldom interferes with peripheral O2 vascular resistance related to sepsis, catecholamine deple-
delivery during anesthesia.49,50 However, trauma or burn tion, or sympathectomy also can interfere with peripheral
patients can be poisoned with much higher levels of great perfusion, either because of low perfusion pressure or
clinical significance before emergency surgery. Patients due to poor distribution of systemic blood flow. Arteriolar
can also be exposed to CO generated when inhalation constriction caused by hypothermia or pressor adminis-
anesthetics interact with dry CO2 -absorbing agents in the tration reduces tissue perfusion, as does redistribution of
circle absorber.78 The risk of intraoperative exposure is blood flow secondary to acute hypovolemia.
low but increases during the first cases performed in
low-use locations. Mitochondrial Abnormalities
Carbon monoxide poisoning is difficult to recognize
clinically and can cause severe peripheral hypoxia in spite Poisoning with cyanide, arsenic, and other heavy metals
of apparently adequate O2 saturation. A standard pulse interferes with the ability of the cytochrome oxidase chain
oximeter does not distinguish carboxyhemoglobin from to combine elemental O2 with a free hydrogen ion to
oxyhemoglobin, so SpO2 reads falsely high when com- form metabolic water. Anaerobic metabolism and lactic
pared with SaO2 measured on a laboratory co-oximeter acidemia occurs, regardless of how much O2 is available
or a newer generation, differential oximeter. Also, the as- at the cellular level, because the O2 cannot be utilized.
sessment of oxygenation from an ABG specimen often
reveals a metabolic acidemia with a relatively normal to Supplemental Oxygen/Hyperoxia
high PaO2 . Carbon monoxide poisoning does not cause
cyanosis. Instead, patients exhibit a characteristic cherry In the absence of misadventure, serious hypoxemia during
red appearance that gives a false sense of adequate intraoperative anesthesia is uncommon because patients
arterial O2 content. Signs of acidemia, such as tachy- usually receive intense monitoring during the almost uni-
cardia, hypertension, or other signs of increased sym- versal use of supplemental O2 . The incidence and risk
pathetic nervous system tone, are nonspecific and often of hypoxemia in postoperative and monitored anesthesia
attributed to light anesthesia or pain during emergence. care (MAC) patients is relatively higher than that seen dur-
In awake patients, symptoms of moderate CO poisoning, ing general anesthesia.39,80 When postanesthesia care unit
such as headache, nausea, vomiting, irritability, and al- (PACU) patients breathe room air, 30% of those younger
tered visual or motor skills, are relatively common and than 1 year of age, 20% aged 1 to 3 years, 14% aged
nonspecific. The administration of 100% O2 will help dis- 3 to 14 years, and 7.8% of adults had hemoglobin satu-
place carbon monoxide from hemoglobin and accelerate rations fall below 90%, with many falling below 85%.28
132 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Clinical observation and assessment of cognitive function located in endothelial cells and red blood cells. Carbonic
do not accurately screen for hypoxemia, so monitoring acid instantaneously dissociates to free hydrogen ion and
with oximetry is essential during MAC and throughout bicarbonate ion. The resulting equilibrium explains how
the PACU admission.7 One cannot predict which patients an increase in PACO2 causes a corresponding increase in
will become hypoxemic or when hypoxemia will occur. Pa- the free hydrogen ion concentration and a decrease in
tients with lung disease or obesity, those recovering from pHa, generating the respiratory component of acid base
thoracic or upper abdominal procedures, and those with abnormalities.
preoperative hypoxemia are at increased risk.81 However,
postoperative hypoxemia frequently occurs in children,
especially those with respiratory infections or chronic
Transport and Exchange
adenotonsillar hypertrophy.82 Hypoxemia also occurs in Carbon dioxide (in the form of dissolved gas, carbonic
awake patients after regional anesthesia.83 acid, bicarbonate ion, and carbamino groupings) is
If a patient qualifies for PACU admission, he or she carried by systemic venous blood into the pulmonary
probably should receive O2 during the initial recovery and circulation. It rapidly diffuses down a relatively small
perhaps during transport to the PACU.84 Supplemental concentration gradient from pulmonary capillary blood
O2 improves PaO2 and helps to prevent or to improve to exchanging airspaces, generating an alveolar partial
hypoxemia, although its efficacy is variable. If hypoxemia pressure of CO2 (PACO2 ). Carbon dioxide-rich gas is then
is caused by areas of low V/Q in the lung, increasing O2 washed out with each tidal exhalation, to be replaced with
concentration improves arterial saturation. Increased O2 fresh gas mixed with a small amount of residual, end-
content in the FRC delays the onset of serious hypoxemia expired alveolar gas left in the nongas-exchanging dead
during airway obstruction or hypoventilation. space. The amount of CO2 remaining in the pulmonary
The cost of supplemental O2 is minimal, and patient mixed venous blood determines the PACO2 . Therefore, CO2
inconvenience is minor. However, supplemental O2 does excretion varies directly, and PACO2 varies inversely with
not address the underlying causes of hypoxemia, does the amount of ventilation that is distributed to perfused
not guarantee that hypoxemia will not occur,51 may delay airspaces in the lung (i.e., the effective ventilation).
the recognition of evolving pulmonary problems,43 and
reduces the value of pulse oximetry as an indicator of
hypoventilation.18,22 Oxygen can cause mucosal drying, Dead Space
and the apparatus may cause corneal abrasion during
emergence. A FiO2 >0.8 promotes resorption atelectasis, Ventilation/perfusion matching also affects CO2 excretion.
because inert nitrogen is replaced in poorly ventilated The extreme diffusability of CO2 minimizes the impact
alveoli, but the actual clinical impact of this problem is of low V/Q units, shunting, and even high V/Q units
small.85,86 Breathing 100% O2 for 24 to 36 hours generates (i.e., areas that receive disproportionate ventilation for the
early signs of pulmonary O2 toxicity. The risk of O2 toxicity amount of blood flow) on PACO2 . However, V/Q mismatch
at ambient pressure may be increased after carmustine will generate hypercapnia when ventilation is distributed
therapy but is not increased after bleomycin therapy. to the dead space, which is comprised of airspaces that
Toxicity is accelerated by hyperbaric O2 therapy. cannot participate in gas exchange because they do not
receive any blood flow.87 A fraction of the tidal volume
is wasted in the normal anatomic and physiologic
dead space (VD/VT). If dead space volume increases
or tidal volume decreases, the VD/VT becomes larger,
How, When, Where, and Why CO2 excretion falls, and hypercapnia ensues. Conversely,
Does Hypercapnia Occur in proportionally more total ventilation is required to meet
CO2 production to maintain a constant PACO2 . An increase
Anesthesia? in dead space may also result in more rebreathing of
CO2 from the previous exhaled breath. Patients with high
dead space ventilation (VDS/VT) are at greater risk for
PHYSIOLOGY OF CARBON postoperative ventilatory failure.
DIOXIDE
Central Nervous System Control
During glucose metabolism, carbon and O2 atoms are se-
quentially cleaved from glucose molecules and released To maintain PACO2 and pHa at a relatively constant level,
from the cell as CO2 , a highly diffusible gas that rapidly the amount of CO2 excreted per unit time must closely ap-
crosses biological barriers. Carbon dioxide diffuses down proximate the amount of CO2 generated in that same time.
a concentration gradient into capillary blood, where it Maintaining a stable pH environment is so physiologically
exerts a partial pressure as it dissolves in the plasma. important that CO2 is the primary variable for control
A small amount of CO2 reversibly combines with un- of ventilation. If PACO2 increases, nearly instantaneous
charged amino groups to form carbamino groupings on diffusion of CO2 across the blood-brain barrier lowers
serum proteins and hemoglobin in red cells. Most CO2 the pH of the cerebrospinal fluid, stimulating chemosen-
combines with water to form carbonic acid in a reac- sitive cells on the floor of the fourth ventricle. These
tion that is catalyzed by the enzyme carbonic anhydrase, cells generate efferent output to the diaphragms through
CHAPTER 8/HYPOXEMIA AND HYPERC APNIA 133
the phrenic nerves and to the intercostal muscles and tone are also depressed. Respiratory acidemia increases
accessory muscles of ventilation. An increased ventilatory cerebral blood flow and intracranial pressure in patients
rate and inspiratory depth ensue. The resulting change in with head injury, intracranial tumors, or cerebral edema.
minute ventilation lowers PACO2 and reduces the efferent At very low pHa, catecholamines may not interact with
outflow to the muscles of ventilation. This centrally reg- adrenergic receptors, so heart rate and blood pressure
ulated, negative feedback loop maintains PACO2 within a can decrease precipitously. However, such pH changes
narrow range, minimizing variation in serum and CNS are seldom seen clinically. Severe acidemia often causes
pH. Tight regulation of PACO2 and its effect on pHa is death from malignant cardiac dysrhythmias.
important, because the kidneys require hours to secrete An increase in PACO2 above the normal of 40 mm Hg
hydrogen ion and retain bicarbonate ions in response to does not always indicate that minute ventilation is in-
a decrease in serum pH. Therefore, the ability to compen- adequate. In many circumstances, hypercapnia is an
sate for an acute respiratory acidemia is limited. acceptable, expected physiologic outcome. For example,
during normal sleep, healthy individuals exhibit minor
ventilatory center depression that increases PACO2 by 3 to
5 mm Hg. Also, both respiratory and metabolic factors
How Is Carbon Dioxide affect the pH in the CNS that drives ventilation. In
Assessed Clinically? the presence of a metabolic alkalemia, minute ventila-
tion is adjusted to generate a compensatory respiratory
acidemia. In both of these examples, ventilatory capac-
A determination of the PACO2 from systemic ABG analysis
ity is completely sufficient to meet the bodys needs for
is the most reliable indicator of the balance between the
CO2 excretion.
production and excretion of CO2 . The simultaneous mea-
Conversely, a low PACO2 does not always indicate that
surement of the pHa is invaluable to assess the etiology
ventilatory capacity is adequate. In some circumstances,
and clinical impact of the PACO2 . To date, a noninvasive
minute ventilation can be woefully lacking, even though
indicator of arterial CO2 content has not yet been widely
the PACO2 is far below normal. For example, consider
deployed. However, the transcutaneous measurement of
a patient with a severe metabolic acidemia. If pHa =
CO2 partial pressures may prove useful.88,89 Monitoring
of end-tidal CO2 (ETCO2 ) using conventional capnometry 7.20 with a PACO2 of 28 mm Hg, minute ventilation is
yields a valuable perspective on airway device placement inadequate to offset the metabolic acidemia, even though
and the qualitative adequacy of ventilation.9093 However, PACO2 is well below normal.
even when a patient is intubated and ventilating in a In surgical patients, assessment of the significance
closed system, estimating PACO2 from ETCO2 as an index of hypercapnia is complex because many diverse fac-
of PACO2 is fraught with inaccuracy. Beyond problems tors interfere with ventilatory drives, alter the mechanics
with the clinical application of capnography, a reliable of ventilation, or affect the rate of CO2 production (see
estimation of the Aa CO2 gradient is difficult, especially Table 8.3). One should suspect inadequate minute ventila-
with the rapid physiologic changes seen in perioperative tion when: (i) Hypercapnia causes a reduction in the pHa
patients. Finally, the assessment of total minute ventila- below 7.25; (ii) hypercapnia and respiratory acidemia oc-
tion using rate and tidal volume calculations, spirometry, cur coincident with tachypnea, anxiety, dyspnea, labored
or ventilator measurements is unreliable90 and does not ventilation, or signs of increased sympathetic nervous sys-
reflect the impact that changes in CO2 production or dead tem activity; or (iii) PACO2 is progressively increasing with
space have on PACO2 . a parallel progressive decrease in pHa.
The acceptable upper limit for PACO2 depends on
specific clinical circumstances, individual patient charac-
teristics and, most importantly, on pH. Carbon dioxide
itself is relatively inert. Within the usual range, the only TABLE 8.3 Causes of Hypercapnia in Perioperative
significant impact of hypercapnia is through its effect on Patients
pH. If PACO2 rises above 90 to 100 mm Hg, it exerts an
independent, direct anesthetic effect on the CNS. An ex- Respiratory center depression
ceptionally high PACO2 may displace O2 from alveoli and Complete airway obstruction
reduce the PACO2 , potentially causing hypoxemia. Increased upper airway resistance
Setting these extremes aside, hypercapnia usually Increased small airway resistance
is innocuous without a threatening reduction in pH or Increased dead space
PaO2 . However, when increased PACO2 causes significant Increased VD/VT
acidemia, the condition can be problematic. Symptoms of Gas trapping
respiratory acidemia in awake patients include agitation, Inadequate mechanical ventilation
confusion, and ventilatory dissatisfaction. The sympa- Neuromuscular or skeletal problems
thetic nervous system response to acidemia causes sweat- Carbon dioxide uptake after insufflation
ing, tachypnea, hypertension, tachycardia, and dysrhyth- Increased CO2 production
mias. Respiratory acidemia secondary to CNS depression Rebreathing of CO2 -enriched gas
often produces somewhat less intense signs because the
medullary centers regulating sympathetic nervous system VD, volume of dead space; VT, tidal volume.
134 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 8.4 Estimated Rate of Decline (in Minutes) of SpO2 After Complete Cessation of Ventilation. (All Patients
Modeled as Previously Denitrogenated with 100% Oxygen, FAO2 = 0.87 at Time 0)
Data from Benumof JL, Dagg R, Benumof R. Critical hemoglobin desaturation will occur before return to an unparalyzed state following
1 mg/kg intravenous succinylcholine. Anesthesiology. 1997;87:979.
is most frequently caused by obstruction at the pharyngeal upper respiratory infections119 or those chronically ex-
level. Obstruction can be caused by posterior tongue dis- posed to secondhand smoke,120 patients with irritable
placement or a change in the anteroposterior and lateral airway conditions or copious secretions, smokers,1,49,50
dimensions secondary to collapse of soft tissues.27 The de- and patients recovering from upper airway surgery are at
gree of obstruction often varies inversely with the level of highest risk.36,113 Laryngospasm can usually be overcome
consciousness. Therefore, deep sedation caused by anes- by providing gentle positive pressure in the oropharynx
thetics, opioids, or sedatives leads to increased upper with 100% O2 . Prolonged laryngospasm is relieved with
airway resistance, hypoxemia, and hypercapnia.39,64,106 a small dose of succinylcholine (e.g., 0.1 mg per kg).
Patients with sleep apnea or with physical findings An intubating dosage of succinylcholine should not be
such as a thick, short neck or large tongue are more used, especially when the PAO2 is decreased or the FRC
prone to obstruction.16,23,37 Soft tissue edema worsens is reduced. If artificial ventilation cannot be achieved,
airway obstruction, especially in children recovering the prolonged duration of paralysis can allow serious
from ENT procedures and in adults recovering from hypoxemia to develop before spontaneous ventilation re-
endarterectomy, thyroid surgery, or other procedures on sumes12,121 (see Table 8.4). Severe laryngeal obstruction
the neck.107 Nebulized vasoconstrictors help somewhat, also can result secondary to laryngeal edema or hypocal-
but steroids have little effect.108 Muscular weakness from cemic tetany of the laryngeal constrictors after parathy-
residual neuromuscular blockade8,109 or diseases such as roid excision.
myasthenia gravis or postpolio syndrome110,111 worsen In the large airways, tracheal stenosis, tracheomala-
upper airway obstruction, but seldom cause primary cia, tracheal webs, or extrinsic compression from edema
airway compromise. Patients with a C1-esterase-inhibitor or tumor can increase flow resistance sufficiently to com-
deficiency can develop severe angioneurotic edema after promise ventilation.122 If obstruction is caused by acute
even slight trauma to the airway.112 extrinsic compression, as occurs with an expanding neck
hematoma, relief of the external pressure is essential.
If the upper airway is clear of vomitus or foreign
Reduction of the cross-sectional area in small air-
bodies, airway obstruction can usually be resolved with
ways increases overall airway resistance. (Resistance to
simple maneuvers, such as lateral positioning, chin lift,
gas flow varies inversely with the fourth power of a tubes
mandible elevation, arousal, or placement of an artificial
radius during laminar flow and with the fifth power dur-
airway.113115 A nasopharyngeal airway is better tolerated
ing turbulent flow.) Smokers, asthmatics, and patients
when the patients gag reflexes are functional. Pathologic,
with other preexisting bronchospastic conditions are at
fixed upper airway obstruction caused by supraglottitis,
highest risk for increased small airway resistance during
retropharyngeal abscess, neck infections, or encroaching the perioperative period.47,49,123 Pharyngeal or tracheal
tumors may require emergency tracheal intubation.116 stimulation from secretions, suctioning, aspiration, or
Great care must be taken during airway manipulations mechanical contacts can trigger a reflex constriction of
because minor trauma can convert a marginal airway into bronchial smooth muscle during intubation, after extu-
a total obstruction. Sedatives or muscle relaxants used to bation, or during emergence. Preoperative spirometric
facilitate tracheal intubation can worsen obstruction by evidence of increased airway resistance predicts intra-
compromising the patients volitional effort to maintain operative or postoperative bronchospasm.50 Histamine
the airway117 and by eliminating spontaneous ventila- release precipitated by medications or allergic reactions
tion. Equipment and personnel required for emergency also increases airway smooth muscle tone. In patients
cricothyroidotomy or tracheostomy should be available. with COPD or decreased lung volume, low radial trac-
Cricothyroidotomy using a 14-gauge intravenous catheter tion on small airways reduces the cross-sectional area and
or a commercially available kit permits oxygenation and increases flow resistance.124 Increased ventilatory require-
marginal ventilation until the airway is secured. ments caused by warming, hyperthermia, or high work of
During emergence, stimulation of the pharynx or breathing generate high flow rates and convert laminar
vocal cords by secretions, foreign matter, or artificial flow to higher resistance, turbulent flow.
airways generates tight apposition of the laryngeal con- During forced vital capacity expiration, a prolonged
strictor muscles, occluding the tracheal inlet and imped- expiratory time or audible turbulent airflow (wheezing) of-
ing gas flow at the laryngeal level.107,118 Children with ten unmasks subclinical airway resistance. Resistance is
136 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
higher during expiration because intermediate diameter decreased compliance in surgical patients can exacerbate
airways are compressed by positive intrathoracic pres- hypercapnia from another cause, it is seldom primarily
sure. High airway resistance does not always result in responsible for ventilatory failure.
audible wheezing because flow might be so impeded that Extrathoracic factors, such as gas in the stomach
no sound is produced. Clinically, it is often difficult to dis- or bowel, ascites, bowel obstruction, intra-abdominal tu-
tinguish increased resistance from decreased pulmonary mor, peritoneal hemorrhage, or pregnancy, will impair
compliance, because the signs are similar. Spontaneously diaphragmatic excursion and increase work of breath-
breathing patients exhibit labored ventilation, accessory ing. Tight chest or abdominal dressings impede thoracic
muscle recruitment, and increased work of breathing with expansion. Obesity can profoundly affect pulmonary com-
either condition, while mechanically ventilated patients pliance, especially when adipose tissue compresses the
exhibit high peak inspiratory pressures. thoracic cage or increases intra-abdominal pressure in
In awake patients, a significant increase in airway supine, lateral, or lithotomy positions. Conditions affect-
resistance generally causes agitation and dissatisfaction ing the lung parenchyma during surgery can also seriously
with ventilation, resulting in labored, tachypneic breath- decrease lung compliance. Reduction of FRC leads to
ing. This finding is particularly prevalent when small air- small airway closure and distal lung collapse, requir-
way resistance is high. Blood gas analysis often will reveal ing greater energy expenditure to reexpand the lungs.
moderate hypocarbia and respiratory alkalemia. If an ABG Pulmonary edema increases the lung weight and inertia
analysis reveals hypercapnia and respiratory acidemia in and elevates surface tension by interfering with surfac-
such a patient, it is likely that he or she is rapidly fa- tant activity, making expansion more difficult. Pulmonary
tiguing, secondary to increased work of breathing, and contusion or hemorrhage interferes with lung expansion,
that ventilatory failure is imminent. The patient should be as do restrictive lung diseases, skeletal abnormalities,
closely observed for progressive respiratory acidemia. intrathoracic lesions, hemothorax, pneumothorax, or sig-
nificant cardiomegaly.
Treatment Resolving problems that reduce compliance will usu-
ally reduce the effort required to maintain ventilation.
Treatment of small airways resistance is directed at an un- Allowing patients to recover in a semisitting position
derlying etiology. Whenever possible, try to eliminate the rather than supine or full sitting reduces work of breath-
offending laryngeal or airway stimulation. Patients with ing. Incentive spirometry and chest physiotherapy helps
preexisting bronchospastic disease often respond to their restore lung volume, as does PEEP or CPAP. However, in
standing regimen of albuterol, pirbuterol, or salmeterol patients with highly compliant lungs, such as those with
inhalers. The administration of inhaled or intravenous end-stage COPD, positive airway pressure might force the
steroid therapy offers little benefit for an acute episode. rib cage and diaphragms toward their excursion limits,
Isoetharine or metaproterenol nebulized in O2 resolves paradoxically increasing the effort required to complete
postoperative bronchospasm with minimal tachycardia. inspiration. Lesser amounts may be well tolerated.
Intramuscular or sublingual terbutaline can be added. Ad-
verse side effects of intravenous theophylline have led to
its abandonment as a mainstream therapy. Bronchospasm NEUROMUSCULAR AND
that is resistant to 2-sympathomimetic medication may
improve with an anticholinergic medication, such as at-
SKELETAL PROBLEMS
ropine or ipratropium. The most obvious cause of impaired ventilation related
If bronchospasm is life threatening, an intravenous to muscular weakness results from the administration of
epinephrine infusion yields profound bronchodilation and neuromuscular relaxation to facilitate surgical exposure.
reduces airway edema. Increased small airway resistance Paralysis mandates the use of positive-pressure ventilation
caused by mechanical factors, such as loss of lung volume, to sustain oxygenation and CO2 removal, creating a host of
retained secretions, or pulmonary edema, usually does not potential causes for intraoperative hypercapnia.19,21 Ne-
resolve with bronchodilator therapy. Restoration of lung glecting to institute positive-pressure ventilation by hand,
volume with incentive spirometry or deep tidal ventilation forgetting to activate the ventilator after paralysis, or in-
increases radial traction on small airways, whereas control terrupting ventilation during intubation or tracheostomy
of left ventricular filling pressures may relieve airway leads to a rapid buildup of CO2 .125 Acute hypoxemia may
resistance secondary to increased lung water. However, also develop rapidly, depending upon the inspired con-
interstitial fluid accumulation can persist, especially if centration of O2 in the mixture. However, with a high
extended contraction of airway smooth muscle obstructs FiO2, it is possible that a patient will passively oxygenate
venous and lymphatic flow, leading to slowly resolving for some period while CO2 accumulates. If the ventilator
airway wall edema. settings for rate and tidal volume are inadequate to match
CO2 production, hypercapnia and respiratory acidemia
Decreased Compliance will develop more gradually. During surgery, leakage of
fresh gas from the circuit, the ventilator, or past the endo-
Reduced pulmonary compliance increases the work of tracheal cuff can waste enough inspired volume to result
breathing and CO2 production. Extremely low compli- in increased PACO2 . Rarely, leakage from a bronchopleural
ance can cause progressive respiratory muscle fatigue, fistula will result in insidious hypoventilation and hyper-
hypoventilation, and respiratory acidemia. Although capnia.
CHAPTER 8/HYPOXEMIA AND HYPERC APNIA 137
Spontaneously breathing patients also can develop be maintained with only one functioning diaphragm,
hypercapnia related to impaired ventilatory mechanics. although marginal ventilation is possible with the external
Extreme positions, such as prone, jackknife, lateral or intercostal muscles alone. However, ventilatory insuffi-
lithotomy, can reduce a patients ability to ventilate, espe- ciency will ensue if the work of breathing or ventilatory
cially if baseline ventilatory reserve is reduced by chronic demand is increased.129,130 Thoracic spinal or epidural
lung disease or obesity. An occasional patient will have blockade interferes with intercostal muscle function and
difficulty maintaining spontaneous breathing in a supine reduces ventilatory reserve, especially in patients with
position. Skeletal problems such as kyphosis and scol- COPD. Conditions that interfere with motor neuron func-
iosis restrict thoracic excursion. Be particularly vigilant tion, such as GuillainBarre syndrome or cervical spinal
for ventilatory impairment in trauma patients who may cord trauma can result in ventilatory failure.
have suffered blunt chest trauma. Disruption of the tho- Simple bedside tests help assess the mechanical
racic cage secondary to costochondral separations or rib component of ventilation. The ability to sustain head
fractures can interfere with chest cavity expansion and elevation in a supine position, a forced vital capacity of
limits minute ventilation. This impairment is accentu- 10 to 12 mL per kg, an inspiratory pressure more negative
ated by pain associated with chest wall expansion and than 25 cm H2 O, and tactile train-of-four assessment
by decreased compliance from an underlying pulmonary imply that ventilatory muscle strength is adequate to
contusion. In a worst-case scenario, sequential rib frac- sustain ventilation. However, failure to meet these clinical
tures will isolate a flail segment that exhibits paradoxical endpoints does not necessarily indicate the need for
inward motion with spontaneous inspiration. Although assisted ventilation. Also, none of these endpoints reliably
the musculoskeletal problems may be significant, the real predicts the recovery of airway protective reflexes after
problem often lies with the underlying pulmonary contu- neuromuscular paralysis, and they do not guarantee the
sion associated with blunt force trauma. ability to sustain spontaneous ventilation over a long
Inadequate reversal of neuromuscular relaxation period of time.75,131,132
compromises a patients ability to maintain airway pa- Occasionally, a clinical picture suggests ventilatory
tency, protect against aspiration, overcome airway resis- insufficiency even when ventilation is adequate. Sponta-
tance, and clear secretions toward the end of surgery neous ventilation with small tidal volumes due to thoracic
and during recovery.74 Shorter-acting muscle relaxants restriction or reduced compliance seems to generate af-
may decrease the incidence of residual paralysis but do ferent input from pulmonary stretch receptors, leading to
not eliminate the risk.126 Marginal reversal can be more dyspnea, labored breathing, and accessory muscle recruit-
dangerous than near-total paralysis. A weak, agitated pa- ment. Blood gas determination usually reveals adequate
tient exhibiting uncoordinated movements and airway ventilation and sometimes even significant hypocarbia
obstruction is more easily identified, whereas a somnolent secondary to hyperventilation. Occasional large, satis-
individual exhibiting mild stridor and shallow ventilation fying lung expansions often relieve these symptoms.
related to marginal neuromuscular function may go un- Similarly, voluntary limitation of thoracic expansion to
noticed. If the latter patient is overlooked, regurgitation avoid painful chest wall or abdominal movement causes
with aspiration or insidious hypoventilation with progres- labored, rapid, shallow breathing characteristic of inad-
sive respiratory acidemia can occur. Residual paralysis is equate ventilation. This splinting seldom causes actual
more likely in patients who have received nondepolariz- hypoventilation and usually improves with analgesia and
ing muscle relaxants but no reversal agents.75 Safety of repositioning. Finally, spontaneous hyperventilation to
techniques designed to avoid the reversal of short- and compensate for a metabolic acidemia can generate tachyp-
intermediate-duration relaxants has not been substanti- nea or labored breathing, which can be mistaken for
ated, and reversal of nondepolarizing relaxants is strongly ventilatory insufficiency.
recommended.84
Patients with neuromuscular abnormalities, such as
myasthenia gravis, EatonLambert syndrome, periodic INCREASED DEAD SPACE
paralysis, muscular dystrophies, or postpolio syndrome,
can suffer postoperative ventilatory insufficiency.107 These Ventilation/perfusion mismatching is a less prevalent
patients can also exhibit exaggerated or prolonged re- cause of perioperative hypercapnia than it is of hypox-
sponses to nondepolarizing muscle relaxants. The du- emia. Occasionally, an acute increase in dead space will
ration and intensity of neuromuscular blockade can be contribute to respiratory acidemia in a perioperative
potentiated by a host of medications used during surgery, patient. Application of PEEP or CPAP expands airway
such as antibiotics, furosemide, propranolol, and pheny- volume and may increase lung anatomic dead space, par-
toin. Hypocalcemia and hypermagnesemia also interfere ticularly in patients with high pulmonary compliance. Al-
with recovery from paralysis. though upper airway dead space is reduced after tracheal
Diaphragmatic contraction is compromised in some intubation or tracheostomy, excessive tubing volume be-
surgical patients, forcing more reliance on intercostal tween the Y piece and the endotracheal tube, or mal-
muscles and reducing the ability to meet increased functioning valves in the anesthesia machine or ventilator
ventilatory demands.127 Impairment of phrenic nerve circuits, can promote CO2 rebreathing. Pulmonary em-
function from interscalene block, trauma, or thoracic bolization with air, thrombus, or cellular debris increases
and neck operations can paralyze one or, rarely, dead space, although the impact on CO2 excretion is often
both diaphragms.128 Adequate ventilation normally can masked by accelerated minute ventilation from hypoxic
138 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
drive or reflex responses. Pulmonary hypotension can during emergence from anesthesia. Shivering, high work
transiently increase dead space ventilation VD/VT by de- of breathing, infection, sympathetic nervous system activ-
creasing perfusion to the well ventilated, nondependent ity, or rapid carbohydrate metabolism from intravenous
lung. Irreversible increases in dead space occur if aspira- hyperalimentation will further accelerate CO2 production.
tion pneumonitis, adult respiratory distress syndrome, or Even mild increases of CO2 production can precipitate res-
systemic inflammatory response syndrome (SIRS) destroy piratory acidemia if neuromuscular paralysis, high airway
the pulmonary microvasculature.133 resistance, or low compliance interferes with ventilation.
Dead space may appear high if an inhalation inter- With the exception of improving work of breathing,
rupts the previous exhalation and spent alveolar gas is reducing shivering, treating hyperthermia, or adjusting
retained in the lungs. This gas trapping occurs when high hyperalimentation, there is little yield from attempting to
airway resistance lengthens the time required to com- manage CO2 production in perioperative patients. Rarely,
pletely exhale. Trapping also can occur when improper when increased dead space precludes the delivery of
inspiratory/expiratory ratios or high ventilatory rates are adequate mechanical ventilation, deliberate hypothermia
used during mechanical ventilation. A common cause of and paralysis may be utilized to reduce CO2 production
gas trapping in the operating room is overzealous positive- in the hope that VD/VT will improve.
pressure ventilation by hand. If a provider does not allow
sufficient time for a patient to exhale before the next
compression of the reservoir bag, tidal volumes are de-
livered with the lungs near full inflation. The resulting THERAPEUTIC HYPERCAPNIA
increase in end-expiratory lung volume traps alveolar gas In a limited number of circumstances, hypercapnia and
and generates hypercapnia similar to overzealous me- respiratory acidemia can be utilized for therapeutic ben-
chanical ventilation. Often the inadequate ventilation and efit. Intraoperatively, a short period of hypercapnia and
rising end-expired CO2 is missed because the capnometer acidemia is sometimes required to restore spontaneous
does not sample actual end-expired gas before the next breathing after the cessation of mechanical ventilation,
inflation. particularly if a degree of hyperventilation was achieved
On rare occasions, a spontaneously breathing patient during surgery. In the past, the addition of CO2 to the
undergoing a surgical procedure with regional anesthesia inspired gas was advocated to speed the return of spon-
or MAC will become hypercapnic from rebreathing the
taneous ventilation and to hasten the washout of inhaled
exhaled CO2 that is trapped within a closed space created
anesthetics. More recently, some advocate combining the
by blankets or surgical drapes near the head.134
rebreathing of CO2 with charcoal filtration of exhaled
volatile anesthetics to achieve shorter emergence intervals
at the end of general anesthesia, although such approaches
INCREASED CARBON are rarely necessary.
DIOXIDE PRODUCTION In an effort to minimize ventilator-induced lung
injury, the application of smaller tidal volumes for
Carbon dioxide production varies directly with metabolic patients requiring mechanical ventilation can be utilized.
rate, body temperature, and substrate availability. During In some patients, this technique mandates a reduction
anesthesia, CO2 production can fall to approximately 60% in overall minute ventilation and the acceptance of
of the normal 2 to 3 mL/kg/minute because hypothermia moderate hypercapnia and respiratory acidemia. These
lowers metabolic activity and neuromuscular relaxation problems can become more pronounced in patients
reduces tonic muscle contraction. with high rates of CO2 production, increased dead
One common cause of increased CO2 production space, or low compliance syndromes. Such permissive
is related to the uptake of CO2 across the peritoneum hypercapnia may also be acceptable during one-lung
after abdominal CO2 insufflation during laparoscopic ventilation137 or during episodes of status asthmaticus.138
procedures.135,136 The flux of CO2 into the bloodstream Hypercapnia may be therapeutic in bronchospasm due
generates a significant increase in PvCO2 and PACO2 . An to the bronchodilator effects of CO2 ;139 however, such
increase in minute ventilation may be required to keep therapy can be problematic.
PACO2 within acceptable limits. Hypercapnia resolves Finally, hypercapnia may improve the delivery of O2
quickly once insufflation is stopped and the abdomen is
to tissues in general,140 to the brain during rewarming,141
deflated. However, some amount of low level CO2 uptake
to the bowel in patients undergoing intra-abdominal
may persist into the postoperative period, especially if
procedures,142 and to the liver during anesthesia.143
CO2 has dissected out along tissue planes.
Conceivably, it might also provide some protection against
During surgery, a hypermetabolic condition, such as
hypotension during anesthetic induction.144
thyroid storm, malignant hyperthermia, or neuroleptic
malignant syndrome, generates CO2 production many
times greater than normal. Severe hypercapnia can rapidly
exceed ventilatory reserve in a spontaneously breathing
patient and can even limit the ability to mechanically
KEY POINTS
ventilate a patient, with a resultant severe acidemia. 1. The most reliable indicator of O2 transfer in the lung
Because metabolism decreases intraoperatively, a pa- is the arterial partial pressure of O2 measured from
tient can exhibit a significant increase in CO2 production an ABG determination.
CHAPTER 8/HYPOXEMIA AND HYPERC APNIA 139
2. Hypoxemia and hypercapnia are the leading causes Perioperative events and postoperative complications. Anes-
thesiology. 1993;78:445.
of morbidity and mortality in patients undergoing
8. Eikermann M, Blobner M, Groeben H, et al. Postoperative
procedures with MAC or deep sedation. upper airway obstruction after recovery of the train of four
3. An acute, global reduction of PAO2 is the most com- ratio of the adductor pollicis muscle from neuromuscular
mon cause of serious hypoxemia in perioperative blockade. Anesth Analg. 2006;102:937.
patients. 9. Thrush DN, Downs JB, Hodges M, et al. Does significant
4. Supplemental O2 offsets hypoxemia caused by hypoxemia alter vital signs? J Clin Anesth. 1997;9:355.
airway obstruction, hypoventilation, or areas of low 10. Casati A, Fanelli G, Pietropaoli P, et al. Continuous
V/Q, but does not address the underlying cause of monitoring of cerebral oxygen saturation in elderly patients
hypoxemia and limits the value of oximetry for the undergoing major abdominal surgery minimizes brain
early recognition of hypoventilation. exposure to potential hypoxia. Anesth Analg. 2005;101:740.
11. Lieberman JA, Weiskopf RB, Kelley SD, et al. Critical
5. Patients often tolerate significant arterial hypox-
oxygen delivery in conscious humans is less than 7.3 ml
emia without injury. However, a higher PAO2 and O2 kg 1 min 1 . Anesthesiology. 2000;92:407.
hemoglobin saturation will delay the onset of life- 12. Benumof JL, Dagg R, Benumof R. Critical hemoglobin de-
threatening hypoxemia should ventilation suddenly saturation will occur before return to an unparalyzed state
cease and allows the provider more time to respond. following 1 mg/kg intravenous succinylcholine. Anesthesi-
6. In surgical patients, the importance of OSA as an eti- ology. 1997;87:979.
ology of hypoxemia and hypercapnia is dangerously 13. DeBenedeto RJ, Craves SA, Gravenstein N, et al. Pulse
underappreciated. oximetry monitoring can change routine oxygen supple-
7. A high PaO2 does not guarantee that sufficient O2 will mentation practices in the postanesthesia care unit. Anesth
reach the vital organs to meet their metabolic needs. Analg. 1994;78:365.
14. Gift AG, Stanik J, Karpenick J, et al. Oxygen saturation in
Many other factors affect peripheral O2 delivery and
postoperative patients at low risk for hypoxemia: Is oxygen
utilization. therapy needed? Anesth Analg. 1995;80:368.
8. The most reliable indicator of adequate CO2 ex- 15. Kabemba AS, Downs JB, Smith RA. Hypoxemiahow low
cretion is the arterial partial pressure of CO2 in can we go? Anesthesiology. 2001;95:A1123.
relationship to the serum pH, as measured by an 16. Frater RAS, Moores MA, Parry P, et al. Analgesia induced
ABG determination. respiratory depression: Comparison of meptazinol and
9. Within the usual ranges, the only significant impact morphine in the postoperative period. Br J Anaesth. 1989;63:
that hypercapnia exerts on human physiology is 260.
through its effect on pH. 17. Grief R, Akca O, Ernst-Peter H, et al. Supplemental
10. Any factor that decreases minute ventilation, in- perioperative oxygen to reduce the incidence of surgical
wound infection. N Engl J Med. 2000;342:161.
creases dead space ventilation, or increases total
18. Witting MD, Hsu Sm Granja CA. The sensitivity of room-air
body CO2 production promotes hypercapnia and pulse oximetry in the detection of hypercapnia. Am J Emerg
respiratory acidemia. Med. 2005;23:497.
11. In most postoperative patients, a moderate degree 19. Peterson GN, Domino KB, Caplan RA, et al. Management of
of hypercapnia and respiratory acidemia is expected the difficult airway: A closed claims analysis. Anesthesiology.
and acceptable. 2005;103:33.
12. Reversal of opioid-induced hypoventilation must be 20. Combes X, LeRoux B, Suen P, et al. Unanticipated diffi-
accomplished gradually, through careful titration of cult airway in anesthetized patients. Anesthesiology. 2004;
naloxone, to avoid a dangerous reversal of analgesia. 100:1146.
21. Caplan RA, Posner KL, Ward RJ, et al. Adverse respiratory
events in anesthesia: A closed claims analysis. Anesthesiol-
REFERENCES ogy. 1990;72:828.
22. Fu ES, Downs JB, Schweiger JW, et al. Supplemental
1. Rose DK, Cohen MM, Wigglesworth DF, et al. Critical oxygen impairs detection of hypoventilation by pulse
respiratory events in the postanesthesia care unit: Patient, oximetry. Chest. 2004;126:1552.
surgical and anesthetic factors. Anesthesiology. 1994;81:410. 23. Kaw R, Michota F, Jaffer A, et al. Unrecognized sleep apnea
2. Macario A, Weinger M, Truong P, et al. Which clinical in the surgical patient: Implications for the perioperative
anesthesia outcomes are both common and important to setting. Chest. 2006;129:198.
avoid? The perspective of a panel of expert anesthesiolo- 24. Andersson L, Lagerstrand L, Thorne A, et al. Effect of CO2
gists. Anesth Analg. 1999;88:1085. pneumoperitoneum on ventilation-perfusion relationships
3. Hines R, Barash PG, Watrous G, et al. Complications during laparoscopic cholycystectomy. Acta Anaesthesiol
occurring in the postanesthesia care unit: A survey. Anesth Scand. 2002;46:552.
Analg. 1992;74:503. 25. Wu A, Drummond G. Sleep arousal after lower abdominal
4. Zelcer J, Wells DG. Anaesthetic-related recovery room surgery and relation to recovery from respiratory obstruc-
complications. Anaesth Intensive Care. 1996;15:168. tion. Anesthesiology. 2003;99:1295.
5. Hambraeus-Jonzon K, Bindslev L, Mellgard A, et al. 26. ASA Task Force on Perioperative Management of Patients
Hypoxic pulmonary vasoconstriction in human beings: with Obstructive Sleep Apnea. Practice guidelines for the
A stimulus-response study. Anesthesiology. 1997;86:308. perioperative management of patients with obstructive
6. Weir EK, Lopez-Barneo J, Buckler KJ, et al. Acute oxygen sleep apnea. Anesthesiology. 2006;104:1081.
sensing mechanisms. N Engl J Med. 2005;353:19. 27. Wu A, Drummond GB. Respiratory muscle activity and
7. Moller JT, Johannessen NW, Espersen K, et al. Ran- respiratory obstruction after abdominal surgery. Br J
domized evaluation of pulse oximetry in 20,802 patients: Anaesth. 2006;96:510.
140 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
28. Mort TC. Emergency tracheal intubation: Complications ambulatory surgery: Smokers vs. nonsmokers. Anesthesi-
associated with repeated laryngoscopic attempts. Anesth ology. 2002;97:842.
Analg. 2004;99:607. 50. Warner DO, Warner MA, Offord KP, et al. Airway obstruc-
29. Warner DO. Laryngeal reflexes. Anesthesiology. 1998;88: tion and perioperative complications in smokers undergo-
1433. ing abdominal surgery. Anesthesiology. 1999;90:372.
30. Motamed C, Spencer A, Farhat F, et al. Postoperative hypox- 51. Ballantyne JC, Carr DB, DeFerranti S, et al. The com-
aemia: Continuous extradural infusion of bupivacaine and parative effects of postoperative analgesic therapies on
morphine vs patient controlled analgesia with intravenous pulmonary outcome: Cumulative meta-analyses of random-
morphine. Br J Anaesth. 1998;80:742. ized, controlled trials. Anesth Analg. 1998;86:598.
31. Cashman JN, Dolan SJ. Respiratory and haemodynamic 52. Eichenberger A, Proietti S, Wicky S, et al. Morbid obesity
effects of acute postoperative pain management: Evidence and postoperative pulmonary atelectasis: An underesti-
from published data. Br J Anaesth. 2004;93:212. mated problem. Anesth Analg. 2002;95:1788.
32. Ferguson LM, Drummond GB. Acute effects of fentanyl on 53. Gander S, Frascarolo P, Suter M, et al. Positive end-
breathing pattern in anaesthetized subjects. Br J Anaesth. expiratory pressure during induction of general anesthesia
2006;96:384. increases duration of nonhypoxic apnea in morbidly obese
33. Bailey PL, Pace NL, Ashburn MA, et al. Frequent hypoxemia patients. Anesth Analg. 2005;100:1854.
and apnea after sedation with midazolam and fentanyl. 54. Coussa M, Proietti S, Schnyder P, et al. Prevention of atelec-
Anesthesiology. 1990;73:826. tasis formation during the induction of general anesthesia
34. Babenco HD, Conard PF, Gross JB. The pharmacodynamic in morbidly obese patients. Anesth Analg. 2004;98:1491.
effects of a remifentanil bolus on ventilatory control. 55. Xue FS, Li BW, Zhang GS, et al. The influence of surgical
Anesthesiology. 2000;92:393. sites on early postoperative hypoxemia in adults undergoing
35. Pandit JJ. The variable effect of low dose volatile anaesthet- elective surgery. Anesth Analg. 1999;88:213.
ics on the acute ventilatory response to hypoxia in humans: 56. Karayiannakis AJ, Makki GG, Mantzioka A, et al. Postop-
A quantitative review. Anaesthesia. 2002;57:632. erative pulmonary function after laparoscopic and open
36. Cartwright CR, Henson LC, Ward DS. Effects of alfentanil cholecystectomy. Br J Anaesth. 1996;77:448.
on the ventilatory response to sustained hypoxia. Anesthe- 57. Andersson LE, Baath M, Thorne A, et al. Effect of
siology. 1998;89:612. carbon dioxide pneumoperitoneum on development of
37. Waters KA, McBrien F, Steward P, et al. Effects of OSA, atelectasis during anesthesia, examined by spiral computed
inhalational anesthesia, and fentanyl on the airway and tomography. Anesthesiology. 2005;102:293.
ventilation of children. J Appl Physiol. 2002;92:1987. 58. Fisher BW, Mujumdar SR, McAlister FA. Predicting pul-
38. Series F, Cormier Y, LaForge J. Influence of apnea type and monary complications after nonthoracic surgery: A system-
sleep stage on nocturnal postapneic desaturation. Am Rev atic review of blinded studies. Am J Med. 2002;112:219.
Respir Dis. 1990;141:1522. 59. Drummond GB. The abdominal muscles in anaesthesia and
39. Bhananker SM, Posner KL, Cheney FW, et al. Injury and after surgery. Br J Anaesth. 2003;91:73.
liability associated with monitored anesthesia care: A closed 60. Gajic O, Gropper MA, Hubmayr RD. Pulmonary edema af-
claims analysis. Anesthesiology. 2006;104:228. ter transfusion: How to differentiate transfusion-associated
40. Bailey PL, Lu JK, Pace NL, et al. Effects of intrathecal circulatory overload from transfusion-related acute lung
morphine on the ventilatory response to hypoxia. N Engl J injury. Crit Care Med. 2006;34(5 Suppl):S109.
Med. 2000;343:1228. 61. Moore SB. Transfusion-related acute lung injury (TRALI);
41. Jaquet Y, Monnier P, Melle GV, et al. Complications clinical presentation, treatment and prognosis. Crit Care
of different ventilation strategies in endoscopic laryngeal Med. 2006;34(5 Suppl):S114.
surgery, a 10 year review. Anesthesiology. 2006;104:52. 62. Looney MR, Gropper MA, Matthay MA. Transfusion-related
42. Tait AR, Malviya S, Voepel-Lewis T, et al. Risk factors for acute lung injury: A review. Chest. 2004;126:249.
perioperative adverse respiratory events in children with 63. Gattinoni L, Caironi P, Cressoni M, et al. Lung recruitment
upper respiratory tract infections. Anesthesiology. 2001;95: in patients with the acute respiratory distress syndrome.
299. N Engl J Med. 2006;354:1775.
43. Downs JB, Smith RA. Increased inspired oxygen concentra- 64. Litnam R, Hayes H, Basco M, et al. Use of dynamic negative
tion may delay diagnosis and treatment of significant deteri- airway pressura (DBAP) to assess sedative-induced upper
oration in pulmonary function. Crit Care Med. 1999;12:2844. airway obstruction. Anesthesiology. 2002;96:342.
44. Duggan M, Kavanagh B. Pulmonary atelectasis: 65. Lasocki S, Sartorius A, Fouillat D, et al. Open and
A pathogenic perioperative entity. Anesthesiology. 2005;102: closed-circuit endotracheal suctioning in acute lung injury:
838. Efficiency and effects on gas exchange. Anesthesiology.
45. Rothen HU, Sporre B, Engberg G, et al. Airway closure, 2006;104:39.
atelectasis and gas exchange during anaesthesia. Br J 66. Kerr ME, Weber BB, Serieka SM, et al. Effect of endo-
Anaesth. 1998;81:68. tracheal suctioning on cerebral oxygenation in traumatic
46. Billard V, Michiels S, Ducret C, et al. Which risk fac- brain-injured patients. Crit Care Med. 1999;27:276.
tors for early postoperative hypoxemia remain in 2003? 67. Moller JT, Wittrup M, Johansen SH. Hypoxemia in the
Anesthesiology. 2003;99:302. postanesthesia care unit: An observer study. Anesthesiology.
47. Schwilk B, Bothner U, Schraag S, et al. Perioperative 1990;73:890.
respiratory events in smokers and nonsmokers undergoing 68. Thomas JA, McIntosh JM. Are incentive spirometry, in-
general anesthesia. Acta Anaesthesiol Scand. 1997;41:348. termittent positive pressure breathing, and deep breath-
48. Xue FS, Huang YG, Tong SY, et al. A comparative study ing exercises effective in the prevention of postoperative
of early postoperative hypoxemia in infants, children, and pulmonary complications after upper abdominal surgery?
adults undergoing elective plastic surgery. Anesth Analg. A systematic review and meta-analysis. Phys Ther. 1994;74:3.
1996;83:709. 69. Overend TJ, Andersom CM, Levy SD. The effect of incentive
49. Myles PS, Iacono GA, Hung JO. Risk of respiratory com- spirometry on postoperative pulmonary complications, a
plications and would infection in patients undergoing systematic review. Chest. 2001;120:971.
CHAPTER 8/HYPOXEMIA AND HYPERC APNIA 141
70. Gan TJ. Risk factors for postoperative nausea and vomiting. 90. Soto RG, Fu ES, Vila H, et al. Capnography accurately
Anesth Analg. 2006;102:1884. detects apnea during monitored anesthesia care. Anesth
71. Borgeat A, Ekatodramis G, Schenker CA. Postoperative Analg. 2004;99:379.
nausea and vomiting in regional anesthesia. Anesthesiology. 91. Burton JH, Harrah JD, Germann CA, et al. Does end tidal
2003;98:530. carbon dioxide monitoring detect respiratory events prior
72. Warner MA, Warner ME, Warner DO, et al. Perioperative to current sedation monitoring practices? Acad Emerg Med.
pulmonary aspiration in infants and children. Anesthesiol- 2006;13:500.
ogy. 1999;90:66. 92. Koniaris LG, Wilson S, Drugas G, et al. Capnographic mon-
73. Ng A, Smith G. Gastroesophageal reflux and aspiration itoring of ventilatory status during moderate (conscious)
of gastric contents in anesthetic practice. Anesth Analg. sedation. Surg Endosc. 2003;17:1261.
2001;93:494. 93. Bhavani-Shankar K, Moseley H, Kumar AY. Capnometry
74. Berg H, Viby-Mogensen J, Roed J, et al. Residual neu- and anaesthesia. Can J Anaesth. 1992;39:617.
romuscular block is a risk factor for postoperative pul- 94. Dahan A, Sarton E, Teppema I, et al. Sex related differences
monary complications: A prospective, randomized, and in the influence of morphine on ventilatory control in
blinded study of postoperative pulmonary complications humans. Anesthesiology. 1998;88:903.
after atracurium, vecuronium, and pancuronium. Acta 95. Etches RC. Respiratory depression associated with patient
Anaesthesiol Scand. 1997;41:1095. controlled analgesia: A review of eight cases. Can J Anaesth.
75. Debaene B, Plaud B, Dilly MP. Residual paralysis in the 1994;41:125.
PACU after a single intubating dose of nondepolarizing 96. Erikson LI. The effects of residual neuromuscular blockade
muscle relaxant with an intermediate duration of action. and volatile anesthetics on the control of ventilation. Anesth
Anesthesiology. 2003;98:1042. Analg. 1999;89:243.
76. Eriksson LI, Sundman E, Olsson R, et al. Functional as- 97. Wheatley RG, Shephard D, Jackson IJB, et al. Hypoxaemia
sessment of the pharynx at rest and during swallowing in and pain relief after upper abdominal surgery: Comparison
partially paralyzed humans: Simultaneous videomanome- of IM and patient controlled analgesia in the postoperative
try and mechanomyography of awake human volunteers. patient. Br J Anaesth. 1992;69:558.
Anesthesiology. 1997;78:1035. 98. Borgbjerg FM, Nielsen K, Franks J. Experimental pain
77. Tramer MR, Fuchs-Buder T. Omitting antagonism of stimulates respiration and attenuates morphine induced
neuromuscular block: Effect on postoperative nausea and
respiratory depression: A controlled study in human volun-
vomiting and risk of residual paralysis. A systematic review.
teers. Pain. 1996;64:123.
Br J Anaesth. 1999;82:379.
99. Karan S, Voter W, Palmer L, et al. Effects of pain and audio-
78. Woehick HJ, Dunning M, Raza T, et al. Physical factors af-
visual stimulation on the opioid-induced depression of the
fecting the production of carbon monoxide from anesthetic
hypoxic ventilatory response. Anesthesiology. 2005;103:384.
breakdown. Anesthesiology. 2001;94:453.
100. Strauss SG, Lynn AM, Bratton SL, et al. Ventilatory re-
79. Firth PG, Head CA. Sickle cell disease and anesthesia.
sponse to CO2 in children with obstructive sleep apnea from
Anesthesiology. 2004;101:766.
adenotonsillar hypertrophy. Anesth Analg. 1999;89:328.
80. Russell GB, Graybeal JM. Hypoxemic episodes of patients
101. Strauser LM, Helikson MA, Tobias JD. Anesthetic care for
in a postanesthesia care unit. Chest. 1993;104:899.
the child with congenital central alveolar hypoventilation
81. Lawrence VA, Dhanda R, Hilsenbeck SG, et al. Risk of
syndrome (Ondines curse). J Clin Anesth. 1999;11:431.
pulmonary complications after elective abdominal surgery.
102. Welborn LG, Hannallah RS, Luban NL, et al. Anemia and
Chest. 1996;110:774.
postoperative apnea in former preterm infants. Anesthesi-
82. Tyler DC, Woodham M, Stocks J, et al. Oxygen saturation
in children in the postoperative period. Anesth Analg. ology. 1991;74:1003.
1995;80:14. 103. Krane EJ, Haberkern CM, Jacobson LE. Postoperative
83. DeLeon-Casasola OA. Postoperative epidural opioid anal- apnea, bradycardia, and oxygen desaturation in formerly
gesia. Anesth Analg. 1996;83:867. premature infants. Anesth Analg. 1995;80:7.
84. American Society of Anesthesiologists. A report by the 104. Gross JB, Blouin RT, Zandsberg S, et al. Effect of flumazenil
American Society of Anesthesiologists Task Force on on ventilatory drive during sedation with midazolam and
postanesthetic care: Practice guidelines for postanesthetic alfentanil. Anesthesiology. 1996;85:713.
care. Anesthesiology. 2002;96:742. 105. Timmers HJ, Wieling W, Karemaker MJ, et al. Denervation
85. Akca O, Podolsky A, Eisenhuber E, et al. Comparable of carotid baroreceotors and chemoreceoptors in humans.
postoperative pulmonary atelectasis in patients given 30% J Physiol. 2003;555:3.
or 80% oxygen during and 2 hours after colon resection. 106. Eastwood PR, Platt PR, Shepherd K, et al. Collapsibility
Anesthesiology. 1999;91:991. of the upper airway at different concentrations of propofol
86. Rusca M, Proietti S, Schnyder P, et al. Prevention of at- anesthesia. Anesthesiology. 2005;103:470.
electasis formation during induction of general anesthesia. 107. Carmichael FJ, Mcguire GP, Wong DT, et al. Computed
Anesth Analg. 2003;97:1835. tomographic analysis of airway dimensions after carotid
87. Hardman JG, Aitkenhead AR. Estimating alveolar dead endarterectomy. Anesth Analg. 1996;83:12.
space from the arterial too end tidal CO2 gradient: 108. Ho LI, Harn HJ, Lien TC, et al. Postextubation laryngeal
A modeling analysis. Anesth Analg. 2003;97:1846. edema in adults: Risk factor evaluation and prevention by
88. Berkenbosch JW, Lam J, Burd RS, et al. Noninvasive hydrocortisone. Intensive Care Med. 1996;22:933.
monitoring of carbon dioxide during mechanical ventilation 109. DHonneur G, Lofaso F, Drummond GB, et al. Susceptibility
in older children: End tidal vs. transcutaneous techniques. to upper airway obstruction during partial neuromuscular
Anesth Analg. 2001;92:427. block. Anesthesiology. 1998;88:371.
89. Vivien B, Marmion F, Roche S, et al. An evaluation of 110. Lambert DA, Giannouli E, Schmidt BJ. Postpolio syndrome
transcutaneous carbon dioxide partial pressure monitoring and anesthesia. Anesthesiology. 2005;103:638.
during apnea testing in brain-dead patients. Anesthesiology. 111. Putnam MT, Wise RA. Myasthenia gravis and upper airway
2006;104:701. obstruction. Chest. 1996;109:400.
142 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
112. Jansen NF, Weiler JM. C1 esterase inhibitor deficiency, 129. Al-Kaisy AA, Chan VWS, Perlas A. Respiratory effects of low
airway compromise, and anesthesia. Anesth Analg. 1998; dose bupivacaine interscalene block. Br J Anaesth. 1999;
87:480. 82:217.
113. Isono S, Tanaka A, Ishikawa T, et al. Sniffing position im- 130. Tobias JD, Del Caomp L, Kenter K, et al. Changes in
proves pharyngeal airway patency in anesthetized patients transcutaneous carbon dioxide, oxygen saturation, and
with obstructive sleep apnea. Anesthesiology. 2005;103:489. respiratory rate after interscalene block. South Med J. 2004;
114. Isono S, Tanaka A, Nishino T. Lateral position decreases 97:21.
collapsibility of the passive pharynx in patients with 131. Kopman AF, Yee PS, Neuman GG. Relationship of the train-
obstructive sleep apnea. Anesthesiology. 2002;97:780. of-four fade ratio to clinical signs and symptoms of residual
115. McCaul CL, Harney D, Ryan M, et al. Airway management paralysis in awake volunteers. Anesthesiology. 1997;86:
in the lateral position: A randomized controlled trial. Anesth 765.
Analg. 2005;101:1221. 132. Kopman AF, Ng J, Zank LM, et al. Residual postoperative
116. Ovassapian A, Tuncibilek M, Weitzel EK, et al. Airway paralysis: Pancuronium versus mivicurium, does it matter?.
management in adult patients with deep neck infections: Anesthesiology. 1996;85:1253.
A case series and review of the literature. Anesth Analg. 2005; 133. Nuckton TJ, Alonso JA, Kallet RH. Pulmonary dead space
100:585. fraction as a risk factor for death in the acute respiratory
117. Drummond GB. Comparison of sedation with midazolam distress syndrome. N Engl J Med. 2002;346:1281.
and ketamine: Effects on airway muscle activity. Br J 134. Schlager A, Luger TJ. Oxygen application by a nasal probe
Anaesth. 1996;76:663. prevents hypoxia but not rebreathing of carbon dioxide in
118. Asai T, Koga K, Vaughan RS. Respiratory complications patients undergoing eye surgery under local anaesthesia.
associated with tracheal intubation and extubation. Br J Br J Ophthalmol. 2000;84:399.
Anaesth. 1998;80:767. 135. Bozkurt P, Kaya G, Yeker Y, et al. Arterial carbon dioxide
119. Schreiner MS, OHara I, Markakis DA, et al. Do children markedly increases during diagnostic laparoscopy in portal
who experience laryngospasm have an increased risk of up- hypertensive children. Anesth Analg. 2002;95:1236.
per respiratory tract infection? Anesthesiology. 1996;85:475. 136. Murdock C, Wolff A, Van Geem T. Risk factors for hy-
120. Skolnick ET, Vomvolakis MA, Buck KA, et al. Exposure percarbia, subcutaneous emphysema, pneumothorax and
to environmental tobacco smoke and the risk of adverse pneumomediastinum during laparoscopy. Obstet Gynecol.
respiratory events in children receiving general anesthesia. 2000;95:704.
Anesthesiology. 1998;88:1144. 137. Morisake H, Serita R, Innami Y, et al. Permissive hypercap-
121. Naguib M, Samarkandi AH, Abdullah K, et al. Succinyl- nia during thoracic anaesthesia. Acta Anaesthesiol Scand.
choline dosage and apnea-induced hemoglobin desatura- 1999;43:845.
tion in patients. Anesthesiology. 2005;102:35. 138. DAngelo E, Calderini IS, Tavola M. The effects of CO2 on
122. Mathru M, Esch O, Lang J, et al. Magnetic resonance respiratory mechanics in anesthetized paralyzed humans.
imaging of the upper airway. Anesthesiology. 1996;84:273. Anesthesiology. 2001;94:604.
123. Warner DO, Warner MA, Barnes RD, et al. Perioperative 139. Mutlu GM, Factor P, Schwartz DE, et al. Severe sta-
respiratory complications in patients with asthma. Anesthe- tus asthmaticus: Management with permissive hypercap-
siology. 1996;85:460. nia and inhalation anesthesia. Crit Care Med. 2003;31:
124. Pelosi P, Croci M, Ravagnan I, et al. The effects of body mass 665.
on lung volumes, respiratory mechanics, and gas exchange 140. Doufas AO, Morioka N, Iscoe S, et al. Hypercapnia improves
during general anesthesia. Anesth Analg. 1998;87:654. tissue oxygenation. Anesthesiology. 2002;97:801.
125. Reilly PM, Sing RF, Giberson FA, et al. Hypercarbia during 141. Hanel F, von Knobelsdorff G, Werner C, et al. Hypercapnia
tracheostomy: A comparison of percutaneous endoscopic, prevents jugular bulb desaturation during rewarming
percutaneous doppler, and standard surgical tracheostomy. from hypothermic cardiopulmonary bypass. Anesthesiology.
Intensive Care Med. 1997;23:859. 1998;89:19.
126. Bevan DR, Kahwaji R, Arsermino JM, et al. Residual block 142. Fleischmann E, Herbst F, Kugener A, et al. Mild hyper-
after mivacurium with or without edrophonium reversal in capnea increases subcutaneous and colonic oxygen tension
adults and children. Anesthesiology. 1996;84:362. in patients given 80% inspired oxygen during abdominal
127. Sharma RR, Axelsson H, Oberg A, et al. Diaphragmatic surgery. Anesthesiology. 2006;104:944.
activity after laparoscopic cholecystectomy. Anesthesiology. 143. Atallah MM, Demian AD, el-DIasty TA, et al. Can we
1999;91:406. increase hepatic oxygen availability? The role of intentional
128. Casati A, Fanelli G, Cedrati V, et al. Pulmonary function hypercarbia. Middle East J Anesthesiol. 2000;15:503.
changes after interscalene brachial plexus anesthesia with 144. Enoki T, Tsuchiya N, Shinomura T, et al. Effect of
0.5% and 0.75% ropivacaine: A double blind comparison hypercapnia on arterial hypotension after induction of
with 2% mepivacaine. Anesth Analg. 1999;88:587. anaesthesia. Acta Anesthesiol Scand. 2005;49:687.
CHAPTER BRONCHOSPASM
CASE SUMMARY postoperatively and tapered over the next few days. He
was discharged 4 days later.
55-year-old, 80 kg, intravenous drug user
A
was scheduled for an incision and drainage
of a right shoulder abscess. He had a long-
standing history of asthma and had been What Baseline Knowledge Is
hospitalized several times in the past for
acute exacerbations of his asthma. He had Relevant?
a small meal 3 hours before the surgery. He received
an intravenous dose of hydrocortisone and nebulized
albuterol in the emergency department. PHYSIOLOGY OF
On physical examination, he was sitting upright, had
sternal retractions and dyspnea with bilateral diffuse BRONCHOMOTOR TONE
wheezing. His right arm, including the axilla and upper
The walls of bronchi and bronchioles are composed
chest, were swollen and inflamed. Laboratory tests showed
mainly of smooth muscle and cartilage plates. The smooth
a serum bicarbonate of 35 mmol per L and a normal elec- muscle bundles encircle the airways obliquely from the
trolyte panel. A chest radiograph showed subcutaneous trachea down to the alveolar ducts. Contraction of these
air in the distribution of the abscess and hyperinflated results in both narrowing and shortening of the airway.
lung fields bilaterally. The parasympathetic nervous system is the dominant neu-
A rapid sequence induction was performed using ronal pathway that controls the airway smooth muscle
ketamine and vecuronium. The trachea was intubated tone. Postganglionic parasympathetic fibers innervate the
and the lungs ventilated with 4% sevoflurane in oxygen, airway smooth muscle down to the level of the terminal
with a tidal volume of 600 mL and an inspiratory bronchioles. Stimulation of the cholinergic nerves causes
to expiratory time ratio (I:E) of 1:4. Bilateral diffuse bronchoconstriction, mucus secretion, and bronchial va-
wheezing was unchanged. His peak inspiratory pressure sodilation. Irritant receptors are found just beneath the
was 36 cm H2 O, and the end-tidal CO2 was 48, with a tight junctions of the epithelial lining of the airway.
prolonged upslope on the capnograph. Shortly thereafter, The afferent and efferent connections of these receptors
the peak inspiratory pressure increased to 60 cm H2 O with travel through the vagus nerve. Acetylcholine admin-
concomitant hypotension. Auscultation revealed a silent istered exogenously or released from parasympathetic
chest. Intravenous boluses of ephedrine were titrated postganglionic nerves induces airway constriction by acti-
to restore the blood pressure, while anesthesia was vating M3 muscarinic receptors on airway smooth muscle
deepened with boluses of ketamine, 40 mg. Sevoflurane (see Fig. 9.1). At rest, the normal human airways have a
was continued in an inspired concentration of 3% to 4%. mild baseline constriction due to vagal activity, which can
The bronchospasm resolved during the next few minutes. be blocked by anticholinergic agents such as atropine, gly-
At the end of the procedure, glycopyrrolate and copyrrolate, or ipratropium.1,2 This resting bronchomotor
neostigmine were administered slowly over 15 minutes tone is believed to serve three different purposes: (i) To
while the patient was deeply anesthetized. A large bore provide a balance between anatomic dead space and air-
gastric tube was then passed, and stomach contents way resistance, and hence optimize the work of breathing
were suctioned out. The trachea was extubated and and gas exchange;3 (ii) prevent collapse of the larger
the patient observed in a left lateral position, applying cartilaginous bronchi during coughing; and, (iii) mini-
cricoid pressure until he was conscious. Intravenous mize the collapse of smaller noncartilaginous bronchi
hydrocortisone and albuterol nebulizers were continued at low lung volumes. Sympathetic nerves may control
143
144 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
CNS
What Are the Differences
M2 receptors inhibit
Vagus release of ACh Between Chronic Obstructive
(negative feedback) Pulmonary Disease and
M1 receptors
M2 receptors Asthma?
Perioperative bronchospasm is frequently, but not ex-
clusively, encountered in patients with reactive airway
ACh ACh
disease, most of which include those with asthma, chronic
obstructive pulmonary disease (COPD), and bronchopul-
monary dysphasia. Although varying degrees of expiratory
airflow limitation and bronchospasm are features of
asthma and COPD, there are some fundamental patho-
physiologic and clinical differences between the two.
These are summarized in Table 9.1 and briefly discussed
M3 receptors in the following text.
Airway smooth muscle
mucus glands
TABLE 9.1 Differences Between Chronic Obstructive Pulmonary Disease (COPD) and Asthma
Asthma COPD
Characteristic
Age at onset Younger (often during childhood) Older (age >40 y)
Allergic etiology Allergies present in >50% of patients None
Smoking status Nonsmokers affected Usually history of heavy smoking
Treatment Response
Bronchodilators Reversible Partial reversibility
Corticosteroids Good Poor
Airflow limitation (FEV1 ) Can normalize after resolution of Cannot normalize; always reduced;
episode deterioriates with advancing disease
Pathology
Airways All Central (bronchitis)
Peripheral (emphysema)
Parenchyma Not involved Destruction
Airway hyperresponsiveness Present May or may not be present
Bronchial smooth muscle Enlarged mass in large airways Enlarged mass in small airways
Epithelium Shedding Metaplasia
Mucus secretion Present Present, heavy
Goals of Therapy
Risk factors Reduce exposure Reduce exposure
Symptoms Control Relief
Pulmonary function/airflow Maintain normal or close-to-normal Prevent progression of pulmonary
limitation pulmonary function; prevent dysfunction
development of irreversible airflow
limitation
Activity and exercise Maintain normal levels of activity and Improve exercise tolerance (strength,
exercise endurance)
Exacerbations Prevent and treat Prevent and treat
Drug related adverse event Avoid Avoid
Disease related mortality Prevent Reduce
CHRONIC OBSTRUCTIVE In the United States, roughly 14.2 million people have
COPD, of whom 12.5 million have chronic bronchitis and
PULMONARY DISEASE 1.7 million have emphysema. It has been estimated that
COPD is defined as a disease state characterized by the the prevalence of chronic airflow obstruction in the United
presence of airflow obstruction due to chronic bronchitis States is 8% to 17% for men and 10% to 19% for women.
or emphysema that is generally progressiveand not fully Pathophysiologic changes characteristic of COPD
reversibleand may be accompanied by airway hyperre- include mucus hypersecretion, ciliary dysfunction, air-
activity and an abnormal inflammatory response of the flow limitation, pulmonary hyperinflation, gas exchange
lungs.5 COPD, generally, is secondary to tobacco use; less abnormalities, pulmonary hypertension, and cor pul-
frequently, cystic fibrosis, -1 antitrypsin deficiency, and monaleand they usually develop in this order over the
bronchiectasis cause similar conditions. Chronic bronchi- course of the disease.5 The airflow limitation character-
tis is defined clinically as the presence of a chronic pro- istic of COPD is that it is primarily irreversible, with a
ductive cough for 3 months during each of 2 consecutive small reversible component. The irreversible component
years (other causes of cough being excluded). Emphysema of airflow limitation is primarily because of remodel-
is defined as an abnormal, permanent enlargement of the ingthat is, fibrosis and narrowingof the small airways
air spaces distal to the terminal bronchioles, accompanied that produces fixed airways obstruction and a conse-
by destruction of their walls and without obvious fibrosis. quent increase in airways resistance. The sites of airflow
146 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
limitation in COPD are the smaller conducting airways, intravenous routes, and some studies suggest greater effi-
including bronchi and bronchioles <2 mm in internal cacy by the inhaled route.
diameter. Loss of elastic recoil and alveolar support to Albuterol is the most commonly used 2 agonist. It is
maintain the patency of small airways, secondary to alve- occasionally administered orally as a 2 to 4 mg per dose,
olar parenchymal destruction, also contributes to the every 6 to 8 hours; but more commonly by a metered dose
irreversible airflow obstruction. More importantly for inhaler that delivers approximately 100 g per puff. The
the anesthesiologist, airway smooth muscle contraction, usual dose is one to two puffs every 4 to 6 hours, not to
ongoing airway inflammation, and intraluminal accumu- exceed 12 puffs per 24 hours; it can also be wet-nebulized
lation of mucus and plasma exudate may be responsible in a dose of 2.5 to 5 mg every 4 to 6 hours. Albuterol and
for the small part of airflow limitation that is reversible terbutaline are short-acting 2 agonists with a duration of
with treatment, especially during acute exacerbations. action of 4 to 6 hours, whereas long-acting agents such
The role of hypoxic pulmonary drive is often overem- as salmeterol and formoterol have a duration of action
phasized in patients with severe COPD. Oxygen therapy exceeding 12 hours. The longer-acting drugs are used
is generally safe in patients with COPD. Although oxygen primarily for prophylaxis rather than rescue therapy.
toxicity from high-inspired concentrations (FiO2 >0.6) is
well recognized, little is known about the long-term effects
of low flow oxygen. The British Medical Research Council STEROIDS
(MRC) study and the National Heart, Lung, Blood In-
stitutes Nocturnal Oxygen Therapy Trial (NOTT) study Anti-inflammatory medication remains the preferred
found that long-term oxygen therapy improves survival in treatment for chronic asthma and prevention of acute
patients with hypoxemia and COPD twofold or more.6 exacerbations of asthma. The mechanism of action of
Hypoxemia in these trials was defined as a PaO2 of steroids is not fully understood but is dependent on the
<55 mm Hg, or SaO2 <90%. Supplemental oxygen was binding of the steroids to cytosolic receptors that translo-
used from 15 to 19 hours per day. The increased survival cate to the nucleus to affect gene transcription/translation.
and quality-of-life benefits of long-term oxygen therapy Hence, any beneficial effects have a typical lag time of
outweigh the possible risks. The increase in PACO2 in pa- at least 6 to 8 hours. The anti-inflammatory effects of
tients with COPD who are on oxygen therapy is more glucocorticoids include decreased inflammatory cell in-
likely a consequence of ventilation/perfusion mismatch- flux, a reduction in the release of mast cell mediators
ing rather than respiratory center depression. Although (and reduced mast cell numbers), and decreased mi-
respiratory center depression as a result of blunting of crovascular leakage, and hence airway edema. Inhaled
hypoxic ventilatory drive with oxygen therapy is not com- corticosteroids used in asthma include beclomethasone,
mon, it is best avoided by titration of oxygen delivery to budesonide, flunisolide, fluticasone, and triamcinolone.
maintain a PaO2 at 60 to 75 mm Hg or SaO2 in the 90% to Although intravenous hydrocortisone (4 mg per kg) may
94% range. be used in an urgent setting, such as in the patient dis-
Patients with COPD may present for COPD-related cussed in our case summary, elective preoperative patients
surgical procedures such as bullectomy, lung volume may benefit from a course of methylprednisone started
reduction, lung transplantation, or elective or emergency 3 to 5 days before surgery.
procedures that are not related to COPD. Given the low morbidity of a short course of steroids,
it is prudent to ensure that asthmatics who are not fully
controlled receive a burst of steroids preoperatively. Sim-
ilarly, any suggestion of intraoperative or postoperative
Which Pharmacological Agents wheezing may justify their use.
Are Relevant for the Treatment
of Reactive Airway Disease? MAST CELL STABILIZERS
Cromolyn inhibits the degranulation of sensitized mast
AGONISTS cells following exposure to specific antigens. It attenuates
the bronchospasm caused by exercise, cold air, aspirin,
The predominant airway smooth muscle receptor is the and environmental pollutants, but is of little use for acute
2 -adrenergic receptor. Agonists combine with the re- episodes of bronchoconstriction.
ceptors and cause an increase in intracellular cyclic adeno-
sine 3 , 5 monophosphate (cAMP)which activates pro-
tein kinases, decreases intracellular levels of calcium, and, ANTICHOLINERGICS
in turn, causes airway smooth muscle relaxation and in-
hibits the release of mediators from the mast cells. Anticholinergic drugs compete with acetylcholine for
Intravenous agonists such as isoproterenol can postganglionic muscarinic receptors, thereby inhibit-
cause serious ventricular arrhythmias and myocar- ing cholinergically mediated bronchomotor tone, result-
dial damage when administered intravenously. Selective ing in bronchodilatation. They block vagally mediated
2 -adrenergic agents, such as terbutaline or albuterol, reflex arcs that cause bronchoconstriction. Treatment
are equally effective when administered by inhaled or with aerosolized anticholinergic agents (e.g., ipratropium
C H A P T E R 9/ B R O N C H O S PA S M 147
bromide and tiotropium) may be more effective than a secretions, and sympathetic stimulation), rather than a
2 agonist in patients with COPD. Ipratropium bromide lack of benefit of the drug.
has bronchodilatory activity with minimal adverse effects Propofol, midazolam, and etomidate all relax airway
and is administered by a metered dose inhaler. Studies in smooth muscle in vitro,9,10 whereas barbiturates may
patients with stable COPD have shown that ipratropium have direct bronchoconstricting effects.9 Propofol may
bromide has equivalent or superior activity when com- also have indirect effects on airway constriction, perhaps
pared with a 2 agonist. In combination with a 2 agonist, through inhibition of vagal tone. Clinically, propofol
an additional 20% to 40% bronchodilation occurs. Ipra- has been shown to be superior to the barbiturates and
tropium has a slower onset (e.g., 30 to 60 minutes) and etomidate in reducing wheezing and airway resistance
longer duration than a 2 agonist and is less suitable for in both asthmatic and nonasthmatic subjects.1113 In
use on an as-needed basis. asthmatics induced with either thiopental, methohexital,
or propofol at equipotent doses, none wheezed following
tracheal intubation when propofol was used, whereas
METHYLXANTHINES both of the barbiturates resulted in a significant incidence
of wheezing.13 In the patient described in this chapter,
Aminophylline and theophylline were extensively used in ketamine had the advantage of vasomotor stability in the
the 1970s and 1980s for their bronchodilating properties. presence of intravascular volume depletion.
The popularity of methylxanthines has decreased dur-
ing the last decade because of the narrow therapeutic
range, large variation in interindividual pharmacokinet- HALOGENATED
ics, and frequent toxicity. The mechanisms of benefi-
cial action may involve increased intracellular calcium HYDROCARBONS (VOLATILE
transport, adenosine antagonism, and prostaglandin E2 ANESTHETICS)
inhibition. Additionally, methylxanthines may improve
diaphragm muscle contractility. The target blood level is All of the volatile anesthetics have direct, and perhaps
10 g per mL, and toxic concentration is >20 g per indirect, relaxant effects on airway smooth muscle in ex-
mL. Many drugs, including alcohol, -blockers, cimeti- perimental models.14 Although differences in the potency
dine, and macrolide antibiotics and quinolones (includ- of these agents are present in vitro, the clinical impor-
ing erythromycin, clarithromycin, ciprofloxacin) decrease tance of these differences is unclear. Sevoflurane is more
aminophylline clearance, thereby raising serum levels and effective than isoflurane, desflurane, and halothane in re-
the potential for increased toxicity. Other agents such as ducing airway resistance after endotracheal intubation
phenytoin, rifampicin, and tobacco and marijuana smok- in patients,15 but does not prevent an increase in air-
ing may increase the clearance of aminophylline, thereby way resistance after intubation of asthmatic children.16
decreasing serum concentrations, risking subtherapeutic A study showed that while sevoflurane and isoflurane
dosing. Toxicity may present in the form of cardiovas- cause a dose-dependent bronchodilatation, desflurane
cular manifestations such as tachycardia, palpitations, decreases airway resistance at 1 MAC, but causes an
extrasystole, flushing, hypotension, circulatory failure, increase in airway resistance at 2 MAC.17 There are no
atrial and ventricular arrhythmia, or central nervous sys- prospective, controlled studies comparing deep inhalation
tem (CNS) manifestations such as headache, irritability, anesthesia to intravenous induction with bronchoprotec-
anxiety, tremor, dizziness, hyperexcitability, and seizures tive agents such as ketamine or propofol in high-risk
(our patient presented with toxic serum levels, atrial fib- patients.
rillation and seizures).
There is no evidence that outcome from an acute
asthmatic attack is improved if theophylline is added to M2 RECEPTORS
a regimen of agonists and steroids, and the potential
toxicity is significant.
ANTAGONISTS
Airway smooth muscle expresses both M2 and M3
muscarinic receptors. In airway smooth muscle, M3 mus-
INTRAVENOUS ANESTHETIC carinic receptors initiate contraction, whereas M2 mus-
AGENTS carinic receptors inhibit further release of acetylcholine
and serve as a negative feedback mechanism (Fig. 9.1).
Among anesthetic induction agents, considerable exper- Under normal circumstances, following vagal stimula-
imental evidence suggests that ketamine has both direct tion, binding of acetylcholine to M3 muscarinic recep-
and indirect relaxant effects on airway smooth muscle tors on the muscle and consequent bronchoconstriction
through non-receptor mechanisms.7 However, the clin- overrides the stimulation of M2 muscarinic receptors.
ical data supporting the use of ketamine for the prevention However, blockade of M3 muscarinic receptors on the
or treatment of bronchospasm is largely anecdotal, and airway smooth muscle inhibits both vagally induced and
in more rigorous trials, unimpressive.8 This could be due exogenously administered acetylcholine-induced airway
to the reluctance to routinely use ketamine at the high constriction. Studies in animal models show that pan-
doses needed to produce bronchodilatation because of curonium and gallamine, at clinically relevant concen-
side effects such as dysphoria, hallucinations, increased trations, can potentiate bronchospasm in the setting of
148 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
vagally induced acetylcholine release. Rapacuronium, a in airway caliber following airway instrumentation are
neuromuscular blocking drug was withdrawn from the thought to result largely from parasympathetic nervous
market due to a high incidence of bronchospastic reac- system activation of airway smooth muscle.24 Such re-
tions associated with its use. A possible mechanism for sponses can be blocked by a muscarinic blockade with
severe bronchospasm following rapacuronium is prefer- either systemic or inhaled anticholinergic agents.
ential blockade of presynaptic M2 muscarinic receptors on Tracheal intubation without adequate neuromuscu-
parasympathetic nerves during a period of marked vagal lar blockade may also induce bronchospasm by inducing
activity due to laryngoscopy and intubation. This would coughing. A cough will reduce the lung volume which, in
result in an enhanced (unopposed) release of acetylcholine turn, markedly increases bronchoconstriction in response
from activated parasympathetic nerves, leading to activa- to a stimulus.25 In a patient with known reactive air-
tion of M3 muscarinic receptors on airway smooth muscle ways, the prevention of coughing at the time of tracheal
and the resultant bronchoconstriction.18 intubation by using either a deep level of anesthetic or
a muscle relaxant may help minimize the likelihood of
bronchospasm.
causation of intraoperative anaphylaxis has been disputed TABLE 9.2 Differential Diagnosis of Wheezing Under
in several studies. Most hypersensitivity reactions are of General Anesthesia
immunologic origin (immunoglobulin E [IgE]-mediated)
or related to direct stimulation of histamine release. Preexisting Bronchospastic Disease
Either of these mechanisms can result in a bronchospastic
Asthma
response that may or may not be accompanied by
Chronic obstructive pulmonary disease
cardiovascular or cutaneous manifestations. In addition
Bronchopulmonary dysplasia
to treating the hypersensitivity reaction with epinephrine,
IV fluids, steroids, and antihistaminics, bronchospasm Bronchospasm Induced Under Anesthesia
may require specific treatment with inhaled 2 agonists.
Tracheal intubation
Surgical stimulation under light plane of anesthesia
Hypersensitivity reactions (histaminoid, anaphylactoid
RECENT UPPER or anaphylactic)
RESPIRATORY INFECTION Aspiration of gastric contents
Mainstem bronchial intubation or carinal irritation by
Data on the incidence of bronchospasm in patients endotracheal tube
(especially children) with an upper respiratory infection Pulmonary edema (cardiogenic or negative pressure)
(URI) is conflicting. Olsson reported a higher frequency
of bronchospasm in patients with an URI.20 Rolf and Conditions Mimicking Bronchospasm (increased
Cote also noted a higher frequency of bronchospasm peak inspiratory pressure, decreased tidal volume,
in intubated patients with an URI (2 in 15) compared wheezing)
to those without an URI (1 in 181).28 However, in a Kinked tracheal tube or breathing system
larger prospective study of children with an active or Secretions
recent URI undergoing elective surgical procedures, Tait Obstruction of the tube by overinflated or herniated
et al. observed the incidence of bronchospasm to be cuff
5.7% in children with an active URI, 2.7% in those Pneumothorax (simple or tension)
with a recent URI (<4 weeks), and 3.3% in those with a
neither active nor recent URI (not a statistically significant
difference).29 Although children with acute and recent wheezes is asthma or bronchospasm. The differential
URIs are at greater risk for respiratory complications, this diagnosis of wheezing under anesthesia is outlined in
study suggests that this subset of children can undergo Table 9.2.
elective procedures without significant increase in adverse
anesthetic outcomes.
TABLE 9.3 Areas of Focus in Preoperative History and TABLE 9.4 Strategies to Prevent Perioperative
Examination of Patients with Hyperreactive Airways Bronchospasm
beneficial effects of LMA over tracheal intubation have in a higher incidence of adverse airway events and carries
been noted in children with URIs,40 and in adults without the risk of severe airway hyperreactivity.50
lung disease.41
STEROIDS
TOPICAL ANESTHESIA
Silvanus et al. compared a preoperative albuterol in-
Gal and Suratt42 demonstrated a two-fold increase in haler with and without oral prednisone (40 mg daily) for
lower airway resistance following tracheal intubation of 5 days in patients with bronchial hyperreactivity under-
awake volunteers under topical anesthesia. However, the going elective surgery. They observed that preoperative
same author also observed a bronchodilator response to treatment with combined corticosteroids and albuterol
aerosolized lidocaine in unanesthetized subjects.43 Given minimizes intubation-evoked bronchoconstriction much
these considerations and the time required to administer more effectively than the inhaled albuterol alone.51
the aerosol compared to the immediacy and efficacy of
intravenous drugs or other inhaled drugs, aerosolized
lidocaine is probably a poor choice for the attenua-
tion of bronchoconstriction associated with endotracheal What Therapeutic Measures Are
intubation.
Taken to Treat Bronchospasm?
TABLE 9.6 Ventilator Management of Patients with rate consistent with effective carbon dioxide elimination,
Severe Bronchospasm 3) fastest inspiratory flow rates to decrease the I:E ratio.
Excessive iPEEP and dynamic hyperinflation may
also increase intrathoracic pressure and decrease right
Slow rates (612 per minute), long expiratory times
and left ventricular preload, leading to a decrease in
(I:E ratio of 1: 31:5)
cardiac output and hypotension; this effect is called vo-
Low to moderate tidal volumes (610 mL/kg)
lutrauma. Hypotension is a common complication of me-
Permissive hypercapnia (accept PACO2 values in the
chanical ventilation in severe bronchospasm52 and should
60100 mm Hg range), ensure adequate
be anticipated and promptly treated as was illustrated in
oxygenation
our patient. The most serious consequence of iPEEP and
No PEEP in the ventilator setting, watch for iPEEP, signs
dynamic hyperinflation is barotraumas, although pneu-
of barotrauma
mothorax, pneumomediastinum, and pneumoperitoneum
Manual emptying of lungs (by squeezing the chest
may also occur. Barotrauma correlates with the degree of
while the tracheal tube is disconnected from the
dynamic hyperinflation. Barotrauma was found to com-
ventilator) to FRC if severe hypotension/PEA
plicate status asthmaticus in 14% to 27% of patients.52
In the authors experience with reviewing adverse
PEEP, positive end-expiratory pressure; iPEEP, intrinsic positive end-
outcomes, circulatory collapse from insufficient venous
expiratory pressure; FRC, functional residual capacity; PEA, pulseless
electrical activity.
return often seems to be the terminal event in patients on
positive-pressure ventilation with severe bronchospasm.
Permissive Hypercapnia
VENTILATORY MANAGEMENT
Mechanical ventilation during an attack of severe bron-
The ventilatory management of bronchospasm under chospasm can be challenging. Permissive hypercapnia is
anesthesia is derived from the experience gained from a relatively safe strategy in the ventilatory management
ventilating patients with status asthmaticus. A low tidal of such patients. High levels of hypercapnia and associ-
volume (8 mL per kg) and a slow respiratory rate (6 to12 ated severe acidosis are well tolerated in the absence of
breaths per minute) are set to provide a prolonged ex- contraindications, such as preexisting intracranial hyper-
piratory time, thereby minimizing air trapping and its tension.53 The role of pH correction during permissive
consequences of hemodynamic compromise and baro- hypercapnia has not been resolved. Aggressive correction
trauma. Aiming for a minute ventilation of 100 mL per of pH is of unproven value, and it is even possible that hy-
kg and plateau pressure <25 cm H2 O will usually prevent percapnic acidosis may have a protective effect on vital or-
severe dynamic hyperinflation. Permissive hypercapnia gans. One should not aim for a normal PACO2 ; an arterial
(with levels of PACO2 up to 90 mm Hg) is generally tol- hemoglobin saturation of approximately 90% is probably
erated when adequate oxygenation is achieved; the main more than adequate. Further attempts to achieve a higher
contraindication is intracranial disease. Table 9.6 summa- saturation may compromise a patients clinical status.
rizes the ventilator management for patients with acute
bronchospasm.
Choice of Ventilator
Positive End-Expiratory Pressure Inhalational anesthetics are potent bronchodilators and
and Auto Positive End-Expiratory Pressure have been successfully used in the management of status
asthmaticus refractory to conventional therapy. Inhala-
Exhalation is a passive event during intermittent positive- tional anesthetics have been shown to decrease airway
pressure ventilation, depending on the elastic properties resistance, dynamic hyperinflation, and iPEEP. Several
of the thorax. In a healthy person, the expiratory time is at case reports have described the successful use of in-
least twice the inspiratory time, resulting in an I:E ratio of halational anesthetics in the management of refractory
1:2. When ventilating a patient with bronchospasm, an I:E asthma. One limitation, however, is that the anesthe-
ratio of 1:3 or 1:4 may be desirable to allow for complete sia ventilators are not capable of generating inspiratory
emptying of the lung units. If the expiration is short, or pressures and flows sufficient to ventilate patients with
if there is premature airway closure, breaths can become severely elevated airway resistance. With increasing air-
stacked and intrathoracic pressure increases, resulting way pressures, the inspiratory flow decreases, which
in inadvertent positive pressure at end-expiration (i.e., limits the tidal volume delivered by anesthesia ventila-
intrinsic positive end-expiratory pressure [iPEEP] or auto tors. Consequently, the delivery of volatile anesthetics
PEEP). This results in increased work of breathing for is also impaired. Although newer anesthesia ventilators
self-ventilating patients and predisposition to barotrauma (e.g., Ohmeda 7810, Datex-Ohmeda, Madison, WI) have
in mechanically ventilated patients. Because tidal volume increased flow capabilities, when faced with severe bron-
is a product of inspiratory time and inspiratory flow rate, chospasm the authors prefer to use an intensive care unit
the inspiratory time can be adjusted by changing the ventilator, and administer total intravenous anesthesia.
inspiratory flow rate. The following conditions tend to Volatile anesthetic agents can also be administered by
reduce iPEEP: 1) largest tidal volume possible without some intensive care unit ventilators, such as the Siemens
excessive peak inspiratory pressure, 2) slowest respiratory Servo 900D (Siemens Medical Systems, Iselin, NJ).
C H A P T E R 9/ B R O N C H O S PA S M 153
REFERENCES
GENERAL MEASURES 1. De Troyer A, Yernault JC, Rodenstein D. Effects of vagal
blockade on lung mechanics in normal man. J Appl Physiol.
Hydration, anxiolysis, adequate pain relief, and correction 1979;46:217.
of fluid and electrolyte imbalances are general measures 2. Douglas NJ, Sudlow MF, Flenley DC. Effect of an inhaled
that can expedite recovery and prevent the stimuli from atropine-like agent on normal airway function. J Appl Physiol.
causing further bronchospasm. 1979;46:256.
3. Widdicombe JG, Nadel JA. Airway volume, airway resistance,
and work and force of breathing: Theory. J Appl Physiol.
1963;18:863.
AWAKE VERSUS DEEP AND 4. van der Velden VH, Hulsmann AR. Autonomic innervation
of human airways: structure, function, and pathophysiology
EARLY VERSUS DELAYED in asthma. Neuroimmunomodulation. 1999;6:145.
EXTUBATION 5. Workshop report, global strategy for the diagnosis, man-
agement and prevention of chronic obstructive pulmonary
Although extubation is not as stimulating as intubation disease, September 2005. Global Initiative for Chronic Ob-
for provoking bronchospasm in susceptible patients, structive Lung Disease. Available at: http://www.goldcopd.org
the possibility of postextubation bronchospasm should /Guidelineitem.asp?l1=2&l2=1&intId=989. Last accessed
always be in the forefront. Deep extubation can be unsafe, November 14, 2005.
6. Cranston JM, Crockett AJ, Moss JR, et al. Domiciliary
especially in patients with a potentially full stomach or
oxygen for chronic obstructive pulmonary disease [Review].
those with difficult airway. Our patient was extubated
Cochrane Database Syst Rev. 2005;(4):CD001744.
under deep anesthesia, but allowed to wake up in the 7. Hill GE, Anderson JL, Whitten CW. Ketamine inhibits
operating room in a left lateral (recovery) position, with agonist-induced cAMP accumulation increase in human
cricoid pressure maintained until he was awake. Patients airway smooth muscle cells. Can J Anaesth. 1999;46:1172.
who develop an acute episode of severe wheezing under 8. Lau TT, Zed PJ. Does ketamine have a role in managing
anesthesia may require continuous sedation with paralysis severe exacerbation of asthma in adults? Pharmacotherapy.
and ventilatory support into the postoperative period 2001;21:1100.
until the bronchospasm has resolved. Alternatively, many 9. Ouedraogo N, Roux E, Forestier F, et al. Effects of
patients will benefit from early removal of the tracheal intravenous anesthetics on normal and passively sensitized
tube, which is the most powerful stimulus for bronchial human isolated airway smooth muscle. Anesthesiology.
1998;88:317.
constriction.
10. Cheng EY, Mazzeo AJ, Bosnjak ZJ, et al. Direct relaxant
effects of intravenous anesthetics on airway smooth muscle.
Anesth Analg. 1996;83:162.
11. Eames WO, Rooke GA, Wu RS, et al. Comparison of the
KEY POINTS effects of etomidate, propofol, and thiopental on respi-
ratory resistance after tracheal intubation. Anesthesiology.
1. Perioperative bronchospasm is a common and poten-
1996;84:1307.
tially fatal complication. 12. Wu RS, Wu KC, Sum DC, et al. Comparative effects of
2. Though it is more likely to be seen in patients with thiopentone and propofol on respiratory resistance after
reactive airway disease (such as asthma and COPD), tracheal intubation. Br J Anaesth. 1996;77:735.
154 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
13. Pizov R, Brown RH, Weiss YS, et al. Wheezing during 34. Kasaba T, Suga R, Matsuoka H, et al. Comparison of epidural
induction of general anesthesia in patients with and with- anesthesia and general anesthesia for patients with bronchial
out asthma. A randomized, blinded trial. Anesthesiology. asthma. Masui. 2000;49:1115.
1995;82:1111. 35. Groeben H, Schafer B, Pavlakovic G, et al. Lung function
14. Wiklund CU, Lim S, Lindsten U, et al. Relaxation by under high thoracic segmental epidural anesthesia with ropi-
sevoflurane, desflurane and halothane in the isolated guinea- vacaine or bupivacaine in patients with severe obstructive
pig trachea via inhibition of cholinergic neurotransmission. pulmonary disease undergoing breast surgery. Anesthesiol-
Br J Anaesth. 1999;83:422. ogy. 2002;96:536.
15. Goff MJ, Arain SR, Ficke DJ, et al. Absence of bron- 36. Groeben H. Effects of high thoracic epidural anesthesia and
chodilation during desflurane anesthesia: A comparison local anesthetics on bronchial hyperreactivity. J Clin Monit
to sevoflurane and thiopental. Anesthesiology. 2000;93: Comput. 2000;16:457.
404. 37. Berry A, Brimacombe J, Keller C, et al. Pulmonary airway
16. Habre W, Scalfaro P, Sims C, et al. Respiratory mechanics resistance with the endotracheal tube versus laryngeal
during sevoflurane anesthesia in children with and without mask airway in paralyzed anesthetized adult patients.
asthma. Anesth Analg. 1999;89:1177. Anesthesiology. 1999;90:395.
17. Dikmen Y, Eminoglu E, Salihoglu Z, et al. Pulmonary 38. Kim ES, Bishop MJ. Endotracheal intubation, but not
mechanics during isoflurane, sevoflurane and desflurane laryngeal mask airway insertion, produces reversible bron-
anaesthesia. Anaesthesia. 2003;58:745. choconstriction. Anesthesiology. 1999;90:391.
18. Jooste E, Klafter F, Hirshman CA, et al. A mechanism 39. Ferrari LR, Goudsouzian NG. The use of the laryngeal mask
for rapacuronium-induced bronchospasm: M2 muscarinic airway in children with bronchopulmonary dysplasia. Anesth
receptor antagonism. Anesthesiology. 2003;98:906. Analg. 1995;81:310.
19. Cheney FW, Posner KL, Caplan RA. Adverse respiratory 40. Tait AR, Pandit UA, Voepel-Lewis T, et al. Use of the
events infrequently leading to malpractice suits. A closed laryngeal mask airway in children with upper respiratory
claims analysis. Anesthesiology. 1991;75:932. tract infections: A comparison with endotracheal intubation.
20. Olsson GL. Bronchospasm during anaesthesia. A computer- Anesth Analg. 1998;86:706.
aided incidence study of 136,929 patients. Acta Anaesthesiol 41. Natalini G, Franceschetti ME, Pletti C, et al. Impact of
Scand. 1987;31:244. laryngeal mask airway and tracheal tube on pulmonary func-
21. Mamie C, Habre W, Delhumeau C, et al. Incidence and risk tion during the early postoperative period. Acta Anaesthesiol
factors of perioperative respiratory adverse events in children Scand. 2002;46:525.
undergoing elective surgery. Paediatr Anaesth. 2004;14:218. 42. Gal TJ, Suratt PM. Resistance to breathing in healthy subjects
22. Groeben H, Schlicht M, Stieglitz S, et al. Both local following endotracheal intubation under topical anesthesia.
anesthetics and salbutamol pretreatment affect reflex bron- Anesth Analg. 1980;59:270.
choconstriction in volunteers with asthma undergoing awake 43. Gal TJ. Airway responses in normal subjects following topical
fiberoptic intubation. Anesthesiology. 2002;97:1445. anesthesia with ultrasonic aerosols of 4% lidocaine. Anesth
23. Kil HK, Rooke GA, Ryan-Dykes MA, et al. Effect of Analg. 1980;59:123.
prophylactic bronchodilator treatment on lung resistance 44. Maslow AD, Regan MM, Israel E, et al. Inhaled albuterol,
after tracheal intubation. Anesthesiology. 1994;81:43. but not intravenous lidocaine, protects against intubation-
24. Hirshman CA. Airway reactivity in humans. Anesthetic induced bronchoconstriction in asthma. Anesthesiology.
implications. Anesthesiology. 1983;58:170. 2000;93:1198.
25. Ding DJ, Martin JG, Macklem PT. Effects of lung volume 45. Groeben H, Silvanus MT, Beste M, et al. Both intravenous
on maximal methacholine-induced bronchoconstriction in and inhaled lidocaine attenuate reflex bronchoconstriction
normal humans. J Appl Physiol. 1987;62:1324. but at different plasma concentrations. Am J Respir Crit Care
26. Covar RA, Macomber BA, Szefler SJ. Medications as asthma Med. 1999;159:530.
triggers. Immunol Allergy Clin North Am. 2005;25:169. 46. Scalfaro P, Sly PD, Sims C, et al. Salbutamol prevents the
27. Mertes PM, Laxenaire MC. Adverse reactions to neuromus- increase of respiratory resistance caused by tracheal intu-
cular blocking agents. Curr Allergy Asthma Rep. 2004;4:7. bation during sevoflurane anesthesia in asthmatic children.
28. Rolf N, Cote CJ. Frequency and severity of desaturation Anesth Analg. 2001;93:898.
events during general anesthesia in children with and with- 47. Groeben H, Silvanus MT, Beste M, et al. Combined li-
out upper respiratory infections. J Clin Anesth. 1992;4:200. docaine and salbutamol inhalation for airway anesthesia
29. Tait AR, Malviya S, Voepel-Lewis T, et al. Risk factors for markedly protects against reflex bronchoconstriction. Chest.
perioperative adverse respiratory events in children with 2000;118:509.
upper respiratory tract infections. Anesthesiology. 2001;95: 48. Elwood T, Morris W, Martin LD, et al. Bronchodilator
299. premedication does not decrease respiratory adverse events
30. Bluman LG, Mosca L, Newman N, et al. Preoperative in pediatric general anesthesia. Can J Anaesth. 2003;50:277.
smoking habits and postoperative pulmonary complications. 49. Koga K, Asai T, Vaughan RS, et al. Respiratory complications
Chest. 1998;113:883. associated with tracheal extubation. Timing of tracheal
31. Myles PS, Iacono GA, Hunt JO, et al. Risk of respiratory extubation and use of the laryngeal mask during emergence
complications and wound infection in patients undergoing from anaesthesia. Anaesthesia. 1998;53:540.
ambulatory surgery: Smokers versus nonsmokers. Anesthe- 50. Pappas AL, Sukhani R, Lurie J, et al. Severity of airway
siology. 2002;97:842. hyperreactivity associated with laryngeal mask airway re-
32. Skolnick ET, Vomvolakis MA, Buck KA, et al. Exposure moval: correlation with volatile anesthetic choice and depth
to environmental tobacco smoke and the risk of adverse of anesthesia. J Clin Anesth. 2001;13:498.
respiratory events in children receiving general anesthesia. 51. Silvanus MT, Groeben H, Peters J. Corticosteroids and
Anesthesiology. 1998;88:1144. inhaled salbutamol in patients with reversible airway ob-
33. Smetana GW. Preoperative pulmonary evaluation. N Engl J struction markedly decrease the incidence of bronchospasm
Med. 1999;340:937. after tracheal intubation. Anesthesiology. 2004;100:1052.
C H A P T E R 9/ B R O N C H O S PA S M 155
52. Williams TJ, Tuxen DV, Scheinkestel CD, et al. Risk factors 54. Crogan SJ, Bishop MJ. Delivery efficiency of metered
for morbidity in mechanically ventilated patients with acute dose aerosols given via endotracheal tubes. Anesthesiology.
severe asthma. Am Rev Respir Dis. 1992;146:607. 1989;70:1008.
53. Mutlu GM, Factor P, Schwartz DE, et al. Severe status 55. Taylor RH, Lerman J. High-efficiency delivery of salbutamol
asthmaticus: Management with permissive hypercapnia and with a metered-dose inhaler in narrow tracheal tubes and
inhalation anesthesia. Crit Care Med. 2002;30:477. catheters. Anesthesiology. 1991;74:360.
CHAPTER ASPIRATION PNEUMONITIS
10 Raymond R. Schultetus
A
woman undergoes cesarean section for a sus- enzymes, and acid. On occasion, large particles of food
pected uterine rupture. General anesthesia are present. Generally, it is accepted that the aspirated
is induced and during a difficult, although gastric contents must exceed 25 mL in volume (0.4 mL
successful, intubation, gastric contents are per kg) and have a pH <2.5 to produce clinically
observed in her posterior pharynx. Her en- significant lung pathology.1 However, gastric contents
dotracheal tube is quickly suctioned; however, gastric con- at a neutral pH, especially if they include particulate
tents are not recovered. The patient is ventilated with 50% matter, can also produce significant and long-lasting lung
oxygen, 50% nitrous oxide, and 0.75% isoflurane. Her oxy- pathology.2
gen saturation by pulse oximetry is 90%. The nitrous oxide
is discontinued with little increase in oxygen saturation. An
arterial blood gas is obtained and sent for analysis. After
the administration of 10.0 cm H2 O positive end-expiratory HISTORIC CONSIDERATIONS
pressure (PEEP), her saturation increases to 98%. At the
completion of the surgery, her FIO2 is 0.5, and she is The syndrome, now known as aspiration pneumonitis,
maintained on mechanical ventilation and PEEP. In the was originally called Mendelson syndrome in recogni-
postanesthesia care unit, a chest radiograph is obtained, tion of the clinician, C. L. Mendelson, who described
which does not reveal acute changes. The arterial blood gas pulmonary aspiration of gastric contents in an obstet-
sample taken during surgery, where the FIO2 is 0.5, reveals ric population. In his study, Mendelson reviewed the
a pH of 7.45, a PACO2 of 34 mm Hg, and a PaO2 of 60 mm records of 44,016 deliveries where, over a 13-year pe-
Hg. During the next several hours, her inspired oxygen is riod (1932 to 1945), ether analgesia was administered
reduced to 30%, and her saturation remains above 95%. to laboring women.3 During this period, the adminis-
A chest radiograph now reveals a right lower lobe infiltrate tration of ether (ether, nitrous oxide, oxygen) by mask
(see Fig. 10.1). In stepwise fashion, PEEP is reduced to 5 cm for delivery analgesia and surgical anesthesia was de
H2 O, and a subsequent arterial blood gas reveals a pH of rigueur. From his review, he identified 66 cases of as-
7.43, PACO2 of 33 mm Hg, and PaO2 of 110 mm Hg. She is piration of gastric contents (0.15%). Forty-five cases of
extubated without incident and administered 40% oxygen aspiration were recognized at the time of occurrence;
by mask. Her oxygen saturation remains above 97%. She however, 21 cases were noted retrospectively on the ba-
is discharged after the third postoperative day. sis of maternal postdelivery signs and symptoms. Of the
45 cases identified at delivery, 5 women aspirated solids.
Of these, three women had complete airway obstruc-
tion and two died of suffocation, whereas one coughed
What Baseline Knowledge Is out a piece of meat and survived. The remaining two
had incomplete airway obstruction, which cleared upon
Relevant? coughing. Forty cases of liquid aspiration remained. From
these cases, Mendelson described a syndrome of tachy-
cardia, cyanosis and dyspnea, and a chest radiograph
DEFINITIONS showing scattered, soft, mottled, confluent densities. Over
half of the patients who had aspirated were not treated
Aspiration pneumonitis is the lungs reaction to the with antibiotics (3 received penicillin, 14 were adminis-
pulmonary aspiration of gastric contents. For aspiration tered sulfonamides, and 2 received both penicillin and
156
C H A P T E R 10 / A S P I R AT I O N P N E U M O N I T I S 157
A B
FIGURE 10.2 Photomicrographs of normal (A) canine lung and (B) canine lung 2 days after the
aspiration of particulate antacid. Most alveoli of the antacid lung are filled with macrophages and
polymorphonuclear leukocytes. (Courtesy of Charles P. Gibbs, M.D.)
From: Landay MJ, Christensen EE, Bynum LJ. Pulmonary manifes- From: Landay MJ, Christensen EE, Bynum LJ. Pulmonary manifes-
tations of acute aspiration of gastric contents. Am J Roentgenol tations of acute aspiration of gastric contents. Am J Roentgenol
1978;131:587. 1978;131:587.
C H A P T E R 10 / A S P I R AT I O N P N E U M O N I T I S 159
A B
FIGURE 10.4 Photomicrographs of (A) canine lung two days after the aspiration of food particles at
a pH of 5.9. There is extensive inflammation with monocytes and polymorphonuclear leukocytes.
(B) canine lung two days after the aspiration of bicarbonate solution at a pH 5.9. The lung tissue is
normal appearing. (Courtesy of Charles P. Gibbs, M.D.)
160 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 10.4 American Society of Anesthesiologists intragastric pressure) actually rises slightly.22 For regurgi-
Preanesthesia Fasting Guidelines tation to occur, the intragastric pressure must exceed the
barrier pressure in the lower esophagus. Because both the
Minimum Recommended stomach and the esophagus are contained within adjacent
Hours of Fasting cavities (abdominal and thoracic), and fasciculation oc-
after Consumption Substance Consumed curs both above and below the diaphragm, pressure rises
within both structures, and barrier pressure is maintained,
2 Clear liquid
preventing regurgitation, even when the stomach is not
4 Breast milk
empty.21 In patients with gastroesophageal reflux disease,
6 Formula and milk
6 Nonfat light meal (eg., tea barrier pressure is low or absent, and the ability to main-
and toast) tain barrier pressure with or without succinylcholine-
induced muscular activity is compromised.
Adapted from: Warner MA, Caplan RA, Epstein BS, et al: Practice
guidelines for preoperative fasting and the use of pharmacologic
agents to reduce the risk of pulmonary aspiration: Application to
healthy patients undergoing elective procedures. Anesthesiology What Is Aspiration Pneumonia?
90;896:1990.
and are maintained in the supine position. Commonly, gastric distension is avoided. Good oral hygiene is nec-
the patients are in coma having suffered stroke or head essary, and patients should be maintained in a semierect
trauma or are heavily sedated. Many are tracheally position (30 degrees), with the head of bed elevated
intubated and have poor oral hygiene. whenever possible to reduce passive regurgitation and the
The relation between poor oral hygiene and aspiration risk of ventilator-acquired pneumonia (VAP). Prevention
pneumonia cannot be overemphasized. A community- is a universal quality measure for VAP, and standing or-
acquired aspiration pneumonia in an edentulous patient ders for head elevation of all eligible intubated patients
(no periodontal flora) should lead the physician to suspect are considered a best practice.29,31,32
and rule out oropharyngeal or pulmonary cancer as an
etiology.
The symptoms of aspiration pneumonia are often
slow in onset, beginning with a feeling of malaise and a low What Is the Acute Respiratory
grade fever, or perhaps chills and fever. Rales, cough, and
wheeze develop, along with increased, sometimes blood- Distress Syndrome?
tinged, sputum production. Hypoxemia, tachypnea, and
leukocytosis develop, and night sweats may occur.28,30 Aspiration pneumonitis and aspiration pneumonia may
Diagnosing aspiration pneumonia may be difficult, progress to the acute respiratory distress syndrome
and the causative organism(s) may be hard to determine. (ARDS). For a pathologic lung process to be designated
Oropharyngeal organisms contaminate sputum samples, as ARDS, certain criteria should be met. The process
making the isolation of anaerobes technically difficult. The must be bilateral on chest radiograph and acute in onset.
best samples are acquired by transtracheal aspirations, There must be a known risk factor such as shock, sepsis,
blood cultures, aspirations of pleural fluid, transthoracic or trauma. The patient must have hypoxemia (PaO2 /FIO2
needle aspiration, or bronchoscopic brush sampling.28,30 <201) and must have a left atrial pressure <19 mm Hg to
Chest radiograph will show a consolidation of depen- differentiate congestive heart failure as a primary cause.33
dent lung segments. If the patient aspirated in a sitting or ARDS begins with an injury to the alveolar-capillary
semisitting position, the posterior segments of the lower membrane that increases permeability. The increase
lobes will be involved. If the patient was supine at the time in permeability is sufficient to overwhelm lymphatic
of aspiration, the superior segments of the lower lobes are drainage. The resulting pulmonary edema is distributed
often involved. In alcoholics, the classic presentation is throughout the lung and causes the appearance of bilateral
consolidation of the right upper lobe28,30 (see Fig. 10.5). lung infiltrates on the chest radiograph (see Fig. 10.6).34
Because diagnosis and treatment may be quite diffi- In computed tomographic studies of the normal lung
cult, prevention of aspiration pneumonitis is important. at FRC, 90% of the lung is 50% to 90% aerated. Less than
When intubation is required, the duration of intubation 1% of the lung is aerated at >90% or <10%. The remainder
and ventilation must be as brief as clinically possible. Air- of the lung has aeration at 10% to 50%. On average at FRC,
way contamination should be minimized, and suctioning lung tissue occupies 30% of the intrapleural space, and
of the airway must be conducted in a sterile manner. An- gas occupies 70%.35
tibiotic use should be minimized to reduce the emergence In ARDS, there is a massive loss of aeration of the
of resistant strains. When tube feeding is administered, lung in combination with an increase in nongaseous
FIGURE 10.5 Right upper lobe aspiration pneumonia. FIGURE 10.6 Adult respiratory distress syndrome. (Courtesy of
(Courtesy of Sarah G. Klein, M.D.) Sarah G. Klein, M.D.)
162 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
pulmonary elements. The lung loses gas exchange function Many ventilation strategies have been applied in the
through the accumulation of pulmonary edema fluid, treatment of ARDS. Among those has been an attempt
inflammation, increased extravascular water, and, in some to allow ventilation of unventilated alveoli by changing
cases, products of lung infection. These changes result patient position. A change in position from supine to
in decreased lung compliance and increased work of semirecumbent will increase the number of aerated
breathing.35 superior segments; however, this is usually attained at the
The loss of aeration in the lungs does not simply cost of having fewer inferior segments aerated. Even the
begin at the lung bases and spread upward toward the prone position has been utilized, with some improvement
apices as the process progresses. In the semirecumbent in oxygenation. Mortality, however, is unaffected.36
position, aeration of up to one third of the lower lung PEEP and CPAP have both been used to treat ARDS.
lobes and two thirds of the upper lung lobes may remain CPAP has been found to recruit more alveoli to gas
at or near normal. In the supine position, the weight of exchange and improve lung compliance and oxygenation,
the abdominal contents presses the diaphragm cephalad. and also appears to reduce lung injury during ventilation
This factor, in combination with the weight of the heart by preventing repetitive opening and closing of alveoli.37
pressing downward and the accumulation of pleural fluid, When low PEEP (8 cm H2 O) was compared with high
causes a massive loss of aeration in the caudal and PEEP (14 cm H2 O), high PEEP was found to improve
dependent portions of the lung. A loss of aerated lung in oxygenation (PaO2 /FIO2 ) and lung compliance. However,
the cephalad portions of the lung is variable (see Fig. 10.7). in spite of improvement in these parameters, there was no
Because some degree of aeration of the cephalad and change in the patients length of mechanical ventilatory
caudad lung is maintained, an anteriorposterior chest support or organ function. On the other hand, the higher
radiograph can be misleading as to the degree of lung level of PEEP (14 cm H2 O) was not shown to be harmful.38
injury.35 The use of high frequency oscillatory ventilation
Hypoxic pulmonary vasoconstriction (HPV) is usually (HFOV) in ARDS has been studied. When a patient
dysfunctional during ARDS. This impairment of HPV with ARDS is converted from controlled mode ventilation
explains why oxygenation is so poor, although significant (CMV) to HFOV, there is a brief (<24 hours) improve-
segments of the lung may remain well ventilated. The ment in oxygenation (PaO2 /FIO2 ). The adverse effects of
impairment of HPV is not universal; and in some patients, ventilation were similar between HFOV and CMV. In these
in spite of a similar radiographic presentation, HPV is studies, there was a nonsignificant trend toward improved
maintained, and oxygenation is much better.35 survival with the use of HFOV. However, when patients
in this study were treated with CMV, a tidal volume of up
to 10 mL per kg body weight was used, and the airway
pressure during CMV averaged 38 cm H2 O. High tidal
volumes during CMV may have caused this trend.39
Lower tidal volumes during CMV have been shown
to improve survival. Traditionally, ventilators were set
to deliver a tidal volume of 10 to 12 mL per kg body
weight. Computed tomography studies of lung aeration
showed that the aerated portion of the lung during ARDS
was actually rather small, leading to the concept of the
baby lungthat is, the aerated portion of the adult
lung with ARDS is reduced in size to that comparable
to a babys lung. The use of high volume ventilation (10
to 12 mL per kg) caused overdistension of the aerated
portions and did not significantly recruit nonaerated por-
tions. Furthermore, the larger tidal volumes caused the
release of cytokinin into the alveolus and the intravas-
cular space, causing further lung injury and increasing
vascular permeability.38,40 It has been theorized that alve-
olar overdistension may predispose to infection of the
blood with lung pathogens, although this does not ap-
pear to be a primary source of bacteremia in ventilated
patients. In many cases of ARDS with sepsis, pathogenic
bacteria are not isolated from the airways. The use of
FIGURE 10.7 Computed tomography of ARDS lung. Extensive smaller ventilating volumes (6 mL per kg of predicted
areas of lung consolidation are evident and represent a marked body weight) can result in a 25% reduction in mortality
increase in lung tissue volume at the expense of aerated lung. (when compared to ventilation at 12 mL per kg predicted
The aerated areas represent a small portion of the lung volume body weight).41
at functional residual capacity and are a mixture of hypoaerated, Recruitment maneuvers (sigh) have been used with
hyperaerated, and normally aerated alveoli. (Courtesy of Sarah ARDS to reexpand collapsed alveoli in the belief that
G. Klein, M.D.) brief periods of high tidal volumes would reexpand alveoli
without exposing the lung to the prolonged high airway
C H A P T E R 10 / A S P I R AT I O N P N E U M O N I T I S 163
pressures that result in alveolar disruption. Unfortunately, nitric oxide or the timing of administration of nitric
a 30-second period of 30 to 40 cm H2 O positive airway oxide. As things stand, the routine use of nitric oxide
pressure does not produce long-lasting effects. Instead, in the treatment of ARDS is not supported by the current
only modest and temporary improvement in oxygenation literature.46
result at a cost of decreased blood pressure.34,38
Because a marked influx of water into the lungs occurs
during ARDS, attempts to limit fluid administration have
resulted. Reductions in lung water do improve oxygena- KEY POINTS
tion and lung compliance; however, many patients with
ARDS are hemodynamically unstable and require fluid 1. Aspiration pneumonitis, although rare, is caused
administration. Studies have shown that prompt fluid re- by the aspiration of gastric secretions and/or food
suscitation improves patient survival, whereas delayed or particles. Because of acidity, the aspirate is usually
restricted fluid resuscitation may be harmful.34 sterile. The more acidic the aspirate, the more
A marked inflammatory response within the alveoli damaging it is to the lung tissue.
is observed in ARDS. For that reason, glucocorticoids 2. The classic signs and symptoms of aspiration are
have been administered to patients early in the course the onset of wheezing, shortness of breath, cyanosis,
of the process. However, results are disappointing. More tachycardia, hypoxemia, and pathologic findings on
recently, there has been an interest in the use of steroids the chest radiograph.
in patients who fail to respond to initial treatments. In 3. The chest radiograph presentation of aspiration
a study of patients unresponsive after 7 days to other pneumonitis is highly variable.
therapy, methylprednisolone was administered. In spite 4. After aspiration, lung injury occurs rapidly. Brief
of an improvement in oxygenation and blood pressure tracheobronchial suctioning is indicated, but usually
and a shortened course of ventilation (as compared to lavage and bronchoscopy are not.
controls), the 28-day and 60-day survival was unchanged 5. If aspiration does not produce signs or symptoms in
when compared with controls receiving similar treatment an otherwise healthy patient undergoing an elective
but without steroids.34 procedure within 2 hours following the procedure,
Because fluid management is important, the pul- further treatment is not required.
monary artery catheter has been used extensively in the 6. Patients who become symptomatic following as-
care of patients with ARDS. Complications resulting from piration most likely will require admission to an
its use have been extensively studied and described. Prob- intensive care unit and ventilatory support.
lems arising directly from catheter use such as infection, 7. Aspiration is best prevented by following preopera-
pulmonary artery rupture, dysrhythmia, and pulmonary tive fasting guidelines and exercising careful airway
embolism readily come to mind. Also, the misinterpre- management. Antacids, histamine blockers, proton
tation and misapplication of data derived from the use pump inhibitors, or gastrointestinal stimulants are
of the catheter can be equally harmful.42 Results from not indicated in healthy patients undergoing an elec-
a retrospective study indicated that the early use of the tive procedure.
pulmonary artery catheter in ARDS actually increased 8. Aspiration pneumonia results from the pulmonary
mortality, length of stay, and costs.43 In more recent, ran- aspiration of infected secretions, usually from the
domized studies, the pulmonary artery catheter was not oropharynx or from nosocomial infection caused by
associated with increased morbidity or mortality, nor did health care workers.
therapy based on use of data derived from this catheter 9. Diminished consciousness and airway reflexes,
result in an improved outcome.44 along with impaired swallowing, are common risk
Nitric oxide is a pulmonary vasodilator. In acute factors for community-acquired pneumonias.
lung injury, its use results in short term increases 10. In community-acquired infections, bacterial flora
in oxygenation and decreases in pulmonary vascular that inhabit the periodontal pocket predominate. In
resistance. A meta-analysis of five randomized controlled hospital-acquired infection, the pathologic bacteria
studies of the use of nitric oxide during acute hypoxic are those commonly contaminating the intensive
respiratory failure showed that only a modest, short care unit.
term (up to 72 hours) improvement in oxygenation, 11. Aspiration pneumonia is slow in onset and usually
and no change in mortality or in the duration of accompanied by malaise and low grade fever.
mechanical ventilation, resulted from its use. In a Rales, cough (possibly blood-tinged), hypoxemia,
subsequent randomized placebo-controlled multicenter and leukocytosis develop.
study, it was again shown that although oxygenation 12. Isolation of the causative organism may be diffi-
improved during the initial administration of nitric cult, as oropharyngeal organisms often contaminate
oxide, there were no changes in any clinically significant the samples, and the infecting organism may be
outcomes.45 In an editorial accompanying this report, anaerobic and difficult to isolate.
the editors observed that most patients who died with 13. Prevention of aspiration is the key to reducing
ARDS died because of multisystem organ failure, and morbidity and mortality.
not hypoxemia. Additionally the study did not require 14. ARDS presents as bilateral chest involvement of
use of low tidal volume ventilation, nor did it consider acute onset following shock, sepsis, or trauma. It is
the possible harm from the long-term administration of caused by injury to the alveolar-capillary membrane.
164 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
and Blood Institute and Food and Drug Administration shock and acute respiratory distress syndrome: A random-
workshop report. JAMA. 2000;283:2568. ized controlled trial. JAMA. 2003;290:2713.
43. Conners AF Jr, Speroff T, Dawson NV, et al. The effectiveness 45. Taylor RW, Zimmerman JL, Dellinger RP, et al. Low-dose
of right heart catheterization in the initial care of critically inhaled nitric oxide in patients with acute lung injury: A
ill patients. JAMA. 1996;276:889. randomized controlled study. J Am Med Soc. 2004;291:1603.
44. Richard C, Warszawski J, Anguel N, et al. Early use of the 46. Adhikari N, Granton JT. Inhaled nitric oxide for acute lung
pulmonary artery catheter and outcomes in patients with injury: No place for NO. JAMA. 2004;291:1629.
CHAPTER PNEUMOTHORAX
11
AND BAROTRAUMA
Bhiken Naik
CASE SUMMARY
What Baseline Knowledge of
28-year-old, 75 kg man is involved in a motor
Pulmonary Physiology Is
A
vehicle accident. His injuries include blunt
head trauma with an admission Glasgow Important?
Coma Scale of 12 and multiple rib fractures
on the left, with radiological evidence of
an underlying pulmonary contusion. On ad-
mission to the emergency room, the arterial blood gas
INTRAPLEURAL PHYSIOLOGY
results are: pH 7.34, PO2 52 mm Hg, PCO2 26 mm Hg, Normal respiration is a cyclical event characterized by an
HCO3 2 to 14 mEq per L, and lactic acid 5 mmol per active inspiration phase followed by expiration, which is
L on a nonrebreathing face mask. Clinical examination passive in nature. During quiet breathing, inspiration is
reveals a flail segment on the left hemithorax with exten- achieved by contraction of the diaphragm, which displaces
sive subcutaneous emphysema, and a chest radiograph the abdominal contents downward and increases the
demonstrates a left hemopneumothorax. A chest tube vertical intrathoracic dimension. During expiration, the
is placed, with drainage of approximately 400 mL of lung and chest wall passively return to their resting
blood; however, the patients respiratory status worsens. position.
He is intubated and transferred to the surgical inten- The elastic forces of the lung and chest wall main-
sive care unit for mechanical ventilation. The patient tain the intrapleural space at subatmospheric pressure
was placed on pressure support and synchronized inter- (see Fig. 11.1). During inspiration, the intrapleural pres-
mittent mandatory ventilation. Initial ventilator settings sure decreases further, and alveolar pressure becomes
were as follows: FiO2 of 60%, pressure support of 10 cm subatmospheric, creating a pressure gradient for airflow
H2 O, positive end-expiratory pressure (PEEP) of 10 cm into the alveoli. In the expiratory phase, the intrapleural
H2 O, synchronized intermittent mandatory ventilation pressure rises and alveolar pressure increases, thereby
rate of 12 breaths per minute and a tidal volume of facilitating airflow out of the lung.
600 mL. During the next 48 hours, the patient developed
acute respiratory distress syndrome (ARDS) and required
escalation of the ventilator parameters. Ventilator set-
tings were readjusted: FiO2 80%, pressure support (PS)
PATHOPHYSIOLOGY OF
20 cm H2 O, PEEP 15 cm H2 O, synchronized intermit- PNEUMOTHORAX AND
tent mandatory ventilation rate of 14 breaths per minute, TENSION PNEUMOTHORAX
and tidal volume of 750 mL. The patients respiratory
status improved during the next 12 hours, and the FiO2 When there is loss of integrity of the visceral or parietal
was weaned to 40%. However, following central venous pleura, air enters the subatmospheric intrapleural space.
line placement through the right subclavian route, the The forces opposing the retraction of the lung are
peak and plateau inspiratory pressures increased to 40 cm reduced, causing the lung to collapse (see Fig. 11.2).
H2 O, accompanied by hemodynamic compromise. A pre- Respiratory compromise occurs from altered pulmonary
sumptive diagnosis of a right-sided tension pneumothorax mechanics, the pendeluft effect, and ventilation-perfusion
was made, and needle decompression was performed, mismatching.
with rapid resolution of the ventilatory and hemodynamic If a flap-valve injury to the pleura develops, air can be
instability. entrained into the intrapleural space during inspiration,
166
C H A P T E R 11/ P N E U M O T H O R A X A N D B A R O T R A U M A 167
FIGURE 11.1 Intrapleural and alveolar pressure during (A) end-expiration, (B) mid-inspiration and
(C) end-inspiration.
with no escape route during expiration. The intrapleural A primary spontaneous pneumothorax affects people
pressure becomes supra-atmospheric, with accompany- who have no clinically apparent lung disorders, whereas
ing severe hemodynamic and respiratory compromise, a secondary pneumothorax occurs in the setting of
resulting in a tension pneumothorax (see Fig.11.3). underlying lung disease.
SPONTANEOUS
PNEUMOTHORAX
Spontaneous pneumothoraces occur in the absence of tho-
racic trauma and are classified as primary or secondary.1
P=0
5 cm H2O 0 cm H2O
FIGURE 11.2 Intrapleural pressures with a left-sided FIGURE 11.3 Right-sided tension pneumothorax with
pneumothorax. mediastinal shift to the opposite side.
168 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 11.1 Classification of Pneumothorax widespread energy transfer to the body. In the thoracic
area, this action can result in alveolar rupture, esophageal
Spontaneous Pneumothorax laceration, bronchial tears, and multiple rib fractures that
can violate the pleura. In addition, there may be soft
Primary
tissue, bone, and possible vascular injuries.
Secondary
Penetrating trauma is classified as high or low ve-
Traumatic Pneumothorax locity. With low velocity-penetrating trauma, the injury is
Blunt injury confined to the anatomic track. With high velocity trauma,
Penetrating injury there may be additional injury distant from the anatomic
Iatrogenic Pneumothorax track, secondary to the accompanying shock wave.
Traumatic pneumothorax can be further subclassified
Neuraxial blockade into closed, open, tension, or hemopneumothorax.
Peripheral nerve blockade
Central venous cannulation
Faulty anesthesia equipment Closed Pneumothorax
Laparoscopic surgery
With closed pneumothorax, there is no violation of the
Miscellaneous chest wall, and the degree of respiratory dysfunction is
Barotrauma related to alterations in atmospheric pressure related to the patients underlying medical condition,
associated injuries, and the size of the accompanying
pneumothorax.
subpleural bullae are seen in approximately 76% to 100%
of the ipsilateral lung and 79% to 96% of the contralateral
lung.5 Open Pneumothorax
If the primary spontaneous pneumothorax is small
and the patient is clinically stable, the American College Open pneumothoraces communicate with the atmosphere
of Chest Physicians recommends observation in the through the chest wall. They can result in severe respi-
emergency department for 3 to 6 hours and discharge ratory embarrassment because of the large amount of
if the repeat chest radiograph does not demonstrate air entrained into the pleural cavity during inspiration.
progression of the pneumothorax.1 If the pneumothorax They are initially managed by placing moist sterile gauze
is large, the lung should be reexpanded with a small bore loosely over the wound, which prevents air from being
catheter or a thoracostomy tube. In the absence of an air entrained into the pleural space during inhalation.
leak, the drainage device can be removed 12 to 24 hours
later, providing the chest radiograph demonstrates no Tension Pneumothorax
evidence of pneumothorax. Persistent air leaks for >4 days
should be evaluated for surgery and possible pleurodesis. Tension pneumothorax is a life-threatening emergency
and requires prompt identification and decompression of
Secondary Spontaneous Pneumothorax the pleural cavity with a needle thoracostomy, followed
by placement of a tube thoracostomy (Fig. 11.3).
Secondary spontaneous pneumothorax can occur be-
cause of a variety of causes. These include chronic Hemopneumothorax
obstructive airway disease, asthma, pulmonary fibro-
sis, pulmonary infarction, and infective disorders of the Hemopneumothorax occurs when there is both air and
lung, with chronic obstructive pulmonary disease being blood within the pleural space. These injuries have to be
the most common. Patients complain of a dramatic in- managed with a large bore tube thoracostomy (e.g., 28 to
crease in dyspnea associated with pleuritic chest pain.6 32 Fr) to allow adequate drainage and decompression
Secondary spontaneous pneumothorax represents a sig- of the pleural cavity. If bleeding from the chest tube is
nificant marker for mortality in patients with chronic persistent, and the rate exceeds 100 to 200 mL per hour
obstructive pulmonary disease, with Videm et al. demon- or the total output is >1,000 mL, a thoracotomy should
strating a fourfold increase in mortality.7 In the absence of be performed.9
pleurodesis therapy, the recurrence rate for pneumotho-
rax is 40% to 50%.8 The current recommended treatment
for secondary spontaneous pneumothorax is chest tho-
racostomy followed by surgical pleurodesis, given the OCCULT PNEUMOTHORAX
potential lethality of recurrent pneumothoraces.1
Occult pneumothorax is a pneumothorax identified by
computed tomography (CT) scan but is not visible by
routine chest radiographs. With the increasing use of CT
TRAUMATIC scanning to evaluate polytrauma patients, this entity is
PNEUMOTHORAX being diagnosed with increasing frequency. Hill et al. in a
retrospective review of 3,121 patients admitted to a level 1
The mechanism of traumatic pneumothorax can be due trauma center, found an incidence of 2.2% for occult pneu-
to blunt or penetrating trauma. Blunt trauma results in mothorax.10 The management of occult pneumothorax
C H A P T E R 11/ P N E U M O T H O R A X A N D B A R O T R A U M A 169
remains controversial. There are no large multicenter TABLE 11.2 Incidence of Pneumothorax with Various
prospective trials to draw clinically relevant conclusions. Nerve Blocks
In a small retrospective review by Collins et al. it appears
that patients who are hemodynamically stable can be Nerve Block Incidence (%)
managed conservatively with interval chest radiographs.11 Intercostal nerve block 0.07319
However, if the patient requires positive pressure ventila- Supraclavicular brachial plexus block 0.56
tion, it is safer to place a tube thoracostomy to avoid the Infraclavicular brachial plexus block 0.20.7
risk of developing a tension pneumothorax.12 Thoracic paravertebral block 0.5
with various patient positions can have a major im- pain, but in the presence of other nonthoracic injuries,
pact on cardiorespiratory function. Pneumothorax during these symptoms may be masked. In penetrating trau-
abdominal laparoscopy can cause paradoxical balloon- matic pneumothorax, there may be either a blowing or
ing of the hemidiaphragm, a decrease in lung compli- sucking chest wound. With multiple fractured ribs, severe
ance, hypo-oxygenation, hypercarbia, and pneumothorax. respiratory compromise may occur from the underlying
A carbon dioxide pneumothorax has a >90% resolution pulmonary contusion, as well as the flail chest segment.
within 2 hours, whereas a helium pneumothorax remains The flail segment can be seen moving inward during in-
essentially unchanged during the same time period;29 spiration and outward during expiration. Patients with
therefore, expectant management (e.g., observation) with tension pneumothorax may have an altered level of con-
asymptomatic carbon dioxide pneumothoraces can be sciousness secondary to profound hypotension. Distended
advocated. On the contrary, helium pneumothoraces do neck veins, tracheal deviation and hyperresonance of the
not resolve rapidly and may require aspiration, even in affected side may also be present.
asymptomatic patients. Under general anesthesia, a pneumothorax can be
potentially difficult to diagnose and can cause significant
hemodynamic changes during positive pressure ventila-
tion. The clinician should maintain a high degree of
PNEUMOTHORAX suspicion for this complication during cannulation of
SECONDARY TO ALTERATION the central veins, peripheral nerve blockade involving the
IN ATMOSPHERIC PRESSURE upper extremity, and placement of a thoracic neuraxial
block. If airway pressures are being monitored graphi-
The use of hyperbaric oxygen (HBO) therapy has evolved cally, a rise in both peak and plateau airway pressure may
during the last 50 years, paralleling an increased un- precede any respiratory or hemodynamic changes (see
derstanding of gas exchange physiology. Currently, HBO Fig. 11.4).
therapy can be lifesaving in several specific syndromes:
Carbon monoxide and cyanide toxicity, air embolism, Electrocardiography
decompression sickness, and clostridial myonecrosis.
During application of HBO therapy, gas-containing A number of nonspecific electrocardiogram ECG changes
cavities within the body either contract or expand during have been described with pneumothorax and tension
compression or decompression, respectively. Clinically pneumothorax. Brock-Utne et al. reported a case of periop-
important gas-containing cavities include the paranasal erative pneumothorax that was preceded by a decrease in
sinuses and the lung. the amplitude of the ECG complex before clinical symp-
During decompression, areas of poor ventilation toms developed.30 PR-segment elevation in the inferior
within the lung fail to reach equilibrium with the ambient leads and reciprocal PR-segment depression in the aVR
pressure and are at risk for barotrauma. Before HBO ther- lead can be seen with left-sided tension pneumothorax
apy, a chest radiograph should be performed and patients secondary to atrial ischemia.31 Other electrocardiographic
with bullous lung disease should be excluded from further changes seen with tension pneumothorax include right-
treatment. If there is a history of chronic obstructive air- ward shift of the mean frontal QRS axis, precordial T-wave
way disease, bronchodilator therapy must be optimized inversion, and electrical alternans (alternating QRS am-
to prevent air trapping and subsequent barotrauma. plitude with each heart beat).32
How Is a Pneumothorax 30
Diagnosed and Managed?
Airway pressure (cm H2O)
Peak pressure
Plateau pressure
20
MAKING THE DIAGNOSIS
Clinical Features
10
The clinical presentation of pneumothorax is determined
by the etiology of the pneumothorax, patient comorbidi-
ties, mechanism of injury, and the degree of hemodynamic
compromise. A history suggestive of pneumothorax in-
cludes sudden onset of chest pain, associated dyspnea, 1 2 3 4 5
cough, and, rarely, hemoptysis. On physical examina- Time (s)
tion, there may be decreased excursion of the affected
side, increased resonance on percussion, and decreased FIGURE 11.4 Elevation in both peak and plateau pressure with
breath sounds on auscultation. Patients with pneumotho- tension pneumothorax.
rax secondary to trauma may complain of pleuritic chest
C H A P T E R 11/ P N E U M O T H O R A X A N D B A R O T R A U M A 171
Radiographic Signs
The radiographic diagnosis of a pneumothorax depends
on the recognition of a visceral pleural line separated
from the parietal pleura by a radiolucent airspace. This
radiolucent area is determined by the volume of air in the
pleural space and the position of the body. Air tends to ac-
cumulate in the nondependent areas of the thoracic cavity
and, therefore, the chest radiograph may be different in
the upright position as compared to the supine position.
Chest Radiograph
In the upright position, air tends to accumulate in
the superior and lateral aspects of the thoracic cavity.
A pneumothorax is visible as a thin, curvilinear opacity
along the lung and is separated from the chest wall by air
in the apical pleural space (see Fig.11.5).
Radiographic techniques for detecting a pneumoth- FIGURE 11.6 Deep sulcus sign with abnormal deepening and
orax in the upright position include: (i) Upright end- lucency of the left costophrenic angle.
expiratory radiograph; (ii) upright inspiratory radiograph;
or (iii) radiograph in the lateral decubitus position with a
horizontal radiographic beam. Seow et al. reported equal
sensitivity for both upright end-expiratory and inspiratory 41% failed to recognize a tension pneumothorax. Over-
chest radiograph in detecting a pneumothorax.33 all scores between attending physicians and anesthesia
In the supine position, air tracks to the anterior residents were not significantly different.35
costophrenic sulcus. The radiographic signs of a pneu- In light of the difficulty in diagnosing a pneumothorax
mothorax in the supine position tend to be more subtle, in the supine patient, other modalities that can increase
with approximately 30% of pneumothoraces going un- diagnostic sensitivity include thoracic ultrasound and CT
detected.34 Radiographic features of a pneumothorax in scan.
the supine position include the deep sulcus sign (see
Fig. 11.6), hypolucency of the hemithorax, depression of Sonography
the ipsilateral diaphragm, and increased sharpness of the
cardiomediastinal border and pericardial fat pads. Focused assessment with sonography in trauma has
Anesthesiologists are deficient in reading chest ra- gained acceptance as a rapid and reliable tool that can
diographs. Kaufman et al. conducted a randomized eval- be used to screen free intraperitoneal fluid associated
uation among anesthesiologists at a New York medical with visceral injury. This modality can also be used to
center. They were asked to assess 10 radiographs, with detect pleural injury. The absence of lung sliding and
no time limit set for these interpretations. Eleven percent comet-tail artifacts suggests the presence of a pneumoth-
of anesthesiologists misdiagnosed a pneumothorax and orax. Rowan et al. in a prospective study demonstrated
that thoracic ultrasound was more sensitive than chest
radiography and as sensitive as CT scanning in detecting
a traumatic pneumothorax. In this study, ultrasound had
an estimated sensitivity and negative predictive value of
100% and a specificity of 94%.36 Thoracic ultrasound is
limited by any condition that prevents the pleural sur-
faces from sliding against each other, such as pleural
adhesions. Furthermore, subcutaneous emphysema can
interfere with the acoustic window and decrease the sen-
sitivity of this device.
Computed Tomography
CT of the chest can assist in diagnosing unsuspected pneu-
mothoraces during the perioperative period and in the crit-
ically ill patient. Management of the chest tube can also be
optimized; Tagliabue et al. reported a 65% incidence of in-
effective tube positioning detected by CT of the chest in pa-
tients with ARDS.37 In the same study, they demonstrated
FIGURE 11.5 Large right-sided pneumothorax. that either pneumothorax or pneumodiastinum seen with
CT was not visible on plain chest radiographs 40% and 80%
172 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Conservative Management
Conservative management of a small pneumothorax
involves placing the patient on 100% oxygen, with strict
clinical and radiological followup to ensure that the
pneumothorax does not enlarge. By administering 100%
oxygen, the nitrogen gradient of the capillary-pleural
space is increased, allowing the air collection to decrease
in size. However, during anesthesia, it is not uncommon
to initiate positive pressure ventilation that can increase
the size of the pneumothorax and potentially precipitate
a tension pneumothorax. Therefore, a pneumothorax
should be drained before induction of anesthesia. If
an expectant management strategy is adopted and a
complication does occur, the clinician should be prepared
to rapidly decompress the pleural space and place a chest
tube.
FIGURE 11.7 Computer tomography of a left-sided
pneumothorax.
Simple Aspiration
Chest tube placement is associated with significant pain
of the time, respectively. Kemper et al. also showed that CT
and discomfort and increases the duration of hospital
scans are more sensitive at detecting pulmonary interstitial
stay. An alternate approach for treating uncomplicated
emphysema, the earliest sign of barotrauma (see Figs. 11.7
spontaneous and iatrogenic pneumothoraces is by sim-
and 11.8).38 However, the enthusiasm for CT for detection
ple aspiration technique. The pneumothorax is aspirated
of pneumothoraces in critically ill patients must be tem-
using a three-way stopcock, with the exit limb placed
pered by the risk associated with transporting critically ill
underwater to ensure the correct direction of airflow.
patients to the radiology suite.
Faruqi et al. reported an 83% and 91% incidence of resolu-
tion using a simple aspiration technique for spontaneous
and iatrogenic pneumothoraces, respectively; whereas in
MANAGEMENT OPTIONS Markos study, the incidence of successful decompres-
sion was 71% and 66%, respectively.39,40 With secondary
The goals of management for a pneumothorax are to: spontaneous and tension pneumothorax, this technique
(i) Prevent further entrainment of air; (ii) drain the should be avoided, and a formal chest thoracostomy must
pneumothorax, allowing for complete reexpansion of the be performed.
lung; and, (iii) prevent any recurrence. The management
strategy is based on the cause of the air leak, the size
of the pneumothorax, the possibility that the patient will Chest Tube Thoracostomy
require positive pressure ventilation, and the associated
comorbidity. Chest tube thoracostomy is the preferred technique for
draining a large pneumothorax, a secondary spontaneous
pneumothorax, or those complicated by hemothorax,
empyema, and persistent air leaks. Chest tubes can
be placed by a trocar technique or by blunt digital
dissection. Extreme caution should be exercised when
using a trocar because uncontrolled entry into the thoracic
cavity can injure the underlying lung. The blunt digital
dissection technique involves making a 3- to 4-cm skin
incision parallel to the chosen interspace. The underlying
fascia and the intercostal muscles are dissected using a
hemostat. Once the parietal pleura is opened, a finger
sweep should be performed, and any adhesions should be
broken down. A hemostat is then used to guide the chest
tube above the superior border of the inferior rib into the
pleural space. The chest tube is placed in the area of air
entrainment. The chest tube is then connected to a pleural
drainage device. Pleural drainage devices should prevent
FIGURE 11.8 Reconstruction computed tomography of a the entrainment of air during the respiratory cycle when
pneumothorax. Air accumulates in the anterior aspect of the the pleural pressure is subatmospheric while allowing
chest as a result of the supine position (see arrows). continuous drainage of air and fluid; this can be achieved
with two devices.
C H A P T E R 11/ P N E U M O T H O R A X A N D B A R O T R A U M A 173
Heimlich Valve level in the water seal chamber determines the negative
The Heimlich valve is a unidirectional flutter valve that is pressure applied to the pleural space. The collection
connected to the chest tube. Because it is unidirectional, bottle allows continuous drainage of the pleural space
it prevents air from being entrained when the pleural without affecting the water seal level, unlike a one-bottle
pressure is subatmospheric and is patent when the pleural drainage device. Additional advantages of the
intrapleural pressure is supra-atmospheric. The limitation single disposable pleural drainage units are that they are
of the Heimlich valve is that it can only be used for simple to use; the pleural drainage can be easily quantified;
drainage of air from the pleural space. and a measurable negative pressure is applied to the
pleural cavity. Baumann et al. tested the accuracy of six
commercially available pleural drainage units regarding
Pleural Drainage Devices negative pressure generation. Their study demonstrated a
In the presence of fluid or blood, a water seal, a drainage maximum mean error of 15.5% at a pressure of 20 cm
chamber, and the ability to apply negative pressure is H2 O; this decrease in delivered negative pressure is of
needed to reexpand the pleural space. Instead of using a little clinical significance.41
cumbersome three-bottle system to achieve these goals,
modern pleural drainage devices incorporate all three Percutaneous Pneumothorax Catheters
components into a single, disposable, molded plastic
unit (see Fig.11.9). The right chamber is the collection Percutaneous pneumothorax catheters are small bore
bottle, the middle chamber is the water seal, and the (<14 Fr) catheters that can be used to treat simple pneu-
left chamber is equivalent to the suction control bottle. mothoraces and uncomplicated pleural effusions. They
Suction control is achieved by a mechanical device are placed through the Seldinger technique and are asso-
without the need for water (dry suction control) or by ciated with a low complication rate. Clinically successful
using a water column. If a water column is used, the reexpansion rates for effusions and pneumothoraces were
water level in the suction control bottle minus the water 86% and 81%, respectively, in a retrospective study con-
ducted by Gammie et al.42 However, because of the small
internal diameter of these catheters, flow rate limitations
may prevent adequate drainage of persistent large air
leaks and can increase the risk of developing a tension
pneumothorax. Baumann et al. demonstrated flow rates
between 2.6 to 5.5 L per minute for 8 Fr catheters and
12.8 to 16.8 L per minute for 14 Fr drainage catheters at
20 cm H2 O.41 Patients with bronchopleural fistulae in
the setting of chest trauma have air leaks ranging from
<1 L per minute to 16 L per minute;43,44 therefore, small
bore (<14 Fr) percutaneous catheters may be inappropri-
ate in the setting of such large persistent air leaks.
TECHNICAL COMPLICATIONS
Technical complications include bleeding and malposi-
20 cm H2O 20 cm H2O 18 cm H2O tion of drainage devices. Bleeding may occur as a result
of damage to the underlying pulmonary tissue, especially
if a trocar technique is being employed, or laceration
of the intercostal vessels. The intercostal vessels are lo-
Suction Control Water-seal Collection cated on the inferior border of the overlying rib, and
Bottle Bottle Bottle in the elderly patient may become tortuous and be
more prone to injury.45 Extrathoracic placement of chest
FIGURE 11.9 Modern pleural drainage device incorporating all tubes, particularly within solid abdominal organs, can be
three components. associated with significant morbidity and mortality. The
risk of malposition is increased in obese patients and those
174 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
REEXPANSION PULMONARY
EDEMA
Reexpansion pulmonary edema is characterized by acute
pulmonary edema following reexpansion of a chronically
collapsed lung. The incidence varies between 0.9% and
20%, and it tends to occur more frequently in the 20 to
39 year age-group.46 Risk factors for developing reexpan-
sion pulmonary edema include untreated pneumothorax
for >72 hours, and complete collapse followed by rapid
reexpansion of the lung. The etiology of reexpansion pul-
monary edema is not fully understood but it appears to
be multifactorial in nature. Decreased levels of surfactant,
increased pulmonary capillary permeability, and sudden
negative intrapleural pressure may unfavorably alter the
fluid balance within the pulmonary vasculature, thereby FIGURE 11.10 Bilateral pulmonary infiltrates seen with the
precipitating pulmonary edema.47,48 An additional factor, acute respiratory distress syndrome.
as shown by Tan et al. is an increase in cardiac output
in patients with reexpansion pulmonary edema,49 which
may lead to higher microvascular pulmonary pressures
and altered pulmonary compliance (see Fig. 11.10) The
and further exacerbation of the capillary leak already
criteria used to define ARDS, as put forward at the first
present. Reexpansion pulmonary edema can be prevented
by slow expansion of the chronically collapsed lung by in- AmericanEuropean Consensus Conference in 1994, is
termittently clamping the chest tube. If pulmonary edema listed in Table 11.3.51
does develop, it is best managed by PEEP therapy de-
livered either by mechanical ventilation or noninvasive
continuous positive pressure ventilation.50 Care should
be exercised when using diuretic therapy, particularly if VENTILATOR-ASSOCIATED
the patients are already hypovolemic. LUNG INJURY
Mechanical ventilation remains the mainstay therapy for
the treatment of ARDS. However, application of positive
What Is the Relation of pressure to the lung can induce a variety of changes
to the lung parenchyma, termed ventilator-associated
Barotrauma to the Acute lung injury (VALI). The four components of VALI are
Respiratory Distress volutrauma, barotrauma, atelectotrauma, and biotrauma
(see Table 11.4). Although volutrauma and barotrauma
Syndrome? cause lung injury at end-inspiration, atelectotrauma is
attributed to the shear stress imposed on the alveoli
during repeated cycles of recruitment and derecruitment.
ACUTE RESPIRATORY Atelectotrauma can be alleviated by the application of
DISTRESS SYNDROME PEEP.
The spectrum of VALI extends from ultrastructural
The ARDS is an acute lung injury characterized by changes of the alveolar epithelium and lung cytoskeleton
increased alveolar capillary permeability, alveolar edema, to macroscopic air leaks.
TABLE 11.3 Criteria for Acute Lung Injury and Acute Respiratory Distress Syndrome
Timing Oxygenation Chest Radiograph Pulmonary Artery Wedge Pressure
Acute Lung Injury Acute PaO2 /FiO2 <300 mm Hg Bilateral infiltrates <18 mm Hg or no clinical evidence
of left atrial hypertension
ARDS Acute PaO2 /FiO2 <200 mm Hg Bilateral infiltrates <18 mm Hg or no clinical evidence
of left atrial hypertension
AIR LEAK
Etiology
With alveolar overdistention, a pressure gradient is estab-
lished which can precipitate alveolar rupture. Air tracks
along the loose connective tissue of the bronchovascu-
lar bundle and can migrate either peripherally or toward
the pulmonary hilum. If air tracks toward the hilum, it
can enter the mediastinum, resulting in pneumomedi-
astinum. This mediastinal air can decompress into the FIGURE 11.11 Pneumomediastinum with continuous
pleural space, soft tissue, the retroperitoneum, or, rarely, diaphragm sign (see arrow).
into the pericardial sac. If air tracks peripherally along the
bronchovascular bundle, it accumulates in the subpleural
connective tissue and forms subpleural cysts. Rupture of and cannot participate in gas exchange. When large tidal
these subpleural cysts can cause a pneumothorax. volumes or high plateau pressures are used to ventilate
The earliest radiographic sign of pulmonary baro- these patients, the functional lung units are preferentially
trauma is pulmonary interstitial emphysema, which oc- ventilated, which results in overdistention and stretch in-
curs when air escapes the alveoli and accumulates in jury to these units. In an attempt to reduce the amount of
the interstitial space. Pulmonary interstitial emphysema stretch injury, the current recommendations are to venti-
is recognized radiologically as rounded or linear lucen- late patients with a tidal volume between 6 mL and 8 mL
cies radiating from the hilum and tracking along the per kg and restrict the plateau pressure to 35 cm H2 O.56
bronchovascular pathway. The ARDS network recently reported an 8.8% reduction in
A pneumomediastinum can be recognized by the mortality and a greater number of ventilator-free days in
continuous diaphragm sign, which occurs when air over patients ventilated with lower tidal volumes (6 mL per kg)
the central portion of the diaphragm allows the top of compared with traditional tidal volumes (12 mL per kg).57
the diaphragm to be seen as a continuous structure (see Despite the statistically significant reduction in mortal-
Fig. 11.11). Additionally, bands of lucency outlining the ity between these groups, the incidence of barotrauma
heart and other mediastinal structures may be seen on was similar. This reiterates the earlier point that baro-
plain chest radiographs. trauma serves only as an indicator of the severity of lung
pathology.
Mortality
The incidence of pneumothorax in patients with ARDS
varies between 7% and 60%.52 In an animal model of What Is the Clinical
ARDS, the presence of an air leak was associated with
an increased mortality; however, in human studies, there Significance of Biotrauma?
does not appear to be a strong correlation between mortal-
ity and the occurrence of an air leak or pneumothorax.53,54 Biotrauma, the release of inflammatory mediators during
Therefore, it appears that air leaks or pneumothorax may mechanical ventilation, is a possible mechanism for the
be a marker for severity of lung injury rather than a cause higher mortality seen with large tidal volume ventilation.
of mortality.55 In an ARDS network study, the interleukin-6 levels
Owing to the heterogeneous distribution of pul- decreased to a greater extent and were lower at day 3
monary edema in ARDS, there are lung units that are in the low-tidal volume compared with the traditional
functional, recruitable, and those that are collapsed tidal volume group.57 In numerous animal models,
176 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
injurious ventilator strategies are associated with a 12. The major risks of treating a pneumothorax are
marked increase in tumor necrosis factor-alpha (TNF-) technical complications related to bleeding and
and other inflammatory cytokines within bronchoalveolar malposition of the drain and reexpansion pulmonary
lavage fluid.58,59 These inflammatory mediators can be edema. The risk of reexpansion pulmonary edema
released into the systemic circulation and potentiate the can be reduced by slow expansion of the lung in
systemic inflammatory response syndrome;60 hence, it is high-risk patients.
possible for injurious ventilatory strategies to cause both 13. The four major aspects of VALI are barotrauma,
pulmonary and extrapulmonary complications. volutrauma, atelectotrauma, and biotrauma.
In summary, there does not appear to be a strong 14. VALI can be reduced by low-tidal volume ventilation
correlation between an air leak and mortality in patients (6 to 8 mL per kg), application of PEEP, and
with ARDS. The presence of an air leak is an indicator of restricting the plateau pressure to 35 cm H2 O.
the severity of the underlying lung pathology. By limiting 15. Biotrauma is the release of inflammatory mediators
tidal volume and plateau pressure and using appropriate during injurious lung ventilation strategies and
levels of PEEP, VALI can be reduced. may cause both pulmonary and extrapulmonary
complications.
REFERENCES
KEY POINTS
1. Baumann MH, Strange C, Heffner JE, et al. Management of
1. The intrapleural pressure is subatmospheric during spontaneous pneumothorax: an American College of Chest
normal respiration; therefore air enters the pleural Physicians Delphi consensus statement. Chest. 2001;119:590.
space with disruption of either the visceral or 2. Melton LJ 3rd, Hepper NG, Offord KP. Influence of height
parietal pleura. on the risk of spontaneous pneumothorax. Mayo Clin Proc.
2. If a flap-valve pleural injury occurs, air is entrained 1981;56:678.
into the pleural space during inspiration and cannot 3. Ohata M, Suzuki H. Pathogenesis of spontaneous pneu-
mothorax. With special reference to the ultra-structure of
be vented during expiration, resulting in a tension
emphysematous bullae. Chest. 1980;77:771.
pneumothorax.
4. Bense L, Eklund G, Wiman LG. Smoking and the in-
3. Primary spontaneous pneumothorax occurs more creased risk of contracting spontaneous pneumothorax.
commonly in tall, thin men with smoking increasing Chest. 1987;92:1009.
the risk approximately 20-fold. 5. Sahn SA, Heffner JE. Spontaneous pneumothorax. N Engl J
4. Secondary spontaneous pneumothorax occurs in a Med. 2000;342:868.
variety of conditions. The occurrence of a spon- 6. Heffner JE, Huggins JT. Management of secondary spon-
taneous pneumothorax in patients with chronic taneous pneumothorax: theres confusion in the air. Chest.
obstructive pulmonary disease represents a signif- 2004;125:1190.
icant marker for mortality. 7. Videm V, Pillgram-Larsen J, Ellingsen O, et al. Spontaneous
pneumothorax in chronic obstructive pulmonary disease:
5. The mechanism of traumatic pneumothorax can
complications, treatment and recurrences. Eur J Respir Dis.
be due to blunt or penetrating trauma. Penetrating
1987;71:365.
trauma is further classified into low or high velocity 8. Light RW, O Hara VS, Moritz TE, et al. Intrapleural
injury. tetracycline for the prevention of recurrent spontaneous
6. Occult pneumothoraces are seen by CT scans but pneumothorax: results of a Department of Veterans Affairs
are not visible by routine chest radiographs. They cooperative study. JAMA. 1990;264:2224.
are managed conservatively if the patient does not 9. Macho JR, Lewis FR, Krupski W. Management of the injured
require positive pressure ventilation. patient. In: Way LW, ed. Current Surgical Diagnosis and
7. There is a small but recognized risk of pneumoth- Treatment, 10th ed. Norwalk: Appleton & Lange, 1994:227.
orax associated with neuraxial, paravertebral, inter- 10. Hill SL, Edmisten T, Holtzman G, et al. The occult
pneumothorax: an increasing diagnostic entity in trauma.
costals, and periclavicular blocks.
Am Surg. 1999;65:254.
8. Clinical features suggestive of pneumothorax in-
11. Collins JC, Levine G, Waxman K. Occult traumatic pneu-
clude chest pain, dyspnea, cough, and occasionally mothorax: immediate tube thoracostomy versus expectant
hemoptysis. In patients with trauma, symptoms of a management. Am Surg. 1992;58:743.
pneumothorax may be masked by other distracting 12. Enderson BL, Abdalla R, Frame SB, et al. Tube thoracostomy
injuries. for occult pneumothorax: a prospective randomized study of
9. Under general anesthesia, a pneumothorax can be its use. J Trauma. 1993;35:726.
difficult to diagnose. Maintain a high index of 13. Sassoon CS, Light RW, OHara VS, et al. Iatrogenic pneu-
suspicion. A rise in both peak and plateau pressures mothorax: etiology and morbidity. Results of a Department
may precede any cardiorespiratory compromise. of Veterans Affair cooperative study. Respiration. 1992;59:
215.
10. In supine patients, air tends to track to the anterior
14. Zaugg M, Stoehr S, Weder W, et al. Accidental pleural punc-
costophrenic sulcus. Chest radiographs fail to detect ture by a thoracic epidural catheter. Anaesthesia. 1998;53:69.
approximately one third of these pneumothoraces. 15. Iida S, Kashimoto S, Matsukawa T, et al. A hemothorax after
11. Pneumothoraces can be treated conservatively with thoracic epidural anesthesia. J Clin Anesth. 1994;6:505.
simple aspiration or chest tube drainage or by 16. Shime N, Shigemi K, Hosokawa T. Intrathoracic migration
placing percutaneous pneumothorax catheters. of an epidural catheter. J Anesth. 1991;5:100.
C H A P T E R 11/ P N E U M O T H O R A X A N D B A R O T R A U M A 177
17. Moore DC. Intercostal nerve blocks for postoperative somatic 41. Baumann MH, Patel PB, Roney CW, et al. Comparison of
pain following surgery of thorax and upper abdomen. Br J function of commercially available pleural drainage units
Anaesth. 1975;47(suppl):284. and catheters. Chest. 2003;123:1878.
18. McCleery RS, Zollinger R, Lenahan NE. A clinical study of 42. Gammie JS, Banks MC, Fuhrman CR, et al. The pigtail
the effect of intercostal nerve block with Nupercaine in oil catheter for pleural drainage: a less invasive alternative to
following upper abdominal surgery. Surg Gynecol Obstet. tube thoracostomy. JSLS. 1999;3:57.
1948;86:680. 43. Rusch VW, Capps JS, Tyler ML, et al. The performance
19. Bartlett RW. Bilateral intercostal nerve blocks for upper of four pleural drainage systems in an animal model of
abdominal surgery. Surg Gynecol Obstet. 1940;71:194. broncho-pleural fistula. Chest. 1988;93:859.
20. Shanti CM, Carlin AM, Tyburski JG. Incidence of pneu- 44. Powner DJ, Cline D, Rodman GH. Effects of chest-tube
mothorax from intercostal nerve block for analgesia in rib suction on gas flow through a bronchopleural fistula. Crit
fractures. J Trauma. 2001;51:536. Care Med. 1985;13:99.
21. Neuburger M, Landes H, Kaiser H. Pneumothorax in vertical 45. Carney M, Ravin CE. Intercostal artery laceration during
infraclavicular block of the brachial plexus. Review of a rare thoracentesis: increased risk in elderly patients. Chest. 1979;
complication. Anaesthesist. 2000;49:901. 75:520.
22. Naja Z, Lonnqvist PA. Somatic paravertebral nerve blockade. 46. Matsuura Y, Nomimura T, Murakami H, et al. Clinical
Incidence of failed block and complications. Anaesthesia. analysis of reexpansion pulmonary edema. Chest. 1991;100:
2001;56:1184. 1562.
23. Mansfield PF, Hohn DC, Fornage BD, et al. Complications 47. Sprung CL, Loewenherz JW, Baier H, et al. Evidence for
and failures of subclavian-vein catheterization. N Engl J Med. increased permeability in reexpansion pulmonary edema.
1994;331:1735. Am J Med. 1981;71:497.
24. Shah KB, Rao TL, Laughlin S, et al. A review of pulmonary 48. Miller WC, Toon R, Palat H, et al. Experimental pulmonary
artery catheterization in 6,245 patients. Anesthesiology. edema following re-expansion of pneumothorax. Am Rev
1984;61:271. Respir Dis. 1973;108:654.
25. Polderman KH, Girbes ARJ. Central venous catheter use. Part 49. Tan HC, Mak KH, Johan A, et al. Cardiac output increases
1: mechanical complications. Intensive Care Med. 2002;28:1. prior to development of pulmonary edema after re-expansion
26. Dorsch JA, Dorsch SE. The anesthesia machine In: of spontaneous pneumothorax. Respir Med. 2002;96:461.
Dorsch JA, Dorsch SE, ed. Understanding Anesthesia Equip-
50. Volpicelli G, Fogliati C, Radeschi G, et al. A case of unilateral
ment. 3rd ed. Baltimore: Williams & Wilkins, 1994:68.
re-expansion pulmonary oedema successfully treated with
27. Russomanno JH, Brown LK. Pneumothorax due to ball-
non-invasive continuous positive airway pressure. Eur J
valve obstruction of an endotracheal tube in a mechanically
Emerg Med. 2004;11:291.
ventilated patient. Chest. 1992;101:1444.
51. Bernard GR, Artigas A, Brigham KL, et al. The American-
28. Silbergleit R, Lee DC, Blank-Reid C, et al. Sudden severe
European consensus conference on ARDS: definitions, mech-
barotrauma from self-inflating bag-valve devices. J Trauma.
anisms relevant outcomes, and clinical trials co-ordination.
1996;40:320.
Am J Respir Crit Care Med. 1994;149:818.
29. Ludemann R, Krysztopik R, Jamieson GG, et al. Pneumoth-
52. Woodside KJ, vanSonnenberg E, Chon KS, et al. Pneumoth-
orax during laparoscopy. Surg Endosc. 2003;17:1985.
orax in patients with acute respiratory distress syndrome:
30. Botz G, Brock-Utne JG. Are electrocardiogram changes
pathophysiology, detection, and treatment. J Intensive Care
the first sign of impending peri-operative pneumothorax?
Med. 2003;18:9.
Anaesthesia. 1992;47:1057.
53. DiRusso SM, Nelson LD, Safcsak K, et al. Survival in patients
31. Strizik B, Forman R. New ECG changes associated with a
tension pneumothorax. Chest. 1999;115:1742. with severe adult respiratory distress syndrome treated with
32. Walston A, Brewer D, Kitchens C. The electrocardiographic high-level positive end-expiratory pressure. Crit Care Med.
manifestation of spontaneous left pneumothorax. Ann Intern 1995;23:1485.
Med. 1974;80:375. 54. Weg JG, Anzueto A, Balk RA, et al. The relation of
33. Seow A, Kazerooni EA, Perricano PG, et al. Comparison of pneumothorax and other air leaks to mortality in the acute
upright inspiratory and expiratory chest radiographs for de- respiratory distress syndrome. N Engl J Med. 1998;338:341.
tecting pneumothoraces. AJR Am J Roentgenol. 1996;166:313. 55. Schnapp LM, Chin DP, Szaflarski N, et al. Frequency and
34. Kong A. The deep sulcus sign. Radiology. 2003;228:415. importance of barotrauma in 100 patients with acute lung
35. Kaufman B, Dhar P, O Neill DK, et al. Chest radiograph injury. Crit Care Med. 1995;23:272.
interpretation skills of anesthesiologists. J Cardiothorac Vasc 56. Boussarsar M, Thierry G, Jaber S, et al. Relationship between
Anesth. 2001;15:680. ventilatory settings and barotrauma in the acute respiratory
36. Rowan KR, Kirkpatrick AW, Liu D, et al. Traumatic distress syndrome. Intensive Care Med. 2002;28:406.
pneumothorax detection with thoracic US: correlation with 57. The Acute Respiratory Distress Syndrome Network. Ventila-
chest radiography and CT-initial experience. Radiology. 2002; tion with lower tidal volumes as compared with traditional
225:210. tidal volumes for acute lung injury and the acute respiratory
37. Tagliabue M, Casella TC, Zincone GE, et al. CT and chest distress syndrome. N Engl J Med. ,2000;342:1301.
radiography in the evaluation of adult respiratory distress 58. Tremblay L, Miatto D, Hamid Q, et al. Injurious ventilation
syndrome. Acta Radiol. 1994;35:230. induces widespread pulmonary epithelial expression of
38. Kemper AC, Steinberg KP, Stern EJ. Pulmonary interstitial tumor necrosis factor-alpha and interleukin-6 messenger
emphysema: CT findings. AJR Am J Roentgenol. 1999; RNA. Crit Care Med. 2002;30:1693.
172:1642. 59. Tremblay L, Valenza F, Ribeiro SP, et al. Injurious ventilatory
39. Faruqi S, Gupta D, Aggarwal AN, et al. Role of simple needle strategies increase cytokines and c-fos m-RNA expression in
aspiration in the management of pneumothorax. Indian J an isolated rat lung model. J Clin Invest. 1997;99:944.
Chest Dis Allied Sci. 2004;46:183. 60. Chiumello D, Pristine G, Slutsky AS. Mechanical ventilation
40. Markos J, McGonigle P, Phillips MJ. Pneumothorax: treat- affects local and systemic cytokines in an animal model of
ment by small-lumen catheter aspiration. Aust N Z J Med. acute respiratory distress syndrome. Am J Respir Crit Care
1990;20:775. Med. 1999;160:109.
CHAPTER PULMONARY EDEMA
A
admitted for mandibular fracture repair af- monary edema after noncardiac surgical procedures, the
ter facial trauma. The patient admitted using mortality rate was 57%, whereas in patients who devel-
heroin, cocaine, and crack. In the emer- oped heart failure without the clinical presentation of
gency room, he appeared to be mildly se- edema, mortality was 15%.2 Another group of patients
dated, but oriented and coherent, without in a major tertiary medical center, with a relatively high
any complaints or symptoms. The patient was brought incidence of edema (7.6%), had a lower mortality rate of
for surgery without completing his full toxicology screen. 11.9%.3 The true incidence of pulmonary edema, however,
A rapid sequence induction with cricoid pressure was per- may be greater because of numerous cases of increased
formed. Following endotracheal intubation, pink frothy extravascular lung water that do not demonstrate an ob-
fluid appeared in the tube, associated with a decrease in vious clinical presentation of edema. It is clear, however,
oxygen saturation from 100% to 90% SO2 despite an FiO2 that this serious complication has a significant implication
of 1.0. Ten cm of positive end-expiratory pressure (PEEP) on surgical outcome; therefore, anesthesiologists must be
were added with some improvement. Surgery was post- familiar with its diagnosis and management. Early diag-
poned, and the patient was taken to the intensive care nosis and appropriate care are the keys for improving
unit (ICU). Chest radiograph demonstrated pulmonary perioperative outcome.
edema with widespread, patchy, bilateral airspace con-
solidations, ill-defined vessels, and peribronchial cuffing.
The diagnosis of heroin-induced pulmonary edema was
made, not associated with renal insufficiency or aspira- What Is the Clinical
tion of gastric contents. Rapid resolution of the infil-
trates was observed within two days, with no parenchymal Presentation of Pulmonary
sequelae. The patient then underwent surgery without Edema?
complications.
Pulmonary edema can be classified into three stages. The
first involves accumulation of extravascular lung water
in the interstitial space. Cuffs of fluid appear around
What Is the Incidence of the bronchi and blood vessels, without severe impair-
Perioperative Pulmonary ment of oxygenation. During the second stage, fluid
enters the alveoli, and hypoxemia appears because of
Edema? increased ventilation/perfusion mismatching. The third
and most dramatic stage is airway flooding. With in-
Pulmonary edema can occur preoperatively, intraopera- creased lung water, there is a reduction in compliance,
tively, or postoperatively. It is defined as an abnormal increased work of breathing, and smaller functional resid-
accumulation of fluids in the extravascular lung space, ual capacity.
and is associated with changes in lung volume, mechan- During the development stages of edema, the patient
ics, and gas exchange. Pulmonary edema can result from becomes tachypneic from attempting to compensate for
various causes or conditions, and is potentially lethal. the reduced lung volume. Clinical signs of increased
The clinical presentation of perioperative pulmonary inspiratory effort will appear: anxiety, alae nasi, and
edema is relatively infrequent, with a calculated incidence suprasternal and intercostal retractions. Patients present
178
C H A P T E R 12 / P U L M O N A R Y E D E M A 179
with rapidly progressive dyspnea, orthopnea, and with or THE BLOOD-GAS BARRIER
without hemoptysis. In rare instances, large amounts of
frothy pink sputum may be seen. Cyanosis and hypoxemia Because of the vulnerability of the blood-gas barrier to
can be present and may remain unresponsive to oxygen stress failure when the capillary pressure rises, it is
therapy. important to know what is responsible for the strength
of the capillary wall. The thin side of the blood-gas
barrier consists of the capillary endothelial layer, alveolar
epithelial layer, and the extracellular matrix (ECM), which
What Are the Anatomic is made up of the fused basement membranes of the two
cellular layers. There is strong evidence that most of the
Considerations? strength comes from the ECM, particularly the type IV
collagen in the basement membranes. This suggests that
The structures involved in the pathogenesis of pulmonary the ECM is the strongest layer.4 The thickness of the
edema are the microvascular endothelium, alveolar ep- capillary basement membranes is frequently related to the
ithelium, pulmonary interstitium, and pulmonary lym- transmural pressure. For example, patients with mitral
phatic system. stenosis who have an increased pulmonary capillary
pressure over several years have thickened basement
membranes.5
MICROVASCULAR
ENDOTHELIUM
The endothelium is composed of a monolayer of cells
What Is the Pathophysiology
situated on the basement membrane (located between of Pulmonary Edema?
the endothelium and epithelium). The junctions between
endothelial cells contain pores that allow the passage of
water and ions. They act as a sieve for protein molecules.
Of these molecules, only the smaller ones (i.e., albumin) GENERAL
can pass normally with relative freedom; large molecules FUNCTIONAL-STRUCTURAL
(i.e., globulin, fibrinogen) cannot.
FORCES AND MECHANISM
Pulmonary edema is defined as an abnormal accumula-
ALVEOLAR EPITHELIUM tion of fluid in the extravascular compartments of the lung.
The relative amounts of intravascular and extravascular
The alveolar epithelial cells, which are situated on the fluid in the lung are mostly controlled by the permeability
other side of the basement membrane, are composed of the capillary membrane, as well as the oncotic pres-
of type 1 and 2 pneumocytes. Type 1 are large, flat sure.6 This relation is described by the Starling equation,
cells that line the alveoli as a continuous sheet and which is used to determine the theoretic amount of fluid
have tight gaps (pores) in their junctions. Therefore, the Qfilt filtered per unit area per unit of time:
alveolar wall is much less permeable than the endothelial
wall. This also explains the cases in which pulmonary Qfilt = Kfilt (HPiv HPev ) t(OPiv OPev ).
edema remains in the interstitium without filling the
alveoli. Type 2 cells are cuboidal stem cells that produce In this equation, HPiv and HPev represent the in-
surfactant. travascular and extravascular hydrostatic pressure, and
OPiv and OPev represent the intravascular and extravas-
cular oncotic pressure, respectively. Kfilt represents the
conductance of the capillary wall and expresses the wa-
PULMONARY INTERSTITIUM ter resistance created by the capillary endothelial cell
The interstitium is the thin layer that lies between junctions with changes in HPiv and HPev , whereas t rep-
the endothelial and alveolar layers. This space contains resents the oncotic reflection coefficient and indicates
fibrin, collagen matrix, pulmonary lymphatic vessels, and the permeability of the capillary membrane to macro-
basement membrane. Its thicker component surrounds molecules. The greater this reflection coefficient, the more
the large blood vessels and airways. the passage of macromolecules will be restricted, thereby
decreasing overall fluid filtration. The net flow, Fnet , is
defined as Qfilt Qlymph , where Qfilt represents fluid tran-
sudation or exudation and Qlymph represents lymphatic
PULMONARY LYMPHATIC absorption. Pulmonary edema develops when the equi-
SYSTEM librium between fluid transudation or exudation Qfilt and
lymphatic absorption Qlymph is disturbed. Therefore, al-
The lymphatic vessels are in the interstitium and drain though under normal conditions the endothelial cells are
the pulmonary microvascular filtrate. relatively impermeable to protein but remain permeable
180 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
to water and solutes, the tight intercellular junctions of The differences between these two types of edema reflect
the alveolar epithelium remain nearly impermeable to the type and extent of injury to the cellular layers of the
water and solutes, thereby constituting an effective bar- blood-gas barrier, whether there is a causative, associated,
rier that is a major factor in preventing the development or ensuing inflammatory process.
of pulmonary edema. Lymphatic drainage (Qlymph ) rep-
resents another way of eliminating excess lung water. A
manyfold increase in lymphatic flow has been observed Site and Distribution of Edema Fluid
with chronically increased hydrostatic pressure. This in-
crease in lymphatic flow is very efficient in eliminating Chest radiographs of patients with early pulmonary
excess water, especially when there is diminished oncotic edema usually manifest a rather inhomogeneous, patchy
pressure due to hypoalbuminemia.7 However, its impact distribution of extravascular fluid accumulation with the
requires time, and therefore, it may not be as effective in apicobasal gradient.8 In addition, the volume of interstitial
acute settings. edema fluid does not correlate with the amount of fluid
Cardiogenic pulmonary edema (CPE) arises when in the adjacent alveoli. The differences can be explained
HPev or the capillary pressure increases secondary to left according to gravity, filtration pressure, and variations in
atrial hypertension, causing transudation of fluid across albumin concentrations in the edema fluid. Apparently,
the interstitium into the alveoli. Noncardiogenic edema the volume of endothelial leaks are not necessarily
most commonly arises when the reflection coefficient (t) is matched by that of the epithelial leaks. These observations
decreased because of endothelial injury. The variable t strongly suggest that there are substantial differences
varies in value from 0 to 1, depending on the restriction of in fluid filtrations and in reflection coefficients for
plasma protein by the endothelium. This injury allows the macromolecules. Therefore, a simple, uniform membrane
passage of normally restricted plasma proteins from the model of passive fluid and solute movements following
vascular space into the interstitium and alveoli, creating pressure gradients does not necessarily explain how
an exudative edema. hydrostatic pulmonary edema occurs. This condition is,
Pulmonary edema can be divided into four main rather, the result of complex alterations in the structure
categories on the basis of its pathophysiology: and function of the central component of the blood-gas
barrier (e.g., endothelial and epithelial cell layers), in
1. Increased hydrostatic pressure edema addition to increased hydrostatic pressure.
2. Permeability edema with diffuse alveolar damage
(DAD)
3. Permeability edema without DAD and Barrier Lesions
4. Mixed edema, due to simultaneous increased hydro-
static pressure and permeability changes There are different types of epithelial lesions on the
thick or thin side of the alveolar-capillary barrier. For
This classification scheme is helpful because pul-
example, sudden stress on the barrier can damage the
monary edema is often seen perioperatively in the ICU
endothelial layer by either opening the intercellular
and emergency department.
junctions or through cellular disruption of the epithelial
layer. In this context, ultrastuctural findings in the lungs
Unusual Findings of patients with hemodynamic pulmonary edema, in
particular neurogenic edema, are of interest in that
The clinical and radiologic manifestations of acute pul- numerous erythrocytes are found in the interstitial and
monary edema are generally well established. However, alveolar edema fluid.8 Therefore, the rusty sputum
pulmonary edema may also demonstrate unusual find- produced by patients with lung congestion and edema
ings. Atypical pulmonary edema is defined as lung edema appears to be of alveolar origin.
with an unusual radiologic appearance but with clinical
findings usually associated with well known etiologies.
Unusual forms of pulmonary edema are defined as lung Spectrum of Cardiogenic to
edema from unusual causes (e.g., rare diseases, or rare
High-Permeability Edema as Capillary
manifestations of common diseases).
Pressure Increases
In practice, the traditional division between hydrostatic
STRUCTURE-FUNCTION and permeability edema does not always meet expecta-
RELATIONSHIP tions. For example, Fein et al.9 pointed out that there is a
substantial overlap between the two groups, even in con-
The formation and morphology of hydrostatic pulmonary ditions where a pure form of hydrostatic or cardiogenic
edema appear to be complex, and fluid and protein edema would be expected. Sprung et al.10 showed that
movement from the microvasculature into the interstitial there is a continuous fall in the ratio of protein in edema
and alveolar spaces cannot be explained solely by uniform fluid-to-serum when plotted against pulmonary artery
membrane models of fluid exchange. Barrier leaks and, in wedge pressure for a large group of patients. They re-
particular, epithelial cell layer leaks play a role not only ferred to it as an intermediate type of pulmonary edema
in permeability edema, but also in hydrostatic edema. on the basis of the protein concentration of the alveolar
C H A P T E R 12 / P U L M O N A R Y E D E M A 181
considerably extended in the last 10 years. Several im- appearance of Kerley lines, and subpleural effusions.18 If
portant discoveries, such as the reversal effects of cyclic the quantity of extravascular fluid continues to increase,
adenosine 3, 5monophosphate (cAMP) and the protec- the edema will migrate centrally, with progressive blur-
tive effect of nitric oxide, may eventually influence the ring of vessels, first at the lobar level and later at the level
clinical therapy. of the hilum. At this point, lung radiolucency decreases
markedly, making identification of the small peripheral
vessels difficult. Peribronchial cuffing becomes apparent,
particularly in the perihilar areas.
What Are the Clinical When transmural pressure exceeds 25 mm Hg,
fluid drainage from the extravascular compartment is at
Implications of Perioperative maximum capacity. The second phase (alveolar flooding)
Pulmonary Edema? then commences, leading to a sudden extension of edema
into the alveolar spaces, and thereby creating tiny nodular
or acinar areas of increased opacity that mature into frank
consolidations (see Fig. 12.2).
PRESSURE/HYDROSTATIC Pulmonary artery catheters are frequently used to as-
PULMONARY EDEMA sess hydrostatic pressure in intensive care patients. The
pulmonary capillary wedge pressure has been shown to
Two pathophysiologic and radiologic phases are recog- reflect left atrial pressure and correlates well with the
nized in the development of pressure edema: interstitial radiologic features of congestive heart failure and pul-
edema and alveolar flooding or edema. These phases are monary venous hypertension (see Table 12.1). However,
virtually identical for left-sided heart failure and fluid over- in acute heart failure, a time lag is often observed be-
load, and are the two most frequently observed causes tween the increased pulmonary capillary wedge pressure
of pressure edema in operating room, intensive care, and the radiologic manifestation of pulmonary edema
and emergency patients. The intensity and duration of due to the relatively slow movement of water through the
both phases are clearly related to the degree of increased widened capillary endothelial cell junctions.8 Similarly,
pressure, which is determined by the hydrostaticoncotic as pulmonary edema resolves, the radiologic findings will
pressure ratio. persist, with decreasing, or even normal, pulmonary cap-
Interstitial edema occurs when mean transmural ar- illary wedge pressure (see Fig. 12.3). In addition, pressure
terial pressure increases 15 to 25 mm Hg. This results in edema may show an asymmetric distribution on chest ra-
early loss of definition of subsegmental and segmental ves- diographs. The most frequent cause is an abnormal lung
sels, mild enlargement of the peribronchovascular spaces, parenchyma, as seen in chronic obstructive pulmonary
A B
FIGURE 12.2 Increased hydrostatic pressure edema in a 33-year-old man with acute myelocytic
leukemia who was admitted for fluid overload with renal and cardiac failure. Successive chest
radiographs demonstrate progressive lobar vessel enlargement, peribronchial cuffing (Left, arrows),
bilateral Kerley lines (Right, arrowheads), and late alveolar edema with nodular areas of increased
opacity. The fluid overload is confirmed by the increasing size of the azygos vein. (From: Gluecker T,
Capasso P, Schnyder P, et al. Clinical and radiologic features of pulmonary edema. Radiographics.
1999;19:15071531.)
C H A P T E R 12 / P U L M O N A R Y E D E M A 183
TABLE 12.1 Correlation between Pulmonary Capillary surgery or immobilization is frequently observed in the
Wedge Pressure and Radiologic Findings lung fields of recumbent patients. This distribution is
typically seen in congestive heart failure but is also
Pulmonary Capillary observed in overhydration.
Wedge Pressure(mm Hg) Radiologic Findings
512 Normal findings
1217 Cephalization of pulmonary Unusual Forms of Hydrostatic Pulmonary
vessels (only in chronic Edema
conditions)
1720 Kerley lines, subpleural Postobstructive Pulmonary Edema
effusions Postobstructive pulmonary edema occurs after relief from
>25 Pulmonary edema an upper airway obstruction and represents a pure form of
hydrostatic edema. Perioperatively, it is most frequently
produced by airway obstruction, laryngospasm, or even
disease (COPD). In cardiac failure, extensive lung emphy- epiglottitis.
sema of the apices (seen in heavy smokers) or marked If the obstruction occurs primarily with forced in-
destruction and fibrosis of the upper and middle portions spiration as the patient struggles to inhale (Muller
of the lungs (seen in end-stage tuberculosis, sarcoidosis, maneuver), the elevated, negative intrathoracic pressure
or asbestosis) will result in pulmonary edema that pre- increases venous return. Subsequently, edema is produced
dominates in the regions less affected by these disease by a sudden, marked decrease in the negative pleural
processes. pressure, which leads to an elevated hydrostatic pres-
Hemodynamic factors can also produce asymmetric sure gradient between the intravascular and extravascular
distribution of pulmonary edema. Edema associated with compartments.21 An obstruction that prevents both in-
severe mitral regurgitation has been shown to favor one spiration and expiration may create an elevated, positive
particular lung, a result of flow impairment provoked intrathoracic pressure that initially prevents edema for-
by the reflux stream directed toward a particular upper mation. However, edema later develops as the obstruction
pulmonary vein.19 Such asymmetric distribution occurs is relieved and the intrathoracic pressure suddenly drops.
in 9% of adults and 22% of children with grade 3 or 4 On chest radiography and computed tomography
mitral regurgitation.20 scan, postobstructive pulmonary edema typically mani-
Finally, the position of the patient also influences fests as septal lines, peribronchial cuffing, and, in more
intravascular and extravascular fluid distribution. In severe cases, central alveolar edema. These findings are
supine patients, axial computed tomography scan usually similar to those seen in pressure edema. Heart size is usu-
demonstrates an anteroposterior gradient, whereas more ally normal, indicating a pressure edema that is not related
asymmetric distribution of edema secondary to prolonged to overhydration. The resolution of clinical symptoms and
FIGURE 12.3 Increased hydrostatic pressure edema in a 53-year-old man with postoperative fluid
overload. Pulmonary capillary wedge pressure was 20 mm Hg. High-resolution computed
tomography scan demonstrates inter- and intralobar septal lines predominating in the anterior
portion of the left lung field with some peribronchial cuffing (arrow). Both lungs display diffuse
ground-glass areas of increased attenuation with a gravitational anteroposterior gradient. (From:
Gluecker T, Capasso P, Schnyder P, et al. Clinical and radiologic features of pulmonary edema.
Radiographics. 1999;19:15071531.)
184 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
radiologic findings is rapid and usually occurs within 2 to of these cells to chemical agents, infectious pathogens,
3 days. gastric fluid, or toxic gas, which destroy or severely
damage the cells. Secondary damage occurs because of
a systemic biochemical cascade that produces oxidat-
Pulmonary Edema with Acute Asthma
ing agents, inflammatory mediators, and enzymes, which
Pulmonary edema with acute asthma is a rare patho- also harm these endothelial cells during sepsis, pancre-
logic condition because the associated trapped air tends atitis, severe trauma, or blood transfusion. On the basis
to maintain a positive intra-alveolar pressure, thereby of these etiologic differences, two major pathophysiologic
decreasing the hydrostatic pressure gradient. Its patho- mechanisms in the development of ARDS have been de-
genesis can be associated with the severity of the Muller scribed: (i) ARDS due to an underlying pulmonary disease,
maneuver (i.e. forced inspiration as the patient struggles which is associated with pulmonary consolidation; and
to inhale). Furthermore, it has been demonstrated that the (ii) ARDS secondary to extrapulmonary disease, which
mean pleural pressure is markedly decreased over the en- manifests as interstitial edema and alveolar collapse.24
tire tidal respiration, reaching 25 cm of water compared These mechanisms are based on physiologic ventilation
with 5 cm in healthy subjects.22 The elevated, nega- mechanics, and, although they have not yet been patho-
tive pleural pressure that occurs during acute asthmatic logically proven, they do have distinct implications for the
episodes helps maintain the patency of the narrowed treatment of affected patients.
airways. However, the lowered pleural pressure results The acute respiratory distress syndrome encompasses
in decreased interstitial pressure, whereas intravascular three, often overlapping, stages.
pressures are only minimally affected. Airway obstruction
in acute asthma is not uniform throughout the lungs, and 1. EXUDATIVE: This stage is characterized by interstitial
is therefore associated with heterogenous accumulation edema with a high protein content that rapidly fills the
of extravascular fluid. The radiologic findings cannot be alveolar spaces. It is associated with hemorrhage and
differentiated from other causes of cardiogenic edema. ensuing hyaline membrane formation (see Fig. 12.4).
The rapid extension of edema into the alveolar spaces
probably explains why findings that are typically seen in
Edema with Acute Pulmonary Embolism interstitial edema (e.g., Kerley lines) are not prominent
Pulmonary edema is seen in <10% of cases of acute in ARDS.
pulmonary embolism.23 It usually appears on computed 2. PROLIFERATIVE: This stage manifests as organization of
tomography scan as heterogeneous areas of increased the fibrinous exudate. Following this organization, one
ground-glass attenuation localized in the territories of the can observe the regeneration of the alveolar lining and
patent segmental or subsegmental arteries. The etiology thickening of the alveolar septa.
is likely to be primarily because of hydrostatic factors 3. FIBROTIC: The last stage is characterized by varying
superimposed on the underlying embolic disease. The degrees of scarring and formation of subpleural and
mechanism of pulmonary edema produced by massive intrapulmonary cysts.
acute pulmonary embolism is believed to be directly re-
lated to pulmonary hypertension, caused by the occlusion
of >50% of the pulmonary arterial bed. Because right-
sided cardiac output is then directed through a reduced
arterial network, capillary hydrostatic pressure increases
markedly, and the increased perfusion of the areas not
involved by the vascular thrombosis leads to edema.
A progression of clinical findings is present in patients approximately 15% of cases of heroin overdose, with
with ARDS. Tachypnea, tachycardia, and respiratory an overall mortality rate of 10%.25 Heroin overdose is
alkalosis usually develop within the first 12 to 24 hours believed to directly depress the medullary respiratory
of the inciting event and may precede the appearance center and lead to hypoxia and acidosis, both of which
of infiltrates on a chest radiograph. The inflammatory can produce permeability edema without DAD.26 This
process and alveolar flooding lead to severe ventilation- absence of DAD can be directly inferred from the rapid
perfusion mismatch; severe hypoxia, increased dead resolution of the disorder observed in all cases that
space, and decreased lung compliance all contribute to the are not complicated by aspiration of gastric contents
development of respiratory failure. Most patients develop or by infection. Unlike cocaine, heroin has no direct
diffuse alveolar infiltrates and progress to respiratory deleterious effect on myocardial function. When heroin-
failure within 48 hours of the onset of symptoms. induced pulmonary edema is not associated with renal
insufficiency or other complications such as aspiration
of gastric contents, rapid resolution of the infiltrates is
PERMEABILITY EDEMA observed within 1 or 2 days with no parenchymal sequelae.
WITHOUT DAD
As the name implies, permeability edema without DAD MIXED EDEMA
refers to pulmonary edema in which permeability changes
are not primarily associated with DAD. The absence of cel- Neurogenic Pulmonary Edema
lular damage is often not proven pathologically, but may
be inferred from the clinical and radiologic course of the Neurogenic pulmonary edema (NPE) is seen in up to 50%
disease; rapid regression is often observed, with venti- of patients who have suffered a severe brain insult such
latory improvements occurring within a short period of as trauma, subarachnoid hemorrhage, stroke, or status
time. Although some degree of DAD may occur, damage epilepticus.27 It is a form of edema that may develop
remains minor and usually only partially affects patient rapidly after central nervous system insults such as head
outcome. injury. These stimuli may produce massive sympathetic
nervous system activation, leading to extreme, but tran-
Heroin-Induced Pulmonary Edema sient, episodes of systemic and pulmonary hypertension,
the latter resulting in edema (see Fig. 12.5). Differenti-
Pulmonary edema directly associated with an overdose ation of NPE from simple fluid overload or aspiration
of opiates occurs almost exclusively with heroin but is edema may be difficult, if not impossible, in patients
also rarely encountered with the use of cocaine and with trauma or immediately following surgery. Therefore,
crack. Heroin-induced pulmonary edema is seen in the diagnosis of NPE is obtained by exclusion. Its cause
A B
FIGURE 12.5 Neurogenic pulmonary edema in a 54-year-old woman who was admitted for
intracranial hemorrhage due to arterial hypertension. A: Chest radiograph obtained at the time of
admission shows airspace consolidations predominantly at the apices. There are no pleural effusions
or Kerley lines, and heart size is normal. B: High-resolution computed tomography scan obtained at
the same time demonstrates confluent alveolar consolidations in the central portions of the lungs.
A few thickened interlobular septa are also seen (arrows). (From: Gluecker T, Capasso P, Schnyder
P, et al. Clinical and radiologic features of pulmonary edema. Radiographics. 1999;19:15071531.)
186 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
remains controversial but probably involves a combina- radiologically during the first 2 days following surgery.
tion of factors associated with hydrostatic edema and The infiltrates progress and are most pronounced on day
factors associated with permeability edema without DAD. 5.30 These signs disappear two weeks after surgery without
It is likely that a combination of factors produce any sequelae, indicating that if DAD is present, it is mild
NPE, and the relative balance of each determines the type and has little or no significance.31
and extent of edema for any given patient. For example,
intravasular volume status, left ventricular compliance,
and perhaps the degree of autonomic neuropathy in
Pulmonary Edema Post Liver
any patient could influence pressure changes in the Transplantation
pulmonary capillaries during a neurologic insult or ictus.
This complex combination of factors influence the degree Liver transplantation is a complex operation involving,
of shunting and oxygenation or PaO2 /FiO2 ratio, and the at times, substantial blood loss, massive transfusion, and
time course for recovery determined by oxygenation. large fluid shifts. Pulmonary complications are common,
Radiologic findings of NPE disappear within 1 to 2 days. with a frequency as high as 75% reported in some stud-
ies,32 and may contribute significantly to perioperative
morbidity and mortality. Protein concentration in the
Reperfusion Pulmonary Edema pulmonary edema fluid produced can be analyzed to de-
Reperfusion pulmonary edema is an acute, mixed, non- termine the mechanism. In addition to diagnostic value,
cardiogenic edema that is observed in up to 90% to 100% the edema fluid/plasma ratio has been shown to have
of patients who have undergone pulmonary thromboen- prognostic significance as well.33 The most common pre-
darterectomy for massive pulmonary embolism or for cipitating factors for noncardiogenic pulmonary edema
segmental stenoses associated with chronic pulmonary are sepsis, pneumonia, gastric aspiration, and multiple
embolism. The main pathophysiologic mechanism of this transfusions.
disorder is directly associated with the rapid increase in
blood flow and blood pressure in the areas distal to the Posttransfusion Pulmonary Edema
recanalized pulmonary arteries. Other mechanisms, such
as mechanical stress due to surgical intervention and bio- Transfusion-related acute lung injury (TRALI) was first
chemical phenomena (e.g., release of oxygen radicals by coined to refer to noncardiogenic pulmonary edema
neutrophils, alterations in surfactant production), must complicating transfusion therapy.34 It has been reported
also be considered.28 after the administration of packed red blood cells,35 fresh
Radiologic findings of pulmonary edema appear frozen plasma,36 and platelets.37 TRALI is most frequently
within the first two days following surgery. Findings at produced by antibodies present in the blood product that
conventional chest radiography usually consist of hetero- are directed against recipient white cell antigens. When
geneous airspace consolidations that predominate in the specific testing for the presence of these antibodies is
areas distal to the recanalized vessels. Recently, however, done, they most commonly occur in multiparous donors
investigators have also found a random distribution of who have been immunized by exposure to paternal
pulmonary edema in up to 50% of cases.29 These authors foreign antigens during pregnancy. In most cases of
hypothesize that the reperfusion pulmonary edema may TRALI, the episode of pulmonary edema is short-lived,
also be due to systemic factors that have not yet been lasting 2 to 6 hours. This was the pattern observed
identified. in patients developing pulmonary edema during liver
transplantation.
Pulmonary Edema Post Lung TRALI is an underreported complication of transfu-
Transplantation sion therapy, and it is the third most common cause of
transfusion-associated death. TRALI is defined as noncar-
Pulmonary edema following lung transplantation is a diogenic pulmonary edema temporally related to transfu-
noncardiogenic form of edema that is observed in sion therapy. The diagnosis of TRALI relies on excluding
up to 97% of patients during the first three days other diagnoses such as sepsis, volume overload, and
following surgery.30 The most important causal factors CPE. Clinical features include chills, fever, tachycardia,
are probably those related to the tissue hypoxia that cough, and various degrees of respiratory distress. Tran-
primarily involves the graft but also the host organs during sient leukopenia may be noted. The chest demonstrates
the procedure, which is performed with extracorporeal bilateral pulmonary infiltrates in the absence of cardiac
circulation. Other factors, such as the disruption of enlargement and pulmonary vascular engorgement. Nor-
pulmonary lymphatic drainage and lung denervation with mal pulmonary capillary wedge pressure on right heart
microvascular changes, are also likely to contribute to the catheterization verifies the noncardiogenic origin of the
process. Pulmonary edema following lung transplantation pulmonary edema formation. The onset of respiratory
is not due to left ventricular failure, fluid overload, distress ranges between a few minutes and 40 hours af-
acute rejection, atelectasis, or infection, although these ter transfusion, with a maximum manifestation at 4 to
conditions can coexist and thereby complicate the clinical 8 hours. In most cases, the symptoms subside within 1 or
picture. 2 days with full recovery. When TRALI is suspected, the
The manifestation of this disease entity is variable. diagnostic algorithm should include the exclusion of other
A mixed hydrostatic and permeability edema can be seen factors and investigation of blood products for antibodies.
C H A P T E R 12 / P U L M O N A R Y E D E M A 187
Treatment is supportive, and the prognosis is substantially a major contributing factor.42 Patients who undergo right
better than most causes of clinical acute lung injury. pneumonectomy are considered to be at higher risk for
postpneumonectomy pulmonary edema than those who
Reexpansion Pulmonary Edema undergo left pneumonectomy, probably due to the smaller
volume of the remaining lung.42
Reexpansion pulmonary edema is an uncommon iatro- The precise etiology and pathophysiology of post-
genic complication that occurs after the rapid reexpansion pneumonectomy pulmonary edema remain controver-
of a collapsed lung following drainage or evacuation of sial and largely unknown. Increased capillary hydrostatic
pleural conditions such as pneumothorax, hydrothorax, pressure and altered capillary permeability are both prob-
or hemothorax. In 64% of cases, reexpansion pulmonary able contributing factors in the development of postpneu-
edema appears suddenly within 1 hour of lung reexpan- monectomy pulmonary edema. Because the remaining
sion. The process usually involves the entire reexpanded lung is subject to increased pulmonary blood flow due
lung,38 although on rare occasions, only a single lobe to the redistribution of cardiac output, an increase in
or segment may be involved. In most cases, reexpan- mean pulmonary arterial blood pressure is observed. On
sion pulmonary edema increases in severity for 24 to the other hand, it has been postulated that the transient
48 hours and then slowly resolves over the next 5 to hypoxia seen perioperatively and immediately following
7 days, indicating that the pathophysiologic process is surgery may induce reflex pulmonary vasoconstriction,
not purely hydrostatic. A prolonged, local hypoxic event, leading to the release of catecholamines. In addition,
the abrupt restoration of pulmonary blood flow, and the these changes in pressure and oxygen levels may result in
sudden, marked increase in negative intrapleural pressure alveolar wall damage and subsequent passage of fluid and
are probably all significant factors in the development of proteins into the alveolar lumina. The latter is suspected
pulmonary edema.39 However, the presence of proteins because of the high protein content of the airspace fluid
and red blood cells in the alveolar fluid, as well as the seen with postpneumonectomy pulmonary edema.
persistence of clinical symptoms and radiologic findings, At conventional chest radiography, severe postpneu-
indicate the presence of a certain degree of DAD. The monectomy pulmonary edema manifests as infiltrates
presence of DAD is also demonstrated by the inefficiency with an appearance identical to that of ARDS. The most
of the reexpanded lung in terms of gas exchange, which frequently seen radiologic findings in milder forms of
leads to a shunt effect that persists for some time.39 postpneumonectomy pulmonary edema are similar to
Patients may be asymptomatic despite findings of those in hydrostatic pulmonary edema without DAD and
pulmonary edema at chest radiography. On the other include Kerley lines, peribronchial cuffing, and ill-defined
hand, they may present with severe symptoms associated vessels. These findings have a tendency to disappear within
with frank respiratory insufficiency. Sometimes, there is a few days, which strongly indicates that lesions of the
little correlation between the extent of the infiltrates at capillary endothelial cells, if present, are mild in this form
radiology and clinical findings. In most cases, patients of the disorder. Pulmonary edema following lobectomy
present with cough, dyspnea, tachypnea, and tachycar- has also been described in some reports but with an
dia, although in rare instances, large amounts of frothy overall prevalence of <1%.
pink sputum may be seen. Early recognition of reex-
pansion pulmonary edema is important, given that the Pulmonary Edema Due to Air Embolism
disease proves fatal in up to 20% of cases.38 Although
reexpansion pulmonary edema manifests unilaterally in Pulmonary edema due to air embolism is seen infrequently
the reexpanded lung, its radiologic appearance is usu- and usually occurs as an iatrogenic complication of an
ally indistinguishable from that of other forms of mixed invasive procedure. Rarely, it may also be associated with
pulmonary edema. open or closed chest trauma.43 Air may enter into the
low-pressure venous system when a pressure gradient
Postpneumonectomy Pulmonary Edema favors such access, which occurs most frequently during
neurosurgical procedures performed with the patient in
Postpneumonectomy pulmonary edema is a life- the sitting position and during manipulation or placement
threatening complication that occurs early in the post- of central venous lines. Other factors can produce massive
operative period following pneumonectomy. It affects the air embolization during orthopedic and cardiac surgical
remaining lung and is usually diagnosed by exclusion. The procedures due to the open vascular system.
prevalence of postpneumonectomy pulmonary edema is The pathophysiologic mechanism of pulmonary
generally reported to range between 2.5% and 5%, with edema due to air embolism is quite simple. The em-
a very high associated mortality rate in all series.40 On bolized air bubbles create a mechanical obstruction of
the other hand, minor postpneumonectomy pulmonary the pulmonary microvasculature due to the relatively
edema has been reported in up to 20% of patients.41 low absorption coefficient of air. These collections of air
Risk factors for postpneumonectomy pulmonary edema create turbulent flow, which favors platelet aggregation,
have classically included excessive administration of fluid fibrin formation, and vasoconstriction, thereby increasing
during surgery, transfusion of fresh frozen plasma, ar- the pressure exerted on the vessel wall. Other nonme-
rhythmia, marked postsurgical diuresis, and low serum chanical factors (e.g., liberation of oxygen radicals from
colloidal osmotic pressure.42 Recently, however, some au- neutrophils) also contribute to the disruption of the cap-
thors no longer consider perioperative fluid overload to be illary endothelium. Macromolecules, proteins, and blood
188 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
cells may then enter the interstitial and alveolar spaces,43 the acute phase of pulmonary edema is associated with
thereby yielding a pathologic picture that ranges from hypoxic metabolites and air embolism.
mild interstitial edema to hemorrhagic airspace consoli- Face mask-administered CPAP in patients on ade-
dations. quate minute ventilation and with an intact sensorium
During intraoperative transesophageal echocardiog- is indicated during preparation for surgery. During this
raphy, indwelling air may be observed within the right- stage, it is important to verify the patients consciousness
sided cardiac chambers. Conventional chest radiography level, adequate reflexes, and respiratory drive. Increas-
initially demonstrates interstitial edema followed by bi- ing the functional residual capacity by keeping airway
lateral, peripheral alveolar areas of increased opacity that pressure positive is often sufficient to reduce the work of
are predominantly found in the lung bases. There is no breathing and squeeze the edema fluid into the walls of
associated cardiac enlargement, and the lesions disappear the tracheobronchial tree.
rapidly in patients who survive the acute event. A study comparing CPAP (continuous pressure of
10 cm H2 O) with bilevel ventilation (inspiratory positive
airway pressure [IPAP] of 15 cm H2 O and expiratory
positive airway pressure [EPAP] of 5 cm H2 O) and
What Is the Treatment conventional (oxygen through face mask) therapy, showed
that patients with acidotic, acute CPE were more likely
of Perioperative Pulmonary to survive to hospital discharge if treated early with
Edema? CPAP and conventional therapy, compared to bilevel
ventilation and conventional therapy, or conventional
therapy alone.44 This was the first study to show a definite,
The therapeutic approach to pulmonary edema includes
short-term survival benefit with CPAP, although other
three major elements:
studies do show a trend toward improved survival.45
1. Normalization of oxygenation and ventilation Pooled data from such studies have also suggested
2. Reduction of excessive extravascular lung water improved survival with CPAP.46 Earlier studies comparing
3. Identification and treatment of the underlying disease the use of CPAP with standard oxygen therapy in CPE
patients have also demonstrated short-term physiologic
Even when the actual cause is unclear, treatment
improvements, but no benefit in terms of survival.47
should begin immediately, since the major pathophysiol-
These studies showed significant reductions in both
ogy of pulmonary edema is well understood and its re-
respiratory rate and PCO2 after 30 minutes with CPAP,
sponse to certain therapeutic measures is well established
and the alveolar-arterial oxygen tension gradient was
and, most important of all, because it is potentially lethal.
significantly reduced after 3 hours of CPAP therapy
The first element of treatment is oxygen therapy,
compared with oxygen alone. However, intubation rates
improvement of pulmonary functional residual capacity
were not affected, and mortality remained high at 30% in
and mechanics with continuous positive airway pressure
both groups. Considering the results of all these studies, it
(CPAP), with or without intubation, and mechanical
is questionable whether early physiologic improvements
ventilation. The second element may include the use of
are important predictors of mortality.
diuretics to achieve a negative water balance, provided
Overall, there is a very low rate of intubation among
that preload can be decreased. Identification of the
patients presenting with acidotic CPE45 (3.4% to 8.7%
cause may be crucial in assessing the natural course
in one study48 ). It is probable that intubation and
and prognosis and directing definitive treatment. Early
mechanical ventilation are useful in selected patients,
diagnosis and appropriate care are the keys for improving
but it is by no means clear that it improves outcome in
perioperative outcome.
terms of mortality in most CPE patients requiring invasive
ventilatory support.
controlled. Invasive monitoring (arterial, central venous criticized because of the risk for the infused colloids to
pressure, or pulmonary catheters) may be required to opti- leak into the interstitium and increase OPev , resulting in
mize hemodynamic and cardiac function perioperatively. delayed edema resolution.
Operative procedures for patients with ARDS may The most useful drugs to reduce HPiv are diuretics.
range from minor (tracheostomy) to major (trauma, Beside reducing total body water, they produce venodi-
vascular, major abdominal). Evaluation of ICU patients lation and increase OPiv . When pulmonary edema is due
with ARDS includes familiarization with the respiratory to congestive heart failure, other measures for preload
status, ventilation parameters, and ventilation need or reduction such as nitroglycerin or neuroaxial block are of
support during transport and surgery. Transport of value. Pulmonary artery pressure reduction with vasodila-
these patients to surgical or imaging procedures can tors may be beneficial as well and have been suggested
be a difficult challenge. The anesthesiologist should be to improve outcome in ARDS patients.51 However, va-
involved in planning the transport and preparing the sodilators can increase lung lymph production, as well as
oxygen supply and appropriate PEEP level to be provided. worsen intrapulmonary shunt and hypoxemia by blunting
The use of a transport ventilator capable of providing hypoxic vasoconstriction.
CPAP and high levels of inspiratory pressure and minute Recently, much attention has been directed to the use
ventilation is crucial. of nitric oxide to treat hypoxemia and pulmonary hyper-
tension, in pulmonary edema patients. Nitric oxide can
reduce pulmonary lymph flow by reducing intravascular
pressure HPiv and permeability.
INTRAOPERATIVE
MANAGEMENT Other Considerations
Ventilation Induction of anesthesia requires agents that maintain
cardiac output, such as narcotics, ketamine, or etomidate.
Intraoperative ventilation should include the administra-
Other drugs that can be useful are those directed at
tion of PEEP. The reduction of preload by PEEP must
specific problems such as sepsis, cardiac decompensation,
be considered in hypovolemic patients. Pulmonary edema
or ischemia. Invasive monitoring is frequently indicated
with reduced compliance and increased resistance re-
to optimize hemodynamic and cardiac function during
quires high minute ventilation and inspiratory pressures.
surgery.
For patients who require a preoperative minute ventila-
tion >15 L and peak inspiratory pressure >15 cm H2 O, an
ICU-type ventilator is recommended intraoperatively.49
During surgery and muscle relaxation, partial ventila- POSTOPERATIVE
tory support measures (synchronized intermittent manda-
tory ventilation, pressure support ventilation) should be
MANAGEMENT
replaced by controlled mechanical ventilation. An arterial
Ventilatory Support
blood-gas analysis should be conducted intraoperatively,
since end-tidal CO2 levels may be different from arterial Ventilatory support should be continued postoperatively,
levels. The current trend in ventilation which has im- with or without intubation, according to the patients
proved the outcome of patients with acute lung injury and needs. Face mask-administered CPAP in patients who
ARDS50 is a strategy of lung protective ventilation. This have adequate minute ventilation and intact sensorium
concept involves the minimization of tidal volumes and may be indicated. During this stage, it is also important
plateau pressures, while ensuring a PEEP higher than the to verify the consciousness level, adequate reflexes, and
lower inflection point on the pressure-volume curve. respiratory drive.
When providing low tidal volumes and low pressures
Fluid Management and Hemodynamic to a patient with ARDS who is breathing spontaneously,
Support inspiratory flow of the ventilator should satisfy the
patients inspiratory flow demand. In cases where high
Fluid management in cases of pulmonary edema has been demand with excessive negative pressures is created by
a subject of controversy. The most common approach the patient, high flow should be delivered to avoid negative
is aggressive therapy of filling pressures and preload, pressure pulmonary edema.
or HPiv reduction. This may result, however, in hemo-
dynamic instability and reduced vital organ perfusion. Noninvasive Ventilation
Intraoperatively, with potential blood loss, adequate fluid
resuscitation should be carried out, even at the expense Noninvasive ventilation for acute CPE has been shown
of worsening the edema. Also, acute postoperative re- to improve oxygenation, increase cardiac output, and
nal failure due to preload reduction will carry a grave reduce the work of breathing. Several studies have evalu-
prognosis, whereas ventilation/perfusion mismatch due ated the use of CPAP in acute CPE,52,53 and others have
to the edema can still be treated adequately with PEEP. evaluated the use of bilevel positive airway pressure.54,55
Another possible measure is to increase OPiv by infus- Collectively, the available data suggest that CPAP is ef-
ing colloid solutions, although this approach has been fective in terms of reducing the intubation rate, and
190 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
confirm the trend towards reduced mortality. Studies Any therapy that substantially improves oxygenation and
evaluating bilevel ventilation have variable conclusions, reduces both mean airway pressure and the duration of
some appearing to show significant benefits, whereas oth- mechanical ventilation may improve survival and neuro-
ers suggest significant disadvantages with this modality.53 logic recovery. Prone ventilation can improve oxygenation
One study compared both types of noninvasive ventilation without worsening the neurologic status.
with standard treatment and concluded that bilevel venti- A recent multicenter study comparing prone and
lation significantly reduced the intubation rate compared conventional ventilation concluded that although prone
with CPAP in patients with CPE.55 ventilation is effective in improving oxygenation, it did
not lead to an improvement in survival from acute lung
Noninvasive Pressure Support Ventilation injury and ARDS.58 Nevertheless, the general belief is that
prone ventilation is an appropriate therapy that benefits
When patients do not respond to conventional medical subgroups of patients, and may improve outcome if used
treatment for acute CPE (e.g., morphine, oxygen mask, early.
diuretics, and nitrates), ventilatory assistance is needed. Although prone ventilation is generally safe, the pres-
Noninvasive pressure support ventilation (NIPSV) has ence of traumatic brain injury is a relative contraindica-
already proven to be effective in the treatment of CPE.54 tion. Anecdotal evidence suggests that in such patients
However, despite the fact that some authors have reported with reduced intracranial compliance, prone position-
a significant improvement of clinical parameters after 15 ing can increase intracranial pressure. Although concerns
to 60 minutes,52,54 NIPSV may dangerously delay, in some about exacerbating intracranial hypertension remain, this
patients, unavoidable tracheal intubation and invasive risk must be weighed against the potential benefits of
mechanical ventilation. improved oxygenation and reduced mean intrathoracic
This ventilatory mode can be administered for a con- pressure. Whether prone ventilation is an appropriate
siderable length of time (e.g., from 2 to 24 hours) to therapy for NPE attributable to subarachnoid hemorrhage
patients already admitted to the ICU or to those admitted should be decided only in conjunction with intracranial
soon after the beginning of NIPSV.56 Accordingly, invasive pressure measurement.
respiratory support may be avoided in a large percentage
of patients. A study in patients with acute myocardial in-
farction showed that NIPSV may be used with reasonable
safety. Only two baseline conditionsmean arterial pres-
KEY POINTS
sure <95 mm Hg and a history of COPDsignificantly 1. Although relatively uncommon, perioperative pul-
predicted the failure of NIPSV.48 The latter condition can monary edema can be associated with high mortality.
at least be partially explained by the sum of long-term and 2. The cardinal symptoms are dyspnea, tachypnea, and
short-term increases in the work of breathing, leading to signs of increased respiratory effort (e.g., accessory
excessive respiratory workload that could not be rapidly respiratory muscle use, anxiety).
managed by NIPSV alone. 3. Although the two principal mechanisms are increased
hydrostatic pressure and permeability, the patho-
Prone Ventilation physiology is more complex with a mixture of both
components.
Mechanical ventilation in the prone position improves 4. The distribution of edema on the radiograph is
oxygenation in 60% of patients with acute lung injury determined by several factors such as gravitational
or ARDS. These patients typically have extensive collapse- forces, hydrostatic pressure, and preexisting lung
consolidation of dorsal lung units when supine, partly pathology.
due to the weight of the overlying heart and the high 5. The duration of pulmonary edema is largely deter-
pleural pressures dorsally.57 When a patient is turned mined by the etiology. Neurogenic or postobstructive
prone, re-aeration of these collapsed lung units occurs. pulmonary edema is transient, whereas diffuse alveo-
Although the newly dependent, ventral lung experiences a lar injury and ARDS can last weeks.
degree of collapse-consolidation, it is much less extensive 6. The three mainstays of treatment for pulmonary
than that seen dorsally. Regional lung perfusion is not edema are (a) normalization of oxygenation and
affected greatly by the change in position, and, as a ventilation; (b) reduction of excessive extravascular
result, ventilation-perfusion matching and oxygenation lung water; and (c) identification and treatment of
are improved. This allows a reduction in the inspired the underlying disease.
oxygen concentration and mean airway pressure, possibly
improving the chance of a successful outcome. REFERENCES
In cases of mixed pulmonary edema such as NPE,
there is a combination of cardiogenic and noncardiogenic 1. Cooperman LH, Price HL, Pulmonary edema in the operative
and postoperative period: A review of 40 cases. Ann Surg.
pathophysiology. As well as being a direct threat to life,
1970;172:883.
the severe hypoxia that results from NPE may worsen the 2. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial
neurologic injury. Positive pressure ventilation and high index of cardiac risk in noncardiac surgical procedures.
levels of PEEP are frequently required, which may worsen N Engl J Med. 1977;297:845.
cerebral perfusion (and therefore outcome) by reducing 3. Arieff AL. Fatal postoperative pulmonary edema: Pathgenesis
cardiac output and impeding cerebral venous drainage. and literature review. Chest. 1999;115:1371.
C H A P T E R 12 / P U L M O N A R Y E D E M A 191
4. West JB, Mathieu-Costello O. Vulnerability of pulmonary 25. Duberstein JL, Kaufman DM. A clinical study of an epi-
capillaries in heart disease. Circulation. 1995;92:622. demic of heroin intoxication and heroin-induced pulmonary
5. Haworth SG, Hall SM, Patel M. Peripheral pulmonary edema. Am J Med. 1971;51:704.
vascular and airway abnormalities in adolescents with 26. Snyder SH. Opiate receptors in the brain. N Engl J Med.
rheumatic mitral stenosis. Int J Cardiol. 1988;18:405. 1977;296:266.
6. Starling EH. On the absorption of fluid from the connective 27. Ell SR. Neurogenic pulmonary edema: A review of the
tissue spaces. J Physiol. 1896;19:312. literature and a perspective. Invest Radiol. 1991;26:499.
7. Ingram RH Jr, Braunwald E. Dyspnea and pulmonary 28. Levinson RM, Shure D, Moser KM. Reperfusion pulmonary
edema. In: Fauci AS, Braunwald E, Isselbacher KJ, et al, edema after pulmonary artery thromboendarterectomy. Am
eds. Harrisons principles of internal medicine, 14th ed. New Rev Respir Dis. 1986;134:1241.
York: McGraw-Hill; 1998:190. 29. Miller WT Jr, Osiason AW, Langlotz CP, et al. Reperfusion
8. Bachofen H, Weibel ER. Structure-function relationships pulmonary edema after thromboendarterectomy: Radio-
in the pathogenesis of pulmonary edema. In: Weir AK, graphic patterns of disease. J Thorac Imaging. 1998;13:178.
Reeves JT, eds. Pulmonary edema. American Heart Associa- 30. Murray JG, McAdams HP, Erasmus JJ, et al. Complica-
tion Monograph Series. Armonk: Futura Publishing; 1998:17. tions of lung transplantation: Radiologic findings. Am J
9. Fein A, Grossman RF, Jones JG, et al. The value of edema Roentgenol. 1996;166:1405.
fluid protein measurement in patients with pulmonary 31. Engeler CE. Heart-lung and lung transplantation. Radiol Clin
edema. Am J Med. 1979;67:32. North Am. 1995;33:559.
10. Sprung CL, Rackow EC, Fein IA, et al. The spectrum of 32. OBrien JD, Ettinger NA. Pulmonary complications of liver
pulmonary edema: Differentiation of cardiogenic, interme- transplantation. Clin Chest Med. 1996;17:99.
diate, and noncardiogenic forms of pulmonary edema. Am 33. Yost CS, Matthay MA, Gropper MA. Etiology of acute
Rev Respir Dis. 1981;124:718. pulmonary edema during liver transplantation. A series of
11. West JB, Tsukimoto K, Mathieu-Costello O, et al. Stress cases with analysis of the edema fluid. Chest. 2001;119:219.
failure in pulmonary capillaries. J Appl Physiol. 1991;70: 34. Looney MR, Gropper MA, Matthay MA. Transfusion-related
1731. acute lung injury a review. Chest. 2004;126:249.
12. Pietra GG, Szidon JP, Leventhal MM, et al. Hemoglobin 35. Barnard RD. Indiscriminate transfusion: a critique of case
as a tracer in hemodynamic pulmonary edema. Science. reports illustrating hypersensitivity reactions. N Y State J
1969;166:1643. Med. 1951;51:2399.
13. Wodopia R, Ko HS, Billian J, et al. Hypoxia decreases 36. Reese EP, McCullough JJ, Craddock PR Jr. An adverse
proteins involved in epthelial electrolyte transport in A549 pulmonary reaction to cryoprecipitate in a hemophiliac.
cells and rat lung. Am J Physiol Lung Cell Mol Physiol. Transfusion. 1975;15:583.
2000;279:L1110. 37. Kopko PM, Marshall CS, MacKenzie MR, et al. Transfusion
14. Vivona ML, Matthay MA, Chabaud MB, et al. Hypoxia related acute lung injury: Report of a clinical look-back
reduces alveolar epithelial sodium and fluid transport in investigation. JAMA. 2002;287:1968.
rats: Reversal by beta-adrenergic agonist treatment. Am J 38. Mahfood S, Hix WR, Aaron BL, et al. Reexpansion pul-
Respir Cell Mol Biol. 2001;25:554. monary edema. Ann Thorac Surg. 1988;45:340.
15. Sartori C, Allemann Y, Duplain H, et al. Salmeterol for the 39. Tarver RD, Broderick LS, Conces DJ Jr. Reexpansion
prevention of high-altitude pulmonary edema. N Engl J Med. pulmonary edema. J Thorac Imaging. 1996;11:198.
2002;346:1631. 40. Turnage WS, Lunn JJ. Postpneumonectomy pulmonary
16. Carpenter TD, Reeves JT, Durmowicz AG. Viral respiratory edema: A retrospective analysis of associated variables.
infection increases susceptibility of young rats to hypoxia- Chest. 1993;103:1646.
induced pulmonary edema. J Appl Physiol. 1998;84:1048. 41. Waller DA, Keavey P, Woofine L, et al. Pulmonary endothelial
17. Durmowicz AG, Nooordeweir E, Nicholas R, et al. Inflam- permeability changes after major lung resection. Ann Thorac
matory processes may predispose children to develop high Surg. 1996;61:1435.
altitude pulmonary edema. J Pediatr. 1997;130:838. 42. van der Werff YD, van der Houwen HK, Heijmans PJM,
18. Staub NC. New concepts about the pathophysiology of et al. Postpneumonectomy pulmonary edema: A retrospec-
pulmonary edema. J Thorac Imaging. 1988;3:8. tive analysis of incidence and possible risk factors. Chest.
19. Schnyder PA, Sarraj AM, Duvoisin BE, et al. Pulmonary 1997;111:1278.
edema associated with mitral regurgitation: Prevalence of 43. Frim DM, Wollman L, Evans AB, et al. Acute pulmonary
predominant involvement of the right upper lobe. Am J edema after low-level air embolism during craniotomy: Case
Roentgenol. 1993;161:33. report. J Neurosurg. 1996;85:937.
20. Gudinchet F, Rodoni P, Sarraj A, et al. Pulmonary oedema 44. Rasanen J, Heikkila J, Downs J, et al. Continuous positive air-
associated with mitral regurgitation: Prevalence of predomi- way pressure by face mask in acute cardiogenic pulmonary
nant right upper lobe involvement in children. Pediatr Radiol. edema. Am J Cardiol. 1985;55:296.
1998;28:260. 45. Kelly CA, Newby DE, McDonagh TA, et al. Randomised
21. Willms D, Shure D. Pulmonary edema due to upper airway controlled trial of continuous positive airway pressure and
obstruction in adults. Chest. 1988;94:1090. standard oxygen therapy in acute pulmonary oedema: Effects
22. Stalcup SA, Mellins RB. Mechanical forces producing pul- on plasma brain natriuretic peptide concentrations. Eur
monary edema in acute asthma. N Engl J Med. 1977;297:592. Heart J. 2002;23:1379.
23. Remy-Jardin M, Remy J, Deschildre F, et al. Diagno- 46. Kelly C, Newby DE, Boon NA, et al. Support ventilation
sis of pulmonary embolism with spiral CT: Comparison versus conventional oxygen. Lancet. 2001;357:1126.
with pulmonary angiography and scintigraphy. Radiology. 47. Pang D, Keenan SP, Cook DJ, et al. The effect of positive
1996;200:699. airway support on mortality and the need for intubation in
24. Gattinoni L, Pelosi P, Suter PM, et al. Acute respiratory cardiogenic pulmonary edema: A systematic review. Chest.
distress syndrome caused by pulmonary and extrapulmonary 1998;114:1185.
disease: Different syndromes? Am J Respir Crit Care Med. 48. Crane SD, Elliott MW, Gilligan P, et al. Randomised con-
1998;158:3. trolled comparison of continuous positive airways pressure,
192 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
bilevel non-invasive ventilation, and standard treatment in pressure in acute pulmonary edema. Crit Care Med. 1997;
emergency department patients with acute cardiogenic pul- 25:620.
monary oedema. Emerg Med J. 2004;21:155. 54. Levitt MA. A prospective, randomised trial of BiPAP in
49. Marks JD, Schapera A, Kraemer ARW. Pressure and severe acute congestive heart failure. J Emerg Med. 2001;
flow limitations of anesthesia ventilators. Anesthesiology. 21:363.
1989;71:403. 55. Park M, Lorenzi-Filho G, Feltrim MI, et al. Oxygen therapy,
50. The Acute Respiratory Distress Syndrome Network. Ventila- continuous positive airway pressure, or noninvasive bilevel
tion with lower tidal volumes as compared with traditional positive pressure ventilation in the treatment of acute
tidal volumes for acute lung injury and acute respiratory cardiogenic pulmonary edema. Arq Bras Cardiol. 2001;
distress syndrome. N Engl J Med. 2000;342:1301. 76:221.
51. Humprey H, Hall J, Sznajder I, et al. Improved survival in 56. Lin M, Yang YF, Chiang HT, et al. Reappraisal of contin-
ARDS patients associated with a reduction in pulmonary uous positive airway pressure therapy in acute cardiogenic
capillary wedge pressure. Chest. 1990;97:1176. pulmonary edema. Chest. 1995;107:1379.
52. Baratz DM, Westbrook PR, Shah PK, et al. Effect of nasal 57. Fletcher SJ, Atkinson JD. Use of prone ventilation in
CPAP on cardiac output and oxygen delivery in patients with neurogenic pulmonary oedema. Br J Anaesth. 2003;90:238.
congestive heart failure. Chest. 1992;103:1397. 58. Gattinoni L, Tognoni G, Pesenti A, et al. Effect of prone
53. Mehta S, Jay GD, Woolard RH, et al. Randomized, prospec- positioning on the survival of patients with acute respiratory
tive trial of bilevel versus continuous positive airway failure. N Engl J Med. 2001;23:568.
CHAPTER PULMONARY EMBOLISM
A
failure and diabetes mellitus presents for
a hip arthroplasty after fracturing her hip
during a fall. The patient undergoes surgical
repair without complications and is admit- What Conditions Predispose
ted to a telemetry bed in the hospital for
postoperative care. The patient remains in the hospital to Venous Thromboembolism?
until postoperative day 5 due to difficulties in titrating the
patients medications for heart failure and diabetes. On
postoperative day 5, the physician caring for the patient is ETIOLOGY
called to emergently evaluate the patient who is exhibiting
tachycardia, hypotension, and tachypnea. An ECG shows Thromboembolic PE are blood clots or thrombi that de-
ST depression in the right-sided leads, and an echocar- velop in the deep veins, typically of the legs or pelvis,
diogram is emergently performed, which shows severe and embolize to the pulmonary arterial system. The spec-
right heart dilation with new severe tricuspid regurgita- trum of patient presentations ranges from completely
tion. The patient is presumed to have a high likelihood asymptomatic to fulminant cardiovascular collapse. The
for a pulmonary embolism (PE) and is immediately an- amount of clot and the patients physiologic reserve are
ticoagulated with heparin. A stat computed tomography the primary factors that dictate the severity of the symp-
(CT) angiogram confirms the diagnosis, and the patient toms and signs seen. A large clot burden can obstruct
remains on anticoagulation for 6 months. the pulmonary artery and cause right heart strain or fail-
ure, as well as severe ventilatory/perfusion mismatch. In
addition, PE is often accompanied by the release of va-
soactive mediators (thromboxane and serotonin), which
What Are the Facts Concerning also increase pulmonary artery pressures. Intraoperative
thromboembolic PE is unusual: While there are multiple
Pulmonary Embolism? case reports of acute massive PE in patients with occult
lower extremity deep vein thromboses (DVT) who under-
PE is one of the most dreaded complications in the peri- went intraoperative manipulation of the lower extremity,1
operative period. Many patients are at increased risk for thromboembolic pulmonary emboli are far more likely to
some form of PE by virtue of their procedure or preexisting present in the postoperative period.
illnesses. There are multiple types of PE, including venous The physiologic factors that increase the risk of devel-
thromboembolic (blood clot), venous air, fat or debris, and oping DVT were originally proposed by Rudolf Virchow in
amniotic fluid. While the source of the emboli and timing 1860 and include venous stasis, damage to the vessel wall,
of symptoms depends on the type of PE, there are many and hypercoagulability. In the surgical patient, venous
similarities in the morbidity seen secondary to PE. Pa- stasis occurs upon the induction of general anesthesia
tients who develop PE typically develop some degree of because of the vasodilatory effects of the drugs adminis-
ventilation/perfusion mismatch and increased pulmonary tered and the immobility of extremities under anesthesia.
arterial pressures, which can lead to right heart strain. A hypercoagulable state is then induced by the stress re-
Because of the high potential for mortality with each of sponse resulting from the surgical intervention.2,3 Finally,
the different types of pulmonary embolic disorders, this patients may experience direct damage to the venous en-
diagnosis should be considered by the anesthesiologist dothelium during surgery. All these conditions can lead to
193
194 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
the formation of clot in the calf veins, which propagates Computed Tomography
proximally to larger veins in the leg (popliteal, femoral,
or iliac). Thrombus in these large proximal leg veins are Helical CT has replaced the ventilation/perfusion scan
more likely to cause physiologically significant PE than (V/Q scan) as the study of choice in many hospitals for
emboli from the smaller calf veins. PE evaluation4 because of the speed of the study and the
ability to concurrently evaluate potential embolic sources
in the legs or pelvis. The results of studies that have
evaluated the helical CT have shown sensitivities up to
DIAGNOSIS 90% with single detector CT scans.8 Multidetector row
CT scans have an increased sensitivity for subsegmental
The diagnosis of PE is confirmed by imaging studies in PE9 but, in postoperative patients, should still be used in
patients with clinical signs and symptoms. The gold stan- conjunction with pretest probability and ultrasonography
dard for diagnoses is a pulmonary angiogram. Because of the lower extremities.10
it is an invasive procedure, and its availability is limited The PIOPED II investigators have recently investi-
in many hospitalsand reasonable alternatives have been gated the use of clinical pretest probability in conjunction
developedthe angiogram is not typically employed. Pa- with either CT angiogram or combined CT angiogram
tients are evaluated by history and physical examination. and CT venography.11 These investigators concluded that
Signs and symptoms aid in formulating a pretest proba- the CT angiogram combined with CT venography has a
bility of the likelihood of PE. Symptoms include dyspnea, higher sensitivity for venous thromboembolism (VTE). In
chest pain, cough, and blood-tinged sputum. Signs in- addition, the pretest probability should still be used in the
clude fever, tachycardia, tachypnea, and coarse breath diagnostic algorithm for PE. The PIOPED II investigators
sounds. Auscultation may yield a new fourth heart sound did note that a negative CT angiogram does not rule out
or accentuation of the pulmonic component of the second a subsegmental PE; studies suggest that it is safe to with-
heart sound. hold anticoagulation in patients with low or intermediate
pretest probability.9,12 The PIOPED pretest probability is
based on the Wells criteria, which uses a scoring system
Laboratory Testing based on signs and symptoms that include DVT, tachycar-
dia, immobilization, and recent surgery among others.7
Laboratory tests (including arterial blood gases) are All postsurgical patients would then, by definition, be
not typically useful for diagnosing or ruling out PE considered at least intermediate pretest probability, and
in the perioperative period. Patients with a normal thus a physicians global judgement should be used in the
PACO2 may still have a PE,4 and patients with a high postsurgical patient. For diagnosing a PE with a CT scan
FiO2 /PaO2 gradient could have one of many diagnoses. (see Fig. 13.1), a pretest probability is assessed, and only
D-dimer, a degradation product of cross-linked fibrin, patients with a high pretest probability and a negative
has been studied for the diagnosis of PE. Two common CT scan should have compression ultrasonography of the
D-dimer tests used are the enzyme-linked immunosorbent lower extremities to rule out VTE.
assays and the latex agglutination assays. Neither test is
helpful in the postoperative period because the D-dimer
is very likely to be positive even in the absence of EPIDEMIOLOGY
PE.5
In addition to the surgical and anesthetic factors, multiple
other preexisting conditions place patients at a higher
Imaging Studies risk for developing a DVT and potential PE in the
perioperative period13 (see Table 13.1). While the lower
Imaging studies are the modality of choice for diagnosing extremities are more frequently the cause of pulmonary
PE. The V/Q scan is useful for patients with renal emboli, patients with upper extremity DVT had a 9%
insufficiency, since nephrotoxic, intravenous contrast is incidence of symptomatic PE in one study.14 Upper
not used in nuclear medicine scans. An algorithm designed extremity venous thrombosis accounts for about 10% of
by the Prospective Investigation of Pulmonary Embolism DVTs and usually occurs as a result of central venous
Diagnosis (PIOPED) investigators has simplified the catheters or in patients with cancer. Rarely, DVT can
diagnostic approach to these patients.6 By assessing occur spontaneously in the upper extremity, known as
the clinical signs and symptoms, the patient is given Paget-Schroetter syndrome. Clinically, patients present
a pretest probability (low, medium, or high) of having with swelling, erythema, and arm pain affected by the
a PE before performing the V/Q scan. The results of a DVT. Ninety percent of pulmonary emboli emanate from
V/Q scan range from a normal to high probability of extremity (upper and lower) DVT, with the remaining
PE. Combining the pretest probability with the results pulmonary emboli arising from pelvic veins, renal veins,
from a V/Q scan help successfully diagnose or exclude inferior vena cava (IVC), and the heart.
PE.7 Previously, patients with a low probability scan Perioperatively, patients are at higher risk for a DVT
(indeterminate) would undergo a pulmonary angiogram with prolonged duration of surgery and immobilization,15
for further workup, but now, their chances of having a PE preexisting thrombophilia, and malignancy. One large
are considered minimal. study used an adminstrative database to estimate the
C H A P T E R 13 / P U L M O N A R Y E M B O L I S M 195
Assess pretest probability and V/Q scan
Bilateral
Diagnose/treat No PE/no
Positive compression
PE /VTE treatment
ultrasound (CUS)
Negative
Positive Negative
No PE/no
Diagnose/treat PE
treatment
FIGURE 13.1 Diagnostic algorithm for patients with suspected pulmonary embolism.
PE, pulmonary embolism; VTE, venous thromboembolism; CUS, compression ultrasound; CT,
computed tomography.
196 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 13.1 Risk Factors for Deep Venous Thrombosis TABLE 13.2 Surgical Procedure Related to Risk
of Venous Thromboembolism
Advancing age
Obesity Surgical Procedure Incidence Vte (%)
Previous venous thromboembolisma NEUROSURGERY
Traumaa
Neoplasma Excision/destruction/biopsy brain 2.3
Respiratory failure Spinal cord surgery 0.5
Infection HEAD AND NECK
Inflammatory bowel disease
Sinus surgery 0.2
Antiphospholipid syndromea
Thyroid or parathyroid surgery 0.1
Dyslipoproteinemia
Nephrotic syndrome CARDIAC OR THORACIC
Paroxysmal nocturnal hemoglobinuria
Coronary artery bypass grafting 1.1
Myeloproliferative diseases
Valve replacement 0.5
Behcets syndrome
Varicose veins VASCULAR
Superficial vein thrombosis Abdominal aortic surgery 1.7
Congenital venous malformation Head and neck endartectomy 0.2
Long distance travel
Prolonged bed rest GASTROINTESTINAL
Immobilization Splenectomy 1.6
Limb paresis Excision of small bowel 1.5
Pregnancy Exploratory laparotomy 0.7
Oral contraceptives
Hormone replacement therapy UROLOGIC
Heparin-induced thrombocytopeniaa Nephrectomy 0.4
Chemotherapy
Tamoxifen GYNECOLOGIC SURGERY
Thalidomide Total abdominal hysterectomy 0.3
Antipsychotics
Central venous cathetera ORTHOPEDIC
Vena cava filter Total hip arthroplasty 2.4
Intravenous drug abuse Total knee arthroplasty 1.9
Shoulder arthroplasty 0.5
a Signifies major risk factor.
Data from: Kyrle PA, Eichinger S. Deep vein thrombosis. Lancet VTE, venous thromboembolism.
2005;365:1163. Data from: White RH, Zhou H, Romano PS. Incidence of symptomatic
venous thromboembolism after different elective or urgent surgical
procedures. Thromb.Haemost. 2003;90:446.
preventative therapy for postoperative VTE should be
based on the patients risk for developing VTE17 as shown
in Table 13.3. to pneumatic compression devices include a diagnosis of
DVT in the extremity to be compressed. Several studies
Antiplatelets support the use of pneumatic compression devices in the
general surgical patient.2023
Aspirin and other antiplatelet agents are not as effective
as fractionated or unfractionated heparin8 for DVT Regional Anesthesia
prophylaxis. There is a risk of significant bleeding with
aspirin, and recent studies supporting its use are limited Regional anesthesia has been studied as a means for
by sample size and methods of screening for DVT.18 lowering postoperative VTE risk. Whereas epidural anes-
thesia does lower the incidence of intraoperative VTE,24,25
Compression Devices epidural analgesia does not appear to decrease this risk.
When considering the choice of epidural or spinal anes-
In addition to anticoagulants, pneumatic compression de- thesia, the patients coagulation status must be evaluated.
vices and elastic stockings may be used to prevent DVT. Patients who are anticoagulated when a neuraxial (spinal
These are especially useful in the patient who is considered or epidural) procedure is initiated are at risk for perispinal
a bleeding risk and in whom anticoagulants are con- hematoma. This rare but serious complication of neurax-
traindicated. Pneumatic compression devices enhance ve- ial procedures is preventable by a working knowledge
nous blood flow in the lower extremities and reduce levels of when to discontinue and restart anticoagulants26 (see
of plasminogen activator inhibitor-1.19 Contraindications Table 13.4).
C H A P T E R 13 / P U L M O N A R Y E M B O L I S M 197
TABLE 13.3 Preventive Therapy Related to Risk of TABLE 13.4 Timing of Anticoagulant Administration in
Developing Postoperative Venous Thromboembolism Relation to Neuraxial Procedures
TABLE 13.5 Intravenous Unfractionated Heparin TABLE 13.6 Contraindications to Thrombolytic Therapy
Treatment Protocola
Absolute
Calculations based on total body weight in Cerebrovascular accident within last 2 months
kilograms (kg) Intracranial procedure within last 2 months
Administer heparin 80 U/kg IV bolus, followed by Known intracranial tumor
infusion at 18 U/kg/h
Relative
Stat activated partial thromboplastin time (aPTT) 6 h
after heparin bolus <10 days postpartum
Infusion adjusted on the basis of the scale below: Surgery within last 7 days
aPTT <35 80 U/kg bolus: Increase infusion by Recent organ biopsy
4 U/kg/h Recent internal trauma
aPTT 3545 40 U/kg bolus: Increase infusion by Bleeding diathesis
2 U/kg/h Recent thoracentesis or paracentesis
aPTT 4670 No change Epidural or lumbar puncture
aPTT 7190 Decrease infusion rate by 2 U/kg/h Recent puncture of noncompressible vessel
aPTT >90 Hold heparin for 1 h, reduce infusion Uncontrolled hypertension (systolic >200 mm Hg or
rate by 3 U/kg/h diastolic >110 mm Hg)
b Institution of aPTT should correspond to plasma Hemorrhagic retinopathy
heparin levels from 0.3 to 0.7 IU/mL anti-Xa activity
by the amidolytic assay
listed in Table 13.6. Recent surgery is included because
a After each dose change, order aPTT and adjust infusion by scale thrombolytics will lyse clots at the surgical sites.
above. When two consecutive aPTTs are therapeutic, may order aPTT Because of the danger of thrombolytics, surgical
daily. embolectomy has been studied as an alternative therapy
b Raschke RA, Reilly BM, Guidry JR, et al. The weight-based heparin
for massive PE. Patients with life-threatening PEs may
dosing nomogram compared with a standard care nomogram.
be placed on extracorporeal membrane oxygenation for
A randomized controlled trial. Ann Intern Med. 1993;119:874881.
IV, intravenous; aPTT, activated partial thromboplastin time. stabilization, and taken to the operating room for open
Data in table from: Buller HR, Agnelli G, Hull RD, et al. Antithrom- thrombus extraction. Surgical embolectomy may have
botic therapy for venous thromboembolic disease: the Seventh lower mortality rates than thrombolyses;35 centers with
ACCP Conference on Antithrombotic and Thrombolytic Therapy. experience have reported acceptable outcomes.36 This
Chest.2004;126:401S. approach allows placement of an IVC filter to prevent
recurrent thromboembolism.
IVC filter placement is recommended for patients with
or subcutaneous LMWH. Subtherapeutic anticoagulation recurrent emboli despite therapeutic anticoagulation, for
increases the risk of recurrent thromboembolism, and those with a contraindication to anticoagulation, and
at least one study supports a weight-based nomogram following surgical embolectomy. The placement of IVC
to achieve therapeutic levels within 24 hours31 (see filters in patients with a high clot burden to prevent
Table 13.5). further emboli and hemodynamic compromise has not
Patients who cannot be anticoagulated should have been rigorously studied.
an IVC filter placed as soon as possible to prevent further
embolization. Patients with a large clot burden may also
be considered for IVC filter placement, although this has
not been rigorously studied. How Is Venous Air Embolism
Patients with hemodynamic instability or life-
threatening hypoxemia due to massive PE may be appro- Detected and Managed?
priate candidates for thrombolytic therapy32 or surgical
embolectomy to decrease clot burden. The thrombolytics
currently available for use in the United States are the PROBLEM ANALYSIS
tissue-plasminogen activator (TPA), streptokinase, and
urokinase, all of which convert plasminogen to plasmin, Etiology
which in turn breaks down fibrin and promotes clot lysis.
Clinical trials of thrombolytic therapy in patients who are Venous air embolism (VAE) typically occurs when air
hemodynamically stable do not demonstrate a mortality enters the venous circulation through an incised or can-
benefit from thrombolysis,33 but have demonstrated a sig- nulated vein. The air eventually travels through the right
nificant risk of intracranial hemorrhage34 and bleeding at heart and into the pulmonary artery; however, it may en-
incision sites. Thrombolysis may be the only option at in- ter the arterial circulation through an atrial or ventricular
stitutions without the ability to perform surgical embolec- septal defect. The introduction of small amounts of air
tomy in a patient with complete cardiovascular collapse through an intravenous catheter is a relatively frequent
due to massive PE. Contraindications to thrombolytics are occurrence during injections or infusions of medication
C H A P T E R 13 / P U L M O N A R Y E M B O L I S M 199
TABLE 13.7 Surgeries and Procedures at Risk of Venous entering at a rapid rate can cause elevated pulmonary pres-
Air Thromboembolism sures, hypotension, and eventual cardiac collapse.38 In the
awake, spontaneously ventilating patient, signs and symp-
Transurethral or radical prostatectomy toms include dyspnea, cyanosis, arrhythmias, hypoten-
Sitting craniotomy sion, increased central venous and pulmonary artery pres-
Spinal surgery including laminectomy sure, decreased cardiac output, mill wheel murmur, and
Laparoscopic surgery cardiac collapse. Physiologic monitors may show signs
Endoscopic bowel procedures of cardiac ischemia, and an acute decrease in end-tidal
Hip replacement CO2 . Patients having high-risk procedures for VAE (e.g.,
Arthroscopic joint procedures sitting craniotomy) should have continuous monitoring
Video-assisted thoracoscopic procedures (see Table 13.8) to allow rapid diagnosis and treatment.
Chest trauma
Any procedure with human or pump delivered infu- Epidemiology
sions (e.g., angiography)
The risk of VAE is dependent on the type of surgical
procedure and the elevation of the vein exposed to the air.
The overall incidence of VAE in sitting neurosurgical
and is typically benign when there is no right-to-left in- procedures appears to be as high as 76%, according
tracardiac shunt. However, great care must be practiced to one study of a group of patients with continuous
with intravenous catheters open to the atmosphere: It has transesophageal echocardiography monitoring.39
been demonstrated that 100 mL of air per second can flow Patient outcome following VAE depends primarily on
into a 14-gauge intravenous catheter with a 5-cm pressure the volume of air that enters the pulmonary circulation
gradient. A variety of surgeries and invasive procedures and, when relevant, the amount of air that crosses into
are associated with an increased risk of VAE; these in- the systemic arterial system and enters the brain. Large
clude procedures in which noncollapsible veins above the amounts of venous air can create an airlock in the right
level of the heart are opened, insufflation of gas into body ventricle and right atrium, which increases right heart
cavities, and venous cannulation. Examples are listed in pressures and may push air through a patent foramen
Table 13.7. ovale into the arterial circulation. Complications are listed
During open surgical procedures, noncollapsible in Table 13.9.
veins that are incised and exposed to subatmospheric
pressure (usually by being elevated above the level of
the heart) increases the likelihood of entrainment of air.
Prevention and Treatment
Noncollapsible veins are held open by surrounding struc- When cannulating veins, it is important to maintain a pos-
tures and do not collapse when venous pressure drops. itive venous-to-atmospheric pressure gradient by keeping
Examples of noncollapsible veins include dural sinuses the vein elevation below the heart. This will maintain a
and prostatic veins. positive pressure in the vein and minimize the entrain-
ment of air. When placing central lines in the internal
Diagnosis jugular or subclavian vein, the patient should be placed
in a Trendelenburg position so that the cannulation site is
The clinical manifestations of VAE depend on the volume below the level of the heart. Similarly, the patient should
of air and the rapidity of entry. A slow entrainment of a be recumbent, and the cannulation site should be com-
large volume of air has been shown to be well-tolerated in pressed during removal of large bore catheters. There are
a dog model of VAE.37 Small amounts of air are usually case reports describing massive air embolism when a cen-
benign in a healthy patient if isolated to the right heart and tral line is removed from a sitting patient.40 It is also im-
pulmonary circulation; but large amounts of air (>50 mL) portant to monitor the stopcock position and intravenous
TABLE 13.8 Monitoring for Patients at High Risk for Venous Air Embolism
Monitors Advantage Disadvantage
Observation (observe surgical field) Noninvasive Not sensitive
End-tidal carbon dioxide Sensitive Nonspecific
Noninvasive
Already widely used
End-tidal nitrogen Specific Hypotension lowers sensitivity
Doppler ultrasound Sensitive Not quantitative
Easy detection of signal Difficult placement in prone, obese, or chest deformity
Multiorifice right atrial catheter Quantitative Not good for continuous screening
Transephoageal echocardiography Sensitive Expensive
Not widely available
200 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 13.9 Complications of Patient Outcome potential to impede cardiac output. In addition, PEEP may
Following Venous Air Embolism increase in the presence of a positive gradient between the
right and left atria, and thereby convert a VAE to a para-
Organ System Complication doxical air embolus through a patent foramen ovale.42
Cardiovascular Hypotension
Arrhythmias
Myocardial ischemia
Right heart failure
What Procedures Place Patients
Cardiac arrest at Risk for Debris Embolism,
Pulmonary Hypoxemia and How Are They Managed?
Pulmonary hypertension
Pulmonary edema
Hypercarbia
V/Q mismatch
ETIOLOGY
Central nervous system Stroke Fat embolism to the venous and pulmonary circulation is
(for air that becomes Brain edema a common occurrence in orthopedic surgical procedures
systemic) and patients who sustain traumatic injury. The mecha-
nism appears to be a disruption of the venous system at
the surgical or injury site, allowing fat from bone marrow
line connection to prevent entry of air into lines. Most or adipose tissue to enter the venous system. In orthope-
modern intravenous pumps now have air detection and dic procedures, intraosseous pressure may be elevated by
alarm systems to alert the practitioner to air in infusions. placement of rod or nails in bones. Fat embolism is most
In surgical cases deemed to be at high risk for VAE commonly seen in hip replacement, knee replacement,
due to surgical exposure of veins, appropriate monitors and intramedullary nailing of the shaft of bones.43 The
should be used, including at least one device sensitive sequela of the emboli can range from a benign course to
enough to detect VAE, such as a precordial Doppler. fulminant pulmonary failure with cardiovascular collapse.
Preventive measures include minimizing the degree of One theory holds that the fat embolism syndrome (FES)
elevation of the surgical site relative to the heart, keeping is not due to mechanical obstruction of the pulmonary
the patient euvolemic to maintain central venous pressure, arterial circulation, but rather because of an immune re-
and avoidance of medicines that increase venous capacity action caused by the breakdown of the embolic fat to free
(e.g., nitroglycerin). The increased awareness and the use fatty acids.44 These fatty acids may cause the release of
of sensitive VAE monitors allows for rapid treatment and toxic intermediaries that promote an inflammatory cas-
is probably the reason for the low mortality rate associated cade resulting in pulmonary and systemic inflammation.
with VAE.
Once the diagnosis is made, treatment should be rapid
(see Table 13.10). DIAGNOSIS
Positioning the patient in the left lateral decubitus
position has been reported.41 This is done to keep the Fat embolism may not be clinically recognized in most
intraventricular air from entering the pulmonary artery patients because only a few patients develop signs and
because the air rises to the nondependent portion of the symptoms. The classic triad of fatty embolism syndrome
ventricle. This position should be used cautiously because is hypoxemia, neurologic abnormalities, and a petechial
chest compressions may be necessary if the patient deteri- rash that occurs 12 to 72 hours after the initial trauma
orates. Positive end-expiratory pressure (PEEP) has been or instrumentation that delivers the fat embolism to the
proposed as a way of increasing right atrial pressures and venous circulation. Signs and symptoms and diagnostic
decreasing VAE. PEEP may also increase cerebral venous findings of FES are listed in Table 13.11.
pressure during seated craniotomies, thereby decreasing
the likelihood of VAE. One disadvantage of PEEP is its
EPIDEMIOLOGY
TABLE 13.10 Treatment of Venous Air Embolism Some degree of fat embolism occurs in up to 90% of
all long bone fractures and orthopedic procedures that
Lower the site of air entry, below the heart if possible instrument bone marrow.45 FES occurs in 0.5% to 2.0%
Flood the field with saline of patients with long bone fractures, and in up to 10% of
Compression of proximal vein (internal jugular vein in patients with multiple long bone fractures or concomitant
the sitting case) pelvic fractures.46 Mortality from FES in retrospective
Aspiration of the right atrial catheter if present studies ranges from 1%47 to 20%.48 The wide range
Discontinue use of nitrous oxide mortality is due to the variability in preexisting comorbid
Intravenous saline bolus conditions of the different patient groups. Many patients
Initiate vasopressor support if necessary with FES are young, multiple trauma patients or elderly
Ionotropic support may help clear the air bubble patients with chronic medical problems and long bone
fractures.
C H A P T E R 13 / P U L M O N A R Y E M B O L I S M 201
TABLE 13.11 Fat Embolism Syndrome: Signs products (squamous cells, lanugo, etc.) is the mechanism
and Symptoms, and Diagnosis of the cardiovascular and pulmonary collapse seen in
this disorder. Fetal squamous cells are thought to be
Signs and Symptoms pathognemonic of AFES but have been found in women
with and without AFES.51 The syndrome has been
Hypoxia-associated tachypnea and dyspnea postulated to be secondary to an immune response to
CNS depression: lethargy, confusion, seizures, focal amniotic fluid contents or an immune reaction stimulated
deficits (in paradoxical fat embolism) by leukotrienes or arachidonic acid within the fluid. This
Petechial rash: head, neck, torso, axilla immune reaction leads to profound multiorgan failure.
Fever A high percentage of patients experience respiratory
Tachycardia failure requiring intubation and mechanical ventilation,
Retinal fat emboli cardiogenic and vasoplegic shock requiring pressors, and
Diagnostic Findings diffuse intravascular coagulation. It has been reported to
occur as early as 20 weeks gestation but is more likely
Lipiduriaurine fat stain
to occur during labor and delivery or in the 48-hour
Fat in alveolifound in bronchoalveolar lavage
postpartum period. There are also reports of AFES after
Fat in bloodmay be found in aspirate of pulmonary
abortions52 and amniocentesis.53
blood or systemic blood
Fat in right ventriclemay be seen on transesophageal
echocardiograph
Ground glass opacitymay be seen on high-resolution
DIAGNOSIS
computed tomography The most common findings in AFES are:
21. Butson AR. Intermittent pneumatic calf compression for 39. Papadopoulos G, Kuhly P, Brock M, et al. Venous and para-
prevention of deep venous thrombosis in general abdominal doxical air embolism in the sitting position. A prospective
surgery. Am J Surg. 1981;142:525. study with transoesophageal echocardiography. Acta Neu-
22. Clagett GP, Reisch JS. Prevention of venous thromboem- rochir (Wien). 1994;126:140.
bolism in general surgical patients. Results of meta-analysis. 40. Roberts S, Johnson M, Davies S. Near-fatal air embolism:
Ann Surg. 1988;208:227. fibrin sheath as the portal of air entry. South Med J. 2003;
23. Coe NP, Collins RE, Klein LA, et al. Prevention of deep 96:10361038.
vein thrombosis in urological patients: a controlled, ran- 41. Coulter TD, Wiedemann HP. Gas embolism. N Engl J Med.
domized trial of low-dose heparin and external pneumatic 2000;342:2000.
compression boots. Surgery. 1978;83:230. 42. Wilkins RH, Albin MS. An unusual entrance site of venous
24. Hills NH, Pflug JJ, Jeyasingh K, et al. Prevention of deep vein air embolism during operations in the sitting position. Surg
thrombosis by intermittent pneumatic compression of calf. Neurol. 1977;7:71.
Br Med J. 1972;1:131. 43. Nazon D, Abergel G, Hatem CM. Critical care in orthopedic
25. Christopherson R, Beattie C, Frank SM, et al. Periopera- and spine surgery. Crit Care Clin. 2003;19:33.
tive morbidity in patients randomized to epidural or general 44. Nixon JR, Brock-Utne JG. Free fatty acid and arterial oxygen
anesthesia for lower extremity vascular surgery. Perioper- changes following major injury: a correlation between
ative Ischemia Randomized Anesthesia Trial Study Group. hypoxemia and increased free fatty acid levels. J Trauma.
Anesthesiology. 1993;79:422. 1978;18:23.
26. Rosenfeld BA, Beattie C, Christopherson R, et al. The 45. Georgopoulos D, Bouros D. Fat embolism syndrome: clinical
effects of different anesthetic regimens on fibrinolysis examination is still the preferable diagnostic method. Chest.
and the development of postoperative arterial thrombosis. 2003;123:982.
Perioperative Ischemia Randomized Anesthesia Trial Study 46. Muller C, Rahn BA, Pfister U, et al. The incidence, pathogen-
Group. Anesthesiology. 1993;79:435. esis, diagnosis, and treatment of fat embolism. Orthop Rev.
27. Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia 1994;23:107.
in the anticoagulated patient: defining the risks (the second 47. Myers R, Taljaard JJ. Blood alcohol and fat embolism
ASRA consensus conference on neuraxial anesthesia and syndrome. J Bone Joint Surg Am. 1977;59:878.
anticoagulation). Reg Anesth Pain Med. 2003;28:172. 48. Robert JH, Hoffmeyer P, Broquet PE, et al. Fat embolism
28. Rogers FB, Shackford SR, Wilson J, et al. Prophylactic vena syndrome. Orthop Rev. 1993;22:567.
cava filter insertion in severely injured trauma patients: 49. Heine TA, Halambeck BL, Mark JB. Fatal pulmonary fat
indications and preliminary results. J Trauma. 1993;35:637. embolism in the early postoperative period. Anesthesiology.
29. Barritt DW, Jordan SC. Anticoagulant drugs in the treat- 1998;89:1589.
ment of pulmonary embolism. A controlled trial. Lancet. 50. Bulger EM, Smith DG, Maier RV, et al. Fat embolism
1960;1:1309. syndrome. A 10-year review. Arch Surg. 1997;132:435.
30. Kanis JA. Heparin in the treatment of pulmonary throm- 51. Lee W, Ginsburg KA, Cotton DB, et al. Squamous and
boembolism. Thromb Diath Haemorrh. 1974;32:519. trophoblastic cells in the maternal pulmonary circulation
31. Carson JL, Kelley MA, Duff A, et al. The clinical course of identified by invasive hemodynamic monitoring during the
pulmonary embolism. N Engl J Med. 1992;326:1240. peripartum period. Am J Obstet Gynecol. 1986;155:999.
32. Raschke RA, Reilly BM, Guidry JR, et al. The weight-based 52. Cromey MG, Taylor PJ, Cumming DC. Probable amniotic
heparin dosing nomogram compared with a standard care fluid embolism after first-trimester pregnancy termination.
nomogram. A randomized controlled trial. Ann Intern Med. A case report. J Reprod Med. 1983;28:209.
1993;119:874. 53. Hasaart TH, Essed GG. Amniotic fluid embolism after
33. Circulation. The urokinase pulmonary embolism trial. A na- transabdominal amniocentesis. Eur J Obstet Gynecol Reprod
tional cooperative study. 1973;47 (Suppl 2):II1108. Biol. 1983;16:25.
34. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary 54. Clark SL, Hankins GD, Dudley DA, et al. Amniotic fluid
embolism: clinical outcomes in the International Cooperative embolism: analysis of the national registry. Am J Obstet
Pulmonary Embolism Registry (ICOPER). Lancet. 1999;353: Gynecol. 1995;172:1158.
1386. 55. Gilbert WM, Danielsen B. Amniotic fluid embolism: de-
35. Gulba DC, Schmid C, Borst HG, et al. Medical compared creased mortality in a population-based study. Obstet
with surgical treatment for massive pulmonary embolism. Gynecol. 1999;93:973.
Lancet. 1994;343:576. 56. Locksmith GJ. Amniotic fluid embolism. Obstet Gynecol Clin
36. Leacche M, Unic D, Goldhaber SZ, et al. Modern surgical North Am 1999;26:435444.vii.
treatment of massive pulmonary embolism: Results in 47 57. Tuffnell DJ. Amniotic fluid embolism. Curr Opin Obstet
consecutive patients after rapid diagnosis and aggressive Gynecol. 2003;15:119.
surgical approach. J Thorac Cardiovasc Surg. 2005;129:1018. 58. Moore J, Baldisseri MR. Amniotic fluid embolism. Crit Care
37. Gottlieb JD, Ericsson JA, Sweet RB. Venous air embolism: Med. 2005;33:S279.
A review. Anesth Analg. 1965;44:773. 59. Hankins GD, Snyder RR, Clark SL, et al. Acute hemodynamic
38. Muth CM, Shank ES. Gas embolism. N Engl J Med. 2000; and respiratory effects of amniotic fluid embolism in the
342:476. pregnant goat model. Am J Obstet Gynecol. 1993;168:1113.
B . C A R D I OVA SC U L A R
CHAPTER EPIDEMIOLOGY AND PREDICTORS
A
history significant for hypertension and a involves the use of known studies and guidelines to predict
55-pack-year history of tobacco use is sched- who may be at risk for developing specific complications
uled for a right femoral-popliteal bypass. The in the perioperative setting. An anesthesiologist considers
patients symptoms include claudication of the factors that determine the safety of an anesthetic and
her right leg and mild shortness of breath. formulates an anesthetic plan utilizing monitors, medica-
She reports being able to walk up two flights of stairs, tions, and perhaps even plans further diagnostic testing to
with only minimal shortness of breath. She takes Lisino- ensure the safety of the patient. The anesthesiologist also
pril for hypertension. Her preoperative laboratory test provides this information to the patient to obtain informed
reports include a hemoglobin (Hgb) of 11.9 g per dL, consent for the planned anesthetic. In this chapter, the fo-
creatinine 1.1 mg per L, glucose 92 mg per dL, Na+ cus will be on the ability to predict which patients may be
140 mEq per L, and K+ 4.3 mEq per L. Preoperative elec- at risk for developing cardiovascular complications in the
trocardiogram shows a normal sinus rhythm at a rate of preoperative, intraoperative, and perhaps most important,
82, with no Q-waves or ST-segment changes. The patient the postoperative period.
undergoes a general anesthetic with endotracheal intuba- Cardiovascular complications most frequently imply
tion and mechanical ventilation. After 1 hour of surgery, perioperative myocardial infarction (MI); however, sud-
the electrocardiogram reveals a new 1.5 mm ST-segment den cardiac arrest, stroke, and myocardial ischemia, as
depression in leads II, III, and aVF. Her vital signs are evidenced by the electrocardiographic changes with or
blood pressure 95/43 mm Hg, heart rate 95 bpm, SpO2 without hemodynamic perturbations, are all forms of car-
98%, and temperature 36.8 C, with an estimated blood diovascular complications.
loss of 600 mL. A bolus of intravenous crystalloid solu-
tion is administered, with a consequent increase in blood
pressure. Subsequently, the patient is treated with intra-
venous metoprolol to reduce the heart rate to 60 bpm. What Measures Have Been
Intraoperative Hgb measurement is 9.6 g per dL, and a
unit of packed red blood cells is transfused. Following Taken to Assess Cardiac Risk
these interventions, the ST segments return to baseline for Noncardiac Surgery?
levels. The surgical procedure is completed. The patient is
extubated uneventfully in the operating room and taken
More than 28 million patients undergo anesthesia for
to the intensive care unit for observation.
surgical procedures each year in the United States. With
the aging population, that number is predicted to reach
40 million in just a few years. Approximately 8 million
of the patients undergoing anesthesia each year have
How Is Epidemiology Defined? known coronary artery disease or coronary risk factors.2
More than 50,000 patients suffer a perioperative MI,
Epidemiology is defined as (i) a branch of medical sci- and approximately 1 million incur perioperative cardiac
ence that deals with the incidence, distribution, and complications.
204
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 205
During the second half of the 20th century, many should all be considered for further cardiac workup. Each
attempts at developing indices to assess perioperative of the above mentioned features or symptoms is asso-
risk were put forward. The development of the now ciated with a higher cardiovascular risk. Patients who
ubiquitous American Society of Anesthesiologists (ASA) have pain while walking or pain of the extremities at rest
Physical Status Classification of surgical mortality by Dr. may have peripheral vascular disease, which may suggest
Robert D. Dripps emerged from this period as one of occult coronary artery disease in the absence of cardiac
the early attempts. Several years later in 1977, Goldman symptoms.
et al. developed the Cardiac Risk Index. Since then, there Several comorbidities have been suggested as fac-
have been many indices offered to stratify the risk of tors that increase the risk of cardiovascular compli-
cardiovascular complications in noncardiac surgery.3 cations in the perioperative setting. A history of di-
As the formal practice of evidence-based medicine abetes mellitus, renal impairment, pulmonary disease,
emerged over the last few decades, numerous studies and hematologic perturbations such as anemia are the
that have examined the predilection to cardiovascular most commonly described. Diabetes mellitus not only has
complications provide a structure from which practicing an increased association with coronary artery disease,
physicians can make reasonable predictions about which but the pathophysiologic effects on the visceral nervous
patients may be at high risk for cardiovascular complica- system may lead to silent ischemia that does not man-
tions. In fact, a task force involving experts throughout the ifest as chest pain. Furthermore, congestive heart failure
medical community set out to review the body of literature is more common among elderly patients with diabetes
pertaining to cardiovascular risk and risk factors in the mellitus than without, even when angiotensin-converting
perioperative setting to develop guidelines for managing enzyme inhibitors are used appropriately. Renal impair-
patients at risk for cardiovascular disease. In 1996, the ment is also associated with increased cardiovascular
American College of Cardiology (ACC) and the American risk. Azotemia alone has an association with cardiovas-
Heart Association (AHA) released Guidelines for Periopera- cular disease and an increase in cardiovascular events.4
tive Cardiovascular Evaluation for Noncardiac Surgery. Re- Lee et al. demonstrated that a creatinine level >2.0 mg
port of the American College of Cardiology/American Heart per dL was an independent predictor of cardiovascular
Association Task Force on Practice Guidelines (Committee complications.5
on Perioperative Cardiovascular Evaluation for Noncar- Both obstructive and restrictive pulmonary disease
diac Surgery) which was subsequently updated in 2002, can lead to perioperative respiratory complications. In
incorporating new information and studies.4 addition, hypoxia, hypercapnia, and acidosis can lead
In this chapter, we will review the current guidelines to worsening cardiovascular performance, and ultimately
of risk stratification for cardiovascular risk, including the contribute to cardiovascular complications. Most signifi-
ACC/AHA guidelines. We will also examine the most cur- cantly, the presence of pulmonary disease may limit the
rent evidence concerning the effectiveness of diagnostic use of -blockers, one of the most commonly used medi-
studies in predicting cardiovascular complications, as well cal therapies for decreasing cardiovascular events. Anemia
as potential interventions aimed at reducing the risk of can further worsen myocardial supply, leading to poten-
cardiovascular complications in the perioperative period. tial worsening of ischemia, as well as heart failure. A
The bulk of this chapter will be dedicated to the exami- hematocrit <28% is associated with an increased risk of
nation of cardiovascular risk for the patient undergoing myocardial ischemia and cardiovascular complications in
noncardiac surgery. At the end of this chapter, we will prostate and vascular surgical patients.4
address specific risk factors for cardiovascular complica- The Revised Cardiac Risk Index identifies six factors
tions in patients undergoing cardiac surgery. This chapter of a patients history that can be used to determine the
is designed to assist the clinical anesthesiologist in mak- risk of major cardiac complications in the perioperative
ing informed decisions about the risk of cardiovascular setting.5 The factors are of approximately equal prognostic
complications and to guide the use of preoperative testing value and include the following:
in evaluating such risk. High-risk surgery
Ischemic heart disease
History of congestive heart failure
History of cerebrovascular disease
What Are the Methods Used to Insulin therapy for diabetes
Preoperative serum creatinine >2.0 mg per dL (see
Establish Risk? Table 14.1).
In a prospective evaluation of 1,422 patients, the
presence of 2 of these factors identified patients with
PATIENT HISTORY moderate (7%) and severe (11%) cardiovascular compli-
cation rates (see Table 14.2). However, according to the
A carefully addressed patient history may quickly identify
ACC/AHA guidelines, only four clinical features serve as
high-risk factors, as well as the necessity and appropriate-
major predictors (see Table 14.3) of cardiac complica-
ness of preoperative testing. Patients who provide a history
tions:
that includes a previous MI, current or recent chest pain,
complaints of worsening shortness of breath, worsen- Unstable coronary syndromes
ing edema, and pacemaker and/or defibrillator placement Decompensated heart failure
206 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 14.1 Factors that Increase the Risk of Perioperative Cardiac Complications in Patients Undergoing Noncardiac
Surgery and Indications for the Use of Perioperative -Blocker Therapy
Odds Ratio
Risk Factor (95% CI)a Perioperative -Blocker Indicated
Ischemic heart diseaseb 2.4 (1.34.2) Yes
Congestive heart failure 1.9 (1.13.5) Yes
High-risk surgeryc 2.8 (1.64.9) Uncertain, but probably
Diabetes mellitus (especially insulin-requiring) 3.0 (1.37.1) Yes
Renal insufficiency 3.0 (1.46.8) Uncertain, but probably if renal insufficiency is
due to diabetes or vascular disease
Poor functional statusd 1.8 (0.93.5) Yes, if poor status is thought to be due to
coronary artery disease or heart failure
a Data are from Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for prediction of cardiac
risk of major noncardiac surgery. Circulation. 1999;100:1043; Reilly DF, McNeely MJ, Doerner D, et al. Self-reported exercise tolerance and
the risk of serious perioperative complications. Arch Intern Med. 1999;159:2185.
b Ischemic heart disease includes angina and prior myocardial infarction.
c High-risk surgery includes intraperitoneal, intrathoracic, and suprainguinal vascular procedures.
d Poor functional status is defined as the inability to walk four blocks or climb two flights of stairs.
TABLE 14.2 Major Cardiac Complication Rates and 95% CIs in Derivation and Validation Cohorts Stratified by Risk
Classification System
a MCRI vs OCRI, & ASA (p < 0.05); RCRI vs MCRI & OCRI (p < 0.0001); RCRI & ASA (p = 0.055).
b RCRI vs OCRI (p = 0.02); RCIR vs MCRI (p < 0.0001); RCRI vs ASA (p = 0.018)
Revised cardiac risk factors include high-risk surgery, ischemic heart disease, congestive heart failure, history of cerebrovascular disease,
insulin therapy for diabetes, and preoperative serum creatinine. CI, confidence interval; ASA, American Society of Anesthesiologists; Pop,
population; ROC, receiver opening characteristic; SE, standard error; OCRI, original cardiac risk index; MCRI, modified cardiac risk index; RCRI,
revised cardiac risk index
Reprinted and adapted from Lee TH, Marcantonio ER, Mangione CM, et al. Derivation and prospective validation of a simple index for
prediction of cardiac risk of major noncardiac surgery. Circulation. 1999;100:10431047.
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 207
TABLE 14.3 ACC/AHA Guidelines: Clinical Predictors of TABLE 14.4 Estimated Energy Requirements of Daily
Increased Perioperative Cardiovascular Risk (Myocardial Activitiesa
Infarction, Heart Failure, and Death)
METs Criteria
Major 1 Can you take care of yourself?
Unstable coronary syndromes Eat, dress, or use the toilet?
Recent myocardial infarctiona with evidence of impor- Walk indoors around the house?
tant ischemic risk by clinical symptoms or noninvasive Walk a block or two on level ground at
study 23 mph or 3.24.8 km/h?
Unstable or severeb angina (Canadian class III or IV)c Do light work around the house like dusting or
Decompensated congestive heart failure washing dishes?
Significant arrhythmias 4 Climb a flight of stairs or walk up a hill?
High-grade atrioventricular block Walk on level ground at 4 mph or 6.4 km/h?
Symptomatic ventricular arrhythmias in the presence Run a short distance?
of underlying heart disease Do heavy work around the house like
Supraventricular arrhythmias with uncontrolled ventric- scrubbing floors or lifting or moving heavy
ular rate furniture?
Severe valvular disease Participate in moderate recreational activities
like golf, bowling, dancing, doubles tennis,
Intermediate or throwing a baseball or football?
Mild angina pectoris (Canadian class I or II) 10 Participate in strenuous sports like swimming,
Prior myocardial infarction by history or pathologic singles tennis, football, basketball, or skiing?
Q-waves
a Adapted from the Duke Activity Status Index and AHA Exercise
Compensated or prior congestive heart failure
Diabetes mellitus Standards MET,metabolic equivalent.
Reprinted with permission from Eagle KA, Berger PB, Calkins H,
Chronic renal insufficiency
et al. ACC/AHA guideline update for perioperative cardiovascular
Minor evaluation for noncardiac surgeryexecutive summary a report of
the American College of Cardiology/American Heart Association
Advanced age Task Force on Practice Guidelines (Committee to Update the
Abnormal ECG (left ventricular hypertrophy, left bundle 1996 Guidelines on Perioperative Cardiovascular Evaluation for
branch block, ST-T abnormalities) Noncardiac Surgery). Circulation. 2002;105:1257.
Rhythm other than sinus (e.g., atrial fibrillation)
Low functional capacity (e.g., inability to climb one flight
of stairs with a bag of groceries)
History of stroke
PHYSICAL EXAMINATION
Uncontrolled systemic hypertension In addition to a detailed medical history, a complete phys-
ical examination should be performed, with particular
a
The American College of Cardiology National Database Library attention to a comprehensive cardiovascular examination.
defines recent MI as >7 d but 1 mo (30 d). Vital signs should be evaluated. Previously unrecognized,
b
May include stable angina in patients who are unusually
untreated, or poorly treated hypertension may be of con-
sedentary.
c Campeau L. Grading of angina pectoris. Circulation. 1976;54: cern. Patients exhibiting stage 3 hypertension (systolic
522523. blood pressure >180 mm Hg and diastolic blood pressure
ECG, electrocardiogram. >110 mm Hg) should be evaluated and treated before
Reprinted with permission from Eagle KA, Berger PB, Calkins H, surgery. The data on which this recommendation is based
et al. ACC/AHA guideline update for perioperative cardiovascular is limited; furthermore, the impact that severe, poorly
evaluation for noncardiac surgeryexecutive summary a report of controlled hypertension (stage 3) has on perioperative
the American College of Cardiology/American Heart Association mortality is unclear.6 Hypertension alone is considered a
Task Force on Practice Guidelines (Committee to Update the minor risk factor; however, it must be considered in con-
1996 Guidelines on Perioperative Cardiovascular Evaluation for
junction with additional clinical risk factors, the patients
Noncardiac Surgery). Circulation. 2002;105:1257.
exercise capacity, and the surgical risk as assigned by the
ACC/AHA guidelines.
A new or worsening heart murmur may also be
workup. In an effort to simplify the determination of of concern, particularly if it is a harsh, crescendo-
a patients functional capacity, the ACC/AHA guidelines decrescendo murmur heard at the left upper sternal
(see Table 14.4) incorporate a method of stratifying func- border. This murmur is characteristic of aortic stenosis
tional capacity based on a calculated metabolic equivalent and may prompt further workup. As indicated in the AHA
(MET). Patients exhibiting <4 METs, coupled with ei- guidelines, severe or critical aortic stenosis has a high
ther the presence of intermediate-risk factors (Table 14.3) risk of cardiovascular complications, including acute MI
or undergoing high-risk surgical procedures, should be and asystole.7 Similarly, both stenotic and regurgitant
referred for noninvasive testing. murmurs of the mitral valve are also associated with
208 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
heart failure. Signs of dependent edema, especially in TABLE 14.5 Absolute and Relative Contraindications to
the lower extremities, with or without accompanying Exercise Treadmill Testing
dyspnea, may be signs of heart failure as well, regardless of
valvular disease. Pulmonary rales, elevated jugular venous Absolute Contraindications
pressure, hepatojugular reflux, and a third heart sound
may also be hallmarks of heart failure.4 Acute myocardial infarction (within 2 days)
Unstable angina not previously stabilized by medical
therapy
Uncontrolled cardiac arrhythmias causing symptoms or
ELECTROCARDIOGRAM hemodynamic compromise
Symptomatic severe aortic stenosis
A routine electrocardiogram is often one of the first diag- Uncontrolled symptomatic heart failure
nostic studies to be performedand arguably one of the Acute pulmonary embolus or pulmonary infarction
simplestin assessing cardiovascular risk. Any abnormal Acute myocarditis or pericarditis
finding (e.g., arrhythmia, Q-waves, ST-segment changes, Acute aortic dissection
etc.) in high-risk patients confers a notable increase in
perioperative risk, as much as 300%.8 However, a nor- Relative Contraindications
mal electrocardiogram in a low-risk patient has very low Left main coronary stenosis
sensitivity and often does not discriminate any further Moderate stenotic valvular heart disease
stratification of risk. Owing to the swift and noninvasive Electrolyte abnormalities
nature of an electrocardiogram, it is frequently performed Severe arterial hypertension
as an early test in assessing perioperative cardiac risk, but Tachyarrhythmias or bradyarrhythmias
are often most valuable when used in conjunction with ad- Hypertrophic cardiomyopathy and other forms of
ditional studies and in patients with clinically determined outflow tract obstruction
intermediate or high risk. Mental or physical impairment leading to inability to
exercise adequately
High-degree atrioventricular block
EXERCISE TREADMILL Reprinted with permission from Gibbons RJ, Balady GJ, Bricker JT,
TESTING et al. ACC/AHA 2002 guideline update for exercise testing: Summary
article: A report of the American College of Cardiology/American
The stress response is the natural reaction of the body Heart Association Task Force on Practice Guidelines (Committee
to surgery. A well designed anesthetic, including the use to Update the 1997 Exercise Testing Guidelines). Circulation.
of narcotics, volatile or intravenous agents, antibiotics, 2002;106:1883.
and regional anesthesia and so on, can reduce the level
of psychologic stress experienced by the patient. In
ischemia should generally undergo exercise testing to
spite of the anesthesiologists best efforts, the conditions
assess the risk of future cardiac events. In fact, the
brought about by surgery elicit a stress response to
exercise treadmill test is designed to produce ischemia
some degree; for the cardiovascular system, this response
in those with significant risk factors for coronary artery
often includes tachycardia, hypertension, and increased
disease. Several studies have demonstrated that a positive
tissue oxygen demand. One of the noninvasive studies,
ischemic response and a low exercise capacity can predict
termed a stress test, is an exercise treadmill test. This is
the outcome following noncardiac surgery.2 An early
a diagnostic study that induces stressful conditions on
study by Mangano et al. demonstrated an incidence of
the heart, pulmonary system, and peripheral vasculature
postoperative MI in 37% of patients who underwent
to ascertain the patients tolerance of increased heart
vascular surgery and had a positive ischemic response by
rate and subsequent cardiac oxygen demand. The exercise
exercise treadmill testing, as opposed to a 1.5% incidence
test typically involves the use of either a treadmill or
of perioperative MI in those who did not. Other studies,
a stationary bicycle, an electrocardiogram, and blood
however, have not demonstrated such definitive results. It
pressure monitoring. In general, exercise testing is a
has been found that a 12-lead resting electrocardiogram
safe procedure; however, MI and even death have been
and exercise capacity were independent predictors of
reported during routine exercise testing, and can be
perioperative cardiac complications, and not necessarily
expected to occur at a rate of up to 1 per 2,500 tests.9
the variables related directly to ischemia.
As a result, the AHA has developed recommendations
for both absolute and relative contraindications for
proceeding with exercise treadmill testing (see Table 14.5).
In addition, patients unable to walk or run on a MYOCARDIAL PERFUSION
treadmill (i.e., patients with severe arthritis, paralysis, STUDIES
or other neurologic conditions, amputees, etc.) require
an alternative means of inducing an increased heart rate, In patients for whom further preoperative testing is de-
which will be discussed in the subsequent text. sired (i.e., high-risk patients) yet who are unable to
Patients with suspected or known coronary artery perform an exercise treadmill test, a myocardial perfu-
disease and new or changing symptoms that suggest sion scan (scintigraphy) may be considered. Patients with
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 209
exercise limitations are often at highest risk, including modality for predicting perioperative risk. However, the
those with peripheral vascular, neurologic, or orthopedic data suggests that the ability of dobutamine stress echocar-
disease. A dipyridamole/thallium or adenosine/thallium diography to predict cardiac complications is not equally
perfusion scan induces coronary vasodilation and assists as effective for all patient groups, and studies have shown
in elucidating regions of redistribution defects. Dipyri- that only certain patient populations are further stratified
damole blocks adenosine reuptake, thereby increasing as to their risk of cardiac complications, similar to dipyri-
adenosine concentration in the coronary vessels. Adeno- damole thallium imaging. Boersma et al. demonstrated
sine acts as a direct coronary vasodilator. By infusing that the additional predictive value of dobutamine stress
these vasodilators, flow is preferentially distributed to ar- echocardiography over clinical risk stratification is lim-
eas distal to normal coronary arteries, and minimizes ited in clinically low-risk patients receiving -blockers.12
flow to areas distal to coronary stenoses. A radioisotope, However, their data suggested that the use of this modality
thallium (99m-Technetium sestamibi is also used), is in- in clinically intermediate- and high-risk patients may dif-
jected. Regions of normal myocardium appear on initial ferentiate those who can safely undergo surgery with the
imaging, whereas areas of myocardial necrosis or areas use of -blockers and those for whom revascularization
distal to significant coronary lesions remain dark. Several should be considered. The presence of stress-induced is-
hours later, a second infusion of radioisotope is injected. chemia during dobutamine stress echocardiography inde-
Areas that remain as defects represent regions of old scar, pendently predicted perioperative cardiac complications
whereas those that reappear as normal suggest areas at in high-risk patients undergoing vascular surgery. Patients
risk for myocardial ischemia. Redistribution defects can in whom extensive ischemia (characterized by >4 left ven-
be quantified; large areas of defect are associated with tricular wall segments) bore a greater risk of cardiac com-
increased cardiac risk. plications than those with 4 ischemic segments, despite
Numerous studies have examined the utility of per- the use of -blockers12 (see Table 14.6).
fusion scans for risk stratification, and much controversy Accordingly, the ACC/AHA guidelines note that the
has arisen over the findings. In the mid-1980s, positive weight of evidence supports the use of dobutamine stress
perfusion scans were found to correlate with adverse pe- echocardiography in properly selected patients, especially
rioperative events. On the other hand, very few adverse those undergoing peripheral vascular revascularization,
events occurred in patients who had no redistribution as a means to determine perioperative cardiac risk.4 It is
defects (e.g., a negative scan). As such, dipyridamole important to note, however, that the guidelines also point
thallium scans found widespread use in the ensuing out that the positive predictive value of dobutamine stress
years. In the early 1990s, however, the former findings echocardiography in patients undergoing vascular surgery
were challenged by a prospective, triple-blinded study has a large range (7%25%), while the negative predictive
which, contrary to previously reported findings, revealed value falls within the range of 93%100%. In spite of the
no association between redistribution defects and adverse low positive predictive value of this modality, one study
outcomes.2,10 Later studies corroborated the fact that thal- suggests that a positive dipyridamole thallium scintigra-
lium redistribution was not significantly associated with phy resulted in lower prognostic value than dobutamine
adverse perioperative cardiac events,10,11 and as a re- stress echocardiography. A meta-analysis by Beattie et al.
sult, the routine use of perfusion scans decreased. More comparing thallium imaging and stress echocardiogra-
recent studies have demonstrated a greater predictive phy concluded that stress echocardiography is superior
value of perfusion scans if used in conjunction with more to thallium imaging in predicting postoperative cardiac
complicated clinical markers and technical examinations. events (see Table 14.7).13
However, the most significant study may have been that Dobutamine stress echocardiography may ultimately
by Baron et al., in which it was reported that myocardial be preferred as a noninvasive test due to the practical
perfusion imaging did not provide independent prognostic advantages over perfusion scintigraphy such as lower
value beyond that of clinical risk stratification.11 costs, reduced imaging time, greater availability, the
absence of exposure to radiation, and the immediacy
of results. Furthermore, this modality may identify
DOBUTAMINE STRESS significant valvular disease, which may represent a
separate risk factor (i.e., severe aortic stenosis). However,
ECHOCARDIOGRAPHY availability and local expertise may be the critical factor
Dobutamine stress echocardiography identifies ventricu- when choosing this test.
lar wall motion abnormalities in patients both at rest and
with increased heart rates brought about by intravenous
injection of dobutamine, designed to increase cardiac oxy-
gen demand. Dobutamine stress echocardiography typi- What Strategies Are Used to
cally involves the description and quantitation of regional Reduce the Risk of
wall motion abnormalities of the 16 left ventricular wall
segments. Because this modality utilizes direct adrener- Cardiovascular Complications?
gic stimulation, it is believed to better approximate the
perioperative stress on the cardiovascular system than Strategies to reduce the incidence of perioperative car-
that imparted by vasodilators such as adenosine or dipyri- diovascular complications can be divided into three
damole. As a result, it has become a widely used testing therapies: medical therapy, preoperative coronary
210 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 14.6 Results of Meta-Analysis Evaluating Ability of Noninvasive Cardiac Tests to Predict Risk of Perioperative
Cardiac Events in Patients Undergoing Vascular Surgerya
revascularization, and intraoperative and postoperative use of coronary stents and coronary artery bypass grafting
monitoring. Medical therapies include the use of peri- (CABG). Proposed monitoring techniques include the use
operative -blockers, 2 agonists, aspirin, statins, cal- of arterial catheters, central venous catheters, pulmonary
cium channel blockers, angiotensin-converting enzyme artery catheters, and transesophageal echocardiography.
inhibitors, and nitrates. Advocated preoperative coronary In this segment, we will discuss the latest evidence for the
revascularization procedures include percutaneous trans- use of both preoperative medical therapy and coronary
luminal coronary angioplasty (PTCA) with or without the revascularization.
TABLE 14.7 Meta-Analytic Comparison of Stress Echocardiography to Thallium Imaging as a Preoperative Screening
Tool
outcome with -blockers.26 Lindenauer et al. utilized apparent in the patient group that had undergone PTCA
an administrative dataset and were able to demonstrate more than 90 days before undergoing noncardiac surgery.
improved perioperative survival in those with at least Their findings also demonstrated that patients who
three risk factors on the Revised Cardiac Risk Index, received PTCA within 90 days of noncardiac surgery had
demonstrating worse survival in those without any risk an incidence of perioperative cardiac events similar to
factors.27 Finally, one cohort study suggested improved those patients with known coronary artery disease who
survival with atenolol compared with metoprolol, which had not undergone revascularization. A closer look at the
the authors suggest may be related to atenolols longer described cardiac events reveals that revascularization led
half-life and lower probability of -blocker withdrawal.28 to a reduction in the incidence of angina pectoris and
Despite evidence that -blockers do not confer the congestive heart failure, not a reduction in the incidence
same level of risk reduction for each type of high-risk of death or nonfatal MI.
feature, the ACC/AHA do recommend (class I) the use The use of coronary stents following PTCA has further
of -blockers in patients previously on -blockers or complicated the picture by posing additional risks of coro-
who have inducible ischemia on a preoperative stress nary thrombosis and bleeding. Complications following
test and are undergoing major vascular surgery. Two PTCA and coronary artery stenting were reported in 40 pa-
ongoing, large-scale studies may help elucidate how these tients who underwent these procedures <6 weeks before
recommendations should be applied to the low- and major noncardiac surgery.33 Eight deaths, seven nonfatal
intermediate-risk populations. MIs, as well as 8 of the 11 bleeding episodes, all occurred
in patients who had surgery <14 days after stenting. In pa-
tients undergoing stent placement, the dilemma involves
whether to interrupt antiplatelet therapy designed to pre-
OTHER AGENTS vent stent thrombosis versus the increased risk of bleeding
2 -Agonists (clonidine, mivazerol), calcium channel block- complications caused by the antithrombotic therapy. In
ers (diltiazem, verapamil, etc.), and nitrates (nitroglyc- this study, stent thrombosis following the discontinuation
erin) have been proposed as both preoperative and of antiplatelet medications 1 to 2 days before surgery ac-
intraoperative therapies to reduce the risk of cardio- counted for most of the fatal events. Unfortunately, those
vascular complications in the perioperative setting for who continued their antithrombotic regimen accounted
noncardiac surgery. Thus far, only small studies have been for most of the bleeding episodes. Wilson et al. reported
performed with mixed results. The use of these agents, similar findings in a group of 207 patients who under-
including intraoperative and postoperative -blockers, in went noncardiac surgery within 60 days of coronary stent
the setting of an acute myocardial ischemia and infarction placement.34 Eight patients had severe cardiac events, six
is discussed in Chapter 15. died, and one suffered a nonfatal MI. Of the six deaths,
two suffered an MI before death. On the basis of these
events, the risk of adverse cardiac complications remains
present at least 6 weeks after stenting. There were no re-
CORONARY ported cardiac complications in patients who had surgery
REVASCULARIZATION after 60 days of coronary stent placement.
These results suggest that noncardiac surgery should
Percutaneous Revascularization be performed at least 6 weeks after PTCA with stent-
ing, to allow completion of coronary reendothelialization
Cardiac risk stratification and preoperative testing as de- and healing, as well as a full course of antiplatelet ther-
scribed in the preceding text may identify patients who apy to be completed. The current practice of poststenting
would benefit from preoperative coronary revascular- antithrombotic therapy uses a combination of aspirin
ization. CABG and PTCA are currently the two major and clopidogrel (Plavix; Sanofi-Aventis Pharmaceuticals,
options for coronary revascularization. To date, there Bridgewater, NJ) for no less than 4 weeks. Most impor-
have been no randomized trials to evaluate the effec- tant, however, the evidence thus far does not support the
tiveness of preoperative PTCA. However, three separate prophylactic use of PTCA to reduce cardiac risk, with
retrospective cohort studies for patients receiving PTCA or without stenting, in the immediate weeks or months
before noncardiac surgery have been published.2931 The before elective noncardiac surgery. In addition, newer
patient populations underwent PTCA to relieve symp- drug-eluting stents require an even longer period before
tomatic angina or to reduce the perioperative risk of noncardiac surgery due to the delay of in-stent endothe-
ischemia identified through noninvasive testing. All three lialization.32
studies demonstrated a low incidence of perioperative
cardiac death and MI. However, no comparison groups Coronary Artery Bypass Grafting
were included in their analysis.
Posner et al. used an administrative database to As with PTCA, there have been no randomized trials
compare adverse cardiac events in patients who did not that have examined the effect of CABG on perioperative
undergo PTCA in the preoperative period versus those who cardiac complications. There are, however, retrospective
did.32 Their findings indicated that those patients who had studies that may shed some light on the potential benefit
undergone PTCA had a lower incidence of perioperative of CABG before noncardiac surgery. A large retrospec-
cardiac complications. The benefit of PTCA was most tive review of 3,368 patients enrolled in the Coronary
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 213
Artery Surgery Study registry suggests a potential protec- in high-risk patients who are scheduled to undergo major
tive effect of preoperative CABG.35 In this study, patients vascular surgery.38 In addition, the ACP guidelines rec-
were assigned to either medical therapy or CABG before ommend the use of perioperative -blocker therapy in all
noncardiac surgery. Their findings suggest that CABG was high-risk patients.
protective in patients undergoing head and neck, abdomi- Both these guidelines, as well as independent and
nal, vascular, and thoracic surgery. When compared with subsequent reviews, recommend that CABG and/or coro-
patients who were given medical therapy, those patients nary revascularization be limited to only those patients
who underwent CABG had a lower incidence of MI (2.7% with a clearly defined need for the procedure that is in-
vs. 0.8%) and perioperative mortality (3.3% vs. 1.7%). The dependent of the need for noncardiac surgery.4,39 Such
greatest reduction in perioperative mortality was found in patients would include, for example, those who have max-
patients who had either advanced angina or multivessel imized medical therapy, yet continue to exhibit poorly
coronary disease. controlled angina pectoris, and/or patients who have
In 1999, Fleisher et al. reviewed Medicare data bases one or more high-risk coronary artery lesions. High-risk
to determine the 30-day and 1-year mortality after non- lesions include the following:
cardiac surgery based on whether patients had undergone
Left main coronary artery disease with >50% stenosis
preoperative cardiac testing and coronary interventions,
Severe two- or three-vessel coronary artery disease with
including CABG and PTCA with or without stenting,
>70% stenosis with involvement of the proximal left
within a year before the noncardiac surgery.36 Their
anterior descending (LAD) artery
findings demonstrated that preoperative revasculariza-
Easily induced myocardial ischemia on preoperative
tion conferred a reduction in 1-year mortality for patients
stress testing
undergoing aortic surgery, but that it had no effect on
Left ventricular systolic dysfunction at rest40
mortality for those undergoing infrainguinal procedures.
Analysis of the Bypass Angioplasty Revascularization In- Several small studies have examined the effectiveness
vestigation addressed the incidence of postoperative car- of these guidelines for stratifying cardiac risk. A small
diac complications following noncardiac surgery among retrospective study by Samain et al., examining the
patients with multivessel coronary artery disease.37 Pa- risk stratification for patients undergoing aortic surgery,
tients with severe angina were randomized to undergo concluded that the ACC/AHA guidelines were effective in
either CABG or PTCA an average of 29 months before stratifying cardiac risk by using clinical predictors and
noncardiac surgery. The incidence of cardiac death and functional capacity.41 Furthermore, a small prospective,
MI were similarly low in both groups (1.6% for both randomized study concluded that preoperative cardiac
groups, n = 250 for each group). These findings iden- stress testing in patients with specific clinical profilesas
tify the low incidence of cardiac complications following defined by the ACC/AHA guidelinesdid not further
CABG, but may also suggest that preoperative testing and identify patients at risk for adverse cardiac events after
subsequent CABG, when appropriate, reduces the risk of vascular surgery.42 This trial was a small pilot study
perioperative cardiac complications for those undergoing involving only 46 patients; however, initial evidence
noncardiac surgery. It is essential to take into considera- suggests that the ACC/AHA guidelines are useful in
tion the cumulative risks of coronary angiography, CABG, identifying patients who need additional preoperative
and subsequent noncardiac surgery. testing, as well as identifying those who can bypass further
In contrast to the Coronary Artery Surgery Study testing and proceed to surgery without increased risk. The
findings, a large, randomized study known as the Coro- latter can ultimately reduce unnecessary and unwarranted
nary Artery Revascularization Prophylaxis Trial reported testing.
that among patients with stable coronary artery dis-
ease, coronary artery revascularization before elective
major vascular surgery does not improve long-term sur-
vival.38 Furthermore, no reduction in early postoperative How Is Cardiovascular Risk
outcomesnamely, MI, death, and length of hospital
staywas reported. These findings are supportive of Assessed In Cardiac Surgery?
the ACC/AHA and American College of Physicians (ACP)
recommendations that reserve the use of CABG or per-
cutaneous revascularization for patients with unstable
cardiac symptoms or advanced coronary artery disease for
PATHOPHYSIOLOGY
whom a survival benefit with CABG have been proved.4,39 Unlike patients undergoing noncardiac surgery, the sin-
gle, most important cause of cardiac complications and
Guidelines death in those experiencing cardiac surgery in the form of
CABG, repair or replacement of valve, repair of congenital
The ACC in cooperation with the AHA offered their guide- defect, or ventricular remodeling is direct myocardial in-
lines for the use of interventions to reduce the incidence jury. In patients undergoing cardiopulmonary bypass and
of perioperative cardiac complications of noncardiac subsequent myocardial ischemia, myocardial injury man-
surgery, first in 1996 and with an updated revision in ifests as transient cardiac contractile dysfunction (known
20024 (see Fig. 14.1). Likewise, the ACP currently supports as myocardial stunning) and MI.43 Myocardial necrosis
the use of preoperative testing and coronary interventions develops within minutes of the interruption of blood flow.
214 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
STEP 3 Recent coronary Yes Recent coronary angiogram Favorable result and Operating
evaluation or stress test? no change in room
symptoms
No Unfavorable result or
Clinical change in symptoms
predictors
STEP 5
STEP 4 Major clinical Intermediate clinical Minor or no
predictorsa predictorsb clinical predictorsc Major Clinical Predictorsa
Consider coronary
Invasive testing angiography
Subsequent cared
dictated by findings
and treatment results
Consider coronary
Invasive testing
angiography
d
Subsequent care may
Subsequent cared include cancellation of
dictated by findings surgery, no further
and treatment results treatment, medical
treatment, or coronary
revascularization
FIGURE 14.1 ACC/AHA Guidelines: Stepwise approach to preoperative cardiac assessment. CHF,
congestive heart failure; MI, myocardial infarction; METs, metabolic equivalents; ECG,
electrocardiogram. Reprinted with permission from Eagle KA, Berger PB, Calkins H, et al. ACC/AHA
guideline update for perioperative cardiovascular evaluation for noncardiac surgeryexecutive
summary a report of the American College of Cardiology/American Heart Association Task Force on
Practice Guidelines (Committee to Update the 1996 Guidelines on Perioperative Cardiovascular
Evaluation for Noncardiac Surgery). Circulation. 2002;105:1257.
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 215
Ultimately, it is the duration of interrupted blood flow, 128 European centers of 19,030 patients undergoing di-
either partial or complete, and the concomitant use of verse cardiac surgical procedures identified multiple risk
cardioprotective techniques that determines the extent of factors associated with increased mortality. The follow-
myocardial necrosis. In virtually all studies, the duration ing factors have all been shown to influence mortality
of both aortic cross-clamp and cardiopulmonary bypass of cardiac surgery: age, female gender, serum creati-
have consistently been demonstrated to be the primary nine, extracardiac arteriopathy (peripheral or cerebral
determinant of postoperative outcomes. vascular disease), chronic airway disease, severe neuro-
Myocardial ischemia that has been limited to <20 logic dysfunction, previous cardiac surgery, recent MI, left
minutes followed by immediate reperfusion has been ventricular systolic ejection fraction, chronic congestive
shown to lead to functional recovery without evidence heart failure, pulmonary hypertension, active endocardi-
of structural injury or biochemical evidence of tissue tis, unstable angina, procedure urgency, critical preoper-
injury.44,45 However, reperfusion after an ischemic time ative condition, ventricular septal rupture, noncoronary
>20 minutes results in irreversible myocardial injury surgery, and thoracic aortic surgery (see Table 14.8).
and/or cellular necrosis. Furthermore, the extent of tissue Although the list of predictors is long, the value
necrosis after reperfusion is related directly to the length of the EuroSCORE lies in the fact that the baseline
of ischemic time. As a result, the combination of ischemic mortality figures were calculated in patients in whom
and reperfusion injury represents the most frequent and none of these risk factors were present. With these patients
serious type of injury that leads to unfavorable outcomes excluded from the calculations, the study demonstrated
in cardiac surgery. Preoperative risk factors may influence very low rates of mortality. For example, 0% in atrial
ischemic reperfusion injury. septal defect repair, 0.4% for CABG, and just above
1% for single valve or replacement (see Table 14.9).
Since its initial publication, the additive EuroSCORE
has been repeatedly validated and has entertained wide
PREDICTORS OF CARDIAC acceptance and use throughout the world. As a result,
RISK AND CARDIAC RISK it has become the primary tool for risk stratification
in cardiac surgery. Unfortunately, the EuroSCORE has
STRATIFICATION been shown to underpredict operative risk in patients
Postoperative mortality remains the principal outcome undergoing combined cardiac procedures.51
of patient injury in the perioperative period. Death can Among the many variables and characteristics found
be both cardiac and noncardiac in origin; in the case to be associated with increased risk of perioperative
of cardiac death, the etiology can be either ischemic or complications during cardiac surgery, several have been
nonischemic in origin. Postoperative mortality is typically consistently found to be major predictors of risk across
reported as either in-hospital or 30-day statistics. The first multiple and diverse study population. These are: age,
attempt to predict postoperative morbidity and mortality female gender, left ventricular systolic function, body
in cardiac surgery was undertaken by the Collaborative habitus, reoperation, type of surgery, and urgency of
Study in Coronary Artery Surgery.46,47 They examined surgery. Interestingly, significant comorbidities, such as
6,630 patients who underwent isolated CABG during the renal insufficiency and diabetes mellitus, have not been
years 19751978. Their findings revealed: (i) significantly shown to be independent risk factors for perioperative
higher mortality in women; (ii) increasing mortality with complications of cardiac surgery. What is clear, however,
advancing age in men (although not in women); and is the importance of the type of surgery, the limitation of
(iii) higher mortality in patients with increasing severity myocardial ischemic time, and utilization and techniques
of angina pectoris, frequency of heart failure and number of myocardial protection on the clinical outcome of the
and extent of coronary artery stenoses. They also identified patient undergoing cardiac surgery.
urgency of surgery as a very strong predictor of outcome;
on the other hand, left ventricular ejection fraction was
not a predictor.
Since that initial study, multiple studies have been How Does Our Case Summary
performedmost of which have looked specifically at
isolated CABG patients as opposed to combined cardiac Fit in This Discussion?
procedures or isolated valvular proceduresand differing
predictors have been proposed by each. The Society of According to the ACC/AHA guidelines, the patient in the
Thoracic Surgeons has developed multiple risk models opening vignette has a single, minor predictor (advanced
based on their own criteria as well as collaborative studies. age) of cardiac risk. By definition, a minor predictor
The most recent data from 2003 reviewed 503,478 CABG recognizes a marker of cardiovascular disease that [has]
procedures and reported the incidence of stroke at 1.63%, not been proven to independently increase perioperative
renal failure at 3.53%, reoperation at 5.17%, prolonged risk.4 The patient also describes activity that is equivalent
ventilation at 5.96%, and sternal infection at 0.63%.48 to 4 METs. Despite the surgery falling under the heading
Perhaps the most useful index for cardiac operative of high risk (5% cardiac risk), the ACC/AHA algorithm
risk evaluation is the European System for Cardiac Opera- recommends that this patient proceed directly to surgery
tive Risk Evaluation (EuroSCORE).49,50 An analysis from without further noninvasive testing (Fig. 14.1).
216 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 14.8 Risk Model Result negative noninvasive test or circumvents noninvasive test-
ing based on clinical predictors and risk factors, it does
Variable Odds Ratio not suggest that a patient is immune from developing car-
diac complications in the perioperative setting. In fact, the
Age (in 10-y increments) 1.640
surgical milieu may pose new stresses to the patient that,
Female gender 1.157
heretofore, were not reproducible through traditional pre-
Noncaucasian 1.249
operative testing, for example, changes in inflammatory
Ejection fraction 0.988
mediators and hemostatic components. As a result, the
Diabetes 1.188
anesthesiologist must be able to recognize the signs and
Renal failure 1.533
symptoms of cardiovascular disease and complications,
Serum creatinine (if renal failure is 1.080
and understand appropriate therapies so as to minimize
present)
the morbidity associated with them. In this featured vi-
Dialysis dependence (if renal failure is 1.381
gnette, the quick use of -blockers, accompanied with
present)
the improved oxygen-carrying capacity brought about by
Pulmonary hypertension 1.185
an increased hematocrit, likely prevented the worsening
Cerebrovascular accident timing 1.198
of ischemia and potentially averted an MI. In this not
Chronic obstructive pulmonary disease 1.296
uncommon scenario, the use of a preoperative -blocker
Peripheral vascular disease 1.487
may be the only recommendable change in management
Cerebrovascular disease 1.244 based on the evidence discussed in this chapter, as well
Acute evolving, extending myocardial 1.282 the recommendations put forth by both the ACC/AHA and
Myocardial infarction timing 1.117 ACP guidelines.
Cardiogenic shock 2.211
Use of diuretics 1.122
Hemodynamic instability 1.747
Triple vessel disease 1.155
Left main disease >50% 1.119
What Conclusions Can Be Made
Preoperative intra-aortic balloon pump 1.480 Concerning Epidemiology and
Status Its Predictors?
Urgent or emergent 1.189
Emergent salvage 3.654
First reoperation 2.738 Perioperative cardiac complications lead to significant
Multiple reoperations 4.282 morbidity and mortality and cost our health care system
Arrhythmias 1.099 billions of dollars each year. As our older population
Body surface area 0.488 continues to increase, the assessment and strategies
Obesity 1.242 to reduce cardiovascular risk for patients undergoing
New York Heart Association class IV 1.098 surgery pose one of the most significant challenges for
Use of steroids 1.214 anesthesiologists now. Familiarity with the algorithm for
Congestive heart failure 1.191 determining risk, understanding the components that
Percutaneous transluminal coronary 1.332 increase risk and, perhaps most important, recognizing
angioplasty within 6 h of surgery the possible interventions that may be implemented for an
Angiographic accident with 1.203 individual patient are all crucial components to providing
hemodynamic instability safe and responsible medical care.
Use of digitalis 1.168 Clinical predictors, alone or in conjunction with
Use of intravenous nitrates 1.088 noninvasive preoperative testing, can stratify patients
according to their risk of perioperative cardiovascular
Reprinted with permission from Shroyer ALW, Plomondon ME, complications. Many risk indices have been proposed
Grover FL, et al. The 1996 coronary artery bypass risk model: The over the years, but the key is the identification of those
Society of Thoracic Surgeons Adult Cardiac National Database. Ann with extensive coronary artery disease, since only a small
Thorac Surg. 1999;67:12051208, with permission from the Society cohort of patients may benefit from prophylactic coro-
of Thoracic Surgenos. nary revascularization. The current guidelines advocate
the continuation of -blocker therapy in all patients previ-
ously on -blockers, and initiation in those patients with
What is important to recognize from this example ischemia on stress testing undergoing vascular surgery.
is the fact that the ACC/AHA guidelines are exactly that, Further study will be required to determine the value of
guidelines, and as such may not be applicable and/or con- -blocker therapy in other patient populations. In the pa-
clusive to each and every individual. In the event that a tient for whom -blockers are contraindicated, the use
patient does not clearly belong to a defined stratification, of 2 agonists may be the most appropriate alternative.
the clinician must use professional judgment and discre- Other therapies such as statins show initial promise in
tion. Even if testing should be considered on the basis of reducing cardiac risk and should be continued periopera-
preoperative risk factors, exercise tolerance, and surgical tively; however, more extensive clinical trials are needed
risk, it should only be performed if the results will change to determine their role. Calcium channel blockers and ni-
management. Secondly, in the event that a patient has a trates may be used in addition to maximal -blockade, yet
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 217
the effect of their individual use has not been thoroughly value in low- and intermediate-risk patients is not
supported in the literature. well established.
Patients with no cardiac risk factors represent a 4. Dobutamine stress echocardiography has been dem-
low-risk population. As a result, they do not warrant onstrated to be the most effective noninvasive test to
further preoperative testing. Patients with three or more stratify patient risk, as well as to identify patients for
risk factors may benefit from noninvasive perioperative whom coronary angiography may be warranted.
testing due to its potential to further stratify this group 5. The ACC/AHA guidelines provide an extensive al-
into moderate- and high-risk classifications. In patients gorithm to help assess a patients risk of cardiac
with three or more risk factors and a positive stress complications, as well as identifying which preoper-
test, the risk of proceeding directly to surgery, even ative tests are warranted on the basis of risk factors,
with the use of -blockade, is substantial, and coronary but may be modified by the results of the Coronary
angiography should be considered to determine whether Artery Revascularization Prophylaxis Trial.
there is a need for coronary revascularization. However, 6. Unlike patients undergoing noncardiac surgery, the
the Coronary Artery Revascularization Prophylaxis Trial major outcome measurement for patients undergoing
clearly questions the value of coronary revascularization, cardiac surgery is postoperative mortality.
except in those with extensive coronary artery disease. 7. Cardiac risk for patients undergoing cardiac surgery
Considerations for cardiac surgery are somewhat is directly related to the extent of myocardial damage
different from the algorithm for those undergoing non- incurred during ischemic time and during reperfu-
cardiac surgery. In cardiac surgery, the focus is placed on sion.
the factors that cause direct myocardial injury and may 8. The EuroSCORE is the most widely used and most
predispose patients to high risks of reperfusion injury extensively validated index for stratifying risk for
following periods of ischemia. Short aortic cross-clamp patients undergoing cardiac surgery.
times, shorter total cardiopulmonary bypass times, and
adequate myocardial protection are the key components
to reducing the perioperative mortality of cardiac surgery. REFERENCES
Other clinical predictors have been shown to influence the 1. Merriam-Webster, Inc. Websters ninth new collegiate dictio-
incidence of perioperative mortality, including age, female nary. Merriam-Webster, Inc; 1991.
gender, left ventricular systolic function, body habitus, 2. Mangano DT. Assessment of the patient with cardiac
reoperation, type of surgery, and urgency of surgery. disease: An anesthesiologists paradigm. Anesthesiology.
1999;91(5):1521.
3. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial
index of cardiac risk in noncardiac surgical procedures.
N Engl J Med. 1977;297(16):845.
KEY POINTS 4. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline
update for perioperative cardiovascular evaluation for non-
1. Cardiac complications, specifically acute MIs, in the cardiac surgeryexecutive summary a report of the Ameri-
perioperative setting lead to significant morbidity can College of Cardiology/American Heart Association Task
and mortality as well, costing the health care system Force on Practice Guidelines (Committee to Update the 1996
billions of dollars each year. Guidelines on Perioperative Cardiovascular Evaluation for
2. The risk of perioperative cardiac complications for Noncardiac Surgery). Circulation. 2002;105:1257.
patients undergoing noncardiac surgery is best deter- 5. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation
mined by the assessment of risk factors for cardiac and prospective validation of a simple index for prediction
of cardiac risk of major noncardiac surgery. Circulation.
disease, a patients functional capacity, and the type
1999;100:1043.
of surgery.
6. Fleisher LA. Preoperative evaluation of the patient with
3. Perioperative -blocker therapy has been shown to hypertension. JAMA. 2002;287(16):2043.
be the most effective medical therapy for reducing 7. Kertai MD, Bountioukos M, Boersma E, et al. Aortic
cardiac risk, especially in high-risk patients, but its stenosis: An underestimated risk factor for perioperative
218 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
complications in patients undergoing noncardiac surgery. the American Heart Association Scientific Sessions 2004. New
Am J Med. 2004;116:8. Orleans: November 710, 2004.
8. Beattie WS. Evidence-based perioperative risk reduction: 26. Brady AR, Gibbs JS, Greenhalgh RM, et al. Perioperative
Can. J Anesth. 2005;52:R1. beta-blockade (POBBLE) for patients undergoing infrarenal
9. Gibbons RJ, Balady GJ, Beasley JW. ACC/AHA guidelines vascular surgery: Results of a randomized double-blind
for exercise testing: Executive summary. A report of the controlled trial. J Vasc Surg. 2005;41(4):602.
American College of Cardiology/American Heart Association 27. Lindenauer PK, Pekow P, Wang K, et al. Perioperative
Task Force on Practice Guidelines (Committee on Exercise beta-blocker therapy and mortality after major noncardiac
Testing). Circulation. 1997;96:345. surgery. N Engl J Med. 2005;353(4):349.
10. Mangano DT, London MJ, Tubau JF, et al. SPI Research 28. Redelmeier D, Scales D, Kopp A. Beta blockers for elective
Group. Dipyridamole thallium-201 scintigraphy as a pre- surgery in elderly patients: Population based, retrospective
operative screening test: A re-examination of its predictive cohort study. Br Med J. 2005;331(7522):932.
potential. Circulation. 1991;84:493. 29. Allen JR, Helling TS, Hartzler GO. Operative procedures not
11. Baron JF, Mundler O, Bertrand M, et al. Dipyridamole- involving the heart after percutaneous transluminal coronary
thallium scintigraphy and gated radinonuclide angiography angioplasty. Surg Gynecol Obstet. 1991;173:285.
to assess cardiac risk before abdominal aortic surgery. N Engl 30. Elmore JR, Hallett JW Jr, Gibbons RJ, et al. Myocardial
J Med. 1994;330:663. revascularization before abdominal aortic aneuysmorraphy:
12. Boersma E, Poldermans D, Bax JJ, et al. Predictors of Effect of coronary angioplasty. Mayo Clin Proc. 1993;68:637.
cardiac events after major vascular surgery: Role of clinical 31. Gottlieb A, Banoub M, Sprung J, et al. Perioperative
characteristics, dobutamine echocardiography, and beta- cardiovascular morbidity in patients with coronary artery
blocker therapy. JAMA. 2001;285:1865. disease undergoing vascular surgery after percutaneous
13. Beattie WS, Abdelnaem E, Wijeysundera DN, et al. A meta- transluminal coronary angioplasty. J Cardiothorac Vasc
analytic comparison of preoperative stress echocardiography Anesth. 1998;12:501.
and nuclear scintigraphy imaging. Anesth Analg. 2006;102:8. 32. Posner KL, Van Norman GA, Chan V. Adverse cardiac
14. Ferguson GG, Sackett DL, Thorpe KE, et al. ASA and Carotid outcomes after noncardiac surgery in patients with prior
Endarterectomy (ACE) Trial Collaborators. Low-dose and percutaneous transluminal coronary angioplasty. Anesth
high-dose acetylsalicylic acid for patients undergoing carotid Analg. 1999;89:553.
endarterectomy: A randomised controlled trial. Lancet. 33. Kaluza GL, Joseph J, Lee JR, et al. Catastrophic outcomes of
1999;353:2179. noncardiac surgery soon after coronary stenting. J Am Coll
15. Robless P, Mikhailidis DP, Stansby G. Systematic review of Cardiol. 2000;35:1288.
antiplatelet therapy for the prevention of myocardial infarc- 34. Wilson SH, Fasseas P, Orford JL, et al. Clinical outcome of
tion, stroke or vascular death in patients with peripheral patients undergoing non-cardiac surgery in the two months
vascular disease. Br J Surg. 2001;88:787. following coronary stenting. J Am Coll Cardiol. 2003;42:234.
16. ONeil Callahan K, Katsimaglis g, Tepper MR, et al. Statins 35. Eagle KA, Rihal CS, Mickel MC, et al. Cardiac risk of
decrease perioperative cardiac complications in patients noncardiac surgery: Influence of coronary disease and type
undergoing noncardiac vascular surgery: The Statins for of surgery in 3368 operations. Circulation. 1997;96:1882.
Risk Reduction in Surgery (StaRRS) study. J Am Coll Cardiol. 36. Fleisher LA, Eagle KA, Shaffer T, et al. Perioperative and
2005;45(3):336. long-term mortality rates after major vascular surgery:
17. Poldermans D, Bax JJ, Kertai MD, et al. Statins are associated The relationship to preoperative testing in the medicare
with a reduced incidence of perioperative mortality in population. Anesth Analg. 1999;89:849.
patients undergoing major noncardiac vascular surgery. 37. Hassan SA, Hlatky MA, Boothroyd DB, et al. Outcomes of
Circulation. 2003;107:1848. noncardiac surgery after coronary bypass surgery or coro-
18. Lindenauer PK, Pekow P, Wang K, et al. Lipid-lowering ther- nary angioplasty in the bypass angioplasty revascularization
apy and in-hospital mortality following major noncardiac investigation (BARI). Am J Med. 2001;110:260.
surgery. JAMA. 2004;291(17):2092. 38. McFalls EO, Ward HB, Moritz TE, et al. Coronary-artery
19. Durazzo AE, Machado FS, Ikeoka DT, et al. Reduction in car- revascularization before elective major vascular surgery.
diovascular events after vascular surgery with atorvastatin: N Engl J Med. 2004;351:2795.
A randomized trial. J Vasc Surg. 2004;39(5):967. 39. Palda VA, Detsky AS. Perioperative assessment and manage-
20. Poldermans D, Boersma E. Beta-blocker therapy in noncar- ment of risk from coronary artery disease. Ann Intern Med.
diac surgery. N Engl J Med. 2005;353:412. 1997;127:313.
21. Mangano DT, Layug EL, Wallace A, et al. Effect of atenolol 40. Fleisher LA, Eagle KA. Lowering cardiac risk in noncardiac
on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med. 2001;345:1677.
surgery. N Engl J Med. 1996;335:1713. 41. Samain E, Farah E, Leseche G, et al. Guidelines for
22. Wallace A, Layug B, Tateo I, et al. Prophylactic atenolol perioperative cardiac evaluation from the American College
reduces postoperative myocardial ischemia. Anesthesiology. of Cardiology/American Heart Association task force are
1998;88:7. effective for stratifying cardiac risk before aortic surgery.
23. Poldermans D, Boersma E, Bax JJ, et al. The effect J Vasc Surg. 2000;31(5):971.
of bisoprolol on perioperative mortality and myocardial 42. Falcone RA, Nass C, Jermyn R, et al. The value of preop-
infarction in high-risk patients undergoing vascular surgery. erative pharmacologic stress testing before vascular surgery
N Engl J Med. 1999;341:1789. using ACC/AHA guidelines: A prospective, randomized trial.
24. Stevens RD, Burri H, Tramer MR. Pharmacologic myocar- J Cardiothorac Vasc Anesth. 2003;17(6):694.
dial protection in patients undergoing noncardiac surgery: 43. Kloner RA, Przyklenk K, Kay GL. Clinical evidence for
A quantitative systematic review. Anesth Analg. 2003;97:623. stunned myocadium after coronary artery bypass surgery.
25. Juul AB. Randomized, blinded trial on perioperative meto- J Card Surg. 1994:9:397.
prolol versus placebo for diabetic patients undergoing non- 44. Heyndrickx GR, Millard RW, McRitchie RJ, et al. Regional
cardiac surgery [abstract]. Late-Breaking Clinical Trials I of myocardial functional and electrophysiological alterations
C H A P T E R 14 / E P I D E M I O L O G Y A N D P R E D I C T O R S 219
after brief coronary artery occlusion in conscious dogs. J Clin 49. Nashef SA, Roques F, Michel P, et al. European system
Invest. 1975;56(4):978. for cardiac operative risk evaluation (EuroSCORE). Eur J
45. Bolli R. Mechanism of myocardial stunning. Circulation. Cardiothorac Surg. 1999;16(1):9.
1990;82:723. 50. Roques F, Nashef SA, Michel P, et al. Risk factors and
46. Coronary Artery Surgery Study (CASS). A randomized trial outcome in European cardiac surgery analysis of the
of coronary artery bypass surgery. Survival data. Circulation. EuroSCORE multinational database of 19030 patients. Eur
1983;68(4):939. J Cardiothoracic Surg. 1999;15(6):816.
47. Alderman EL, Fisher LD, Litwin P, et al. Results of coronary 51. Karthik S, Srinivasan AK, Grayson AD, et al. Limitations of
artery surgery in patients with poor left ventricular function additive EuroSCORE for measuring risk stratified mortality
(CASS). Circulation. 1983;68(4):785. in combined coronary and valve surgery. Eur J Cardiothrorac
48. Shroyer AL, Coombs LP, Peterson ED, et al. The Society Surg. 2004;26(2):318.
of Thoracic Surgeons: 30-day mortality and morbidity risk
models. Ann Thorac Surg. 2003;75(6):1856.
CHAPTER MYOCARDIAL ISCHEMIA
15
AND INFARCTION
A
popliteal bypass surgery. He has a history nary artery disease (CAD) are presenting for noncardiac
of hypertension, diabetes, and hypercholes- surgery. These patients are at increased risk for periopera-
terolemia, for which he is on metoprolol, tive myocardial events. In fact, of the estimated 27 million
lisinopril, insulin, and atorvastatin. He also patients undergoing noncardiac surgery in the United
has a 30-pack-year smoking history. His States each year, 1 million experience perioperative car-
preoperative laboratory values are normal, except for a diac complications, ranging from congestive heart failure
hematocrit of 36%, creatinine of 1.6 mg per dL, and some and MI to death, costing an estimated $20 billion.2,3 These
nonspecific ST-T wave changes on his ECG. risks are even greater in patients undergoing vascular pro-
After discussion with the patient and surgeon, it is cedures.4 Patients who suffer postoperative MI or death
decided to proceed with the case and administer the on average incur $15,000 to more than $20,000 in ad-
patient his prescribed metoprolol, atorvastatin, and one ditional hospital costs due to prolonged hospitalizations,
half his dose of insulin. During a general anesthetic, his compared with similar patients who do not suffer MI, and
blood pressure is often labile. He also requires two units are at increased risk for other noncardiac complications
of packed red blood cells due to loss of fluid volume and as well.5
blood. Upon emergence, he is noted to be tachycardic, with The reduction of cardiac events in the perioperative
a heart rate of 110 bpm, which is treated with esmolol. period could potentially reduce cardiac (as well as
In the recovery room, he is noted to have ST-T wave overall) morbidity and mortality, in addition to decreasing
depressions on electrocardiogram (ECG) monitoring. A hospital stay and overall costs. Attempts to improve the
12-lead ECG is obtained showing 1.5 mm ST-T depression
perioperative outcome of patients at risk for CAD have
on leads V1 through V3. He denies chest pain, and
historically focused on three approaches: (i) preoperative
his vital signs are back to baseline values (150/85 mm
identification of high-risk patients who may benefit from
Hg, heart rate 88 bpm). Five milligrams of intravenous
myocardial revascularization; (ii) improved detection of
metoprolol are administered with partial resolution of his
perioperative myocardial ischemia to allow for prompt
ST-T depression. Serum troponin levels are sent to the
therapeutic intervention; and (iii) the prophylactic use
laboratory, and a cardiologist is consulted. The patient
of anesthetic and anti-ischemic techniques to decrease
subsequently rules in for an acute myocardial infarction
the prevalence and severity of postoperative myocardial
(MI). Coronary angiography demonstrates an occluded
ischemia.
left anterior descending artery, for which he is stented.
Recently, there has been considerable debate, not
He is placed on clopidogrel, as well as aspirin, and his
only over the level of preoperative assessment that is
preoperative medications are continued with increased
necessary for patients, but also the value of routine
doses. The rest of his hospitalization is unremarkable,
preoperative workups in patient populations at risk for
and he is discharged on the seventh postoperative day.
perioperative myocardial insults. This chapter outlines
some of the recent data regarding preoperative assess-
ments and focuses on the perioperative management of
these patients. The pathophysiology, demographics, and
What Is the Incidence and prognosis of postoperative myocardial ischemia and in-
Epidemiologic Impact of farction in patients undergoing noncardiac surgery were
Perioperative Cardiovascular also reviewed. In addition, patients with a history of my-
ocardial revascularization, either by coronary stenting or
Complications? angioplasty, or surgical intervention through coronary
artery bypass grafts (CABG), merit special consideration
Cardiovascular disease continues to be the leading cause in our anesthetic approach, and are discussed in detail in
of death in the United States.1 However, recent advances this chapter.
220
C H A P T E R 15 / M Y O C A R D I A L I S C H E M I A A N D I N F A R C T I O N 221
ECG, electrocardiogram; PA, pulmonary artery; TEE, transesophageal echocardiography; PAC, pulmonary artery catheter.
OTHER MODALITIES
What Are the Proposed
In light of these limitations of ECG monitoring, other
modalities for identifying myocardial ischemia have
Mechanisms of Perioperative
been presented as potentially beneficial in the peri- Myocardial Ischemia?
operative period. These techniques include the use of
pulmonary capillary wedge pressure tracings to detect
v-waves, which have not been proved to be particu- ISCHEMIC SYNDROMES
larly sensitive or specific, and the monitoring for re-
gional wall motion abnormalities with transesophageal Stable ischemic syndromes presumably occur because
echocardiography (TEE). These techniques, however, of the increased oxygen demand on the myocardium
are not without their respective drawbacks either; for in the presence of fixed coronary plaques that reduce
example, the predictive value of pulmonary capillary oxygen supply. Unstable syndromes are thought to be
wedge pressure tracings for monitoring ischemia is rather the result of endothelial dysfunction and inflammation,
poor. plaque rupture with local thrombus, and vasoreactivity
that produces intermittent critical decreases in coronary
oxygen supply.12 Patients with elevated coronary calcium
Transesophageal Echocardiography levels on computed tomography scan have greater rates
of perioperative MI after vascular surgery.20
Although TEE, another monitoring tool, is extraordinarily
Endothelial function is impaired in conditions such
sensitive for detecting regional wall motion abnormalities
as CAD, hypertension, hypercholesterolemia, diabetes,
associated with ischemia (which occur before surface ECG
and tobacco abuse, resulting in exaggerated vasoconstric-
changes), it is not practical for continuous monitoring.17
tion. Poor endothelial function is also associated with
The correct use of TEE as a monitor requires a higher
poor outcome after vascular surgery. The treatment used
level of training and expertise.
to heal the endothelium, often with agents designed
to combat hypercholesterolemia such as the inhibitors
Troponin Levels of hydroxymethylglutaryl-coenzyme A (HMG-CoA) re-
ductase or statins, improves perioperative outcome,
With respect to laboratory values, troponin levels tend to although for therapy to be effective, it may have to begin
be more specific in detecting perioperative MI than CK- weeks before surgery.21
MB isoenzyme measurements; troponin elevations corre-
late with lower survival rates after vascular surgery.13,14 A
recent study advocating the perioperative surveillance of ST-SEGMENT LEVELS
troponin levels found that patients undergoing abdominal
aortic surgery, who had abnormal but low troponin levels, In patients with LVH, diminished coronary vasodilator re-
were still at risk for MI and increased mortality.18 In pa- serve results in poor subendocardial perfusion. Ischemia
tients undergoing CABG surgery, however, troponin levels in the early postoperative period after noncardiac surgery
may not be of diagnostic value in the immediate postop- typically correlates with ST-segment depression rather
erative period, secondary to the nature of the surgical than ST elevation; ST-segment depression usually pre-
intervention itself.19 cedes postoperative cardiac complications. It is important
C H A P T E R 15 / M Y O C A R D I A L I S C H E M I A A N D I N F A R C T I O N 223
for the anesthesiologist to appreciate that most periopera- More emphasis is being placed on the aggressive man-
tive MIs are detected by the non-Q wave variety, although agement of these patients throughout the perioperative
MIs identified by ST elevation can also occur and are period. There are several classes of pharmacologic agents
associated with a higher mortality rate. that have enabled patients with CAD to enjoy not only
a longer lifespan, but also a better quality of life. These
agents may play vital roles in the prophylaxis against
TACHYCARDIA perioperative ischemia, as well as in its treatment (see
Table 15.3). It is imperative for the anesthesiologist to de-
AND HYPOTENSION termine which of the agents that a patient is taking should
The postoperative period is characterized by adrenergic be continued through the perioperative period and which
stress, which can induce myocardial ischemia in pa- may be initiated to facilitate the patients anesthetic.
tients with CAD, cause coronary vasoconstriction, and
facilitate platelet aggregation.12 Tachycardia not only in-
creases oxygen demand, but limits diastolic time and
coronary perfusion to the left ventricle, and it can paradox- What Is the Role of -Blockers
ically reduce coronary artery diameter. Hypertension and and Other Antianginal Agents
tachycardia in the postanesthesia care unit (PACU) have
been shown in a large study to correlate with increased in Reducing Adverse Cardiac
mortality and unplanned intensive care unit (ICU) ad- Events?
missions (although association does not necessarily mean
causation).22
-BLOCKERS
SURGICAL STRESS adrenergicblocking drugs, through their ability to sup-
In addition, surgery itself causes significant changes to press perioperative tachycardia, appear most efficacious
the hematologic system. Surgical stress can induce a hy- clinically and economically in preventing perioperative
percoaguable response as a result of increased platelet myocardial ischemia.2428 They are well tolerated by
number and function, diminished fibrinolysis, decreases most surgical patients and may reduce long-term cardiac
in natural anticoagulants (including protein C and an- events. adrenergicblocking drugs have been approved
tithrombin III), and increases in procoagulants (including for the treatment of hypertension, supraventricular tachy-
fibrinogen, factor VIII coagulant, and von Willebrand fac- cardias, ventricular arrhythmias, angina, and MI. They
tor).23 These postoperative changes may contribute to an are the cornerstone of acute and chronic post-MI therapy
increased likelihood of coronary artery thrombosis or rup- and are recommended by the AHA, as they are thought to
ture of preexisting coronary plaques in the postoperative reduce episodes of reinfarction.8
period; however, their relative importance in predicting The antihypertensive effects of -blockers can be
postoperative coronary events remains speculative. very useful during adrenergic activation such as occurs
in endotracheal intubation, extubation, electroconvulsive
therapy (ECT), and sternotomy. They also blunt tachycar-
dia during these events, which is likely the predominant
Does Pharmacologic mechanism of their anti-ischemic effects.
Several trials that document the ability of -blockers
Prophylaxis Work? to improve perioperative cardiac outcomes have been
published, although recent trials have questioned this con-
Cardiologists and internists are increasingly using ag- clusion in certain patient populations, notably diabetics.29
gressive, long-term pharmacologic means to reduce risk A recent meta-analysis of several randomized, controlled
in patients with CAD (Table 15.2). These strategies in- trials demonstrated that perioperative -blockade reduced
clude cholesterol reduction with statin agents, which myocardial ischemia and infarction, as well as short-term
stabilize coronary plaques; antihypertensive therapy with and long-term cardiac mortality.28 Another retrospective
angiotensin-converting enzyme (ACE) inhibitors, which study examining a large cohort of patients found that
also reduce sympathetic tone; -adrenergicblocking perioperative -blockers reduced the risk of in-hospital
drugs for decreasing myocardial workload; and strict death among high-risk patients, but not low-risk patients,
glucose control in diabetics. These treatments, although undergoing major noncardiac surgery.27 The benefit in
improving patient symptoms, quality of life, and pro- outcome from perioperative blockade in high-risk patients
longing lifespan, require the anesthesiologist to consult may persist for up to 2 years after vascular surgery.24
with the patients surgeon and primary physician to en- There are, however, several limitations to consider
sure that, perioperatively, medications are continued as when using perioperative adrenergic blockade. 1 se-
necessary. Many of these treatments have demonstrated lective drugs are less likely to cause bronchospasm, even
not only benefits for patients on a daily basis, but also in in patients with reactive airway disease. Nevertheless,
the acute perioperative setting and will be discussed in asthma and chronic obstructive pulmonary disease are
this chapter. relative contraindications to -blockade. Additionally,
224 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 15.3 Pharmacologic Agent Actions that May Have Benefit with Respect to Myocardial Ischemia
Perioperative
Agent Potential Role Drawbacks Recommendation
-adrenergic Benefit in reducing myocardial Caution with use in patients Continue throughout
blockers ischemia by reducing with underlying chronic perioperative period
myocardial work; also lung disease, as can lead to
demonstrated reduced bronchoconstriction
perioperative morbidity and
mortality in patients with CAD
Antiplatelet Aspirin decreases platelet activity May contribute to Continue aspirin through
agents and is beneficial in acute perioperative bleeding and day of surgery
coronary ischemia and for be contraindication to Clopidrogel and ticlopidine
chronic management regional anesthetic should be held 7 d
techniques before surgery
Statins Reduce cholesterol levels and Most perioperative data is Continue throughout
progression of atherosclerosis, retrospective; may cause perioperative period;
and are plaque-stabilizing hepatic side effects check liver function tests
preoperatively
Volatile Theoretic benefit of myocardial Mostly animal data with Useful if general
anesthetics preconditioning that may limit minimal human studies anesthesia is chosen
ischemic damage if it occurs demonstrate benefit
2 agonists Reduce norepinephrine release Data on newer agent Continue clonidine
and intraoperative ischemia dexmedetomidine lacking through perioperative
for perioperative ischemia period
prevention
ACE inhibitors Potential for plaque stabilization Data not well established for Debatable; some withhold
after MI benefit; may be associated morning of surgery,
Decrease remodeling of left with profound hypotension whereas others continue
ventricle post MI after induction of general it throughout
anesthesia
CAD, coronary artery disease; MI, myocardial infarction; ACE, angiotensin converting enzyme.
there is a very small subset of patients with severe CAD (CABG) surgeryhave found that intravenous nitroglyc-
(markedly positive stress tests in multivessel distributions) erin given prophylactically failed to reduce the prevalence
in whom -blockade or medical management has not of perioperative myocardial ischemia or infarction.33,34
reduced cardiac events, but rather may be considered While reducing preload and afterload helps diminish the
candidates for myocardial revascularization.30 workload for the myocardium, oftentimes the use of ni-
Although the validity of the data regarding the use troglycerin is accompanied by a compensatory increase in
of perioperative -blockers has recently been questioned, heart rate. This tachycardia is presumed to be the reason
their use is still advocated in most patients.8,31,32 Most that nitroglycerin was not found to be of benefit, as even
of the debate around the purported evidence is in small increases in heart rate are associated with great
regard to the factors used in the individual studies increases in myocardial oxygen demand.
such as power, analysis, or exclusion criteria. There
is a large, multicenter, randomized, double-blinded,
placebo-controlled, prospective study being undertaken
that should elucidate more information on the benefits of
perioperative -blockade.32 Given that the vast majority Is There a Role for Calcium
of the evidence is favorable, the use of -blockers in the Channel Blockers in the
perioperative period is currently widely advocated.
Perioperative Period?
Calcium channel blockers are often used for patients with
OTHER ANTIANGINAL hypertension; however, their perioperative use remains
DRUGS controversial. An initial study from the mid-1990s noted
an increased mortality in patients with CAD who were tak-
Other antianginal drugs appear less promising. Two ing nifedipine orally in a nonsurgical setting.35 This led to
studiesone in noncardiac surgery and one in fast-track what has been deemed a North American bias against
C H A P T E R 15 / M Y O C A R D I A L I S C H E M I A A N D I N F A R C T I O N 225
the widespread use of calcium channel blockers. Two first-line agents in patients with a history of CAD or MI.8,40
recent meta-analyses in patients undergoing noncardiac Statins also reduce the progression of atherosclerosis in
surgery, published in the same issue of the same journal, patients with CAD and have plaque-stabilizing properties
arrived at differing conclusions regarding the periopera- that may diminish the incidence of perioperative cardiac
tive benefits of these agents.36,37 Although both involved events.42 This plaque-stabilizing property is thought to be
the retrospective analysis of the medical literature, they key to the positive effect of statins, given the perioperative
had different numbers of studies and patients. The larger stressors that can destabilize coronary plaques.21 They
analysis demonstrated a potential advantage in the reduc- can also reduce coronary artery calcium deposits, a pos-
tion of myocardial ischemia, congestive heart failure and sible predictor of perioperative cardiac events in vascular
death, whereas the other analysis demonstrated no dis- surgery patients.
cernible benefit with the use of calcium channel blockers Several large observational studies have demon-
or nitrates.36,37 In addition, an accompanying editorial strated that the use of statin agents in the perioperative
touted the use of perioperative -blockade as a means of period result in lower cardiac-related morbidity and mor-
risk reduction, but questioned the findings regarding the tality.4345 Although these studies were retrospective in
use of calcium channel blockers.38 All three advocated nature, a more recent prospective trial also demonstrated
the need for more randomized, controlled trials to bet- that short-term treatment with atorvastatin significantly
ter clarify the situation, and therefore, at this time, the reduced the incidence of major adverse cardiac events
support for the use of calcium channel blockers is not as after vascular surgery.21 Most recently, a review article
robust as for -blockade in reducing ischemia in patients suggested several recommendations for the perioperative
undergoing noncardiac surgery. use of statins, including timing of administration and
therapeutic targets.46 Patients prescribed statin agents
preoperatively should have them continued throughout
the perioperative period. Postrevascularization patients
Is Aspirin Indicated for (either surgical or interventional) should have statin
Improving and Preventing agents started after the procedure if they were not taking
them before.
Myocardial Ischemia
and Infarction?
EPIDURAL ANALGESIA
Of the antiplatelet drugs, aspirin has the longest record
of not only relative safety, but also efficacy in prevent- Epidural anesthetics reduce cardiac preload and afterload
ing and improving outcomes in myocardial ischemia and postoperative adrenergic and coagulation responses,
and infarction.39 Its antiplatelet function is potentiated and produce coronary vasodilatation (thoracic epidurals
by irreversibly acetylating cyclooxygenase and inhibiting only).47 These effects suggest that they may play a role
thromboxane synthesis. Newer agents, such as ticlopidine in reducing perioperative myocardial ischemia. However,
or clopidogrel (inhibit ADP-induced platelet aggregation) substantiation that epidural anesthetics play a positive
or the IIA-IIIB glycoprotein inhibitors, have so far been role in cardiac outcome has been limited and conflict-
shown to decrease acute MIs and are recommended by ing in individual trials.4850 While a study examining
the AHA in this setting.40 However, these antiplatelet cardiac events in elderly patients undergoing orthopedic
agents are associated with an increased risk of intraop- surgery found decreased cardiac morbidity with regional
erative bleeding and may preclude regional anesthetic techniques, another study comparing general anesthesia
techniques.41 There are currently no studies examining with either spinal or epidural regional techniques in pa-
the use of the newer antiplatelet agents with respect to tients undergoing vascular surgery found no difference in
perioperative ischemia. cardiac morbidity or mortality.48,49
Concerns about respiratory depression, neuroaxial
hematomas, and the expense of monitoring have limited
the use of epidural narcotics in greater numbers of
What Are Other Pharmacologic patients.51,52 The risk of neuroaxial hematoma in patients
undergoing vascular or cardiac surgery, where large
Strategies that Can Be amounts of heparin are administered, needs to be carefully
Used to Improve Cardiac weighed. Although this complication continues to be rare,
it can be devastating for patients and their families. The
Outcomes? risk of a hematoma going undetected is particularly
increased in postoperative patients with the potential
for prolonged intubation and/or sedation and in whom
neurologic evaluations can be difficult.
STATINS Overall, although epidural anesthesia may improve
the outcome of other organ systems, its ability to
Statins reduce cholesterol levels in patients with hyper- reduce MI remains speculative.50,53 Two recent meta-
cholesterolemia and decrease the likelihood of reinfarc- analyses suggest that regional anesthesia may, in-
tion in patients with coronary disease; they are considered deed, be associated with a one third reduction in
226 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
perioperative MI, especially if thoracic epidurals are analgesic and antiplatelet properties; however, substantial
used.54,55 data is lacking. Ketorolac may reduce the stress response
The use of regional anesthetics, such as spinal or to surgery without increasing bleeding times or produc-
epidural, should be done on an individual case basis. ing renal insufficiency. A randomized trial demonstrated
Although there may be benefit, the risks involved are real that the addition of ketorolac to morphine administered
as well, and confirmation that anesthetic technique affects by patient-controlled analgesia can reduce the duration
outcomes continues to be elusive. The anesthesiologist and severity of myocardial ischemia following total joint
should thoroughly discuss options with patients and their arthroplasty.62 Whether this is a result of improved anal-
families and tailor an anesthetic to maximize patient gesia or antiplatelet effects is not clear at this time;
safety, comfort, and surgical outcome. however, concerns about increased postoperative hemor-
rhage make the use of these therapies in surgical patients
controversial. The newer generation of cyclooxygenase in-
hibitors were initially touted to have analgesic potency
VOLATILE ANESTHETICS superior to their predecessors, without the potential neg-
While most approaches that reduce myocardial ischemia ative side effects of gastrointestinal bleeding. However,
perioperatively are targeted at modulating the my- as of this writing, two major agents from this class of
ocardial oxygen supply-demand curve (with -blockers drugsvaldecoxib (Pfizer Inc, New York, NY) and rofe-
for example), volatile anesthetics may protect the my- coxib (Merck & Co, Inc, Whitehouse Station, NJ) have
ocardium from ischemia and reperfusion injury and been pulled from the market, secondary to concern with
reduce myocardial infarct size.5660 The mechanism of potential cardiac-related morbidity and mortality. There
action is termed preconditioning. Ischemic precondition- are plans under way for a large trial with celecoxib to
ing originally referred to the protective benefit of short gauge its potential benefit and safety.
periods of ischemia before longer, more damaging peri-
ods of ischemia. Anesthetic-induced preconditioning was
coined when studies demonstrated that the administra- 2 -AGONISTS
tion of a volatile anesthetic before a period of myocardial
ischemia resulted in a similar degree of cardioprotection, 2 adrenergic receptors act at prejunctional sites to medi-
as observed with ischemic preconditioning.61 ate a reduction in norepinephrine release from presynap-
Volatile anesthetics may also be cardioprotective tic terminals, thereby decreasing noradrenergic central
nervous system transmission and producing sedation,
when administered during myocardial reperfusion. Their
anxiolysis, and analgesia.
mechanism of action is complex; they affect the balance
of oxygen supply and demand in the myocardium by
dilating coronary arteries, preserving energy-dependent Clonidine
cellular function, and attenuating the action of reactive
Premedication with clonidine reduces hypertension,
oxygen species that has been implicated in myocardial
tachycardia, and norepinephrine levels in patients un-
ischemic injury.57 One study that compared volatile
dergoing surgery.63 Clonidine also suppresses the nor-
anesthetics with propofol, in cardiac surgery employing
mal postoperative increase in fibrinogen levels and an-
cardiopulmonary bypass, found that patients receiving
tagonizes epinephrine-induced platelet aggregation. In
volatile anesthetics had better cardiac performance, less
addition, it has demonstrated an ability to reduce
need for inotropic support, and decreased concentrations intraoperative myocardial ischemia.64,65
of plasma troponins postoperatively. Although most of
this data is from animal studies that have proven to
be difficult to reproduce in human models, given the
Dexmedetomidine and Mivazerol
near-impossibility of reproducing a predictable model The more specific 2 agonists, dexmedetomidine and
of ischemia, the preconditioning abilities of volatile mivazerol, may also reduce postoperative myocardial
anesthetics have led authors to suggest that they should ischemia but, again, substantial evidence thus far has
be incorporated into general anesthetic techniques for been lacking. A meta-analysis examining the role of all
patients with known or suspected CAD.56 It should be common 2 agonists suggested that they reduced mortal-
noted, however, that most studies with respect to volatile ity and MI after vascular surgery, in addition to reducing
anesthetics have been done in cardiac surgical patient ischemia in cardiac surgery patients.66 Although most of
populations, and there have been no proven benefits in the data was from one large study, and most data involved
outcome with the use of volatile anesthetics. clonidine, the results thus far for justifying the use of
dexmedetomidine have been favorable but still await large
prospective trials. Overall, the use of 2 agonists has a role
NONSTEROIDAL in the possible prophylaxis against perioperative ischemia.
ANTI-INFLAMMATORY
DRUGS GLUCOSE MANAGEMENT
Nonsteroidal anti-inflammatory drugs may be particu- Hyperglycemia appears to impair preconditioning mech-
larly useful in surgical patients with CAD due to their anisms67 and has been shown to correlate with poor
C H A P T E R 15 / M Y O C A R D I A L I S C H E M I A A N D I N F A R C T I O N 227
2. Of the estimated 27 million patients undergoing 5. Fleisher L, Corbett W, Berry C, et al. Cost-effectiveness of
differing perioperative beta-blockade strategies in vascular
noncardiac surgical procedures each year in the
surgery patients. J Cardiothorac Vasc Anesth. 2004;18:7.
United States, 1 million will suffer a perioperative 6. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation
myocardial event. and prospective validation of a simple index for prediction of
3. Predictors for perioperative complications include, cardiac risk of major noncardiac surgery. Circulation. 1999;
but are not limited to, advanced age, peripheral 100:1043.
vascular disease, chronic renal insufficiency, uncon- 7. Eagle KA, Brundage BH, Chaitman BR, et al. Guidelines
trolled hypertension, and diabetes mellitus. for perioperative cardiovascular evaluation for noncardiac
4. Recent studies have suggested that aggressive preop- surgery. Report of the American College of Cardiology/
erative evaluation and revascularization of patients American Heart Association Task Force on Practice Guide-
with CAD may not confer a benefit in outcome, lines. Committee on Perioperative Cardiovascular Evalua-
tion for Noncardiac Surgery. Circulation. 1996;93:1278.
which has led to emphasis being placed on periop-
8. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide-
erative management. line update for perioperative cardiovascular evaluation for
5. Patients who present to surgery may be taking a host noncardiac surgeryexecutive summary. A report of the
of pharmacologic agents. Some may benefit from American College of Cardiology/American Heart Association
the continuation of these agents (e.g., -blockers, Task Force on Practice Guidelines (Committee to Update the
cholesterol-lowering drugs (statins), 2 agonists, and 1996 Guidelines on Perioperative Cardiovascular Evaluation
aspirin or antiplatelet drugs) during the periopera- for Noncardiac Surgery). Anesth Analg. 2002;94:1052.
tive period. It is imperative for the anesthesiologist to 9. Mcfalls EO, Ward HB, Moritz TE, et al. Coronary-artery
identify which drugs their patient is currently taking, revascularization before elective major vascular surgery.
make appropriate decisions regarding their periop- N Engl J Med. 2004;351:2795.
10. Lustik S, Eichelberger J, Chhibber AK, et al. Preoperative
erative use, and be aware of possible complications
stress testing: New guidelines. J Clin Anesth. 2002;14:375.
from their use. 11. Devereaux PJ, Goldman L, Yusuf S, et al. Surveillance and
6. There is strong evidence that -adrenergicblocking prevention of major perioperative ischemic cardiac events
drugs and, more recently, statin agents reduce in patients undergoing noncardiac surgery: A review. CMAJ.
perioperative ischemic events. 2005;173:779.
7. Although some studies have demonstrated that 12. Landesberg G. The pathophysiology of perioperative my-
epidural techniques may reduce ischemia for certain ocardial infarction: Facts and perspectives. J Cardiothorac
surgical procedures, there are many contraindica- Vasc Anesth. 2003;17:90.
tions and potential complications that may limit 13. Kim LJ, Martinez EA, Faraday N, et al. Cardiac troponin I
their use. Overall, no one anesthetic technique has predicts short-term mortality in vascular surgery patients.
Circulation. 2002;106:2366.
shown to be clearly beneficial for the prevention of
14. Martinez EA, Nass CM, Jermyn RM, et al. Intermittent car-
perioperative ischemia.
diac troponin-I screening is an effective means of surveillance
8. Intraoperative and postoperative monitoring for is- for a perioperative myocardial infarction. J Cardiothorac
chemia should be based on the individual patient Vasc Anesth. 2005;19:577.
and surgical procedure. There are advantages and 15. Charlson M, Peterson J, Szatrowski TP, et al. Long-term
disadvantages to all of the more commonly em- prognosis after perioperative cardiac complications. J Clin
ployed modalities. Epidemiol. 1994;47:1389.
9. Conditions that can trigger ischemic episodes during 16. Charlson ME, Mackenzie CR, Ales K, et al. Surveillance
the intraoperative and postoperative course are often for postoperative myocardial infarction after noncardiac
under the direct control of the anesthesiologist and operations. Surg Gynecol Obstet. 1988;167:404.
17. Fleisher LA. Real-time intraoperative monitoring of myocar-
should be aggressively treated; they include tachy-
dial ischemia in noncardiac surgery. Anesthesiology. 2000;
cardia, hypothermia, anemia, hypoxia, and pain.
92:1183.
10. Patients presenting for surgery within 6 weeks of 18. LeManach Y, Perel A, Coriat P, et al. Early and delayed
revascularization (surgically or percutaneously) are myocardial infarction after abdominal aortic surgery. Anes-
at increased risk for morbidity and mortality and, thesiology. 2005;102:885.
therefore, should only have emergent procedures. 19. Noora J, Ricci C, Hastings D, et al. Determination of troponin
I release after CABG surgery. J Card Surg. 2005;20:129.
20. Mahla E, Vicenzi MN, Schrottner B, et al. Coronary artery
REFERENCES
plaque burden and perioperative cardiac risk. Anesthesiology.
1. Kochanek KD, Murphy SL, Anderson RN, et al. Deaths final 2001;95:1133.
data for 2002. National Vital Statistic Reports, Vol. 53. October 21. Durazzo AE, Machado FS, Ikeoka DT, et al. Reduction in car-
12, 2004. diovascular events after vascular surgery with atorvastatin:
2. Mangano D. Perioperative cardiac morbidity. Anesthesiology. A randomized trial. J Vasc Surg. 2004;39:967.
1990;72:153. 22. Rose DK, Cohen MM, DeBoer DP. Cardiovascular events in
3. Mangano D, Goldman L. Preoperative assessment of patients the postanesthesia care unit: Contribution of risk factors.
with known or suspected coronary disease. N Engl J Med. Anesthesiology. 1996;84:772.
1995;333:1750. 23. Devereaux PJ, Goldman L, Cook DJ, et al. Perioperative
4. LItalien GJ, Paul SD, Hendel RC, et al. Development and cardiac events in patients undergoing noncardiac surgery:
validation of a Bayesian model for perioperative cardiac risk A review of the magnitude of the problem, the pathophysiol-
assessment in a cohort of 1,081 vascular surgical candidates. ogy of the events and methods to estimate and communicate
J Am Coll Cardiol. 1996;27:779. risk. CMAJ. 2005;173:627.
230 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
24. Mangano DT, Layug EL, Wallace A, et al. Effect of atenolol ASRA Consensus Conference on Neuraxial Anesthesia and
on mortality and cardiovascular morbidity after noncardiac Anticoagulation). Reg Anesth Pain Med. 2003;28:172.
surgery. N Engl J Med. 1996;335:1713. 42. Nissen SE, Tuzcu EM, Schoenhagen P, et al. Statin therapy,
25. Poldermans D, Boersma E, Bax JJ, et al. The effect LDL cholesterol, C-reactive protein, and coronary artery
of bisoprolol on perioperative mortality and myocardial disease. N Engl J Med. 2005;352:29.
infarction in high-risk patients undergoing vascular surgery. 43. Poldermans D, Bax JJ, Kertai MD, et al. Statins are associated
N Engl J Med. 1999;341:1789. with a reduced incidence of perioperative mortality in
26. Fleisher LA, Corbett W, Berry C, et al. Cost-effectiveness of patients undergoing major noncardiac vascular surgery.
differing perioperative beta-blockade strategies in vascular Circulation. 2003;107:1848.
surgery patients. J Cardiothorac Vasc Anesth. 2004;18:7. 44. Lindenauer PL, Pekow P, Wang K, et al. Lipid-lowering ther-
27. Lindenauer PK, Pekow P, Wang K, et al. Perioperative apy and in-hospital mortality following major noncardiac
beta-blocker therapy and mortality after major noncardiac surgery. JAMA. 2004;291:2092.
surgery. N Engl J Med. 2005;353:349. 45. ONeil-Callahan K, Katsimaglis G, Tepper MR, et al. Statins
28. McGory M, Maggard M, Ko C. A meta-analysis of periop- decrease perioperative cardiac complications in patients
erative beta-blockade: What is the actual risk reduction? undergoing noncardiac vascular surgery: The statins for risk
Surgery. 2005;138:171. reduction in surgery (StaRRS) study. J Am Coll Cardiol. 2005;
29. Juul AB, Wetterslev J, Kofoed-Enevoldsen A, et al. The 45:336.
diabetic postoperative mortality and morbidity (DIPOM) 46. Biccard B, Sear JW, Foex P. Statin therapy: A potentially
trial: Rationale and design of a multicenter, randomized, useful peri-operative intervention in patients with cardiovas-
placebo-controlled, clinical trial of metoprolol for patients cular disease. Anaesthesia. 2005;60:1106.
with diabetes mellitus who are undergoing major noncardiac 47. Meissner A, Rolf N, Van Aken H. Thoracic epidural anesthe-
surgery. Am Heart J. 2004;147:677. sia and the patient with heart disease: Benefits, risks, and
30. Davies R, Goldberg A, Forman S, et al. Asymptomatic cardiac controversies. Anesth Analg. 1997;85:517.
ischemia pilot (ACIP) study two-year follow-up: Outcomes of 48. Bode R, Lewis K, Zarich P, et al. Cardiac outcome after
patients randomized to initial strategies of medical therapy peripheral vascular surgery. Comparison of general and
versus revascularization. Circulation. 1997;95:2037. regional anesthesia. Anesthesiology. 1996;84:3.
31. Devereaux PJ, Beattie WS, Choi PT, et al. How strong is 49. Matot I, Oppenheim-Eden A, Ratrot R, et al. Preoperative
the evidence for the use of perioperative B-blockers in non- cardiac events in elderly patients with hip fracture random-
cardiac surgery? Systematic review and meta-analysis of ized to epidural or conventional analgesia. Anesthesiology.
randomized controlled trials. Br Med J. 2005;331(7512):313. 2003;98:156.
32. Devereaux P, Yusuf S, Yang H, et al. Are the recommenda- 50. Dodds TM, Burns AK, DeRoo DB, et al. Effects of anesthetic
tions to use perioperative beta-blocker therapy in patients technique on myocardial wall motion abnormalities during
undergoing noncardiac surgery based on reliable evidence? abdominal aortic surgery. J Cardiothorac Vasc Anesth. 1997;
CMAJ. 2004;171:245. 11:129.
33. Dodds T, Stone J, Coromillas J, et al. Prophylactic nitro- 51. Chaney MA. How important is postoperative pain after
glycerin infusion during noncardiac surgery does not reduce cardiac surgery?. J Cardiothorac Vasc Anesth. 2005;19:705.
perioperative ischemia. Anesth Analg. 1993;76:705. 52. Chaney MA. Cardiac surgery and intrathecal/epidural tech-
34. Ali I, Buth K, Maitland A. Impact of preoperative intravenous niques: At the crossroads? Can J Anaesth. 2005;52:783.
nitroglycerin on in-hospital outcomes after coronary artery 53. Garnett RL, MacIntyre A, Lindsay P, et al. Perioperative
bypass grafting for unstable angina. Am Heart J. 2004; ischaemia in aortic surgery: Combined epidural/general
148:727. anaesthesia and epidural analgesia vs general anaesthesia
35. Furberg CD, Psaty BM, Meyer JV. Nifedipine: Dose-related and i.v. analgesia. Can J Anaesth. 1996;43:769.
increase in mortality in patients with coronary heart disease. 54. Beattie WS, Badner NH, Choi PT. Epidural analgesia reduces
Circulation. 1995;92:1326. postoperative myocardial infarction: A meta-analysis. Anesth
36. Wijeysundera DN, Beattie WS. Calcium channel blockers Analg. 2001;93:853.
for the reduction of cardiac morbidity following noncardiac 55. Rodgers A, Walker N, Schug S, et al. Reduction of post-
surgery: A meta-analysis. Anesth Analg. 2003;97:634. operative mortality and morbidity with epidural or spinal
37. Stevens RD, Burri H, Tramer MR. Pharmacologic myocar- anaesthesia: Results from overview of randomised trials. Br
dial protection in patients undergoing noncardiac surgery: Med J. 2000;321(7275):1493.
A quantitative systematic review. Anesth Analg. 2003;97: 56. De Hert SG, Cromheecke S, ten Broecke PW, et al. Effects
623. of propofol, desflurane and sevoflurane on recovery of
38. Butterworth J, Furberg CD. Improving cardiac outcomes myocardial function after coronary surgery in elderly high-
after noncardiac surgery. Anesth Analg. 2003;97:613. risk patients. Anesthesiology. 2003;99:314.
39. Antithrombotic Trialists Collaboration. Collaborative meta- 57. Tanaka K, Ludwig L, Kersten J, et al. Mechanisms of car-
analysis of randomised trials of antiplatelet therapy for dioprotection by volatile anesthetics. Anesthesiology. 2004;
prevention of death, myocardial infarction, and stroke in 100:707.
high risk patients. Br Med J. 2002;324(7329):71. 58. De Hert SG, ten Broecke PW, Mertens E, et al. Sevoflurane
40. Braunwald E, Antman E, Beasley J, et al. ACC/AHA 2002 but not propofol preserves myocardial function in coronary
guideline update for the management of patients with surgery patients. Anesthesiology. 2002;97:42.
unstable angina and non-ST-segment elevation myocardial 59. Garcia C, Julier K, Bestmann L, et al. Preconditioning with
infarctionsummary article: A report of the American sevoflurane decreases PECAM-1 expression and improves
College of Cardiology/American Heart Association task force one-year cardiovascular outcome in coronary artery bypass
on practice guidelines (Committee on the Management of graft surgery. Br J Anaesth. 2005;94:159.
Patients With Unstable Angina). J Am Coll Cardiol. 2002;40: 60. Jenkins DP, Pugsley WB, Alkhulaifi AM, et al. Ischaemic
1366. preconditioning reduces troponin T release in patients
41. Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia undergoing coronary artery bypass surgery. Heart. 1997;77:
in the anticoagulated patient: Defining the risks (the second 314.
C H A P T E R 15 / M Y O C A R D I A L I S C H E M I A A N D I N F A R C T I O N 231
61. Cason BA, Gamperi AK, Slocum RE, et al. Anesthetic- 77. Miller JS, Dodson TF, Salam MM, et al. Vascular compli-
induced preconditioning: Previous administration of isoflu- cations following intra-aortic balloon pump insertion. Am
rane decreases myocardial infarct size in rabbits. Surg. 1992;58:232.
Anesthesiology. 1997;87:1182. 78. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004
62. Beattie WS, Warriner CB, Etches R, et al. The addition of guideline update for coronary artery bypass graft surgery:
continuous intravenous infusion of ketorolac to a patient- A report of the American College of Cardiology/American
controlled analgetic morphine regime reduced postoperative Heart Association Task Force on Practice Guidelines (Com-
myocardial ischemia in patients undergoing elective total mittee to Update the 1999 Guidelines for Coronary Artery
hip or knee arthroplasty. Anesth Analg. 1997;84:715. Bypass Graft Surgery). Circulation. 2004;110:e340.
63. Dorman T, Clarkson K, Rosenfeld BA, et al. Effects of 79. Stein PD, Schunemann HJ, Dalen JE, et al. Antithrom-
clonidine on prolonged postoperative sympathetic response. botic therapy in patients with saphenous vein and internal
Crit Care Med. 1997;25:1147. mammary artery bypass grafts: The Seventh ACCP Confer-
64. Ellis JE, Drijvers G, Pedlow S, et al. Premedication with ence on Antithrombotic and Thrombolytic Therapy. Chest.
oral and transdermal clonidine provides safe and efficacious 2004;126(3 Suppl):600S.
postoperative sympatholysis. Anesth Analg. 1994;79:1133. 80. Flaker GC, Warnika JW, Sacks FM, et al. Cholesterol and
65. Wallace AW, Galindez D, Salahieh A, et al. Effect of clonidine Recurrent Events CARE Investigators. Pravastatin prevents
on cardiovascular morbidity and mortality after noncardiac clinical events in revascularized patients with average
surgery. Anesthesiology. 2004;101:284. cholesterol concentrations. J Am Coll Cardiol. 1999;34:
66. Wijeysundera DN, Nail JS, Beattie WS. Aplha-2 adrenergic 106.
agonists to prevent perioperative cardiovascular complica- 81. Okrainec K, Platt R, Pilote L, et al. Cardiac medical therapy
tions: A meta-analysis. Am J Med. 2003;114:742. in patients after undergoing coronary artery bypass graft
67. Ishihara M, Inoue I, Kawagoe T, et al. Effect of acute surgery: A review of randomized controlled trials. J Am Coll
hyperglycemia on the ischemic preconditioning effect of Cardiol. 2005;45:177.
prodromal angina pectoris in patients with a first anterior 82. Coriat P, Richer C, Douraki T, et al. Influence of chronic
wall acute myocardial infarction. Am J Cardiol. 2003;92:288. angiotensin-converting enzyme inhibition on anesthetic
68. Van den Berghe G, Wouters P, Weekers F, et al. Intensive induction. Anesthesiology. 1994;81:299.
insulin therapy in the critically ill patients. N Engl J Med. 83. Wilson SH, Fasseas P, Orford JL, et al. Clinical outcome of
2001;345:1359. patients undergoing non-cardiac surgery in the two months
69. Ouattara A, Lecomte P, Le Manach Y, et al. Poor intraop- following coronary stenting. J Am Coll Cardiol. 2003;42:234.
erative blood glucose control is associated with a worsened 84. Kaluza GL, Joseph J, Lee JR, et al. Catastrophic outcomes of
hospital outcome after cardiac surgery in diabetic patients. noncardiac surgery soon after coronary stenting. J Am Coll
Anesthesiology. 2005;103:687. Cardiol. 2000;35:1288.
70. Nelson AH, Fleisher LA, Rosenbaum SH. Relationship 85. Sharma AK, Ajani AE, Hamwi SM, et al. Major noncardiac
between postoperative anemia and cardiac morbidity in surgery following coronary stenting: When is it safe to
high-risk vascular patients in the intensive care unit. Crit operate? Catheter Cardiovasc Interv. 2004;63:141.
Care Med. 1993;21:860. 86. Reddy PR, Vaitkus PT. Risks of noncardiac surgery after
71. Hogue CW Jr, Goodnough LT, Monk TG. Perioperative coronary stenting. Am J Cardiol. 2005;95:755.
myocardial ischemic episodes are related to hematocrit level 87. Ong AT, van der Giessen WJ. Drug-eluting stents for
in patients undergoing radical prostatectomy. Transfusion. interventional revascularization of coronary multivessel
1998;38:924. disease. J Interv Cardiol. 2005;18:447.
72. Spiess BD. Choose one: Damned if you do/damned if you 88. Nebeker JR, Virmani R, Bennet CL, et al. Hypersensitivity
dont! Crit Care Med. 2005;33:1871. cases associated with drug-eluting coronary stents: A review
73. Spiess BD. Risks of transfusion: Outcome focus. Transfusion. of available cases from the research on adverse drug events
2004;44(12 Suppl):4S. and reports (RADAR) project. J Am Coll Cardiol. 2006;47:
74. Elmore JR, Franklin DP, Youkey JR, et al. Normothermia 175.
is protective during infrarenal aortic surgery. J Vasc Surg. 89. Auer J, Berent R, Weber T, et al. Risk of noncardiac surgery
1998;28:984. in the months following placement of drug-eluting coronary
75. Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative stent. J Am Coll Cardiol. 2004;43:713.
maintenance of normothermia reduces the incidence of 90. Satler LF. Recommendations regarding stent selection in
morbid cardiac events. A randomized clinical trial. JAMA. relation to the timing of noncardiac surgery postpercu-
1997;277:1127. taneous coronary intervention. Catheter Cardiovasc Interv.
76. Masaki E, Takinami M, Kurata Y, et al. Anesthetic manage- 2004;63:146.
ment of high-risk cardiac patients undergoing noncardiac 91. Grines C, Bonow RO, Casey DE, et al. Prevention of Prema-
surgery under the support of intraaortic balloon pump. J Clin ture Discontinuation of Dual Antiplatelet Therapy in Patients
Anesth. 1999;11:342. With Coronary Artery Stents. Circulation 2007:115;813.
CHAPTER HEART FAILURE
A
a transurethral resection of the prostate. His our population, suboptimal implementation of disease-
medical history is pertinent for diabetes mel- prevention strategies, and increase in the prevalence of
litus, hypertension, hypercholesterolemia, risk factors for cardiac disease.
and heart failure. His medications in- While the morbidity and mortality of patients with
clude furosemide, metoprolol, enalapril, and some acquired cardiovascular conditions, specifically post
digitalis. acute myocardial infarction, has declined, more patients
Currently, he has dyspnea on moderate exertion, two- are actually presenting in the endstages of their diseases
pillow orthopnea, and mild ankle edema. On physical
with ventricular dysfunction and heart failure.
examination, his blood pressure is 144/85 mm Hg, heart
Heart failure is the final common pathway for all
rate 64 bpm, and respiratory rate of 18. There is no jugular
cardiac diseases, and represents one of the biggest
venous distension or gallop, but faint bibasilar rales are
epidemiologic challenges facing industrialized countries
heard in both lung bases. His electrocardiogram (ECG)
today.
reveals nonspecific ST-T changes. A recent transthoracic
In the perioperative period, heart failure has signifi-
echocardiogram reports an ejection fraction of 30%
cant implications, both as a risk factor for adverse cardiac
and mild impaired ventricular filling. A cardiologist
events following cardiac and noncardiac surgery,2,3 and
recommends increasing the dose of diuretic, and places
as a complication from other processes such as ischemia,
him at intermediate risk for perioperative cardiac events.
hypoxia, and fluid overload. There is a general belief that
The patient undergoes the procedure a couple of days
perioperative decompensated heart failure carries a poor
later under spinal anesthesia. There are no intraoperative
prognosis.
complications, but he receives a total of 3 L of Ringers
lactate. In addition, 10 L of glycine is used for irrigation,
followed by intravenous furosemide. In the postanesthesia
care unit, the patient is dyspneic, restless, and his SpO2
is 94% on 4 L of oxygen per minute. His heart rate is What Is the Epidemiologic
100 bpm, and his blood pressure is 170/100 mm Hg. Rales
are heard in both midlung fields. A chest radiograph
Impact of Heart Failure?
shows redistribution of the vasculature and interstitial
pulmonary edema. Supplemental oxygen is increased, Improvements in the knowledge and care of conditions
and topical nitroglycerin and additional intravenous such as coronary artery disease (CAD), high blood
furosemide are administered, followed by brisk diuresis. pressure, diabetes mellitus, and alarming increasing rates
He is then admitted to the intensive care unit (ICU). There of obesity have led to an increase in the prevalence
are no electrocardiographic changes, and biomarkers for of chronic heart failure (CHF).4 For example, there is
acute myocardial infarction are negative. The patient an estimated 70% increase in heart failure due to CAD
improves, and he is discharged from the ICU 24 hours during the first decade of this century.5 In the past, the
later, following reinstitution of his chronic medications. term congestive has been widely used to describe the
syndrome, but current trends have recommended deleting
such terminology because not all patients with heart
INTRODUCTION failure exhibit signs and/or symptoms of volume overload.
In the United States and Europe, respectively, 5
Cardiovascular disease is currently the number one killer and 10 million people carry a diagnosis of heart fail-
of men and women, claiming more lives than the next four ure. Therefore, heart failure is one of the most common
232
C H A P T E R 16 / H E A R T F A I L U R E 233
conditions encountered in clinical medicine and a great 3. STAGE C: Structural disease with previous or current
public health care problem in developed countries.6,7 The signs or symptoms of heart failure
precise epidemiology of CHF is elusive because the term 4. STAGE D: Presence of end-stage heart failure that
heart failure has been used to describe a wide spec- requires specialized interventions
trum of clinical and pathophysiologic conditions, ranging
This classification is intended to complement the New
from asymptomatic systolic and diastolic dysfunction to
York Heart Association (NYHA) functional capacity. The
life-threatening acute pulmonary edema and cardiogenic
latter was initially proposed in 1928 to classify patients
shock.8 with heart disease based on severity of clinical symptoms,
CHF is not a primary condition but rather a multi- but suffers from several limitations such as being sub-
factorial clinical syndrome, resulting from a structural or jective, changing frequently over time, and not including
functional disorder that leads to poor myocardial perfor- patients in presymptomatic stages13,14 (see Fig. 16.1).
mance. In the United States, CHF occurs as a complication
from diseases such as coronary atherothrombosis, hyper-
tension, and valvular heart disease, and less commonly
from primary structural conditions such as cardiomy-
opathies.
What Is the Importance
CHF is predominately a condition of the elderly; and Impact of Acute
with a prevalence estimated at 0.8% in the sixth decade, Decompensated Heart Failure?
increasing dramatically to 10% in octogenarians.9 Ap-
proximately one half million new cases are diagnosed each
year; these numbers are expected to rise due to the current Current available information on acute decompensated
trends of aging in our population and improvement in the heart failure (ADHF) is derived from a variety of sources
treatment of other cardiovascular diseases.10 After age 40, and includes data on patients participating in clinical
the lifetime risk of developing heart failure is one in five for trials or registries and patients admitted to cardiology
both men and women.1 Therefore, even if one estimates wards and ICUs; therefore, it should be interpreted
that only a small percentage of such patients will require with caution. Because ADHF encompasses a broad
surgery, it becomes clear that this represents a formidable spectrum of clinical presentations and conditions, the
epidemiologic challenge for perioperative physicians. term acute heart failure syndromes is occasionally used
National guidelines in the United States for the di- to describe the condition; a key factor to emphasize is
agnosis and management of CHF have been established the heterogeneity of the condition and its varied clinical,
pathophysiologic, and prognostic implication. Although
and published. A new classification system was created
the terms, acute and decompensated, have been used
in 2001, emphasizing risk factors and previous conditions
to describe AHF, it is apparent that both are not the same.
that lead to heart failure, and the importance of early in-
More descriptive terms found in the literature include new
tervention before the clinical syndrome becomes manifest.
onset (de novo heart failure) and acutely exacerbated heart
This new classification scheme identifies four stages:11,12
failure.15
1. STAGE A: Patients at risk for heart failure but without For our purpose and simplicity of terms, we will
structural heart disease or signs or symptoms consider ADHF as part of the overall syndrome of AHF.
2. STAGE B: Structural disease (e.g., old myocardial in- Literature on the syndrome of perioperative ADHF is
farction, left ventricular hypertrophy) but no clinical rather scarce, and many issues remain unresolved. For
signs/symptoms of heart failure example, it is currently unknown if ADHF in the setting
Class IV
Patients with cardiac disease resulting in inability to do physical
activity without symptoms. Symptoms may occur at rest
Class III
Patients with cardiac disease that results in marked limitation of physical
activity. Comfortable at rest. Less than ordinary activity results in symptoms
Class II
Patients with cardiac disease that results in slight limitation of physical activity.
Comfortable at rest. Ordinary activity results in symptoms
Class I
Patients with cardiac disease but no limitations of physical activity
FIGURE 16.1 New York Functional Capacity and Objective Assessment. (Adapted from: Fleg JL,
Pina IL, Balady G, et al. Assessment of functional capacity in clinical and research applications: An
advisory from the Committee on Exercise, Rehabilitation, and Prevention, Council on Clinical
Cardiology, American Heart Association. Circulation. 2000;102:1591.)
234 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
of CHF is worse than that of most cancers.37 Mortality Temporal relationship and speed of clinical presenta-
remains high, with some studies reporting an overall 50% tion (acute vs. chronic)
reduction of life expectancy in elderly patients with CHF Presence of high filling pressures (decompensated vs.
compared with individuals without the condition.38 Data compensated)
from the Framingham study reveals that survival of CHF Site of cardiac damage (left side vs. right side)
patients at 5 years is only 25% for men and 38% for Hemodynamic impact (low vs. high cardiac output)
women; an overall one year survival was 57% in men and
64% in women, regardless of the etiology.39,40 The overall
prognosis depends mainly on the underlying condition SYSTOLIC HEART FAILURE
that leads to CHF in the presence of treatable precipitating
The predominant abnormality in this form of heart failure
factors and the response to therapy.
is the inability of the heart to contract normally and eject
sufficient blood in systole. The clinical presentation in-
cludes weakness, fatigue, progressive dyspnea, decreased
exercise tolerance, and tissue hypoperfusion. It is impor-
What Is the Pathophysiology tant to recognize that approximately 3% to 6% of the
of Chronic Heart Failure? adult population have left ventricular dysfunction in the
absence of clinical symptoms,44,45 and that the degree of
A simple working definition considers heart failure as a ventricular dysfunction is directly related to the risk of
multisystem disorder with abnormalities of cardiac, skele- adverse events.46,47
tal muscle, and renal dysfunction.41 Early conception
regarded heart failure as a syndrome caused by an ab-
normal pump mechanism unable to meet the metabolic DIASTOLIC HEART FAILURE
requirements of the body (hemodynamic model) or as a In patients with heart failure and preserved or normal
syndrome with abnormal salt and water retention (cardio ejection fraction (HFnlEF), the predominant physiology
renal model).42 These concepts were subsequently ex- is impaired ventricular filling with preservation of systolic
panded; it is now accepted that heart failure is a complex function. Diastolic dysfunction and subsequent decompen-
blend of structural, functional, neuroendocrine changes sated heart failure occur secondary to abnormal ventricu-
and compensatory responses that lead to abnormal ejec- lar relaxation, increase in ventricular stiffness, and reduced
tion of blood (systolic dysfunction) or a predominant compliance. The most common causes include CAD, hy-
alteration in the filling of the heart (diastolic dysfunc- pertension, amyloidosis, and hypertrophic cardiomyopa-
tion). This results in a progressive clinical syndrome thy. The importance of diastolic dysfunction and normal
of fatigue, dyspnea, fluid retention, peripheral edema, left ventricular systolic function (HFnlEF) is twofold:
and, in severe cases, pulmonary edema and hemody- (i) It is rather common (up to 40% cases of heart fail-
namic shock. The underlying condition and predispos-
ure) affecting predominantly the elderly; and (ii) mortality
ing factors may be cardiac or extracardiac, transient or
has remained unchanged for the last two decades.4850
permanent.
A key concept behind the current working theories
on the pathophysiology of heart failure is that before this
becomes clinically manifest, three main conditions must RIGHT VENTRICULAR
be met:
FAILURE
1. Intrinsic myocardial damage secondary to an insult or
index event such as myocardial infarction or volume Right ventricular dysfunction is responsible for 50% of
overload from valvular dysfunction cardiac complications and 19% of mortality post heart
2. Activation of compensatory mechanisms in an attempt transplant surgery.51 Postcardiotomy right ventricular
to correct the molecular, structural, and hemodynamic failure carries a substantial mortality.52 Similarly, right
alterations ventricular failure that occurs post acute myocardial
3. Myocardial remodeling, which modifies ventricular size, infarction carries a fivefold higher mortality than patients
shape, and function.43 without right ventricular involvement.53,54
When right ventricular failure is due to functional
When compensatory mechanisms become over- or structural pulmonary diseaseeither in acute fashion
whelmed and secondary structural and biochemical de- such as in pulmonary embolism and ARDS or in chronic
rangements occur, a downward spiral course occurs, and fashion, usually from chronic obstructive pulmonary dis-
these molecular, biochemical, and anatomic responses easeit is termed cor pulmonale.55,56 The underlying
turn into a vicious cycle that ultimately leads to dis- mechanism of cor pulmonale is an association between
ease progression and heart failure, and, despite adequate
the development of pulmonary hypertension and dysfunc-
treatment, sometimes death.
tion of the right ventricleeither hypertrophy, dilatation,
There are various clinical classifications for heart
or both.57,58
failure based on a particular parameter being measured:
Right ventricular preload, afterload, and contractility
Ventricular function (systolic and diastolic heart can be affected by acute changes in afterload and
failure) pulmonary vascular resistance, which may lead to acute
236 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
right ventricular failure and, ultimately, cardiovascular TABLE 16.2 Initial Assessment of the Chronic Heart
collapse. Owing to the interdependence of both ventricles, Failure Patient
left ventricular failure follows due to the ensuing reduction
in preload of the left ventricle, decreased cardiac output Complete history and physical examination
and coronary perfusion, thereby setting a vicious cycle of 12-lead ECG and chest radiograph
biventricular failure. Laboratory evaluation: CBC, urinalysis, electrolytes,
BUN, creatinine, glucose, lipid profile, LFTs, and thyroid-
stimulating hormone
2-D echocardiogram
HIGH OUTPUT HEART Radionuclide ventriculography may be performed to
FAILURE assess left ventricular ejection fraction and chamber
volumes
High output heart failure occurs as a result of an elevation
Coronary angiogram if the patient has angor pectoris
of left ventricular diastolic volume and pressure in the
or signs of significant ischemia.
presence of normal, or even increased, pump function.
The causes of high output heart failure are varied,
ECG, electrocardiogram; CBC, complete blood cell count; BUN, blood
but the consequences are the same: Pulmonary conges-
urea nitrogen; LFTs, liver function tests; 2-D, two dimensional.
tion with or without appropriate cardiac output to supply Adapted from Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA
the particular metabolic demand of the body. Examples 2005 guideline update for the diagnosis and management of
include the excessive administration of intravenous flu- chronic heart failure in the adult: A report of the American College
ids, low vascular resistance, or exaggerated venous blood of Cardiology/American Heart Association Task Force on Practice
flow such as in atrioventricular fistulas, cirrhosis, hy- Guidelines (Writing Committee to Update the 2001 Guidelines
perkinetic states such as sepsis, anemia, thyrotoxicosis, for the Evaluation and Management of Heart Failure). Circulation.
acromegaly, Pagets disease, and pregnancy. High output 2005;112:e154235.
HF is characterized by marked vasodilatation, increased
pulse amplitude, warm extremities, water and salt reten- congestion. Elevation of jugular venous pressure is one
tion, decreased renal blood flow, and activation of the of the most sensitive signs of right heart failure. On
neuroendocrine system.59 pulmonary auscultation, rales may be heard initially in
both bases, but as the disease progresses, it may actually
be manifest as cardiogenic pulmonary edema with frothy
(usually bloody) sputum and marked signs of respiratory
How Is Chronic Heart Failure distress. Ventricular gallops can be present and carry a
Diagnosed? poor prognosis.
Several scoring systems for the diagnosis of heart
failure have been developed. A recent and widely utilized
The diagnosis of heart failure is based on clinical find- classification, the Minnesota heart failure criteria, incor-
ings as well radiological, laboratory, and invasive and porates the left ventricular ejection fraction (LVEF) as
noninvasive cardiac tests11 (see Table 16.2). Respira- part of their criteria (see Table 16.3). This classification
tory symptomatology is usually the earliest, and most scheme, however, is geared more for epidemiologic stud-
frequent, presenting manifestation of the condition. Dys- ies of heart failure rather than as a clinical classification
pnea as a symptom dominates the clinical picture and as scheme for its diagnosis.61
the myocardial dysfunction persists, dyspneic symptoms Electrocardiographic findings in patients with CHF
progress and begin to occur with minimal exertion and are usually nonspecific. They include electric signs of
even at rest. As the situation worsens, paroxysmal noctur- chamber enlargement, intraventricular conduction de-
nal dyspnea ensues, caused by the mobilization of fluids fects, sinus tachycardia, and isolated pulmonary vascular
and increase in pulmonary venous pressure while in the congestions (PVCs). The presence of atrial fibrillation
recumbent position, and orthopnea follows, a step further
in escalation of symptoms due to the same factors, and
correlates with an increased pulmonary capillary wedge TABLE 16.3 Minnesota Heart Failure Criteria
pressure (PCWP).60 Because many patients with CHF also
have chronic obstructive pulmonary disease, it is very Dyspnea
difficult clinically to differentiate between the respiratory Pulmonary rales
and cardiovascular causes of the respiratory symptoma- Cardiomegaly on chest radiograph
tology. Progressive fatigue and generalized weakness is Interstitial pulmonary edema
commonly found in patients with CHF and is due to S3 heart sound
decreased cardiac output and decreased distal perfusion. Tachycardia (>120 beats per minute)
Findings on physical examination depend on the Left ventricular ejection fraction <40%
stage and severity of myocardial dysfunction. Usually,
patients with CHF appear ill, with pale skin and cold and Adapted from: Kim J, Jacobs DR Jr, Luepker RV, et al. Prognostic value
edematous extremities; frequently they are diaphoretic of a novel classification scheme for heart failure: The Minnesota heart
and may not be able to lie flat in bed due to pulmonary failure criteria. Am J Epidemiol. 2006;164:184193.
C H A P T E R 16 / H E A R T F A I L U R E 237
TABLE 16.4 Cutoff Values for Natriuretic Peptides LVEF <40% has been shown to be a predictor of poor
prognosis and adverse cardiac events.63
BRAIN NATRIURETIC PEPTIDE (BNP) Serum levels of brain natriuretic peptide (BNP), a
<100 pg/mL: CHF highly unlikely peptide secreted from both ventricles in response to
100400 pg/mL: CHF possible volume or pressure expansion, and its biologic inactive
>400 pg/mL: CHF is highly likely precursor (NT-proBNP) have shown to be consistently
NT-PROBNP associated with decompensated states and poor outcome
<300 pg/mL: CHF highly unlikely in patients with heart failure.6466
300450 pg/mL: <50 y As with other biomarkers, there are false results that
300900 pg/mL: >50 y possible can be misleading in some circumstances. BNP increases
>450 pg/mL: <50 y normally with age, female gender, chronic renal failure,
>900 pg/mL: >50 y highly unlikely for CHF and obesity. In some patients with acute cardiogenic
pulmonary edema, BNP may be normal due to a delay
CHF, chronic heart failure; NT-proBNP, N-terminal probrain natri- in the release of BNP from the left ventricular wall. Cutoff
uretic peptide. concentrations of BNP in normal and CHF states are
Adapted with data from: Nohria A, Lewis E, Stevenson LW. Medical shown in Table 16.4.
management of advanced heart failure. JAMA. 2002;287:628640
and Felker GM, OConnor CM. Inotropic therapy for heart failure: An
evidence-based approach. Am Heart J. 2001;142:393401.
How Is Chronic Heart Failure
Treated?
and left bundle branch block are poor prognostic ECG
findings.62 On chest radiograph, common findings include No single modality or agent exists for the management
cardiomegaly, and pulmonary congestion. Echocardio- of the patient with CHF, and recommended guidelines
graphy is extremely useful, not only as a method of include combination of agents and a multidisciplinary ap-
establishing the severity of myocardial dysfunction, but proach. Current therapeutic management is based on the
also to establish the cause and determine the predomi- American College of Cardiology (ACC)/AHA guidelines
nant pathophysiologic mechanism (systolic vs. diastolic for CHF and the stage of progression of the condition
dysfunction), which has prognostic and therapeutic im- (see Fig. 16.2).11 Therefore, therapeutic goals vary from
plications. A common echocardiographic measurement is prevention and reversal of myocardial remodeling to man-
the resting LVEF. Although not universally accepted, an agement of low output syndrome and cardiogenic shock.
Hospice
VAD, Transplantation
Stage D
Inotropes
Stage B
ACE inhibitors or ARBs (angiotensin receptor blocker) + b-blockers in some patients
Stage A
Treat hypertension, diabetes, dyslipidemia
Risk factor reduction, patient and family education
FIGURE 16.2 Stepwise approach to the evaluation and management of heart failure. (Adapted
from: Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 guideline update for the diagnosis and
management of chronic heart failure in the adult: A report of the American College of
Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to
Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). Circulation.
2005;112:e154.)
238 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
The most common presentation of perioperative heart 1. PROFILE A (DRY AND WARM): Patients in this category
failure involves a patient, with chronic, preexisting left are characterized by adequate perfusion and normal
ventricular dysfunction, who suffers an acute decompen- filling pressures. In this profile, cardiac output is ad-
sation. This is manifested by increased filling pressures, equate and PVC/edema is absent. Therapy is adjusted
with or without decreased cardiac output, and tissue hy- to maintain normal volume status and avoid precipi-
poperfusion. It is important to emphasize that the etiology tating factors. Evaluation should include a search for
of perioperative ADHF is frequently multifactorial when other causes of dyspnea, the most common presenting
compared to medical exacerbations, with various el- symptom of ADHF.
ements occurring simultaneously or in sequence (see 2. PROFILE B (WET AND WARM): These patients are charac-
Table 16.5). Under special circumstances, perioperative terized by an acute increase in filling pressure, leading
heart failure may not involve the left heart but rather to PVCs and edema; however, tissue perfusion is nor-
the right ventricle. Diagnosis and treatment of such a mal or only slightly impaired. The predominant mech-
condition are discussed in detail in Chapter 17. anism is that of impaired diastolic function. Patients
Acute heart failure symptoms develop rapidly and in this category may or may not have decreased sys-
vary in intensity from effort fatigue to shocklike signs tolic function, but cardiac output and systemic blood
and symptoms, neurologic manifestations, cyanosis, low pressure are usually maintained.
blood pressure, oliguria, and death. The diagnostic strate- 3. PROFILE L (DRY AND COLD): These patients exhibit low
gies are similar than those for the chronic state. Clinical cardiac output without clinical evidence of elevated
evaluation, electrocardiography, echocardiography, chest filling pressures. Determining optimal preload guided
radiographs, and determination of serum markers should by frequent reassessment of cardiac output and filling
C H A P T E R 16 / H E A R T F A I L U R E 239
FIGURE 16.3 Clinical mode and severity of acute heart failure. FIGURE 16.4 Hemodynamic assessment and therapeutic
(Modified from: Forrester JS, Diamond GM, Swan HJ. rational for acute heart failure. (Modified from: Forrester JS,
Correlative classification of clinical and hemodynamic function Diamond GM, Swan HJ. Correlative classification of clinical and
after acute myocardial infarction. Am J Cardiol. 1977;39:137 and hemodynamic function after acute myocardial infarction. Am J
Nohria A, Tsang SW, Fang JC, et al. Clinical assessment identifies Cardiol. 1977;39:137 and Nohria A, Tsang SW, Fang JC, et al.
hemodynamic profiles that predict outcomes in patients Clinical assessment identifies hemodynamic profiles that predict
admitted with heart failure. J Am Coll Cardiol. 2003;41:1797.) outcomes in patients admitted with heart failure. J Am Coll
Cardiol. 2003;41:1797.)
IV, intravenous; CBF, cerebral blood flow; CHF, chronic heart failure; SAH, subarachnoid hemorrhage; ICP, intracranial pressure
Unlike -adrenergic agents, PDEIII inhibitors do 3. PROFILE L (DRY AND COLD): In patients with severe sys-
not require interaction with membrane receptors, and tolic dysfunction or those with hypotension, inotropic
their use is associated with minimal increases in agents may be required to accommodate intravenous
myocardial oxygen consumption.70 These agents also fluid therapy.
produce pulmonary and systemic vasodilatation, and 4. PROFILE C (WET AND COLD): The mainstay of treatment
hypotension may result, particularly in patients with includes the use of inotropes and perhaps intra-aortic
decreased preload. Preliminary evidence suggests that counterpulsation. Frequently, the combination of in-
PDEIII inhibitors increase cerebral blood flow and may otropes with different mechanisms of action (e.g., a
provide beneficial effects against cerebral vasospasm PDEIII inhibitor and a -adrenergic agent) produces
due to their direct effects on cerebral vessels.71 better results, allowing for lower doses of each agent,
C H A P T E R 16 / H E A R T F A I L U R E 241
and thereby minimizing side effects.72 Frequently, high 2. Froehlich JB, Eagle KA. Anaesthesia and the cardiac patient:
doses of adrenergic agents with - receptor properties The patient versus the procedure. Heart. 2002;87:91.
are necessary if patients are on preexisting -blockers 3. Kertai MD, Klein J, van Urk H, et al. Cardiac complications
after elective major vascular surgery. Acta Anaesthesiol
or if PDEIII inhibitors are not indicated because of
Scand. 2003;47:643.
preexisting hypotension. In some cases, intra-aortic 4. Butler J. More risk factors affecting heart failure outcomes!
counterpulsation is useful because it improves systolic Time for hope or despair? J Am Coll Cardiol. 2005;46:
function by decreasing the impedance to left ventricu- 1027.
lar ejection and enhances diastolic coronary perfusion. 5. Bonneux L, Barendregt JJ, Meeter K, et al. Estimating clinical
Finally, in a very small subset of cases, placement of a morbidity due to ischemic heart disease and congestive heart
ventricular assist device may be considered. The deci- failure: The future rise of heart failure. Am J Public Health.
sion of foremost importance in these circumstances is 1994;84:20.
whether to place a temporal device (e.g., expectation of 6. Aronow WS, Ahn C. Frequency of congestive heart failure
cardiac recovery or as a bridge to transplantation) or in older persons with prior myocardial infarction and serum
low-density lipoprotein cholesterol > or = 125 mg/dl treated
permanent mechanical support.
with statins versus no lipid-lowering drug. Am J Cardiol.
2002;90:147.
7. Kannel WB. Incidence and epidemiology of heart failure.
CONCLUSION Heart Fail Rev. 2000;5:167.
8. Zannad F, Adamopoulos C, Mebazaa A, et al. The challenge
Perioperative heart failure is a significant complication of acute decompensated heart failure. Heart Fail Rev. 11:135,
that has an adverse impact on mortality, length of 2006.
hospital stay, and cost. Furthermore, owing to the 9. Owen A. Life expectancy of elderly and very elderly patients
exponential increase in patients at risk, it is imperative with chronic heart failure. Am Heart J. 2006;151:1322, e1e4.
that perioperative clinicians perfect their diagnostic and 10. Ahmed A. American College of Cardiology/American Heart
monitoring skills, because the clinical presentation can Association chronic heart failure evaluation and manage-
ment guidelines: Relevance to the geriatric practice. J Am
vary from mild exacerbation to florid pulmonary edema
Geriatr Soc. 2003;51:123.
and cardiogenic shock. Although strong outcome data 11. Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005
is lacking, we believe that prompt diagnosis, combined guideline update for the diagnosis and management of
with early and aggressive treatment, will lead to a better chronic heart failure in the adult: A report of the American
prognosis. College of Cardiology/American Heart Association Task
Force on Practice Guidelines (Writing Committee to Update
the 2001 Guidelines for the Evaluation and Management of
Heart Failure). Circulation. 2005;112:e154.
KEY POINTS 12. Hunt SA, Baker DW, Chin MH, et al. ACC/AHA guidelines for
the evaluation and management of chronic heart failure in
1. Heart failure in the perioperative setting is considered the adult: Executive summary. A report of the American
a strong predictor for cardiac complications. College of Cardiology/American Heart Association Task
2. Owing to the increased population with CHF, the Force on Practice Guidelines (Committee to Revise the 1995
incidence of perioperative ADHF is expected to rise. Guidelines for the Evaluation and Management of Heart
3. Despite the lack of strong data on diagnosis and Failure). Circulation. 2001;104:2996.
management of perioperative ADHF, clinicians must 13. Miller-Davis CS, Marden S, Leidy NK. The New York Heart
Association Classes and functional status: What are we really
be knowledgeable of the risk factors and diagnostic
measuring? Heart Lung. 2006;35:217.
modalities. 14. van den Broek SA, van Veldhuisen DJ, de Graeff PA, et al.
4. Perioperative ADHF requires prompt diagnosis and Comparison between New York Heart Association classi-
therapy, and successful management frequently re- fication and peak oxygen consumption in the assessment
quires a multidisciplinary approach. of functional status and prognosis in patients with mild to
5. Presently, perioperative ADHF is considered part of moderate chronic congestive heart failure secondary to ei-
AHF syndrome, and diagnostic and recommended ther ischemic or idiopathic dilated cardiomyopathy. Am J
therapeutic options are similar as for nonsurgical Cardiol. 1992;70:359.
settings. 15. Siirila-Waris K, Lassus J, Melin J, et al. Characteristics,
6. The use of a hemodynamic model to guide therapy outcomes, and predictors of 1-year mortality in patients
hospitalized for acute heart failure. Eur Heart J. 2006;27:
provides clinicians with reasonable therapeutic goals
3011.
and prognostic indicators
16. Fonarow GC, Corday E. Overview of acutely decompensated
7. Perioperative echocardiography provides invaluable congestive heart failure (ADHF): A report from the ADHERE
information in many patients, and should always be Registry. Heart Fail Rev. 2004;9:179.
considered in the presence of hypotension. 17. Hugli O, Braun JE, Kim S, et al. United States emergency
department visits for acute decompensated heart failure,
1992 to 2001. Am J Cardiol. 2005;96:1537.
REFERENCES
18. Cuffe MS, Califf RM, Adams KF Jr, et al. Short-term intra-
1. Rosamond W, Flegal K, Friday G, et al. Heart disease and venous milrinone for acute exacerbation of chronic heart
stroke statistics2007 update: A report from the American failure: A randomized controlled trial. JAMA. 2002;287:1541.
Heart Association Statistics Committee and Stroke Statistics 19. Zannad F. Acute heart failure syndromes: The Cinderella
Subcommittee. Circulation. 2007;115:e69. of heart failure research. Eur Heart J. 2005;7(Suppl B):b8.
242 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
20. DiDomenico RJ, Park HY, Southworth MR, et al. Guidelines 36. Hernandez AF, Whellan DJ, Stroud S, et al. Outcomes in
for acute decompensated heart failure treatment. Ann heart failure patients after major noncardiac surgery. J Am
Pharmacother. 2004;38:649. Coll Cardiol. 2004;44:1446.
21. Madias JE. Killip and Forrester classifications: Should 37. Stewart S, MacIntyre K, Hole DJ, et al. More malignant
they be abandoned, kept, reevaluated, or modified? Chest. than cancer? Five-year survival following a first admission
2000;117:1223. for heart failure. Eur J Heart Fail. 2001;3:315.
22. Cotter G, Moshkovitz Y, Milovanov O, et al. Acute heart 38. Ho KK, Anderson KM, Kannel WB, et al. Survival after the
failure: A novel approach to its pathogenesis and treatment. onset of congestive heart failure in Framingham Heart Study
Eur J Heart Fail. 2002;4:227. subjects. Circulation. 1993;88:107.
23. Zannad F, Mebazaa A, Juilliere Y, et al. Clinical profile, con- 39. Ahmed A, Aronow WS, Fleg JL. Predictors of mortality and
temporary management and one-year mortality in patients hospitalization in women with heart failure in the Digitalis
with severe acute heart failure syndromes: The EFICA study. Investigation Group trial. Am J Ther. 2006;13:325.
Eur J Heart Fail. 2006;8:697. 40. Ho KK, Pinsky JL, Kannel WB, et al. The epidemiology of
24. Fonarow GC, Adams KR Jr, Abraham WT, et al. Risk strat- heart failure: The Framingham Study. J Am Coll Cardiol.
ification for in-hospital mortality in acutely decompensated 1993;22(4 Suppl A):6A.
heart failure: Classification and regression tree analysis. JAM. 41. Jackson G, Gibbs CR, Davies MK, et al. ABC of heart failure.
293:572, 2005. Pathophysiology. Br Med J. 2000;320(7228):167.
25. Kumar R, McKinney WP, Raj G, et al. Adverse cardiac events 42. Mann DL. Mechanisms and models in heart failure: A com-
after surgery: Assessing risk in a veteran population. J Gen binatorial approach. Circulation. 1999;100:9991008.
Intern Med. 2001;16:507. 43. Bristow MR. Why does the myocardium fail? Insights from
26. Nohria A, Tsang SW, Fang JC, et al. Clinical assessment basic science. Lancet. 1998;352(Suppl 1):SI8.
identifies hemodynamic profiles that predict outcomes in 44. Davies M, Hobbs F, Davis R, et al. Prevalence of left-
patients admitted with heart failure. J Am Coll Cardiol. ventricular systolic dysfunction and heart failure in the
2003;41:1797. Echocardiographic Heart of England Screening study: A pop-
27. Goldman L, Caldera DL, Southwick FS, et al. Cardiac risk ulation based study. Lancet. 2001;358(9280):439.
factors and complications in non-cardiac surgery. Medicine 45. Rodeheffer RJ. Epidemiology and screening of asymptomatic
(Baltimore). 1978;57:357. left ventricular dysfunction. J Card Fail. 2002;8(6 Suppl):
28. Lee TH, Marcantonio ER, Mangione CM, et al. Derivation S253.
and prospective validation of a simple index for prediction 46. Gavazzi A, Berzuini C, Campana C, et al. Value of right ven-
of cardiac risk of major noncardiac surgery. Circulation. tricular ejection fraction in predicting short-term prognosis
1999;100:1043. of patients with severe chronic heart failure. J Heart Lung
29. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2006 Transplant. 1997;16:774.
guideline update on perioperative cardiovascular evaluation 47. Richards AM, Nicholls MG, Espiner EA, et al. B-type
for noncardiac surgery: Focused update on perioperative natriuretic peptides and ejection fraction for prognosis after
beta-blocker therapy: A report of the American College myocardial infarction. Circulation. 2003;107:2786.
of Cardiology/American Heart Association Task Force on 48. Nishimura R. Understanding Diastolic Heart Failure: The
Practice Guidelines (Writing Committee to Update the 2002 tip of the iceberg. J Am Coll Cardiol. 2007;49:695.
Guidelines on Perioperative Cardiovascular Evaluation for 49. Hans Persson M, Lonn E, Edner M, et al. Diastolic dysfunc-
Noncardiac Surgery): Developed in collaboration with the tion in heart failure with preserved systolic function: Need
American Society of Echocardiography, American Society for objective evidence results from the CHARM echocardio-
of Nuclear Cardiology, Heart Rhythm Society, Society of graphic substudycharmes. J Am Coll Cardiol. 2007;49:687.
Cardiovascular Anesthesiologists, Society for Cardiovascular 50. Swedberg K, Cleland J, Dargie H, et al. Guidelines for the
Angiography and Interventions, and Society for Vascular diagnosis and treatment of chronic heart failure: Executive
Medicine and Biology. Circulation. 2006;113:2662. summary (update 2005): The Task Force for the Diagnosis
30. Auerbach A, Goldman L. Assessing and reducing the cardiac and Treatment of Chronic Heart Failure of the European
risk of noncardiac surgery. Circulation. 2006;113:1361. Society of Cardiology. Eur Heart J. 2005;26:1115.
31. Auerbach AD. Perioperative cardiac risk reduction: Doing it 51. Stobierska-Dzierzek B, Awad H, Michler RE. The evolving
right. Cleve Clin J Med. 2006;73(Suppl 1):S25. management of acute right-sided heart failure in cardiac
32. Park KW. Preoperative cardiac evaluation. Anesthesiol Clin transplant recipients. J Am Coll Cardiol. 2001;38:923.
North America. 2004;22:199, v. 52. Moazami N, Pasque MK, Moon MR, et al. Mechanical sup-
33. Eagle KA. American College of Cardiology/American Heart port for isolated right ventricular failure in patients after car-
Association. Surgical patients with heart disease: Summary diotomy. J Heart Lung Transplant. 2004;23(12):13711375.
of the ACC/AHA guidelines. Am Fam Physician. 1997;56:811. 53. Mehta SR, Eikelboom JW, Natarajan MK, et al. Impact of
34. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide- right ventricular involvement on mortality and morbidity
line update for perioperative cardiovascular evaluation for in patients with inferior myocardial infarction. J Am Coll
noncardiac surgeryexecutive summary. A report of the Cardiol. 2001;37:37.
American College of Cardiology/American Heart Association 54. Berger PB, Ryan TJ. Inferior myocardial infarction. High-
Task Force on Practice Guidelines (Committee to Update the risk subgroups. Circulation. 1990;81:401.
1996 Guidelines on Perioperative Cardiovascular Evaluation 55. MacNee W. Pathophysiology of cor pulmonale in chronic
for Noncardiac Surgery). Anesth Analg. 2002;94:1052. obstructive pulmonary disease. Part two. Am J Respir Crit
35. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide- Care Med. 1994;150:1158.
line update for perioperative cardiovascular evaluation for 56. MacNee W. Pathophysiology of cor pulmonale in chronic
noncardiac surgeryexecutive summary. A report of the obstructive pulmonary disease. Part one. Am J Respir Crit
American College of Cardiology/American Heart Association Care Med. 1994;150:833.
Task Force on Practice Guidelines (Committee to Update the 57. Via G, Braschi A. Pathophysiology of severe pulmonary
1996 Guidelines on Perioperative Cardiovascular Evaluation hypertension in the critically ill patient. Minerva Anestesiol.
for Noncardiac Surgery). Circulation. 2002;105:1257. 2004;70:233.
C H A P T E R 16 / H E A R T F A I L U R E 243
58. Weitzenblum E. Chronic cor pulmonale. Heart. 2003;89:225. 65. Bettencourt P. NT-proBNP and BNP: Biomarkers for heart
59. Anand IS, Chandrashekhar Y, Ferrari R, et al. Pathogenesis failure management. Eur J Heart Fail. 2004;6:359.
of oedema in chronic severe anaemia: Studies of body water 66. Doust JA, Glasziou PP, Pietrzak E, et al. A systematic review
and sodium, renal function, haemodynamic variables, and of the diagnostic accuracy of natriuretic peptides for heart
plasma hormones. Br Heart J. 1993;70:357. failure. Arch Intern Med. 2004;164:1978.
60. Stevenson LW, Perloff JK. The limited reliability of physical 67. Nohria A, Lewis E, Stevenson LW. Medical management of
signs for estimating hemodynamics in chronic heart failure. advanced heart failure. JAMA. 2002;287:628.
JAMA. 1989;26:884. 68. Felker GM, OConnor CM. Inotropic therapy for heart
61. Kim J, Jacobs DR Jr, Luepker RV, et al. Prognostic value of a failure: An evidence-based approach. Am Heart J. 2001;142:
novel classification scheme for heart failure: The Minnesota 393.
heart failure criteria. Am J Epidemiol. 2006;164:184. 69. Nieminen MS, Bohm M, Cowie MR, et al. Executive
62. Murkofsky RL, Dangas G, Diamond JA, et al. A prolonged summary of the guidelines on the diagnosis and treatment of
QRS duration on surface electrocardiogram is a specific acute heart failure: The Task Force on Acute Heart Failure
indicator of left ventricular dysfunction [see comment]. J Am of the European Society of Cardiology. Eur Heart J. 2005;26:
Coll Cardiol. 1998;32:476. 384.
63. Marantz PR, Tobin JN, Wassertheil-Smoller S, et al. The 70. Rettig GFJ, Schieffer HJ. Acute effects of intravenous
relationship between left ventricular systolic function and milrinone in heart failure. Eur Heart J. 1989;10(Suppl C):39.
congestive heart failure diagnosed by clinical criteria. 71. Drexler H, Hoing S, Faude F, et al. Central and regional vas-
Circulation. 1988;77:607. cular hemodynamics following intravenous milrinone in the
64. Maisel A, Hollander JE, Guss D, et al. Primary results of conscious rat: Comparison with dobutamine. J Cardiovasc
the Rapid Emergency Department Heart Failure Outpatient Pharmacol. 1987;9:563.
Trial (REDHOT). A multicenter study of B-type natriuretic 72. Royster RL, Butterworth JF IV, Prielipp RC, et al. Com-
peptide levels, emergency department decision making, and bined inotropic effects of amrinone and epinephrine after
outcomes in patients presenting with shortness of breath. cardiopulmonary bypass in humans. Anesth Analg. 1993;77:
J Am Coll Cardiol. 2004;44:1328. 662.
CHAPTER ACUTE PULMONARY
17
HYPERTENSION
CASE SUMMARY counterparts, with a wall thickness only one third that of
the aorta due to a thinner media and fewer smooth muscle
67-year-old man with coronary artery dis- cells (SMCs).2
A
ease is scheduled for myocardial resvas- The pulmonary circulation is a low pressure, low
cularization under cardiopulmonary bypass resistance circulation capable of handling large increases
(CPB). He has a history of diabetes mellitus in flow. Normal PAP is approximately 25/8 mm Hg, with a
and requires daily subcutaneous NPH in- mean pulmonary arterial pressure (MPAP) of 15 mm Hg,
sulin. Intraoperative monitoring includes a or approximately one sixth of the systemic circulation,
pulmonary artery catheter and transesophageal echocar- and is maintained even with several-fold increase in flow.
diography (TEE). Surgical revascularization is uneventful, Although inherently imperfect, the following formula is
and separation from CPB is accomplished without the use frequently used in the clinical arena to derive pulmonary
of inotropic agents. TEE findings shows good biventricu- vascular resistance (PVR):
lar function, and his pulmonary artery pressure (PAP) is
25/16 mm Hg. The initiation of intravenous protamine is PVR = (MPAP LAP)/CO
associated with a sudden increase of PAP (65/40 mm Hg)
and severe systemic hypotension (50/30 mm Hg). TEE where PVR = pulmonary vascular resistance, MPAP =
demonstrates a dilated and hypokinetic right ventricle mean pulmonary artery pressure, LAP = left atrial pres-
(RV) with severe tricuspid regurgitation. The infusion sure (or its surrogate, pulmonary artery occlusion pres-
of protamine is stopped, and CPB is reinstituted. A bo- sure), and CO = cardiac output. The normal values are
lus of intravenous epinephrine, followed by an infusion, 1 to 1.5 Wood units or 80 to 120 dynes per cm3 (multiply-
restores systemic blood pressure and improves right ven- ing Wood units by 80). Other measurements include the
tricular function. Milrinone and nitroglycerin are initiated transpulmonary gradient (TPG) (MPAPLAP; nL 5 to
and adjusted to avoid systemic hypotension. The patient 10 mm Hg) and the mean systemic arterial-to-pulmonary
experiences a gradual recovery over several minutes, with artery pressure ratio (normal 6 mm Hg). PAP increases
significant improvement in right ventricular function and minimally with age, exercise, or high output states (e.g.,
decreased PAP within baseline values. Subsequently, he pregnancy, anemia). However, the healthy lung has sev-
is weaned from CPB without any incident. After trans- eral compensatory mechanisms to attenuate an increase
fusion of multiple blood products and vigorous surgical in intravascular pressure.
hemostasis, he is transferred to the intensive care unit Of foremost importance is the presence of a large,
where he exhibits an uncomplicated recovery. recruitable vasculature, which is responsible for main-
taining a constant PAP and PVR in response to increased
pulmonary flow during exercise. Additionally, active pul-
monary vasodilation occurs, which is regulated by endoge-
What Is the Normal Physiology nous nitric oxide (NO) and prostacyclin (PGI2).3 When
pulmonary endothelial cells are exposed to chemical or
of the Pulmonary Circulation? mechanical stress, they synthesize NO and prostaglandin
I2 (PGI2), which then diffuse toward SMCs. Nitric oxide
The lungs are the only organs that receive the entire car- mediates an increase in SMC cyclic guanosine monophos-
diac output (CO), and are able to accommodate this high phate (cGMP), whereas PGI2 increases cyclic adenosine
flow by their low resistance.1 Anatomically, pulmonary monophosphate (cAMP). Both intracellular mediators are
arterial vessels have larger diameters than their systemic known to facilitate smooth muscle relaxation.
244
C H A P T E R 17 / A C U T E P U L M O N A R Y H Y P E R T E N S I O N 245
MPAP, mean pulmonary arterial pressure; LAP, left atrial pressure; TPG, transpulmonary gradient.
246 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 17.2 WHO Classification of Chronic Pulmonary endothelial thickening and hyperplasia, causing an in-
Hypertension crease in PVR. Vascular endothelial cell dysfunction or
injury leads to an imbalance between vasodilators and
Pulmonary Arterial Hypertension (PAH) vasoconstrictors, prothrombotic and antithrombotic ele-
ments, growth inhibitor, and mitogenic factors.10
Idiopathic (IPAH) Idiopathic pulmonary hypertension (IPH), formerly
Familial (FPAH) known as primary pulmonary hypertension, is extremely
Associated (APAH): rare, associated with very high PAPs (>60 mm Hg), and
Collagen vascular disease carries a poor prognosis.9
Congenital systemic-to-pulmonary shunts Secondary pulmonary hypertension is more common
Portal hypertension and, in general, is less severe than IPH. It is more likely to
HIV infection be encountered in the perioperative setting.11 Symptoms
Drugs and toxins and signs usually do not reflect the underlying disease but
Other rather the impact of the abnormal pulmonary vasculature
Associated with significant venous or capillary on the RV, pulmonary pressures, and oxygen saturation.12
involvement: Secondary pulmonary hypertension can be divided into
Pulmonary veno-occlusive disease (PVOD) the following three categories:
Pulmonary capillary hemangiomatosis (PCH)
Persistent pulmonary hypertension of the 1. Pulmonary venous hypertension
newborn 2. Pulmonary vascular obstruction
3. Hypoxemia
Pulmonary Hypertension with Left Heart Disease Pulmonary venous pressure elevation usually results
from left ventricular pathology (e.g., mitral valve disease,
Left-sided atrial or ventricular heart disease diastolic dysfunction) or pulmonary veno-occlusive dis-
Left-sided valvular heart disease ease (e.g., drug related, invasive tumors). With pulmonary
vascular obstruction, one has to differentiate between
Pulmonary Hypertension Associated with Lung
those disease processes from parenchymal disease (e.g.,
Diseases and/or Hypoxemia
interstitial lung disease, cystic fibrosis) and those without
Chronic obstructive pulmonary disease (e.g., pulmonary embolus, vasculopathy). Lastly, hypox-
Interstitial lung disease emia can be a result of living at high altitude, having an
Sleep-disordered breathing underlying central neurologic ventilatory problem, or hav-
Alveolar hypoventilation disorders ing airway disease (e.g., asthma, obstructive sleep apnea).
Chronic exposure to high altitude Often, secondary pulmonary hypertension is multifacto-
Development abnormalities rial rather than the result of one isolated disease process.
eccentric hypertrophy. Further progression of the disease, collapse. The topic of pulmonary embolism is discussed
or the presence of a stimulus triggering acute pulmonary in detail in Chapter 13.
vasoconstriction, may result in decompensated RV failure. Those undergoing CPB represent a unique subset
of patients, in whom the development of pulmonary
hypertension is actually a predictor of long-term mor-
tality and myocardial infarction.11 Organ damage after
What Is the Etiology of Acute CPB is produced by two closely related mechanisms:
Systemic inflammatory response syndrome (SIRS) and
Pulmonary Hypertension? ischemia/reperfusion injury. The SIRS is triggered by
the exposure of blood to the large synthetic contact
The pulmonary vasculature is composed 40% of capil- surfaces of the extracorporeal circulation, whereas is-
laries, 50% of arteries, and 10% of veins, whereby only chemia/reperfusion injury triggers effects mainly in the
arteries and veins are able to actively dilate and con- heart and lungs.1416 The pulmonary endothelium has a
strict SMCs. Physiologically, the release of endothelial central role in the pathophysiology of APH.17 The en-
cell-mediated vasodilators (NO, PGI2) maintains a low dothelial cells modulate pulmonary vascular tone through
PVR. According to Poiseuilles law, PVR is inversely re- the release of endothelium-derived, constricting fac-
lated to the fourth power of the radius of the pulmonary tor (EDCF) and endothelium-derived relaxing factors
artery or arterioles. Therefore, even mild structural or (EDRFs), including endothelin-1, NO, and prostacyclin.18
functional narrowing of the pulmonary vasculature can Injury will cause an imbalance between these factors and
increase PAP. Mediator-induced vasoconstriction or a loss an increased production of the very potent vasoconstric-
of physiologic vasodilation will increase PVR and subse- tor, endothelin-1.
quently PAP. Important etiologic factors responsible for Another mechanism of acute pulmonary vasocon-
perioperative APH are shown in Table 17.3 striction, first described by Euler and Liljestrand in 1946,
Thrombi can be microscopic or macroscopic in size. is through activation of the physiologic hypoxic pul-
Microscopic thrombi are a result of an imbalance be- monary vasoconstriction. This reflex, aimed at improving
tween procoagulatory and anticoagulatory factors, often oxygenation, aids in matching ventilation with perfusion
a consequence of endothelial damage or systemic coagu- by redirecting blood flow from underventilated areas to
lation. Sepsis and the acute respiratory distress syndrome regions with relatively higher ventilation.19 The subse-
(ARDS) are classic examples of microscopic thrombi, in quent increase in PVR by pulmonary arteriolar SMC is
which the increase in PAP is more gradual. This is in sharp mediated by oxygen-sensitive, voltage-dependent potas-
contrast to macroscopic thrombi or emboli that can cause sium (K+ ) channels.20 Their inhibition by decreased PaO2
acute catastrophic obstruction and vasoconstriction, lead- blocks outward K+ currents, with subsequent calcium en-
ing to acute right ventricular failure and cardiovascular try and vasoconstriction. The main determinant of this
mechanism is the PaO2 .
Endothelial cell swelling or narrowing can occur
TABLE 17.3 Etiology of Perioperative Pulmonary as a result of direct trauma or indirectly through is-
Hypertension chemia/reperfusion injury, infection, or even hemorrhagic
shock.21 Luminal narrowing can happen acutely, sec-
Embolism ondary to the disease process, or gradually, with vascular
Clots remodeling in a chronic stage associated with intimal
Deep vein thrombosis thickening.
Opening tourniquet Increased alveolar pressure vessels (e.g., high levels
Labor and pregnancy of positive end-expiratory pressure [PEEP]) can lead to
Fat (bone cement) increased PVR if it exceeds intravascular pressure due to
Air embolism mechanical compression of alveolar vessels.
Iatrogenic Protamine can rarely be associated with catastrophic
Negative venous pressure (e.g., sitting craniotomy) pulmonary vasoconstriction.2224 During this reaction,
CO2 embolism patients experience a precipitous increase in PAP and
Laparoscopy right ventricular dysfunction within a few minutes of
Amniotic fluid the injection.25 Human and animal studies have shown
Drug-induced pulmonary vasoconstriction activation of complement (C5a) and the cyclooxygenase
Protamine pathway (thromboxane A2 and B2).2628 Additionally,
Ischemia-reperfusion thrombocytes and leukocytes become trapped in the
Congenital heart disease surgery pulmonary circulation. Interestingly, this reaction is not
Aortic declamping observed if protamine is given in the absence of heparin.
Post cardiopulmonary bypass Some anesthetic agents (NO, atracurium, ketamine,
Diffuse lung injury sodium thiopental) are known to increase PAP. Although
Loss of pulmonary parenchyma these in general do not pose a risk to a healthy patient,
they can have clinically significant effects in patients with
248 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
b agonists
Prostacyclin
Prostaglandin E1
receptors Extracellular
PKA
Contractility
FIGURE 17.2 cAMP pathway on a pulmonary vascular smooth muscle cell. ATP, adenosine
triphosphate; cAMP, cyclic adenosine monophosphate; AMP, adenosine monophosphate; PKA,
protein kinase A.
parenchyma. Therefore, monitoring for exhaled concen- Similarly, it appears that intracellular phosphodi-
trations of NO2 is routinely performed. Approximately esterase type I (PDE1) modulates vascular tone and the
70% of NO is excreted as nitrate by the kidneys after development of tolerance to NO-releasing drugs in the sys-
48 hours.41 temic circulation. Evgenov et al. have demonstrated that
Sildenafil, a PDEV inhibitor that increases endoge- the selective inhibition of PDE1 augments the therapeutic
nous cGMP concentrations, has recently been approved effects of iNO in an animal model of acute lung injury.44
for the treatment of chronic pulmonary hypertension. It
has also been studied in models of APH, in which it
was able to lower PAPs without causing systemic hy-
potension. While other vasodilators might be effective in What Are the Important
lowering PAPs in hemodynamically stable patients, only Anesthetic Management
iNO, inhaled prostacyclin, and the PDEV inhibitor, silde-
nafil, have shown to target predominantly the pulmonary Principles for the Patient with
circulation.42,43 A partial list of drugs that have been uti- Chronic Pulmonary
lized to treat APH in the perioperative setting is given in
Table 17.4. Hypertension?
Recently, endothelin-receptor antagonists have been
studied and were found to be a useful tool in the setting Pulmonary vessels constrict with hypoxia and hyper-
of APH, especially when used concomitantly with NO.32 capnia, and relax with hyperoxia and hypocapnia.
Nesiritide
Extracellular
Membrane Guanylyl cyclase
i NO GTP Intracellular
Phosphodiesterase V
NO donors cGMP 5 AMP
Soluble Mg++ Inhibitors (sildenafil)
Phosphorylated
Guanylyl cyclase Inactive protein kinase
active kinase
PKG
Contractility
FIGURE 17.3 The cGMP pathway on a pulmonary smooth muscle cell. GTP, guanosine
triphosphate; cGMP, cyclic guanosine monophosphate; PKG, protein kinase G; AMP, adenosine
monophosphate.
250 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 17.4 Drugs Utilized in the Treatment of pulmonary vasoconstriction. Continuous CO, along with
Perioperative Pulmonary Hypertension continuous mixed venous oxygen saturation monitoring,
is helpful. In patients with moderate or severe elevation
Nitrosodilators (Increase Nitroglycerin of PAP, continuous TEE monitoring may be invaluable
Intracellular cGMP) Sodium nitroprusside to assess ventricular function, optimize intravenous fluid
Inhaled nitric oxide replacement, and guide inotropic therapy. In addition,
Nesiritide detection of right-to-left shunts may reveal the etiology of
Sildenafil hypoxemia in the presence of APH (e.g., patent foramen
-Adrenergic agonists Dobutamine ovale).
(Increase Intracellular Isoproterenol
cAMP) Epinephrine
Prostaglandins (Increase Prostacyclin
Intracellular cAMP) Epoprostenol What Is the Mechanism
-Adrenergic blockers Tolazoline of Right Ventricular Failure?
Phentolamine
Calcium channel Nicardipine
blockers Factors affecting RV preload, afterload, or contractility
Direct smooth muscle Hydralazine can negatively influence ventricular function, causing
relaxants ischemia and right ventricular failure. The thin walls
and crescent shape of the RV enable it to withstand wide
cGMP, cyclic guanosine monophosphate; cAMP, cyclic adenosine fluctuations in volume, but acute changes in pressure are
monophosphate. poorly tolerated.
When RV afterload increases, the peristaltic contrac-
tion of the RV is lost, and it takes on the characteristics
Additionally, changes in CO, gravity, and airway pressures of the LV. To compensate for the lost volume ejected,
that are common in the perioperative period affect the the ventricle dilates to maintain stroke volume. Therefore,
pulmonary circulation to a much greater degree than the the isovolumic contraction time, as well as the ejection
systemic circulation. Many patients have the preexisting time, is prolonged, and with increases in right ventricu-
diagnosis of pulmonary hypertension; therefore, preoper- lar end-diastolic pressure (RVEDP), perfusion of the right
ative knowledge of the status of the pulmonary vascular coronary artery can no longer occur during systole, hence
bed, degree of valvular pathology, lung function, right- decreasing the oxygen supply during a time of increased
and left-sided heart pressures, and intracardiac shunts demand.
are imperative in many patients to optimize perioperative If these changes happen slowly over an extended
management. period of time, the RV has time to adapt, but as in our
Intraoperatively, one must prevent increases in PAP case study, a sudden increase in PVR is poorly tolerated,
and PVR by avoiding vasospastic stimuli known to induce leading to right ventricular systolic and diastolic failure,
and exacerbate pulmonary vasoconstriction,3 such as: and ultimately cardiovascular collapse. Chronically, the
RV is able to generate systemic pressures, but when
Hypoxia
presented with acute elevations in afterload, the RV
Hypercarbia
cannot usually generate systolic pressures >50 mm Hg.
Acidosis
There is a high degree of ventricular interdependence
Prolonged CPB
that is vital for normal ventricular function. Although
Ischemia-reperfusion injury
always present, ventricular dependence is most apparent
Inflammatory mediators
during sudden volume shifts.45 An increase in RV pressure
Pulmonary leukosequestration
produces a leftward shift of the interventricular septum.
Microemboli
The restriction imposed by the pericardium reduces
Excessive thromboxane or endothelin production
left ventricular preload and compliance, increases wall
Commonly used medications that will increase PVR stiffness, and subsequently decreases CO. The same
are ketamine, NO, thiopental, and histamine-releasing mechanism holds true for acute dilation of the LV, which
muscle relaxants (e.g., atracurium). Moreover, adding high decreases RV preload and CO. The loss of the contribution
levels of PEEP may divert blood flow away from well ven- of the left ventricular contraction after insertion of a left
tilated areas of the lung, thereby worsening oxygenation. ventricular assist device can precipitate right ventricular
Similarly, avoidance of hypoxia, hypercarbia, and aci- failure.46
dosis is still vital in the postoperative period, particularly The summation of these mechanisms predisposes the
in the presence of residual effects of general anesthet- RV to rapid failure in the presence of an acute increase in
ics or muscle relaxants. Inadequate tidal volumes will afterload. The increased right atrial and right ventricular
predispose patients to atelectasis and hypoxic pulmonary pressures mediating an increase in volume cause tricuspid
vasoconstriction, with consequent increases in PVR. annular dilation, with subsequent regurgitation. This
Invasive monitoring is frequently indicated, partic- leads to hepatic venous congestion, and the concomitant
ularly in patients with severe preexisting disease or in decrease in LV preload reduces overall CO. The reduced
those suffering from prolonged exposure to stimuli for CO, in turn, decreases blood pressure and coronary
C H A P T E R 17 / A C U T E P U L M O N A R Y H Y P E R T E N S I O N 251
4. Lee FA. Hemodynamics of the right ventricle in normal and 26. Conzen PF, Habazettl H, Gutmann R, et al. Thromboxane
disease states. Cardiol Clin. 1992;10:5967. mediation of pulmonary hemodynamic responses after
5. Mebazaa A, Karpati P, Renaud E, et al. Acute right neutralization of heparin by protamine in pigs. Anesth Analg.
ventricular failurefrom pathophysiology to new treatments. 1989;68:2531.
Intensive Care Med. 2004;30:185196. 27. Hobbhahn J, Conzen PF, Habazettl H, et al. Heparin rever-
6. Bristow MR, Zisman LS, Lowes BD, et al. The pressure- sal by protamine in humanscomplement, prostaglandins,
overloaded right ventricle in pulmonary hypertension. Chest. blood cells, and hemodynamics. J Appl Physiol. 1991;
1998;114:101S106S. 71:14151421.
7. Santamore WP, Gray L. Jr Significant left ventricular con- 28. Park KW. Protamine and protamine reactions. Int Anesthe-
tributions to right ventricular systolic function. Mechanism siol Clin. 2004;42:135145.
and clinical implications. Chest. 1995;107:11341145. 29. Kowalewski J, Brocki M, Dryjanski T, et al. Right ventricular
8. Bove AA, Santamore WP. Ventricular interdependence. Prog morphology and function after pulmonary resection. Eur J
Cardiovasc Dis. 1981;23:365388. Cardiothorac Surg. 1999;15:444448.
9. Rich S, Dantzker DR, Ayres SM, et al. Primary pulmonary 30. Reed CE, Spinale FG, Crawford FA Jr. Effect of pulmonary
hypertension. A national prospective study. Ann Intern Med. resection on right ventricular function. Ann Thorac Surg.
1987;107:216223. 1992;53:578582.
10. Farber HW, Loscalzo J. Pulmonary arterial hypertension. 31. Hurford WE, Zapol WM. The right ventricle and critical
N Engl J Med. 2004;351:16551665. illness: A review of anatomy, physiology, and clinical evalu-
11. Reich DL, Bodian CA, Krol M, et al. Intraoperative hemo- ation of its function. Intensive Care Med. 1988;14(Suppl 2):
dynamic predictors of mortality, stroke, and myocardial 448457.
infarction after coronary artery bypass surgery. Anesth Analg. 32. Lee JH, Kim S, Park BK, et al. The effect of a combination
1999;89:814822. of inhaled nitric oxide and an endothelinA-receptor antag-
12. Ricciardi MJ, Rubenfire M. How to manage primary pul- onist on hemodynamic dysfunction in experimental acute
monary hypertension. Giving hope to patients with a life- pulmonary thromboembolism. Lung. 2005;183:139149.
threatening illness. Postgrad Med 1999;105:4548, 5156. 33. McLean RF, Prielipp RC, Rosenthal MH, et al. Vasodilator
13. Boxt LM, Katz J, Kolb T, et al. Direct quantitation of right and therapy in microembolic porcine pulmonary hypertension.
left ventricular volumes with nuclear magnetic resonance Anesth Analg. 1990;71:3541.
imaging in patients with primary pulmonary hypertension. 34. Priebe HJ. Efficacy of vasodilator therapy in canine model
J Am Coll Cardiol. 1992;19:15081515. of acute pulmonary hypertension. Am J Physiol. 1988;255:
14. Morariu AM, Loef BG, Aarts LP, et al. Dexamethasone: H1232H1239.
benefit and prejudice for patients undergoing on-pump 35. Dent G, Magnussen H, Rabe KF. Cyclic nucleotide phospho-
coronary artery bypass grafting: A study on myocardial, pul- diesterases in the human lung. Lung. 1994;172:129146.
monary, renal, intestinal, and hepatic injury. Chest. 2005;128: 36. Bhorade S, Christenson J, OConnor M, et al. Response
26772687. to inhaled nitric oxide in patients with acute right heart
15. Ng CS, Wan S, Yim AP. Pulmonary ischaemia-reperfusion syndrome. Am J Respir Crit Care Med. 1999;159:571579.
injury: Role of apoptosis. Eur Respir J. 2005;25:356363. 37. Griffiths MJ, Evans TW. Inhaled nitric oxide therapy in
16. Ng CS, Wan S, Yim AP, et al. Pulmonary dysfunction after adults. N Engl J Med. 2005;353:26832695.
cardiac surgery. Chest. 2002;121:12691277. 38. Moncada S, Palmer RM, Higgs EA. Nitric oxide: Physiol-
17. Riedel B. The pathophysiology and management of periop- ogy, pathophysiology, and pharmacology. Pharmacol Rev.
erative pulmonary hypertension with specific emphasis on 1991;43:109142.
the period following cardiac surgery. Int Anesthesiol Clin. 39. Hogman M, Frostell CG, Hedenstrom H, et al. Inhalation of
1999;37:5579. nitric oxide modulates adult human bronchial tone. Am Rev
18. Chen YF, Oparil S. Endothelial dysfunction in the pulmonary Respir Dis. 1993;148:14741478.
vascular bed. Am J Med Sci. 2000;320:223232. 40. Sato Y, Walley KR, Klut ME, et al. Nitric oxide reduces the
19. Newman JH. Pulmonary hypertension. Am J Respir Crit Care sequestration of polymorphonuclear leukocytes in lung by
Med. 2005;172:10721077. changing deformability and CD18 expression. Am J Respir
20. Mauban JR, Remillard CV, Yuan JX. Hypoxic pulmonary Crit Care Med. 1999;159:14691476.
vasoconstriction: Role of ion channels. J Appl Physiol. 2005; 41. Young JD, Sear JW, Valvini EM. Kinetics of methaemoglobin
98:415420. and serum nitrogen oxide production during inhalation of
21. Mazzoni MC, Borgstrom P, Intaglietta M, et al. Lumenal nar- nitric oxide in volunteers. Br J Anaesth. 1996;76:652656.
rowing and endothelial cell swelling in skeletal muscle capil- 42. Dias-Junior CA, Vieira TF, Moreno H, Jr, et al. Silde-
laries during hemorrhagic shock. Circ Shock. 1989;29:2739. nafil selectively inhibits acute pulmonary embolism-induced
22. Hakim TS, Picone A, Oleary CE, et al. Protamine-induced pulmonary hypertension. Pulm Pharmacol Ther 2005;18:
pulmonary venoconstriction in heparinized pigs. Anesth 181186.
Analg. 1995;81:3843. 43. Souza-Silva AR, Dias-Junior CA, Uzuelli JA, et al. Hemody-
23. Montalescot G, Fischman AJ, Strauss HW, et al. Imaging the namic effects of combined sildenafil and l-arginine during
ovine heparin-protamine interaction with 111In-protamine. acute pulmonary embolism-induced pulmonary hyperten-
J Appl Physiol. 1993;75:963971. sion. Eur J Pharmacol 2005;524:126131.
24. Viaro F, Dalio MB, Evora PR. Catastrophic cardiovascular 44. Evgenov OV, Busch CJ, Evgenov NV, et al. Inhibition of
adverse reactions to protamine are nitric oxide/cyclic guano- phosphodiesterase 1 augments the pulmonary vasodilator
sine monophosphate dependent and endothelium mediated: response to inhaled nitric oxide in awake lambs with acute
Should methylene blue be the treatment of choice? Chest. pulmonary hypertension. Am J Physiol Lung Cell Mol Physiol.
2002;122:10611066. 2006;290:L723L729.
25. Morel DR, Zapol WM, Thomas SJ, et al. C5a and throm- 45. Santamore WP, DellItalia LJ. Ventricular interdependence:
boxane generation associated with pulmonary vaso- and Significant left ventricular contributions to right ventric-
broncho-constriction during protamine reversal of heparin. ular systolic function. Prog Cardiovasc Dis. 1998;40:289
Anesthesiology. 1987;66:597604. 308.
254 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
46. Santamore WP, Austin EH III, Gray L Jr. Overcoming right 56. Yoo KY, Kim H, Jeong CW, et al. Effects of inotropic
ventricular failure with left ventricular assist devices. J Heart drugs on mechanical function and oxygen balance in postis-
Lung Transplant 1997;16:11221128. chemic canine myocardium: Comparison of dobutamine,
47. Boxt LM. Radiology of the right ventricle. Radiol Clin North epinephrine, amrinone, and calcium chloride. J Korean Med
Am. 1999;37:379400. Sci. 2005;20:732739.
48. Shah MR, Hasselblad V, Stevenson LW, et al. Impact 57. Lobato EB, Florete O, Bingham HL Jr. A single dose of mil-
of the pulmonary artery catheter in critically ill patients: rinone facilitates separation from cardiopulmonary bypass
Meta-analysis of randomized clinical trials. JAMA. 2005;294: in patients with pre-existing left ventricular dysfunction. Br
16641670. J Anaesth. 1998;81:782784.
49. Simini B. Pulmonary artery catheters in intensive care. 58. Lobato EB, Janelle GM, Urdaneta F, et al. Comparison of
Lancet. 2005;366:435436. milrinone versus nitroglycerin, alone and in combination, on
50. Zwissler B. Acute right heart failure. Etiology grafted internal mammary artery flow after cardiopulmonary
pathophysiologydiagnosistherapy. Anaesthesist. 2000; bypass: Effects of alpha-adrenergic stimulation. J Cardiotho-
49:788808. rac Vasc Anesth. 2001;15:723727.
51. Goldstein JA. Pathophysiology and management of right 59. Lobato EB, Urdaneta F, Martin TD, et al. Effects of milrinone
heart ischemia. J Am Coll Cardiol. 2002;40:841853. versus epinephrine on grafted internal mammary artery flow
52. Goldstein JA. Right heart ischemia: Pathophysiology, natural after cardiopulmonary bypass. J Cardiothorac Vasc Anesth.
history, and clinical management. Prog Cardiovasc Dis. 2000;14:911.
1998;40:325341. 60. Haikala H, Nissinen E, Etemadzadeh E, et al. Troponin
53. Goldstein JA, Harada A, Yagi Y, et al. Hemodynamic C-mediated calcium sensitization induced by levosimen-
importance of systolic ventricular interaction, augmented dan does not impair relaxation. J Cardiovasc Pharmacol.
right atrial contractility and atrioventricular synchrony in 1995;25:794801.
acute right ventricular dysfunction. J Am Coll Cardiol. 1990; 61. Slawsky MT, Colucci WS, Gottlieb SS, et al. Acute hemo-
16:181189. dynamic and clinical effects of levosimendan in patients
54. Topol EJ, Goldschlager N, Ports TA, et al. Hemodynamic with severe heart failure. Study Investigators. Circulation.
benefit of atrial pacing in right ventricular myocardial 2000;102:22222227.
infarction. Ann Intern Med. 1982;96:594597. 62. Yokoshiki H, Sperelakis N. Vasodilating mechanisms
55. Nikolaidis LA, Trumble D, Hentosz T, et al. Catecholamines of levosimendan. Cardiovasc Drugs Ther. 2003;17:111
restore myocardial contractility in dilated cardiomyopathy at 113.
the expense of increased coronary blood flow and myocardial 63. Ukkonen H, Saraste M, Akkila J, et al. Myocardial efficiency
oxygen consumption (MvO2 cost of catecholamines in heart during levosimendan infusion in congestive heart failure.
failure). Eur J Heart Fail. 2004;6:409419. Clin Pharmacol Ther. 2000;68:522531.
CHAPTER CARDIAC DYSRHYTHMIAS
CASE SUMMARY
What Is the Importance of
54-year-old man (90 kg, 181 cm, body mass
Cardiac Dysrhythmias?
A
index [BMI] 27.7 kg per m2 ) presented
for wide local excision of a melanoma
on his thigh, along with intraoperative Cardiac dysrhythmia (also arrhythmia) comprises any
lymphatic mapping and sentinel lymph node abnormality or perturbation in the normal activation se-
biopsy. He had no medical allergies, took quence of the myocardium. Cardiac dysrhythmias can
no medicines, and reported excellent exercise tolerance produce too slow a ventricular rate (bradydysrhythmia)
(he walks his treadmill at 5 mph and 15% grade for or too fast a ventricular rate (tachydysrhythmia). These
30 to 60 minutes nearly every day). His preoperative abnormalities frequently occur in the perioperative pe-
examination was significant only for sinus bradycardia riod. Although some are benign and require only watchful
with a rate of 54 bpm. His heart and lung examination waiting or assurance of no biochemical derangements,
findings were normal. His laboratory values (the surgeon others result from developing or ongoing malignant pro-
ordered a complete blood count, electrolytes, blood urea cess(es). Some dysrhythmias represent a harbinger of a
nitrogen [BUN], creatinine, and lactic dehydrogenase) more serious condition (e.g., bradycardia that develops in
the face of arterial hypoxemia). According to Atlee, the
and electrocardiogram (ECG) demonstrated no abnor-
first recorded death during anesthesia, that of Hannah
malities.
Greener in 1848,1 was most likely because of ventricu-
Anesthetic induction was carried out with fentanyl
lar fibrillation (VF) (a malignant cardiac dysrhythmia)
200 g, propofol 200 mg, and cisatracurium 14 mg. His
resulting from the sensitizing action of chloroform.2
trachea was intubated with a 7 Fr endotracheal tube,
Although lethal cardiac dysrhythmias remain a rare
and anesthesia was maintained with desflurane 5.8% in
occurrence, any abnormal cardiac rhythm represents a
50% oxygen per air. Cefazolin (1 gm) was given. Shortly
potentially unstable condition. Some dysrhythmias are
after anesthetic induction, but before incision, his sinus
dangerous because they provoke inappropriate medical
rate fell into the 40s and he developed isorrhythmic intervention (such as the treatment of benign prema-
atrioventricular dissociation (see Fig. 18.1). His blood ture ventricular contractions with antiarrhythmic agents
pressure (which had been 120/50 mm Hg) fell to 80/30. A as documented in the Cardiac Arrhythmia Suppression
decision was made to insert a transesophageal pacemaker Trial [CAST] study3 ), whereas other abnormal rhythms
(TAPSCOPE, Cardiocommand, Inc, Tampa, FL), and can threaten cardiovascular homeostasis. Both brady-
pacing was begun at 60 bpm. When the pacing rate dysrhythmias and tachydysrhythmias can produce an
was increased to 90 bpm, the patient developed a imbalance between myocardial oxygen supply (by reduc-
second degree, Mobitz type I (Wenckebach) block (see ing cardiac output or shortening diastole) and demand
Fig. 18.2). (by increasing rate), and some dysrhythmias can progress
After completion of his surgery, he admitted to feeling to life-threatening situations (e.g., supraventricular tachy-
occasional pounding in his chest and neck, especially at cardia producing myocardial ischemia, leading to ventric-
night. A subsequent Holter monitor revealed significant ular tachycardia [VT] and death).
sinus bradycardia with ventricular escapes. A permanent A cardiac dysrhythmia should always be considered
pacemaker was suggested to increase his overall heart in the differential diagnosis of any sudden hemodynamic
rate, because his postoperative echocardiogram showed imbalance. For example, an abrupt reduction in blood
moderate aortic regurgitation in a structurally normal pressure associated with little change in heart rate might
valve. result from an atrioventricular (atrioventricular [AV])
255
256 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 18.1 Isorhythmic atrioventricular nodal dissociation. As noted in the text, this 54-year-old
man with outstanding exercise tolerance and preexisting sinus bradycardia developed an accelerated
idioventricular rhythm that overtook his sinus node pacemaker, resulting in the rhythm strip shown
here. The PP interval at complexes 3 to 4 (1,395 ms) represents a rate of 43 bpm. The idioventricular
escape interval was 1,250 ms (rate = 48 bpm). Note that QRS complex 8 is fusion beat, wherein the
wide complex behavior of the ventricular escape was overtaken (and, therefore, narrowed) by the
sinus event (P wave) that can be seen immediately preceding it.
nodal junctional rhythm, and the hemodynamics in this well. The authors have assumed that that the reader has
case might be further compromised by the sympathetic basic electrocardiographic knowledge.
discharge associated with an isorhythmic AV dissociation.
Sometimes, merely reducing the depth of the inhalation
anesthetic agent, or the substitution of another balanced What Are the Basic Facts About
anesthetic technique, may end the dysrhythmia and Bradydysrhythmias, and How
improve blood pressure.
This chapter focuses on the origins, recognition, Are They Managed?
and treatment of the common atrial and ventricular
perioperative dysrhythmias. His-Purkinje system (HPS)
conduction blockade (heart block) will be discussed as
SINUS NODE DYSFUNCTION
The sinoatrial (SA) node and the atrium are intimately
involved in the initiation of a cardiac cycle, and therefore
any failure of these tissues can result in bradycardia.
Conditions that lead to failure of heart beat initiation
include:
Sinus node arrest (no spontaneous depolarization)
Sinus node exit block (SA node depolarizes but electric
signal is not propagated within the region of the SA
node)
Atrial tissue failure (the propagating depolarization fails
1 2 3 4 5 6 7 8 9 10 11 12 to reach the AV node)
Often, without electrophysiologic study, differentia-
FIGURE 18.2 Second degree atrioventricular Mobitz type tion of these conditions is difficult if not impossible. Ab-
I block (Wenckebach) during transesophageal pacing. An
normal electrolytes, preoperative -blocker use, and many
esophageal stethoscope with two stainless steel rings for atrial
of the intraoperative drugs have the potential to aggravate
pacing was introduced to 36 cm (from the lips). Initially, atrial
bradycardia and bradycardia-dependent arrhythmias.4
capture was obtained with an indicated output of 8 mA and a
Probably the most common bradycardia results from
rate of 60 bpm. When the pacing rate was increased to 90 bpm,
the slowing of the sinus node, as in our case summary.
Wenckebach block developed. The two top traces are
In the operating room, it can be caused by drugs,
electrocardiogram (ECG) leads II and V5, respectively, and the
especially dexmedetomidine and vagotonic agents such as
bottom trace is the pulse oximeter plethysmogram. The numbers
fentanyl, sufentanil, and remifentanil. In a retrospective
(added) are shown below the pacing artifacts, which are large
analysis of 6,663 electronically recorded cases of neuraxial
signals, initially downward, and cause considerable distortion of
anesthesia, Lesser et al. found that a baseline heart rate
the ECG baseline. The smaller, upward deflections are the QRS
<60 bpm and male gender were risk factors for severe
complexes. QRS complexes are absent after pacing artifacts 3, 6,
(<40 bpm) bradycardia.5
9, and 11. Note that the interval from the pacing stimulus to the
QRS lengthens in sequences 13, 46, and 79. Because there
are P-wave deflections at pacing stimulus 3 and 11, the lack of a
QRS indicates AV block. There is no P-wave deflection after ATRIOVENTRICULAR
pacing stimulus 6 and 9, and therefore these QRS failures might NODAL BLOCK
have resulted from failure to obtain atrial capture with the
transesophageal pacing device. Electric events that initiate cardiac contraction generally
start in the sinus node, spread (or arborize) over the atrial
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 257
tissue, activate the AV node, and then traverse the HPS Second Degree
to activate the ventricles. In the presence of P waves, but
without ventricular activation, AV nodal block is present. Second degree block represents disease along the HPS.
Atrioventricular block can be a temporary or permanent In second degree AV block, some impulses are blocked.
disturbance of AV impulse conduction due to anatomic Classification depends upon the PR interval stability. In
or functional impairment of conduction. AV block is Mobitz I (also called Wenckebach) block, the PR interval
classified as first-, second-, or third-degree (complete) progressively lengthens until the ventricles fail to activate.
block. The level of AV block can also be defined as supra-, Assuming hemodynamic stability, Wenckebach block is
intra, or infra-Hisian block. treated with watchful waiting. In Mobitz II block, the
PR interval is stable. Mobitz II represents more serious
AV nodal/HPS disease and often requires pacemaker
First Degree
placement. Distinguishing Mobitz I from Mobitz II can
First-degree AV block is a prolonged PR interval exceeding be quite challenging, especially in the presence of a 2:1
200 ms. In general, this condition is benign, although it block (see Fig. 18.3).
has been associated with significant bradycardia during
spinal anesthesia.6 Progression to higher grades of AV
block is rare in the general population,7 but it has been Third Degree
reported with spinal8 and general9 anesthesia. With a
prolonged first-degree block exceeding 400 ms, apparent In third-degree AV block, complete failure of the HPS
AV dyssynchrony can be present, and higher grade block results in no atrial event being conducted to the ventricles.
can develop if the atrial rate increases with metabolic Ventricular systoles continue only in the presence of
demand from exercise, trauma, anemia, as the HPS fails junctional or ventricular escape activity. The presence
to conduct all of the atrial impulses to the ventricles. of P waves without a clear relation to the QRS complexes
Treatment of first-degree heart block will depend upon (rate and apparent PR interval) confirms the diagnosis of
symptomatology. this problem (see Fig. 18.4).
FIGURE 18.3 Bradycardia from a 2:1 second degree atrioventricular block. This 12-lead
electrocardiogram (ECG) was obtained from a 71-year-old woman. Although a permanent dual
chamber pacemaker was present (DDD mode, lower rate = 60 bpm), a complete lead fracture on the
ventricular lead failure prevented ventricular sensing and pacing. The ECG shows a sinus rhythm
with rate of 82 bpm and the ventricular rate is 41 bpm. The PR interval from the conducted events is
195 ms. The P waves occur at the end of each T wave. As a result, the QT interval was reported as
680 ms, owing to the distortion of the T wave from the next P wave. It does not exceed 510 ms (QTc
would be 426 ms). Differentiating Mobitz I from Mobitz II with this ECG is not possible. However,
atrial pacing at a rate of 100 bpm demonstrated a ventricular rate of 33 bpm with a PR interval of
195 ms, confirming the presence of Mobitz II block. Therapy consisted of ventricular lead revision.
258 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 18.4 Third-degree atrioventricular block. This 12-lead electrocardiogram (ECG) was
obtained from an asymptomatic 83-year-old man. It shows a sinus rhythm with a rate of 77 bpm
(note that P waves, marked by downward arrows, are hidden in the QRS complexes as well as the
T waves). The ventricular rate is 43 bpm and the QRS complexes are 150 ms in duration and in a left
bundle branch pattern. There is no relation between any P wave (arrows on the rhythm strip lead II)
and an ensuing QRS. Also present on this ECG is left atrial enlargement. The patient underwent
placement of a permanent pacemaker followed by uneventful parotidectomy.
Whether treatment (pharmacologic or pacing) be- time required to arrange the equipment, establish cen-
comes necessary depends upon the patients hemody- tral access, and determine the appropriate position of the
namic stability and medical condition. Pharmacologic ventricular catheter to provide ventricular capture. Flow-
therapy can include atropine, ephedrine, or epinephrine. directed catheters and a right internal jugular approach
Some practitioners might also consider glycopyrrolate, afford the shortest insertion times.13 The reported inci-
but it is indicated only for vagally induced bradycar- dence of successful capture in urgent situations without
dia.10 Use of these chronotropic drugs can sometimes fluoroscopy ranges from 30% to 90%.14 Also, temporary
lead to uncontrolled sinus tachycardia.11 pacing in the patient with a permanently placed pace-
maker or implantable cardioverter-defibrillator (ICD) may
be contraindicated without reprogramming, because the
PACING temporary pacing equipment can interfere with the per-
In this setting, temporary cardiac pacing also can be manent cardiac generator.
considered. Pacing may be carried out through transcuta- Most transvenous, flow-directed pacing catheters
neous, transvenous, transthoracic (introduction of pacing offer only ventricular pacing. The pulmonary artery AV
wire[s] directly into the thorax), and transesophageal pacing catheter, described by Zaidan in 1983,15 allows for
modalities. Transcutaneous and ventricular-only transve- AV sequential pacing through electrodes attached to the
nous pacing, even if feasible, may exacerbate hemody- outside of the catheter, as well as routine pulmonary artery
namic problems in patients with cardiomyopathy, as these catheter functions. Combination of the two functions into
pacing modalities do not preserve AV synchrony (i.e., they one catheter eliminates the need for separate insertion
produce only ventricular or global myocardial activation). of temporary transvenous pacing electrodes. However,
Although perioperative temporary pacing has been com- several potential disadvantages exist with this catheter:
pletely reviewed elsewhere,12 a few important aspects will Varying success in initiating and maintaining capture15
be discussed. External electrode displacement from the catheter16 and
Relatively high cost when compared with standard
Transvenous pulmonary artery catheters
Transvenous cardiac pacing provides the most reliable The Paceport PAC provides ventricular pacing with
means of temporary pacing, albeit at the expense of the a separate bipolar pacing lead (Chandler probe), which
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 259
allows more stable ventricular pacing as well as pul- be used (with a special adapter) to record the intraatrial
monary artery catheter function.17 This catheter has been electrogram. Problems with esophageal pacing include:
used for successful resuscitation in cardiac arrest during
The necessity for special generators that provide an
closed chest cardiac massage when transcutaneous and
output of 20 to 30 mA, with pulse width of 10 to 20 ms
simple bipolar pacing had failed. A newer AV Paceport
(typical temporary generators have a maximum output
PAC adds another lumen to allow placement of another
of 20 mA with pulse width duration of 1 to 2 ms)
pacing lead for atrial pacing. The atrial wire can also
The ability to pace only the left atrium and not the
be used to diagnose supraventricular tachydysrhythmias
left ventricle, which can be a significant problem in
(supraventricular and ventricular tachycardia [SVT]) by
emergency situations23
atrial electrograms and to overdrive pace of atrial flutter
Phrenic nerve stimulation with significant diaphrag-
(AFL) and reentrant SVT.18
matic movement24 and
Induction of ventricular tachydysrhythmias during
Transcutaneous rapid atrial pacing has been noted. No long-term
complications with this modality have been described,
Transcutaneous pacing, first described by Zoll,19 is readily and no significant esophageal trauma has been reported
available and can be rapidly implemented in emergency despite long-term therapy of up to 60 hours.28
situations. Capture rate is variable, and the technique
often causes pain in awake patients, but usually is
tolerated until temporary transvenous pacing can be
instituted. It may be effective even when endocardial
pacing fails.20 It is now considered by many to be
How Are Supraventricular
the method of choice for prophylactic and emergent Tachydysrhythmias Detected
applications.21 and Managed?
Transesophageal
Esophageal pacing is the newest technique available, and ATRIAL PREMATURE
it has been shown to be quite reliable.2225 Esophageal COMPLEX
pacing is relatively noninvasive, well tolerated even in
most awake patients, and it appears to be devoid of An atrial premature complex (APC) results from inap-
serious complications. It is contraindicated in the pa- propriate early depolarization of atrial tissue, not under
tient with atrial disease (e.g., atrial fibrillation (AF) or control of the sinus node, which might cause some form of
flutter), AV nodal disease, or any patient with a perma- P wave to be inscribed on the surface ECG (see Fig. 18.5).
nently implanted cardiac generator, because the electric Subsequent HPS activation results in ventricular depo-
output from the esophageal pacemaker can inhibit the larization, inscribing a QRS complex often identical to
output from the permanent device. This modality is use- that of the normal sinus beat. APCs frequently reset the
ful for heart rate support of cardiac output, overdrive sinus node timing, so the interval from the APC to the
suppression of reentrant SVT, and for diagnostic atrial next true P wave might be less than fully compensatory.29
electrograms. Ventricular capture must be excluded be- Frequently, the P wave from the APC remains hidden
fore attempts are made at rapid atrial pacing for overdrive in the prior T wave. If a P wave is present, it usually
suppression to prevent potential VT or VF. Some sur- differs in morphology from the normal sinus P wave,
gical positions (e.g., prone) can increase the chance of because it originates from a site different from the SA
unintentional ventricular capture, and esophageal atrial node. Because the distance from this aberrant atrial focus
pacing should be followed very carefully.26 Typically, the might be different from that of the SA node to the AV
pacing stimulus is delivered using a modified esophageal node, the PR interval also might be different from that of
stethoscope, with the distal end of esophageal stethoscope a sinus event. Although these morphologic features can
inserted to a depth of 30 to 40 cm from the teeth. Cap- help differentiate APCs from premature ventricular com-
ture should be confirmed using the peripheral pulse (i.e., plexes (PVCs), none of these features is absolutely reliable.
from the pulse oximeter plethysmogram or an invasive For instance, as noted in the preceding text, the aberrant
hemodynamic monitor), because the pacing stimulus of- P wave may fall upon the preceding T wave and becomes
ten is large relative to the QRS and frequently fools the difficult to identify. The postextrasystolic pause may ap-
electrocardiographic counting algorithm on the monitor. pear fully compensated if there is a delay in the sinus
Atrial capture is obtained in virtually all patients using node discharge of the following beat. When aberrant ven-
an indicated output of 8 to 20 mA; the output should tricular conduction occurs, the QRS complex may appear
be set to two to three times the threshold for capture. widened.
Thresholds are not influenced by weight, age, atrial size, If the HPS is, in fact, redepolarized before complete
or previous cardiac surgery.25 Because there is no sens- repolarization of the conduction system or ventricular
ing element involved, esophageal pacing is AOO <mode tissue, a bizarre, wide complex QRS can be inscribed on
pacing. Transesophageal ventricular pacing is generally the surface ECG. Most commonly, this QRS will appear
unreliable, yet the optimal site appears to be 2 to 4 cm dis- in a right bundle branch patternthis event is termed the
tal to the atrial site.27 The esophageal stethoscope can also Ashman Phenomenon.30
260 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 18.5 An atrial premature contraction (PAC) resets the sinus node. In this strip, the top
trace is lead V5, the middle trace is the pulse oximeter plethysmogram, and the bottom trace is from
a noninvasive arterial pressure device (Tensys Corporation, San Diego, CA). This strip was obtained
from an awake, normal 30-year-old woman with a sinus rate of 44 bpm (interval of 1,365 ms) and
PR interval of 200 ms. It shows a PAC (the third QRS complex on the top trace) preceded by a
P wave that is morphologically different from the remaining P waves. The next true sinus event (fifth
complex) takes place 1,390 ms from the abnormal P wave, and the remainder of the sinus events
follow this new timing cycle. The fourth QRS most likely represents an AV junctional escape beat
(narrow complex, similar axis as remaining QRS events) that did not reset the sinus node timer.
Owing to the profound sinus bradycardia, there was sufficient time after each QRS for complete
ventricular repolarization.
A few simple criteria often help distinguish a wide diseases of the mitral valve such as mitral stenosis and mi-
complex QRS inscribed by an aberrantly conducted tral valve prolapse, ischemic heart disease, and congestive
APC from a PVC. Generally, the initial deflection of an heart failure. Noncardiac medical conditions associated
aberrantly conducted QRS is identical in direction to the with APCs include acute and chronic pulmonary diseases,
sinus-induced QRS, and its configuration is similar to the chronic renal failure, and metabolic abnormalities.
right bundle branch block (RBBB) pattern with a duration
<0.14 seconds. On the other hand, the QRS inscribed by
a PVC shows an initial deflection to be opposite to that of
Management
sinus rhythm QRS, and its configuration is different from In asymptomatic cases, no treatment is required for oc-
the RBBB pattern with a duration >0.14 seconds. casional APCs. For patients with frequent symptomatic
APCs, management begins with simple reassurance, along
Hemodynamic Significance with identification and avoidance of precipitating factors
such as stress or excess caffeine. If these measures fail
In most cases, APCs are completely asymptomatic. When to alleviate symptoms, drug therapy can be started with
they occur frequently in the conscious patient, they may -blockers,32 which may also help prevent APCs from trig-
cause palpitations or an unpleasant feeling of irregular gering other more serious tachyarrhythmias such as AF.
heart beats. APCs very early in the hyperexcitable phase
of the cardiac cycle may precipitate tachyarrhythmias, in
particular AF.31
PAROXYSMAL REENTRANT
Prevalence SUPRAVENTRICULAR
TACHYCARDIA
Occasional APCs are very common, even in patients with-
out underlying heart disease. In normal subjects, stress, Paroxysmal reentrant supraventricular tachycardias are
physical exhaustion, heavy smoking, alcohol, and caffeine characterized by abrupt onset and regularity. The patho-
may induce APCs. The frequency of APCs increases with physiology of these arrhythmias involves two tissues that
increasing age and in the presence of structural heart have different conduction velocities and refractory periods
disease. The incidence of APCs is higher in patients with (slow and fast pathways). The impulse travels down one
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 261
FIGURE 18.6 Atrial fibrillation. This 12-lead electrocardiogram (ECG) was obtained from a
72-year-old woman. It demonstrates an irregularly irregular rhythm without any clear relation
between small deflections that might be called P waves and the following QRS complexes. In fact,
this ECG was misdiagnosed by both the ECG machine and a number of physicians, who believed
this patient to be in sinus rhythm with APCs. Also seen is a leftward axis (35 degrees), poor R wave
progression, left ventricular hypertrophy, and T waves consistent with strain.
pathway while the second is in the refractory period, then the sinus node as the primary cardiac pacemaker, being
travels up the second, thereby perpetuating the arrhyth- replaced by totally disorganized atrial activity with rapid
mia. The most common paroxysmal reentrant supraven- fibrillatory waves with varying morphology and an ir-
tricular tachycardia is AV nodal reentrant tachycardia, regularly irregular ventricular rhythm on the ECG (see
which involves reentry within the AV node. AV nodal reen- Fig. 18.6). The QRS complex is usually narrow but may
trant tachycardia usually demonstrates a regular, narrow be wide in cases of coexisting bundle branch blocks or
complex tachycardia of 160 to 180 bpm. It is generally aberrant conduction. With QRS complexes of varying am-
benign unless structural heart disease is present. Patients plitudes and total irregularity of the arterial pulse, this
typically present with palpitations and shortness of breath. rhythm has often been classically referred to as delirium
cordis. This arrhythmia has important clinical implica-
Management tions because patients with AF have increased risk for
morbidity and mortality. AF can often lead to symptoms
Vagal maneuvers, such as Valsalva maneuver or carotid impairing patients functional status and quality of life.
massage (after ensuring there is no carotid bruit) usually AFL is sometimes seen with AF. It is characterized by
terminate the tachycardia. Adenosine can be used with more regular atrial activity, with a saw-tooth pattern on
good success rates. AV nodal blocking agents, such ECG, and the ventricular response is generally regular
as -blockers and calcium channel blockers, often are because of a 2:1, 3:1, or greater atrial-to-ventricular
effective in terminating and preventing the recurrence of conduction pattern (see Fig. 18.7). Sustained AFL is less
AV nodal reentrant tachycardia. Patients with refractory common than AF, and AFL generally degenerates to AF.
AV nodal reentrant tachycardia or those who do wish to The evaluation and management of AFL is identical to
take medication can undergo catheter ablation. that of AF and will be discussed in the subsequent text.
Epidemiology
ATRIAL FIBRILLATION
AND ATRIAL FLUTTER The overall prevalence of AF is estimated to be one percent
of the total population. The prevalence increases with ad-
AF is the most common sustained atrial arrhythmia en- vancing agefrom 0.1% for people younger than 55 years
countered in anesthesia practice.33 AF represents loss of to 9% in those older than 80 years34 Prevalence is higher
262 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 18.7 Atrial flutter with a variable ventricular response. This 12-lead electrocardiogram
(ECG) was obtained from a 58-year-old man on postoperative day 3 following a right thoracotomy
with right upper lobectomy. Note the notching of the baseline, primarily showing in lead II. The
variable block in this setting has produced the irregularly irregular ventricular rhythm. Left
ventricular hypertrophy is present with considerable T-wave flattening. (Courtesy of Daniel Lenihan,
MD, FACC.)
in men than in women, and higher in whites than African Differential Diagnosis
Americans. The risk of AF increases with cardiovascu-
lar diseases such as hypertension, ischemic heart disease, Usually, AF is relatively easy to distinguish from other
valvular heart disease, and sick sinus syndrome.35 electrocardiographic rhythms with a wavy baseline (such
as AFL or Parkinsons disease), because the ventricular
systoles in AF occur at an irregularly irregular interval.
Classification Often, it is difficult to recognize P waves in patients with
AF has been classified based upon its morphology and ap- Parkinsons disease or other forms of tremors because of
pearance on ECG. The baseline undulations may be clearly the fine motion artifacts.
distinct and visible (coarse AF), intermediate (medium Multifocal atrial tachycardia is often misdiagnosed as
AF), or barely discernable (fine AF). To the extreme, there AF. It is usually observed in critically ill, elderly patients,
may be no perceptible undulation of the baseline. There often in the presence of chronic obstructive pulmonary
is no consistent correlation of these different types of AF disease, and is characterized by the absence of one
with either the severity of AF or its associated underly- dominant atrial pacemaker, the presence of three or more
ing cardiac conditions. Currently, the American College of distinct P wave morphologies on the surface ECG, and
Cardiology/American Heart Association/European Society varying PP, PR and RR intervals. These P waves are usually
of Cardiology (ACC/AHA/ESC) classifies AF into four types tall and peaked, resembling P waves seen in chronic
obstructive pulmonary disease and right heart failure.
as follows:36
Although multifocal atrial tachycardia may present with
1. PAROXYSMAL: Sudden AF that spontaneously reverts; an irregularly irregular rhythm, it can be differentiated
that is, the abnormal rhythm self-terminates in <7 days from AF by the presence of distinct P waves and a well-
(usually <24 hours) and may be recurrent, defined by defined isoelectric baseline.
two or more episodes
2. PERSISTENT: AF lasting longer than 7 days Pathophysiology
3. PERMANENT: AF lasting longer than 1 year
4. LONE: AF in people younger than 60 years without clin- AF typically occurs in patients with underlying cardiac
ical or echocardiographic evidence of cardiopulmonary disease, often complicated by heart failure, resulting in
diseases or hypertension elevated atrial pressure and chamber enlargement. The
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 263
pathophysiology of AF involves factors that trigger the example, in the patient with hypertensive cardiomyopathy
arrhythmia and perpetuate it. Factors triggering the onset and decreased left ventricular compliance (diastolic ven-
of AF usually involve foci of cells in the left atrium in tricular dysfunction), the loss of the atrial contraction can
the vicinity of the pulmonary veins.37 A sudden increase significantly impair stroke volume. In the patient with
in adrenergic discharge causes rapid firing of these cells already poor ventricular function, the additional loss of
followed by a marked vagal response. The perpetuation stroke volume, decreased diastolic time for ventricular
of AF involves the generation of a multitude of atrial filling, and an irregular rhythm can significantly reduce
wavelets that encounter nonhomogeneous conduction in cardiac output. Finally, compared to the same heart rate
the left atrium, causing intraatrial reentry and AF.38 The but with regular rhythm, the irregular rhythm of AF sig-
incidence of AF has been directly correlated with left nificantly decreases cardiac output and coronary blood
atrial volume index as determined by echocardiography,39 flow.41,42 These mechanisms explain why many patients
emphasizing the importance of the abnormal stretch with the acute onset of AF experience symptoms of angina
of the left atrium as a cause of AF. The Framingham and dyspnea, even without previous history of coronary
Heart Study found several echocardiographic predictors artery disease or congestive heart failure.
of AF in patients without rheumatic heart disease.
These included left atrial enlargement, increased left
ventricular wall thickness, and reduced left ventricular
Etiology
fractional shortening.40 In individuals with paroxysmal Preoperatively, AF often is associated with mitral valve
AF who have structurally normal or near-normal hearts, disease, congestive heart failure, ischemic heart disease,
atrial premature beats have been shown to be the most cardiomyopathy, hyperthyroidism, and lung disease. It
important trigger of AF. can arise without evidence of myocardial disease (lone
AF) and may also result from effects of drugs and
Hemodynamic Effects recreational substances. When evaluating these patients,
any precipitating causes such as severe anxiety and
Adverse effects of AF include decreased cardiac output sympathomimetic effects of substances such as alcohol,
owing to the loss of atrial volume augmentation, an in- caffeine, and cocaine ingestion and/or withdrawal should
creased ventricular rate that further precludes adequate be identified. A drug history is needed, because agents
diastolic filling, an irregularly irregular cardiac rhythm, such as theophylline, adenosine, and digitalis all have the
and, sometimes, hypotension (see Fig. 18.8). These conse- ability to cause AF.
quences often exacerbate poor cardiac performance when Mitral valve diseases, including mitral stenosis, mitral
found in conjunction with coexisting heart disease. For regurgitation, and mitral valve prolapse are associated
FIGURE 18.8 Hypotension from atrial fibrillation and inadequate diastolic filling times. This strip
is from an 81-year-old man who developed atrial fibrillation during a composite head and neck
resection with a fibular free flap. The top tracing is electrocardiogram (ECG) lead I. The bottom
tracing is the invasive arterial pressure. The arrows show the relation between the QRS and the pulse
waveform. Note that with a ventricular rate of 125 bpm (RR interval of 480 ms), the resulting
arterial waveform reflects a poor stroke volume. Treatment consisted of metoprolol, which resulted
in a ventricular rate of 3540 bpm, so a transvenous pacemaker was placed with a rate of 75 bpm.
264 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
with increased incidence of AF, likely because of a dilated any precipitating causes. Searching for reversible precipi-
and poorly functional left atrium. Although rheumatic tating factorssuch as the recent use of caffeine, alcohol,
heart disease is now uncommon in the United States, or marijuanais important, as episodes of AF incited
it is associated with a high prevalence of AF. One study by these causes will usually abate once the factors are
evaluating the frequency of AF in more than 1,000 patients removed. In older patients, coexisting medical problems
with rheumatic heart disease found a prevalence of 70% should be ruled out, such as untreated hyperthyroidism
in those with mitral stenosis, mitral regurgitation, and and chronic lung disease. Patients with structural heart
tricuspid regurgitation. Patients with mitral stenosis and disease usually develop AF from underlying cardiac con-
mitral regurgitation had a 52% prevalence of AF, whereas ditions, such as mitral stenosis from rheumatic heart
isolated mitral stenosis had a 29% prevalence.43 disease, hypertension, ischemic heart disease, and car-
Cardiomyopathy and congestive heart failure are diomyopathy.
found in up to 30% of AF patients.44 AF has been The physical examination should focus on findings
reported in 10% to 28% of patients with hypertrophic associated with the conditions mentioned in the preced-
cardiomyopathy.45 Surprisingly, stable coronary artery ing text. Electrocardiography will verify the presence of
disease remains an uncommon cause of AF. In the AF, as well as other abnormalities such as prior MI, left
Coronary Artery Surgical Study involving 18,000 patients ventricular hypertrophy, bundle branch block, or pre-
with chronic, stable coronary artery disease documented excitation. A chest radiograph can aid in assessment of
by angiography, AF was present in only 0.6%.46 In acute lungs and cardiac silhouette. Transthoracic echocardio-
severe ischemia, however, AF can be precipitated by graphy can be used to evaluate atrial size, ventricular
hypoperfusion of the left atrium, especially during an function, and valvular heart disease. It may also identify
acute inferior wall myocardial infarction (MI);47,48 AF has a thrombus in the left atrium, although sensitivity is su-
been shown to occur transiently in 6% to 10% of patients perior with transesophageal echocardiography (TEE) for
with an acute MI. The development of AF during an MI identification of thrombi in the left atrium or left atrial
portends a worse prognosis due to comorbidities such as appendage. Exercise stress testing might be indicated
older age and heart failure. to investigate exercise-induced AF or detect underly-
Hyperthyroidism appears to be a significant pre- ing ischemic heart disease. Holter monitoring will help
cipitating factor in AF. In one population-based study document intermittent AF, poor rate control, or other as-
involving 40,000 patients with clinical hyperthyroidism, sociated arrhythmias. Finally, electrophysiologic studies
8% had AF.49 In patients older than 60 years, AF occurred may be needed to identify the focus of AF that may be
in 10% to 20%, but in patients younger than 40 years, the amenable to possible catheter ablation.
risk was <1%. The underlying pathophysiology usually in-
cludes a hyperdynamic circulation secondary to increased Management
sympathetic stimulation, hypersensitivity of -receptors,
and dilated cardiomyopathy. Hyperthyroidism should al- The principle objectives in the management of AF include
ways be suspected in cases of AF in the absence of cardiac ventricular rate control and prevention of thromboem-
causes. Subclinical hyperthyroidism, defined by a low bolic complications. In any patient, a logical approach
serum thyroid-stimulating hormone (TSH) concentration includes the following questions:
with a normal serum thyroid hormone level, appears to
increase the risk of AF fivefold.50 Spontaneous reversion Is the patient hemodynamically stable or unstable?
to sinus rhythm often occurs within 6 weeks in patients Has the duration of AF been more than 48 hours?
who achieve a euthyroid state. Patients older than 60 years Is AF associated with a preexcitation syndrome such as
often demonstrate an age-related decline in the frequency Wolff-Parkinson-White (WPW) syndrome?
of spontaneous reversal.51 Does the patient have ongoing ischemia, cardiomyopa-
Chronic lung disease, especially chronic obstructive thy, or congestive heart failure?
pulmonary disease and obstructive sleep apnea, appears For unstable patients with a rapid ventricular re-
to produce right ventricular dysfunction as a result of sponse resulting in deteriorating hemodynamics, emer-
chronic hypoxia and pulmonary hypertension. Longstand- gent cardioversion is indicated. Primary indications for
ing pulmonary hypertension ultimately causes chronically urgent cardioversion include significant arterial hypoten-
elevated atrial pressures and dilation, leading to AF. Pa- sion and poor perfusion of vital organs, particularly in pa-
tients with untreated obstructive sleep apnea have a higher tients with severe underlying cardiopulmonary and cere-
frequency of AF recurrence if it remains untreated.52 brovascular disease. Manifestations of life-threatening
conditions brought on by acute AF include congestive
Assessment and Evaluation heart failure and coronary or cerebral ischemia. AF in pa-
tients with an underlying preexcitation syndrome may
Assessment of patients with AF should include a his- result in extremely rapid ventricular rates and severe
tory and physical examination, ECG, chest radiograph, hypotension, which also requires urgent cardioversion.
echocardiogram, and thyroid function tests. Further in- Following cardioversion, atrial contraction may be im-
vestigation, when indicated, includes Holter monitoring, paired by the stunning effect of electric discharge, and
exercise stress testing, and electrophysiologic studies. The therefore the risk of thrombus formation remains high. As
intent of these studies is to define symptoms, the clinical a result, these patients should remain on anticoagulation
type, the frequency and duration of AF episodes, as well as therapy for at least 1 month.
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 265
In cases of stable AF, the practice guidelines of the Patients with AF who have an accessory AV pathway
ACC/AHA/ESC recommend rate control and chronic anti- with a pre-excitation syndrome (e.g., WPW syndrome)
coagulation for most patients. For patients with preserved can present with a ventricular rate exceeding 250 bpm,
left ventricular function, -blockers (e.g., atenolol) or cal- associated with a widened QRS complex due to abnormal
cium channel blockers (e.g., diltiazem) are recommended. conduction over their accessory pathway. This pattern
-blockers remain the preferred choice in patients with may lead to the erroneous diagnosis of VT. Appropriate
ischemic heart disease, and calcium channel blockers may drug therapy for these patients can include procainamide
be preferable in patients with lung disease. Digoxin should or ibutilide (although reported to terminate AF in the
be considered for patients with a history of heart failure presence of WPW, this is an unlabeled indication for
or in elderly patients with poor exercise tolerance. ibutilide.10 )
Even with a well controlled ventricular rate, chronic Electric cardioversion can also be used. Drugs that
AF is associated with increased thromboembolic strokes slow normal AV conduction without slowing the accessory
ranging from 3% to 5% per year.53 The risk of stroke in- pathway (e.g., -blockers, digoxin, or calcium channel
creases with advancing age >70 years and in the presence blockers) are contraindicated. Adenosine and lidocaine
of underlying diseases such as congestive heart failure, are ineffective and are also contraindicated, because their
hypertension, diabetes, rheumatic heart disease, valvular use will delay appropriate therapy. Once the rhythm is
disease, and history of prior thromboembolic events.54 In stabilized, patients with WPW and AF should be evaluated
most cases, chronic anticoagulation is achieved with war- for catheter ablation of the accessory pathway.
farin. Aspirin can be substituted when a contraindication Finally, caution is warranted in the acute treatment
to warfarin exists, but has less effectiveness for preventing of patients with significant cardiomyopathy or history
thromboembolic complications.55 of congestive heart failure when using calcium channel
In stable patients who present with new onset AF blockers. These drugs can further depress myocardial
function, which could aggravate the cardiac failure.
of <48 hours duration, the risk of thromboembolism is
In these patients, IV amiodarone and digitalis (or -
very low. Conversion to sinus rhythm can be achieved
blocker in the patient with stable cardiomyopathy) are the
either with pharmacologic agents or electric direct cur-
preferred drugs for rate control. Sotalol is contraindicated
rent cardioversion. Success rates vary from 30% to 60%
in patients with impaired left ventricular function.
using drugs such as dofetilide, flecainide, ibutilide and
propafenone.56 Amiodarone is usually less effective and
often is reserved for patients with left ventricular dysfunc- Intraoperative Atrial Fibrillation
tion. Electric cardioversion usually achieves better results
The acute intraoperative onset of AF during general anes-
(75% to 93% success rate) and is free of proarrhythmia
thesia should be considered a serious cardiac event with
complications.57
potentially life-threatening consequences. As with any in-
In patients with AF lasting more than 48 hours of
traoperative arrhythmia, mechanical irritation of the atria
unknown onset, history of pulmonary embolism, known
should be identified and eliminated. For instance, the in-
preexisting atrial thrombi, and attempted conversion to
sertion of a guide wire during the placement of a central
sinus rhythm should be delayed until the patient has been
venous catheter has been associated with both atrial and
adequately anticoagulated for four weeks. If cardiover-
ventricular dysrhythmias, including APCs, PVCs, AF, and
sion is needed before completion of the anticoagulation
VT. Sympathetic or parasympathetic discharge, which can
therapy, TEE evaluation should take place immediately
occur during manipulation of the trachea (intubation or
before cardioversion to rule out the presence of intracar- tractions), heart, or brainstem, or during peritoneal trac-
diac thrombi. TEE studies in patients with AF longer than tion, can result in AF. Once these surgical manipulations
48 hours showed the presence of left atrial appendage are recognized and discontinued, the arrhythmia usually
thrombus in approximately 15% of patients with AF. resolves. In cases of severe hemodynamic deterioration
The Assessment of Cardioversion Using Transesophageal caused by AF, immediate cardioversion is indicated.
Echocardiography trial randomized more than 1,200 pa- Even in the stable patient, the onset of new AF should
tients to either TEE arm (TEE and cardioversion and be carefully investigated and properly managed. Any
4 weeks of warfarin therapy) or through a conventional underlying cardiac diseases that may have precipitated
approach (3 weeks of therapeutic anticoagulation fol- the AF, such as myocardial ischemia and congestive
lowed by electric cardioversion and 4 weeks of warfarin heart failure should be identified and treated. An arterial
therapy). The rate of embolic events were similar in the blood gas should be obtained to rule out hypoxemia,
groups, but patients in the TEE arm had a shorter dura- hypercarbia, acidosis, or alkalosis. Other laboratory tests
tion of AF and higher cardioversion success rates, with should include an electrolyte panel. Specifically, serum
less bleeding.58 potassium and serum magnesium should be obtained, as
After successful cardioversion, more than 70% of pa- hypokalemia and hypomagnesemia are common in the
tients without maintenance antiarrhythmic therapy will perioperative period and may contribute to the AF. These
experience recurrence of AF.59 Amiodarone is generally electrolyte abnormalities are also common in patients
believed to be the most effective agent for this pur- who have been on diuretic therapy. If patients have been
pose, followed by sotalol and flecainide.60 Unfortunately, on digoxin preoperatively, a digoxin level determination
long-term prophylactic therapy can be associated with may be needed to confirm a therapeutic level or rule
significant side effects. out toxic levels. Placement of a central venous pressure
266 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 18.9 Torsade de pointes (TdP). This strip was obtained from a 64-year-old woman with
acquired long QT syndrome owing to moxifloxacin administration. Before the antibiotic, her QTc
was 450 ms, afterward it was 515 ms. She underwent preoperative chemotherapy with paclitaxel.
Immediately following her segmental mastectomy and axillary dissection with general anesthesia,
she experienced her first episode of TdP at extubation. Her underlying rhythm is sinus at 80 bpm
(P waves are marked with the bars). The top trace is electrocardiogram (ECG) lead II and the bottom
trace is the invasive arterial pressure. Treatment consisted of electric cardioversion and phenytoin,
because her electrolytes were within normal range. (Courtesy of Joseph Swafford, MD, FACC.)
268 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
The majority (80%) of patients presenting with a sustained SVT with aberrant conduction, 24-hour Holter monitoring
wide QRS tachycardia are diagnosed with VT. SVT with may be prudent to document the frequency of VT episodes
aberrant conduction is relatively rare. and their association with potential precipitating factors.
When in doubt, always assume the dysrhythmia Signal-averaged ECG is useful to identify patients at high
represents VT, and not SVT with aberrant conduction, risk for developing VT.74 In patients with substrates pre-
especially in patients with ischemic heart disease. How- disposing to VT, the conduction of the cardiac impulse
ever, the following criteria are helpful to distinguish VT is slowed by areas of abnormal myocardium affected by
from SVT with aberrant conduction: necrosis, fibrosis, and inflammation. These areas produce
small electric potentials (late potentials) that arrive later
Compare the atrial rate to the ventricular rate. In SVT,
than the normally conducted action potential. These ven-
the dysrhythmia arises from the atria and propagates to
tricular late potentials are on the order of microvolts,
the ventricles. There will likely be some type of AV block
too small to be detected on the surface ECG, although
present (owing to the high rate), so the ventricular rate
they can be identified with signal-averaged ECG while
will be lower than, or at best equal to, the atrial rate.
the patient is in sinus rhythm. Exercise testing can help
On the other hand, the VT originates in the ventricles
detect coronary artery disease and may be useful to in-
and can propagate retrograde to the atria. Therefore,
duce catecholamine-sensitive VT.75 An echocardiogram
in VT, the ventricular rate will be at least equal to, if
will help detect structural heart diseases. Finally, electro-
not higher than, the atrial rate. Unfortunately, in many
physiologic studies constitute the most reliable method of
cases of VT, it is difficult to discern the P waves without
confirming the diagnosis of VT.76 They also allow docu-
use of special leads such as esophageal leads.
mentation of the hemodynamic consequences of the VT,
In general, a QRS complex >140 ms is present with VT
and identification of patients at high risk for occurrence
and <140 ms in SVT with aberrant conduction.
of VF who would require an ICD.77
The QRS morphology in SVT with aberrant conduction
Because the VTs comprise a very heterogeneous
usually follows a RBBB pattern. On the other hand,
group of arrhythmias, each particular type of VT will
because of its ventricular origin, the QRS morphology
be discussed individually, including etiology, associated
in VT often assumes a more bizarre appearance, diseases, and treatment.
typically resembling a ventricular beat originating from
a ventricular pacemaker.
In a patient with a history of ischemic heart disease,
especially with low LVEF, and without previous history UNSTABLE VENTRICULAR
of paroxysmal tachycardia, a wide complex tachycardia TACHYCARDIA
is more likely to be VT instead of SVT with aberrant
conduction.71 The degree of hemodynamic compromise induced by VT
depends not only on the rapidity of the ventricular rate,
but also on the presence and severity of underlying car-
Pathophysiology
diac diseases and left ventricular function.78,79 The word
The ventricular substrate most favorable to the genesis unstable confers serious signs and symptoms of hemo-
of VT consists of an area of abnormal myocardium dynamic deterioration, such as mental status change,
next to an area of healthy myocardium, such as areas dyspnea, or angina. Signs include rales, rhonchi, pul-
of myocardial fibrosis following a myocardial infarct, monary edema, hypotension, and acute ECG changes
or areas of distended myocardium due to congestive consistent with ischemia. In this setting, unstable VT
heart failure. This juxtaposition creates a setting whereby requires immediate therapy with synchronized cardiover-
slowed conduction, unidirectional block, reentry, and sion to restore normal cardiac rhythm. In addition to
perpetuation of reentrant circuits will precipitate VT.72,73 supplemental oxygen, continuous ECG monitoring and
On the other hand, a homogeneous, completely healed, oxygen saturation by pulse oximetry should be instituted.
fibrotic scar from an old infarct is much less likely to Intravenous access should be secured, and equipment
cause reentry and VT. for endotracheal intubation should be ready, as further
patient deterioration may be imminent.
Evaluation
Because VT can deteriorate to VF and sudden cardiac STABLE VENTRICULAR
arrest, reversible conditions that precipitate or sustain VT TACHYCARDIA
should be investigated, identified, and rapidly corrected.
These causes include electrolyte imbalances (especially In the hemodynamically stable patient with VT, an at-
potassium, magnesium and calcium), drug toxicities, and tempt should be made to differentiate monomorphic from
serious cardiac disease such as acute coronary ischemia polymorphic VT. In monomorphic VT, the QRS complexes
and congestive heart failure. Even after the acute episode are wide, regular, and stable, with a uniform configura-
of VT has reverted to sinus rhythm, a comprehensive tion appearing almost identical in shape. The morphology
investigation including a careful history, physical exam- does not change from one beat to another. Monomorphic
ination, and cardiac workup is indicated. Although a VT can be further divided into two types, depending on
12-lead ECG is usually adequate to differentiate VT from the presence or absence of underlying heart disease.
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 269
In the patient with monomorphic VT and underlying acute myocardial ischemia secondary to coronary artery
structural heart diseases, the most common pathologies stenosis or coronary artery spasm. In many instances, the
include ischemic coronary artery disease and dilated conventional history of angina on exertion or obvious ST
cardiomyopathy. In the ischemic heart, the origin of VT abnormalities may not be present before the attack of
typically involves an extensive fibrotic scar following a MI. VT.82 Treatment is directed primarily towards relieving
The juxtaposition of scar tissue and healthy myocardium myocardial ischemia with nitroglycerin, -blockers, or
serves as an anatomic basis for slowed conduction and calcium channel blockers. In the acute setting, electric
reentry. The risk is highest in the first months following cardioversion (unstable hemodynamics) or amiodarone
an acute MI. Even when the infarct is well healed and therapy (stable hemodynamics) should be employed.
the ischemia is well controlled, these patients still carry a Amiodarone administration is indicated following electric
significant risk for occurrence of VT many years later.80 cardioversion. For chronic refractory cases, coronary
Many patients with VT have underlying, nonis- revascularization may be required.83
chemic, dilated cardiomyopathy. This heterogeneous Polymorphic VT with a prolonged baseline QT inter-
group includes patients with decompensated valvular val often presents as TdP, in which the initial deflection
heart disease, alcoholic and nutritional cardiomyopathies, of the QRS complexes follow a periodic pattern of change
hypertensive cardiomyopathy, chemotherapy-induced from an upward or positive direction to a downward
cardiomyopathy, and viral myocarditis. Here again, the or negative direction. This condition can either be con-
presence of patchy areas of dilated myocardium next to genital or acquired,84 and a prolonged QTc (QT interval
healthy myocardial tissue serves as substrates for conduc- corrected for heart rate, see the subsequent text) portends
tion blocks and reentry. Monomorphic VT in the patient an increased incidence of sudden cardiac death.85 In the
without underlying heart disease (idiopathic VT) carries congenital type, mutations in seven different genes desig-
a better prognosis than VT in the presence of structural nated as LQT 1 to 7 have been identified. These genotypes
disease. Examination of the QRS morphology and its axis usually express their clinical manifestations through two
offers clues to the site of VT origin and the treatment of principal phenotypes: the Romano-Ward syndrome and
choice. Monomorphic VT presenting with RBBB config- the Jervell-Lange-Nielsen syndrome. The former is more
uration and a superior axis usually originates from the common, transmitted through an autosomal dominant
apex of the left ventricle. It often responds to verapamil, mechanism, and involves only the heart. The latter is more
and, for this reason, is referred to as verapamil-sensitive rare, transmitted through an autosomal recessive mech-
VT. It occurs most commonly in adolescents and young anism, and is associated with congenital sensorineural
adults with no detectable structural heart disease. It can deafness.86 Excessive adrenergic stimulation caused by
be induced by atrial pacing and can be readily terminated physical exertion or emotional stress may precipitate VT,
by verapamil, which also is effective in preventing recur- with resultant syncope or cardiac arrest. blockade, aim-
rence. Because of its relatively narrow QRS complexes and ing to reduce sympathetic stimulation of the myocardium,
its favorable response to verapamil, this type of VT has is the treatment of choice. Left stellate ganglion sympa-
often been misdiagnosed as supraventricular tachycardia thectomy for refractory cases has been done with variable
with aberrant conduction. success. The rationale behind this treatment is the belief
Monomorphic VT with left bundle branch block that VT is caused by unbalanced predominance of the left
(LBBB) morphology and an inferior axis is referred to sympathetic system over the right.
as adenosine-sensitive VT. It is usually precipitated by In the acquired type, TdP can be precipitated by se-
excessive caffeine, strenuous exercise, and psychologic vere electrolyte abnormalities such as hypomagnesemia
stress. It usually affects young to middle-aged adults. and hypokalemia. A large number of classes of drugs also
An exercise stress test or infusion of a catecholamine known to cause prolongation of the QT interval include
such as isoproterenol will induce this arrhythmia and antiarrhythmic agents, antihistamines, antibiotics, and
an adenosine infusion can be used to abolish it. After tricyclic antidepressants. Class IA antiarrhythmic drugs
the acute episode, -blockers are the drug of choice for such as quinidine, procainamide, and disopyramide and
maintenance therapy to prevent recurrences. class III drugs such as sotalol, dofetilide, and amio-
darone have been most often implicated. Furthermore,
the risk of TdP increases dramatically in the setting of
combined effects of hypokalemia, hypomagnesemia, and
POLYMORPHIC antiarrhythmic drugs. Finally, other significant contribut-
VENTRICULAR TACHYCARDIA ing factors to the occurrence of TdP include congestive
heart failure and stroke.
In polymorphic VT, the morphology of the QRS complexes Depending on the severity of the hemodynamic
changes from one QRS complex to the next. Polymorphic compromise, the modalities of treatment may include
VT can be subdivided into two types: Those with a normal urgent cardioversion or ventricular overdrive pacing.
baseline QT interval and those with a prolonged QT In more stable patients, intravenous magnesium is the
interval. These QT intervals may be obtained from the drug of first choice. It has the advantages of being
patients 12-lead ECG recorded before the onset of current effective in both reverting the TdP and preventing
VT.81 its recurrence. Other effective modalities of treatment
Polymorphic VT with normal baseline QT interval include -blockers, overdrive pacing, and implantation of
remains the most common presentation of VT following an ICD.70,87
270 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Measurement of the QT interval can be difficult, and digitalis toxicity. The administration of digoxin-immune-
ECG-interpreted QT intervals can be incorrect.88 Using FAB (Digibind), as well as correction of hypokalemia
the standard Bazett formula to correct for the normal and hypomagnesemia (if present), remain crucial to the
shortening of the QT interval with increasing heart rate, management of bidirectional VT due to digitalis toxicity.93
Phenytoin has also been used (unlabeled) in this setting.
Further, cardioversion is not indicated in this situation
QTc = QT(measured)/square root of RR interval because it carries a high risk of precipitating refractory
[in seconds] VF in the setting of digitalis toxicity.
38. Moe GK, Abildskov JA. Atrial fibrillation as a self-sustaining of prosthetic valve thrombosis: Results of the International
arrhythmia independent of focal discharge. Am Heart J. PRO-TEE Registry. J Am Coll Cardiol. 2004;43:77.
1959;58:59. 59. Pritchett EL. Management of atrial fibrillation. N Engl J Med.
39. Tsang TS, Gersh BJ, Appleton CP, et al. Left ventricular 1992;326:1264.
diastolic dysfunction as a predictor of the first diagnosed 60. Roy D, Talajic M, Dorian P, et al. Canadian Trial of Atrial
nonvalvular atrial fibrillation in 840 elderly men and women. Fibrillation Investigators. Amiodarone to prevent recurrence
J Am Coll Cardiol. 2002;40:1636. of atrial fibrillation. N Engl J Med. 2000;342:913.
40. Vaziri SM, Larson MG, Benjamin EJ, et al. Echocardio- 61. Maisel WH, Rawn JD, Stevenson WG. Atrial fibrillation after
graphic predictors of nonrheumatic atrial fibrillation. The cardiac surgery. Ann Intern Med. 2001;135:1061.
Framingham heart study. Circulation. 1994;89:724. 62. Vaporciyan AA, Correa AM, Rice DC, et al. Risk factors
41. Kochiadakis GE, Skalidis EI, Kalebubas MD, et al. Effect of associated with atrial fibrillation after noncardiac thoracic
acute atrial fibrillation on phasic coronary blood flow pattern surgery: Analysis of 2588 patients. J Thorac Cardiovasc Surg.
and flow reserve in humans. Eur Heart J. 2002;23:734. 2004;127:779.
42. Clark DM, Plumb VJ, Epstein AE, et al. Hemodynamic effects 63. Crystal E, Connolly SJ, Sleik K, et al. Interventions on
of an irregular sequence of ventricular cycle lengths during prevention of postoperative atrial fibrillation in patients
atrial fibrillation. J Am Coll Cardiol. 1997;3:1039. undergoing heart surgery: A meta-analysis. Circulation. 2002;
43. Diker E, Aydogdu S, Ozdemir M, et al. Prevalence and 106:75.
predictors of atrial fibrillation in rheumatic valvular heart 64. Sedrakyan A, Treasure T, Browne J, et al. Pharmacologic
disease. Am J Cardiol. 1996;77:96. prophylaxis for postoperative atrial tachyarrhythmia in
44. Stevenson WG, Stevenson LW, Middlekauff HR, et al. general thoracic surgery: Evidence from randomized clinical
Improving survival for patients with atrial fibrillation and trials. J Thorac Cardiovasc Surg. 2005;129:997.
advanced heart failure. J Am Coll Cardiol. 1996;28:1458. 65. Bigger JT Jr. Relation between left ventricular dysfunction
45. Robinson K, Frenneaux MP, Stockins B, et al. Atrial and ventricular arrhythmias after myocardial infarction. Am
fibrillation in hypertrophic cardiomyopathy: A longitudinal J Cardiol. 1986;57:8B.
study. J Am Coll Cardiol. 1990;15:1279. 66. Kennedy HL, Whitlock JA, Sprague MK, et al. Long-term
46. Cameron A, Schwartz MJ, Kronmal RA, et al. Prevalence and follow-up of asymptomatic healthy subjects with frequent
significance of atrial fibrillation in coronary artery disease and complex ventricular ectopy. N Engl J Med. 1985;312:193.
(CASS Registry). Am J Cardiol. 1988;61:714. 67. Buxton AE, Lee KL, Fisher JD, et al. Multicenter Unsustained
47. Crenshaw BS, Ward SR, Granger CB, et al. Atrial fibrillation Tachycardia Trial Investigators. A randomized study of the
in the setting of acute myocardial infarction: The GUSTO- prevention of sudden death in patients with coronary artery
I experience. Global Utilization of Streptokinase and TPA disease. N Engl J Med. 1999;341:1882.
for Occluded Coronary Arteries. J Am Coll Cardiol. 1997;30: 68. Cleland JG, Ghosh J, Freemantle N, et al. Clinical trials
406. update and cumulative meta-analyses from the American
48. Wong CK, White HD, Wilcox RG, et al. New atrial fibrillation College of Cardiology: WATCH, SCD-HeFT, DINAMIT,
after acute myocardial infarction independently predicts CASINO, INSPIRE, STRATUS-US, RIO-lipids and cardiac
death: The GUSTO-III experience. Am Heart J. 2000;140:878. resynchronisation therapy in heart failure. Eur J Heart Fail.
49. Frost L, Vestergaard P, Mosekilde L. Hyperthyroidism and 2004;6:501.
risk of atrial fibrillation or flutter: A population-based study. 69. Bardy GH, Lee KL, Mark DB, et al. Amiodarone or an
Arch Intern Med. 2004;164:1675. implantable cardioverter-defibrillator for congestive heart
50. Auer J, Scheibner P, Mische T, et al. Subclinical hyperthy- failure. N Engl J Med. 2005;352:225.
roidism as a risk factor for atrial fibrillation. Am Heart J. 70. Khan IA. Long QT syndrome: Diagnosis and management.
2001;142:838. Am Heart J. 2002;143:7.
51. Woeber KA. Thyrotoxicosis and the heart. N Engl J Med. 71. Akhtar M, Shenasa M, Jazayeri M, et al. Wide QRS complex
1992;327:94. tachycardia. Reappraisal of a common clinical problem. Ann
52. Kanagala R, Murali NS, Friedman PA, et al. Obstructive sleep Intern Med. 1988;109:905.
apnea and the recurrence of atrial fibrillation. Circulation. 72. Josephson ME, Horowitz LN, Farshidi A. Continuous local
2003;107:2589. electrical activity. A mechanism of recurrent ventricular
53. Frost L, Engholm G, Johnsen S, et al. Incident stroke tachycardia. Circulation. 1978;57:659.
after discharge from the hospital with a diagnosis of atrial 73. de Bakker JM, van Capelle FJ, Janse MJ, et al. Reentry as a
fibrillation. Am J Med. 2000;108:36. cause of ventricular tachycardia in patients with chronic
54. Atrial Fibrillation Investigators. Risk factors for stroke and ischemic heart disease: Electrophysiologic and anatomic
efficacy of antithrombotic therapy in atrial fibrillation. correlation. Circulation. 1988;77:589.
Analysis of pooled data from five randomized controlled 74. Gomes JA, Winters SL, Ip J, et al. Identification of patients
trials. Arch Intern Med. 1994;154:1449. with high risk of arrhythmic mortality. Role of ambulatory
55. Go AS, Fang MC, Singer DE. Antithrombotic therapy for monitoring, signal-averaged ECG, and heart rate variability.
stroke prevention in atrial fibrillation. Prog Cardiovasc Dis. Cardiol Clin. 1993;11:55.
2005;48:108. 75. Stevenson WG, Brugada P, Waldecker B, et al. Clinical,
56. Nichol G, McAlister F, Pham B, et al. Meta-analysis angiographic, and electrophysiologic findings in patients
of randomised controlled trials of the effectiveness of with aborted sudden death as compared with patients
antiarrhythmic agents at promoting sinus rhythm in patients with sustained ventricular tachycardia after myocardial
with atrial fibrillation. Heart. 2002;87:535. infarction. Circulation. 1985;71:1146.
57. Gallagher MM, Guo XH, Poloniecki JD, et al. Initial 76. Volgman AS, Zheutlin TA, Mattioni TA, et al. Reproducibility
energy setting, outcome and efficiency in direct current of programmed electrical stimulation responses in patients
cardioversion of atrial fibrillation and flutter. J Am Coll with ventricular tachycardia or fibrillation associated with
Cardiol. 2001;38:1498. coronary artery disease. Am J Cardiol. 1992;70:758.
58. Tong AT, Roudaut R, Ozkan M, et al. Transesophageal 77. Moss AJ, Hall WJ, Cannom DS, et al. Multicenter Automatic
echocardiography improves risk assessment of thrombolysis Defibrillator Implantation Trial Investigators. Improved
C H A P T E R 18 / C A R D I A C D Y S R H Y T H M I A S 273
survival with an implanted defibrillator in patients with 85. Straus SM, Kors JA, De Bruin ML, et al. Prolonged QTc
coronary disease at high risk for ventricular arrhythmia. interval and risk of sudden cardiac death in a population of
N Engl J Med. 1996;335:1933. older adults. J Am Coll Cardiol. 2006;47:362.
78. Hamer AW, Rubin SA, Peter T, et al. Factors that predict 86. Chiang CE, Roden DM. The long QT syndromes: Genetic
syncope during ventricular tachycardia in patients. Am basis and clinical implications. J Am Coll Cardiol. 2000;36:1.
Heart J. 1984;107(5 Pt 1):997. 87. Passman R, Kadish A. Polymorphic ventricular tachycardia,
79. Lima JA, Weiss JL, Guzman PA, et al. Incomplete filling long Q-T syndrome, and torsades de pointes. Med Clin North
and incoordinate contraction as mechanisms of hypotension Am. 2001;85:321.
during ventricular tachycardia in man. Circulation. 1983;68: 88. Charbit B, Samain E, Merckx P, et al. QT interval mea-
928. surement: Evaluation of automatic QTc measurement and
80. Mukharji J, Rude RE, Poole WK, et al. Risk factors for new simple method to calculate and interpret corrected QT
sudden death after acute myocardial infarction: Two-year interval. Anesthesiology. 2006;104:255.
follow-up. Am J Cardiol. 1984;54:31. 89. Yildirim H, Adanir T, Atay A, et al. The effects of sevoflurane,
81. Nguyen PT, Scheinman MM, Seger J. Polymorphous ventric- isoflurane and desflurane on QT interval of the ECG. Eur J
ular tachycardia: Clinical characterization, therapy, and the Anaesthesiol. 2004;21:566.
QT interval. Circulation. 1986;74:340. 90. Owczuk R, Sawicka W, Wujtewicz MA, et al. Influence of
82. Wolfe CL, Nibley C, Bhandari A, et al. Polymorphous spinal anesthesia on corrected QT interval. Reg Anesth Pain
ventricular tachycardia associated with acute myocardial Med. 2005;30:548.
infarction. Circulation. 1991;84:1543. 91. Paventi S, Santevecchi A, Ranieri R. Effects of sevoflurane
83. Eisenberg SJ, Scheinman MM, Dullet NK, et al. Sudden versus propofol on QT interval. Minerva Anestesiol. 2001;67:
cardiac death and polymorphous ventricular tachycardia 637.
in patients with normal QT intervals and normal systolic 92. Morris SN, Zipes DP. His bundle electrocardiography during
cardiac function. Am J Cardiol. 1995;75:687. bidirectional tachycardia. Circulation. 1973;48:32.
84. Khan IA. Clinical and therapeutic aspects of congenital and 93. Kelly RA, Smith TW. Recognition and management of
acquired long QT syndrome. Am J Med. 2002;112:58. digitalis toxicity. Am J Cardiol. 1992;69:108G.
CHAPTER HYPOTENSION AND SHOCK STATES
CASE SUMMARY provide the energy for the blood to circulate through
the system. This is a rather complex system with several
properties:
64-year-old man presented to the emergency
A
room with fever of 39.44 C, vomiting, and ab- It is elastic.
dominal pain. His medical history was signif- It has a mean pressure, which is determined by more
icant for hypertension and congestive heart than just the pumping action of the heart.
failure, with a left ventricular ejection frac- It fills via passive and active mechanisms.
tion of 40%. His vital signs revealed a blood Because of passive filling, mean cardiovascular pres-
pressure of 120/90 mm Hg and a heart rate of 110 bpm. The sure is dependent upon volume and compliance. In
serum hemoglobin was 14 g per dL and a white cell count of addition, the pumping action of the heart provides the
20,000 per mm3 . The intraoperative course was character- blood with a significant amount of kinetic energy while
ized by several episodes of hypotension, which responded consuming vastly different amounts of chemical energy.
to intravenous fluids and repeated doses of ephedrine. A The physics of the circulatory system in its most basic
sigmoid diverticular perforation was found and repaired. elements can be described by Ohms Law, which mathe-
Anticipating prolonged mechanical ventilation, the patient matically relates flow, resistance and pressure. As applied
was left intubated. In the intensive care unit (ICU), his to the flow of electricity it is generally expressed as:
blood pressure decreased to 70/50 mm Hg without ade-
quate response to a 500 cc bolus of intravenous normal V=IR (1)
saline solution. A dopamine infusion produced a mild
response. A transesophageal echocardiography (TEE) ex- where V is the voltage, or electromotive force, I is the
amination revealed significant hypokinesia of the lateral current, and R is the resistance. In physiologic terms, the
and inferior walls and an estimated ejection fraction of voltage is equivalent to the mean intravascular hydrostatic
25%. A pulmonary artery catheter (PAC) was then inserted. pressure (P) and the current is the flow (Q), with R as the
His pulmonary artery pressure was 50/28 mm Hg, wedge expression of vascular resistance.
pressure of 24 mm Hg, and a cardiac output of 6.5 L. Nore-
pinephrine was substituted for dopamine, which increased P = Q R, or Q = P/R (2)
the blood pressure to 110/60 mm Hg. Subsequently, the ad-
dition of milrinone and nesiritide improved the ejection Therefore, if there is no pressure differential along the
fraction and urinary output. Following triple antibiotic circuit, there is no flow, and if there is a high resistance
therapy, the patients temperature decreased and his leuko- without increase in pressure, there is little flow. In prac-
cytosis improved. During the next several days, he was tical terms, the pressure (P) is either the systemic mean
weaned from the ventilator and intravenous infusions. Af- arterial pressure (MAP) or mean pulmonary artery pres-
ter a long and protracted hospitalization, he was eventually sure (MPAP), with the flow (Q) as cardiac output and the
discharged home. resistance (R) as either systemic vascular resistance (SVR)
or pulmonary vascular resistance (PVR).
Much of the basic research into the physiology of
the heart was performed a century ago by German
What Is the Basic Hemo- physiologist, Otto Frank, and English physiologist, Ernest
dynamic Physiology Relevant to Starling. More recently, extensive work has been done by
Kiichi Sagawa et al. at Johns Hopkins.
Hypotension and Shock States? Otto Frank appreciated the necessity of applying
physics to the study of biology. He also recognized the
The cardiovascular system is a circuit with two hydraulic need for measurement systems, which could respond
pumps placed in series. The pumps (right and left heart) with speed and precision. Franks work, Zur Dynamik des
274
C H A P T E R 19 / H Y P O T E N S I O N A N D S H O C K S TAT E S 275
D Ejection
ESPVR
EAE
C (Contractility)
(Afterload)
Pressure
Pressure
Isovolumic Isovolumic
relaxation contraction
EDPVR
(Compliance
Stroke
B
A Filling volume
EDV
ESV (Preload)
Volume
Volume
FIGURE 19.2 The pressure-volume loop with inclusion of
FIGURE 19.1 A diagrammatic representation of the parameters of contractility, compliance, preload, and resistance
pressure-volume loop. A: Opening of mitral valve and filling of (afterload). EDPVR, end-diastolic pressure-volume relation;
ventricle. B: Beginning of ventricular contraction while both ESPVR, end-systolic pressure-volume relation; EAE, effective
mitral and aortic valves remain closed; C: Opening of aortic arterial elastance; ESV, end-systolic volume; EDV, end-diastolic
valve with ejection of stroke volume; D: Closure of aortic valve at volume.
end-systole.
Normal
therefore related to afterload, for example, the cardiac
ventricular wall tension during ejection. The x-intercept
of the EAE reflects preload or ventricular end-diastolic
volume (VEDV). The intercept of the EAE and ESPVR Hypodynamic
is the pressure-volume at which the aortic valve closes,
and is the volume at end-systole (ESV). The area within
the loop is cardiac work, and the distance between the
two limbs of the loop (upward = isovolumic contraction,
downward = isovolumic relaxation) is the stroke volume Preload
(SV). Figure 19.2 demonstrates the pressure-volume loop
with the inclusion of controlling parameters. FIGURE 19.3 Starling ventricular function curves
Ernest Henry Starling spent a great deal of his demonstrating relation of preload to ventricular output at
professional career defining the relation of the heart different levels of contractility.
and its function.3 He examined methods to vary blood
276 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
(PAWP) or pulmonary artery occlusive pressure (PAOP) response syndrome (SIRS).5 These patients frequently
for the left ventricle. This, which will be further discussed exhibit manifestations of hypovolemia (increased capil-
in the subsequent text, creates difficulty when examining lary permeability and vasodilation). They are also in a
the role of ventricular compliance and diastolic function hyperdynamic state (vasodilation from microbial toxins
on cardiac performance using Starling curves. and circulating cytokines) and suffer cardiogenic depres-
For each heart, there is typically a family of sion (negative inotropic effects of cytokines, microbial
curves, rather than a single curve. This family, then, toxins, and coronary hypoperfusion).
allows for variations in contractility: Higher curves
typically indicate higher contractility, with a higher SV
for a given VEDV. Because Starling did not believe that
vascular resistance had any negative impact on ventricular
HYPOVOLEMIC SHOCK
function, changes in resistance and afterload cannot be Hypovolemic shock encompasses a range of entities in
characterized by the Starling curves. which the effective circulating blood volume is inade-
quate. Although there are some rather obvious causes of
hypovolemic shock, as shown in Table 19.1, including
acute blood loss, there are also less obvious causes in-
What Are the Diagnostic cluding acute vasodilation (e.g., loss of sympathetic tone),
Categories and Pathophysiology increased capillary permeability, third-space fluid shifts,
and obstruction of venous return (tension pneumothorax
of Shock States? and cardiac tamponade).
Because the underlying problem is essentially inad-
Shock, and its manifestations, can be defined as a syn- equate preload, supportive treatment usually consists of
drome of failure of the heart to pump volume in suffi- intravascular fluids: Crystalloids, colloids, blood prod-
cient quantity and under sufficient pressure, to maintain ucts, and so on. The reduction in preload (VEDV) leads
the pressure-flow relation for adequate tissue perfusion to a decrease in SV, triggering compensatory tachycar-
and the maintenance of aerobic metabolism. Traditional dia and vasoconstriction. Figure 19.4 shows the graphic
teaching includes three primary categories of shock: representation of hypovolemic shock as depicted by the
(i) Hypovolemic, (ii) cardiogenic, and (iii) septic. This Starling ventricular function curves (see Fig. 19.4A) and
classification fails to recognize other etiologies, including the pressure-volume loop (see Fig 19.4B). The benefit of vi-
anaphylactic, inflammatory, obstructive, and neurogenic. sualizing the pathophysiology of this shock state using the
Whereas obstructive shock results in a decrease in preload pressure-volume loop is that not only can the reduction
leading to a low cardiac output and high resistance, the in preload and SV be depicted, but also the compensatory
others mentioned exhibit a similar pathophysiology to sympathetic and adrenergic-mediated vasoconstriction.
septic shock; that is, a low resistance and elevated cardiac
output. This has led to the use of the term, hyperdy-
namic, which is used to characterize any shock state CARDIOGENIC SHOCK
manifested by vasodilation with a compensatory increase
in cardiac output. A simplistic approach to any given clin- Cardiogenic shock is commonly due to acute cardiac
ical syndrome of hypoperfusion often leads to a single ischemia, which can result from primary coronary hy-
pathophysiologic diagnosis and treatment related to a sin- poperfusion (hypotension, coronary obstruction) or an
gle form of shock. However, a better understanding of the acute increase in oxygen demand (e.g., hypertensive cri-
pathophysiology shows that frequently, there are mixed sis). Other causes, as shown in Table 19.1, include negative
shock states, with coexisting components of hypovolemic, inotropic agents, cytokines, cardiomyopathies and my-
cardiogenic, and hyperdynamic shock. A prime example ocarditis, structural cardiac defects, and dysrhythmias.
are patients suffering from the systemic inflammatory Most of these etiologies are characterized by systolic
Normal
Normal
Ventricular output
Ventricular output
A Preload
A Preload
Pressure
Pressure
B Volume
ESPVR
Vasoconstriction
EAE
Pressure
Pressure
Vasodilation
EDPVR
Volume
Volume
FIGURE 19.7 The effects of vasoconstriction () and
vasodilation (. . . . . .) on the effective arterial elastance (EAE,) and
FIGURE 19.6 Diastolic (lusitropic) dysfunction with decreased
ventricular compliance as represented by the pressure-volume resultant effects on ventricular output at a constant preload and
loop. The rise in the end-diastolic pressure-volume relation contractility, as represented by the pressure-volume loop.
EDPVR, end-diastolic pressure-volume relation; ESPVR,
(EDPVR: lower arrows) represents the lusitropic dysfunction,
resulting in a decrease in ventricular compliance. Compensatory end-systolic pressure-volume relation; EAE, effective arterial
vasoconstriction is demonstrated as an increase in the slope of elastance.
the effective arterial elastance (EAE: upper arrow). Note that true
preload or end-diastolic volume is decreased, although
end-diastolic pressure often measured as a surrogate of true degree A-V block). Compensatory tachycardia is asso-
preload is increased. ciated with increased myocardial oxygen demand and
decreased supply. The decrease in myocardial oxygen
delivery to the myocardium is caused by a decrease in
diastolic time. With 70% of coronary perfusion occur-
reduction in preload; moreover, there is a concomitant ring during diastole (predominantly to the left ventricle),
increase in heart rate and SV. A high SV is difficult to shortening of the diastolic time may prove critical and
conceptualize using the ventricular function curves of
Starling, because circulating cytokines and toxins depress
contractility and systolic function. The evaluation of
myocardial performance in most hyperdynamic shock
states will demonstrate either no change in contractility
or a negative inotropic effect. How then does one explain
the apparent paradox of increased SV despite reduced
preload? The answer lies in the analysis of the pressure-
volume loop. Figure 19.7 demonstrates the changes
Pressure
Volume
How Are Heart Rate and Shock FIGURE 19.8 Hyperdynamic (septic) shock as represented by
the pressure-volume loop. Note the vasodilatation demonstrated
Related? by the decrease in slope of the effective arterial elastance (EAE),
and reduced preload (arrows) without change in contractility.
Tachycardia is the most common dysrhythmia in all forms The end result is an increase in ventricular output (stroke
of shock, with the obvious exception of bradydysrhyth- volume and therefore, cardiac output).
mias that can occur with cardiogenic shock (e.g., third
C H A P T E R 19 / H Y P O T E N S I O N A N D S H O C K S TAT E S 279
result in mismatch of myocardial oxygen supply/demand, acute respiratory distress syndrome (ARDS), seen during
with resultant coronary ischemia and cardiac failure. hemodynamic resuscitation, is an often observed conse-
However, treatment directed primarily towards reducing quence of unmonitored fluid administration during shock.
heart rate with negative chronotropic agents (-blockers Maintenance of CVP at 12 to 15 mm Hg, or a pulmonary
or Ca++ channel antagonists) will often not be tolerated artery occlusive pressure (PAOP) in the same range, can
due their coexisting negative inotropic effects. Instead, be helpful in minimizing fluid overload. However, correct
efforts should always be predominantly directed towards interpretation of these values must take into account the
eradication of the underlying process and improvement influence of ventricular compliance and the impact of
of SV. selective left or right heart dysfunction.
A discussion on the choice of volume expanders
is beyond the scope of this discussion. However, the
importance of maintaining proper preload throughout re-
suscitation is emphasized, and other issues are addressed
How Is Shock Managed? such as providing adequate oxygen-carrying capacity, nor-
malizing coagulation, maintaining intravascular oncotic
The management of shock requires a series of both di- pressure, and understanding the impact of various crys-
agnostic and therapeutic maneuvers, as well as serial talloid preparations on electrolyte and acidbase balance.
assessments to evaluate the response to therapy. Diag- One of the many controversial issues in the treat-
nostic maneuvers include physical examination, as well ment of hemodynamic instability is the management of
as more invasive monitoring such as continuous arterial coexisting acidbase disequilibrium. Excessive lactic acid
pressure monitoring, central venous or pulmonary arte- formation due to accompanying tissue hypoxia may re-
rial pressure monitoring, central venous or pulmonary sult in critical metabolic acidosis. Severe reductions in
artery assessment of mixed venous oxygen saturation, blood pH (7.00) may further compromise tissue per-
and echocardiography. Concomitant information derived fusion and cerebral and cardiac function, and interfere
from a PAC and echocardiography can be particularly with the effectiveness of inotropes and vasopressors. In
helpful in evaluating the relation of VEDV and VEDP in the last few decades, aggressive attempts to normal-
real time. Therapeutic maneuvers, which may also aid ize the acid pH have been replaced by a less intense
in the diagnosis, include fluid challenges and a variety approach: Recognizing the hazards of alkalosis and bi-
of pharmacologic agents. The initial choice of pharma- carbonate administration. Untoward effects seen with the
cology is influenced by the physicians assessment of the rapid, intravascular administration of buffer agents in-
adequacy of cardiac output and preload. clude: Decreased ionized calcium, decreased tissue oxygen
The management of shock syndromes into the middle availability because of a left shift in the oxyhemoglobin
of the 20th century was often predicated on providing dissociation curve, and decreased intracellular pH due to
adequate perfusion pressure. Little regard was given for the rapid intracellular transit of nonionized carbon diox-
the adverse effects of persistent organ hypoperfusion, ex- ide produced by the buffering of bicarbonate. Because
acerbated by the use of vasoconstrictor agents such as of these sequelae, physicians allow for mild to moderate
norepinephrine. This frequently led to bowel and renal acidosis and rely on reperfusion and tissue oxygenation to
ischemia and a protracted course of organ dysfunction, correct it. Profound acidosis, however, still warrants treat-
with eventual demise. Eventually, the need to provide ment, particularly in patients who depend on vasopressor
adequate flow to peripheral tissues and vital organs was and inotropic support. Although predominantly guided
recognized. Clinicians then became aware of the impor- by empiric observations, most clinicians consider treating
tance of insuring adequate intravascular volume as the profound acidosis when the pH is <7.25, whereas the ad-
initial and critical phase of resuscitation. Pharmacologic ministration of bicarbonate without acidbase evaluation
support was also revised, with emphasis on improving should be discouraged.
cardiac output and relegating the use of vasoconstric- General considerations when choosing pharmaco-
tor agents to situations of persistent hypotension despite logic agents for the treatment of shock include mechanism
adequate flow. of action, duration of action, ease of titration, preferen-
Optimizing preload relative to cardiac function re- tial effects on specific organ systems, and agent-specific
mains the cornerstone of hemodynamic resuscitation. toxicity. In addition, receptor responsiveness is an im-
Selection of volume expanders or the use of venodilators portant property that should guide the choice of a given
and diuretics to reduce fluid overload can be challeng- drug, particularly in conditions where the response to
ing, particularly in the presence of diastolic dysfunction. adrenergic stimulation may be attenuated, such as with
Although assessment of preload with central or PAOP elderly patients or those with widespread sepsis and/or
monitoring can be of considerable value in many situa- inflammation. Most agents that are used to increase per-
tions, reliance on these indices of VEDP as a surrogate fusion pressure and blood flow ultimately rely on the
for volume can be misleading. The addition of echocar- release of free ionized calcium into the intracellular space,
diography to evaluate cardiac volume can be of benefit whether by the sarcoplasmic reticulum or through cal-
in identifying the role that cardiac pathology plays in cium channels, thereby facilitating the bond between
influencing ventricular pressure-volume relations. Car- the contractile proteins. The maintenance of normal
diogenic pulmonary edema and/or the increased tendency ionized calcium is, therefore, critical to the effective ac-
towards extravascular lung water accumulation in the tion of both exogenous and endogenous inotropic agents
280 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
and vasopressors. The measurement of ionized calcium liberation of norepinephrine at the sympathetic nerve
levels and administration of calcium chloride should be terminal.
considered part of the resuscitation protocol in shock. The relevance for understanding these mechanisms
However, the excessive administration of calcium may in- is the often observed failure to achieve the goals for car-
duce cardiac dysrhythmias, as well as vasospasm-induced diac output augmentation by relying solely on -receptor
end-organ dysfunction, and should be avoided. stimulation, thereby requiring additional agents to pro-
vide synergistic activity. The necessity for using combined
inotropic stimulation to achieve an adequate response
is most commonly observed in older patients in whom
Which Inotropic Agents Can Be -receptor responsiveness is decreased, and in those with
inflammatory septic processes in which cytokines impair
Used to Treat Hypotension? -receptor performance.6 In these situations, the selec-
tion of inotrope is critical if one is to avoid an undesired
The use of positive inotropes should be considered vasoconstrictor effect. The use of dopamine to achieve an
when SV is inadequate, despite optimal levels of preload increase in contractility often results in a vasoconstrictor
and/or evidence of decreased myocardial contractility. response, as the dose required to stimulate -receptors by
This decision should be based on direct measurement of this weak stimulant often exceeds 10 mg/kg/minute, with
cardiac output and index, and determination of mixed resultant -receptorinduced vasoconstriction and end-
venous oxygen saturation or visual assessment of systolic organ hypoperfusion. Conversely, dobutamine, another
function (e.g., echocardiography). When choosing an of the weak stimulants, often results in vasodilation
inotrope, the effects on contractility and vascular tone as its predominant effect, thereby worsening hypoten-
should be considered. Table 19.2 summarizes both the sion. One proposed pharmacologic option to achieve a
mechanism of action and the accompanying vascular desired contractility response is to take advantage of the
effects of commonly used inotropic pharmacology. opposed vascular effects of norepinephrine and dobu-
Although all inotropic agents ultimately perform their tamine to improve myocardial inotropy and minimize
task by increasing free intracellular ionized calcium, the change in vascular resistance.7,8 Epinephrine is the initial
mechanism to attain that response may differ. Adrenergic inotrope preferred by many clinicians due to its strength
agonists stimulate sarcolemmal -receptors, thereby acti- of -receptor stimulation and the predictability of its dose
vating adenyl cyclase, whereas drugs such as milrinone or response. Epinephrine, in doses up to 170 ng/kg/minute,
enoximone act predominantly by inhibition of phophodi- will demonstrate increased inotropic stimulation with
esterase type III. These two pathways increase intracellu- minimal change in vascular resistance, increasing car-
lar cyclic adenosine monophosphate (cAMP), which leads diac index, and oxygen delivery. Doses above 170 to
to increased intracellular calcium. Other agents (e.g., 200 ng/kg/minute may increase vascular resistance, which
digoxin) will inhibit membrane Na-K ATPase, thereby may lead to a decrease in both cardiac index and oxy-
opening calcium channels in the cell membrane. Within gen delivery.9 This doseresponse effect may account for
these categories of agents, those directly stimulating the the reported failure of epinephrine to improve splanch-
-receptor include epinephrine, norepinephrine, isopro- nic oxygen delivery when compared to the combination
terenol, dopamine, dobutamine, glucagon, and ephedrine. of norepinephrine and dobutamine, as several of these
Ephedrine also acts indirectly through the increased studies utilized doses of epinephrine far in excess of
200 ng/kg/minute.7,8 A later study using epinephrine
dosage in the 100 to 200 ng/kg/minute range demon-
TABLE 19.2 Pharmacologic Agents Available strated improved splanchnic perfusion with epinephrine
for the Treatment of Shock with Associated Inotropic compared with norepinephrine and dobutamine.10 If a
and Vascular Effects -receptor stimulant is selected to achieve an inotropic
response, and a favorable result is not achieved, additional
Adenyl Cyclase Stimulants pharmacologic therapy should include a synergistic drug
that acts through a different mechanism, for example,
Isoproterenolinotrope/vasodilator
milrinone or digoxin.11
Dobutamineinotrope/vasodilator
Table 19.2 also shows the different vascular effects
Epinephrineinotrope/vasodilator
of these agents. Afterload reduction, as a result of
Norepinephrinevasoconstrictor/inotrope
vasodilation with dobutamine and milrinone, has been
Phenylephrinevasoconstrictor
shown to benefit some patients with cardiogenic shock,
Glucagoninotrope
whereas epinephrine, norepinephrine, and dopamine may
Phosphodiesterase Inhibitor be expected to increase myocardial work and oxygen
Milrinoneinotrope/vasodilator demand as vasoconstriction develops. The management
Na-K ATPase Inhibitor of ischemic-mediated cardiogenic shock can create a
dilemma for the clinician who recognizes the potential
Digoxininotrope/splanchnic constrictor
exacerbation of coronary ischemia secondary to diastolic
Endothelin Stimulation hypotension arising from vasodilation. Awareness of the
Vasopressinvasoconstrictor underlying state of the vasculature in patients with
coronary ischemia and shock is critical in determining
C H A P T E R 19 / H Y P O T E N S I O N A N D S H O C K S TAT E S 281
the preferable inotrope when one is required. Occasional Examination of the noncardiac effects of these agents
success with intra-aortic balloon counterpulsation in often raises the question of renal benefit from strategies
cardiogenic shock, and in septic shock complicated by such as low dose dopamine. The dopaminergic-1 (DA-1)
cardiac failure, is likely because of the increase in coronary receptor stimulation leading to vasodilation of the af-
perfusion by diastolic augmentation while decreasing ferent renal arteriole, produced by administration of 2 to
afterload, favoring the myocardial oxygen supply relation 3 g/kg/minute of dopamine, gained significant popularity
in a manner difficult to achieve with pharmacologic as a theoretic means for renal protection and improvement
cardiac support. in renal function in patients at risk for ischemic-mediated
Digoxin is rarely chosen for acute inotropic interven- renal insufficiency and oliguria. Recent evidence, how-
tion. Its narrow therapeutic index makes it a risky choice ever, has failed to demonstrate this benefit, and the
in patients subjected to rapid changes in potassium, cal- routine use of dopamine for renal protection has now been
cium, acidbase balance, and metabolism. Digoxin also nearly abandoned.13 There remains, however, a potential
produces splanchnic vasoconstriction which can lead to beneficial effect of dopaminergic stimulation in patients
bowel ischemia. The reason it still occupies a place on receiving agents that stimulate the -receptor. Both ani-
the therapeutic armamentarium is that its mechanism of mal and human data have demonstrated that the increase
action may prove useful in the presence of ineffective in renal vasoconstriction and decrease in renal blood
stimulation, or tolerance to the effect of drugs such as flow (RBF) accompanying norepinephrine infusion may
milrinone. be reversed by the simultaneous administration of 2 to
Most of the agents discussed in the preceding text 3 g/kg/minute of dopamine, demonstrating an increase
have the potential for tachydysrhythmias. The cost of in RBF, paralleling an increase in perfusion pressure, and
uncoupling myocardial oxygen supply and demand with minimizing any effect of norepinephrine on renal vascu-
increasing heart rate creates justifiable concern, partic- lar resistance.14,15 Consideration may therefore be given
ularly in patients with known or suspected myocardial to the administration of low dose, renal dose dopamine
ischemia. This response, coupled with profound vasodi- in patients receiving stimulants, including epinephrine,
lation, has restricted the utilization of isoproterenol, the norepinephrine, or phenylephrine.
classic -receptor agonist, solely for the pharmacologic
support of severe bradycardia. Norepinephrine, because
of its concomitant vasoconstriction, and epinephrine in
doses within its predominant inotropic range, tend to What Role Do Vasopressor
produce less tachycardia than may be anticipated from Agents Play in the Treatment
their strong stimulation. Clinicians often recognize that
avoiding tachycardia is often best accomplished with aug- of Hypotension and Shock?
mentation of contractility, and therefore the chronotropic
effects of these agents may not manifest if the inotropic The use of vasoconstrictor, vasopressor pharmacology
benefit is realized. was the basis for much of hemodynamic resuscitation
Much of this discussion has involved inotropic ther- into the 1960s when, with the increasing recognition of
apy and focused on the pharmacologic support of con- the negative impact of augmenting vascular resistance
tractility and systolic function. Less commonly, patients to maintain perfusion pressure without attention to car-
will present in cardiogenic failure and shock with diastolic diac output (preload and contractility), clinicians realized
or lusitropic dysfunction. In these states, functional im- that tissue hypoperfusion resulting from this therapy led
pairment is one of myocardial relaxation, and therefore to delayed morbidity and mortality, secondary to end-
a decrease in compliance is manifested by low cardiac organ ischemia and necrosis, most notably in the renal
output with low VEDV (preload), yet with elevated VEDP and splanchnic circulations. With improved monitoring
and pulmonary edema. A physiologic approach to ther- by central venous and pulmonary artery catheterization,
apy would be to increase the diastolic uptake of ionized indicator dilution cardiac output, mixed venous oxime-
calcium into the sarcoplasmic reticulum and facilitate try, and echocardiography, a more physiologic approach
the pumping of calcium out of the cell. Because pure to hemodynamic insufficiency has appropriately led to
lusitropic agents are not clinically available, chronic ther- far less use of vasoconstrictor agents as first-line ther-
apy is often predicated on decreasing the liberation of apy for shock. Recent evidence has been enthusiastically
free ionized calcium during systole with calcium chan- received which demonstrated that early emergency de-
nel blockers or -receptor antagonists. Acutely, however, partment, goal-directed therapy of shock within the first 6
such an approach would likely worsen the clinical picture hours following admission could improve outcome.16 This
by adding systolic dysfunction to an already compromised approach emphasizes the early administration of fluid
cardiac state. Evidence has shown benefit from the use therapy to optimize intravascular volume. Once sufficient
of dobutamine.12 This is most likely a consequence of its preload has been achieved, the recommendations include
vasodilation which reduces impedance to ejection while the administration of a vasopressorspecifically nore-
maintaining or augmenting systolic function. Care of pinephrineif hypotension persists. There is little argu-
these patients requires close attention to the adequacy of ment with recognizing that a delay in therapy to improve
preload and caution in any administration of diuretic ther- perfusion of vital organs mitigates against success and that
apy for pulmonary edema. Echocardiography can prove improving myocardial and cerebral perfusion with vaso-
quite valuable in assessing ventricular size and VEDV. pressors should be beneficial. Also, the short-term use of
282 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
any agent that improves perfusion pressure has at least observed in patients who are in shock, chronically ill, or
some potential for minimizing ischemic damage to vital abuse cocaine.
organs. Concern, however, must be expressed as to the im- Controversy and uncertainty exists over the correct
plication some have taken from this and other studies to timing for vasopressor therapy. Studies involving criti-
rely on vasoconstrictor therapy for treating hypotension cally ill, surgical patients demonstrated an optimal level
during a prolonged course of resuscitation.17 The result of cardiac index of 4.50 L/min/m2 , beyond which fur-
of such an approach fails to recall the untoward con- ther elevation did not improve the systemic uptake of
sequences of end-organ necrosis that was demonstrated oxygen.19,20 This supernormal hemodynamic approach
into the second half of the 20th century. Early therapy gained modest popularity in the treatment of shock, par-
in the emergency department to improve intravascular ticularly in critically ill, surgical patients, with emphasis
volume and increase perfusion pressure was the objective on the utilization of fluid and inotropic therapy to achieve
of these studies, and not advocacy for a new approach a CI of 4.50 L/min/m2 . Subsequent studies in nonspe-
to overall shock management. Examination of the pub- cific, critically ill patients were unable to demonstrate
lished evidence supporting early goal-directed therapy in a benefit for the routine use of supernormal values of
shock included the often forgotten admonition to assess CI and oxygen delivery, with one showing a trend to-
mixed venous oxygen saturation with oximetry-capable, wards increased mortality in the supernormally treated
central venous catheters to determine the need for in- group.21,22 Where, then, does that leave the clinician in
otropic therapy.16 determining the proper timing for vasoconstrictor ther-
Although fewer in number compared to inotropic apy in the treatment of shock? Most often, consideration
agents, some of the same considerations should prevail, for such an approach will arise in patients with hyper-
namely mechanism of action and nonvascular effects of dynamic shock. One empiric approach favored by these
the drug. Table 19.2 outlines the mechanism of action authors is to reserve the use of vasoconstrictor agents to
and vascular and inotropic effects of these agents. As with patients with persistent hypotension, having achieved a
inotropic agents, vasoconstrictor selection offers differing CI of 4.50 L/min/m2 . Whether this value is accepted or
mechanisms of action. The agents commonly selected as other considerations adopted for the use of vasopressors
first options to increase vascular resistance and perfusion agents, it is imperative that the clinician continuously and
pressure do so by stimulation of -1 adrenergic receptors. carefully assesses the adequacy of tissue and end-organ
The agents in this category are phenylephrine and nore- perfusion while administering agents whose mechanism
pinephrine. As is the case with -receptor responsiveness, of action is to elevate vascular resistance.
-receptors have also shown decreased responsiveness in
patients with sepsis and systemic inflammation, likely
because of cytokine release, as manifested by failure to
respond to escalating doses of stimulants. To gain a What Types of Diagnostic
synergistic effect, adding additional -adrenergic drugs Therapy Can Be Used in the
would provide little benefit. Instead, the addition of a va-
sopressor with an alternative mechanism of action can Management of Shock?
provide added benefit; such is the case for vasopressin,
whose vasoconstriction results from the stimulation and Although the discussions of fluid therapy as in the preced-
release of endogenous endothelin. Recent work with va- ing text, acidbase balance, and pharmacologic support
sopressin (antidiuretic hormone) has indicated both a are all consistent with the management of hemodynamic
relative lack of the hormone in septic shock and a signif- insufficiency and hypoperfusion syndromes, there are
icant clinical improvement with its use in this setting.18 a number of causes of shock that require intervention
Despite low endogenous levels, vasopressin receptor sen- directly and are specifically related to the identified etiol-
sitivity appears to be increased, with an organ-specific ogy. Hypovolemic shock secondary to acute hemorrhage
heterogeneity leading to vasodilation in some distribu- and/or trauma often requires immediate surgical inter-
tions and vasoconstriction in others. This inconsistent vention to control the source of blood loss. Cardiogenic
response can lead to coronary vasoconstriction, thereby shock resulting from acute myocardial infarction may
requiring caution and increased vigilance when adminis- require immediate therapy with anticlotting agents or in-
tered to patients with known or suspected coronary artery vasive therapies including angioplasty or coronary artery
disease. Dosing of vasopressin, as inferred from experi- bypass. Emergent surgical intervention may also be nec-
mental data, is recommended between 0.01 and 0.10 U per essary in cases of acute cardiac and septal perforation or
minute. acute papillary muscle rupture with mitral valve failure.
Phenylephrine is a direct-acting, pure 1 agonist. It ac- Pneumothorax and cardiac tamponade are acute causes
tivates -adrenergic receptors in vascular smooth muscle, of obstructive shock with low cardiac output that may
causing vasoconstriction. Norepinephrine is a potent ag- require emergent invasive intervention.
onist and moderate agonist, and peripheral vascular re- The area of shock receiving the most attention
sistance is increased through vasoconstriction. Ephedrine both experimentally and clinically today is that of hy-
exhibits both a direct and indirect agonist activity and perdynamic shock resulting from sepsis and systemic
stimulates the release of endogenous epinephrine and inflammation. Antibiotic selection and the surgical man-
norepinephrine. As such, it is a poor choice in the patient agement of septic foci are among the mainstays of current
who has exhausted their natural catechols, as is often therapy. Attempts to identify common etiologic factors
C H A P T E R 19 / H Y P O T E N S I O N A N D S H O C K S TAT E S 283
have included examination of the role of bacterial toxins and also exerts an anti-inflammatory effect by inhibiting
and cytokines, including tumor necrosis factor (TNF), the production of inflammatory cytokines. A study in
proinflammatory interleukins (IL-1 and IL-6), platelet ac- 2001 demonstrated a significant reduction in mortality
tivating factor (PAF), complement activation, nitric oxide in patients treated with APC.35 The recommended rou-
(NO, endothelial derived relaxant factor [EDRF]), to name tine, indiscriminate use of this agent in all patients with
just a few. Therapies directed towards the individual septic physiology cannot be currently supported. Hem-
factors named earlier have provided mixed results. An- orrhagic complications are well recognized and, limit
tibodies to endotoxin, complement, and TNF have not patient selection. Subsequent evaluation suggests that
resulted in considerable success.23,24 A similar lack of APC be restricted to those patients exhibiting greater
positive results have been shown for receptor antago- morbidity and risk of death according to acute physiol-
nists to proinflammatory interleukins25 and the use of the ogy and chronic health evaluation (APACHE) criteria.36,37
anti-inflammatory interleukin, IL-10.26 Where the inhibi- More extensive use and further evaluation will help define
tion of production of all NO resulted in increased blood the patients in whom it will be most helpful.
pressure and decreased need for pharmacologic support,
the resultant end-organ hypoperfusion proved unaccept-
able.27 These efforts involved the inhibition of all forms
of NO, including constitutive (naturally occurring) and What Is the Impact of Shock
inducible (produced as a result of septic or inflammatory
stimulation). Recent enthusiasm has developed over the
States on Organ Function?
ability to selectively inhibit inducible nitric oxide syn-
thase (INOS), thereby maintaining the baseline-naturally Although well beyond the scope of this discussion, clin-
occurring EDRF and normal vascular tone.28 Much of icians must be aware of the common vital organ distur-
this work, along with other studies that have examined bances that must be addressed, often at the same time
adrenergic receptor responsiveness, have arisen through that resuscitation is taking place. Multiorgan dysfunction
the application of genomic physiology and pharmacology, syndrome (MODS) is the manifestation of prolonged hy-
and the study of gene expression. poperfusion that most correlates with mortality.38 Main-
In further examining the currently available adjunc- tenance of oxygenation and acidbase balance in shock
tive therapy in hyperdynamic and septic shock, the role is often complicated by the development of respiratory
of steroids must be considered. In the early 1970s, insufficiency due to cardiogenic pulmonary edema, fluid
steroid therapy in supraphysiologic doses were consid- overload, ARDS, exhaustion, aspiration, or other coex-
ered critical to the successful treatment of septic shock. isting pulmonary processes. Oxygen therapy, intubation,
A controversial study recommended 30 mg per kg of and ventilatory support are all therapeutic options that
methylprednisolone, and claimed a reduction in mortality may have to be instituted with the inception of treatment.
from 40% to 10%.29 Eventually, two separate multicen- Fluid and electrolytes, as well as acidbase balance, may
ter studies demonstrated a lack of effect of the routine be complicated by renal insufficiency. Most evaluations
administration of these high dosages of steroids, and have suggested that the development of renal failure, with
the practice was soon abandoned.30,31 Recently, however, the necessity for dialysis as a complication of shock, is the
depressed baseline cortisol levels and corticotrophin stim- single most ominous organ failure contributing to mortal-
ulation have been identified in septic patients, and steroids ity in MODS. The evaluation of renal protective therapies
have been shown to decrease the duration of vasopressor are ongoing. For the present time, proper physiologic-
dependency and duration of shock.32,33 The mechanism based, hemodynamic support to optimize tissue perfusion
for this beneficial response is still debated. Steroids may provides the best opportunity to avoid this complication.
increase the responsiveness of - and -adrenergic recep- Other organ dysfunction, including hepatic, pancreatic,
tors, decrease the gene expression of INOS,34 or mediate cerebral and gastrointestinal, must all be anticipated, with
as an anti-inflammatory agent against proinflammatory treatment as necessary when it occurs.
cytokines. Because baseline cortisol levels in septic and
inflammatory shock patients are expected to be elevated, Summary
current recommendations are that, if a presteroid, serum
cortisol level is <25 g per dL, therapy should be initi- The management of hemodynamic instability and shock
ated with hydrocortisone, 100 mg every 8 hours. As these presents one of the greatest acute challenges for the clini-
laboratory studies may take considerable time, following cian. Recognition of individual and environmental limita-
blood draw for assessment, steroid therapy with the above tions should always be considered, with rapid stabilization
dosage should be considered in patients requiring escalat- and transfer when necessary and feasible. Mortality in the
ing pharmacologic support, with continuation dependent best of hands approaches 35% to 40% in septic shock,39
on laboratory results. and even higher with cardiogenic shock. The emphasis in
Activated protein C (APC, drotrecogin alfa) was ap- this discussion has focused on an understanding of basic
proved in 2002 by the Food and Drug Administration physiology and the recognition of individualized patho-
(FDA) for use in sepsis. Down-regulation of thrombo- physiology as a guide to therapy. The choice of fluid and
modulin causes a decrease in the activation of protein drug must be driven by a clear understanding of the phar-
C during sepsis. Activated protein C inhibits the gener- macophysiologic relations in the patient and their disease.
ation of thrombus by inactivating factor Va and VIIIa Research must and will continue to exert a positive impact
284 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
on the high mortality in these hypoperfusion syndromes 10. Sequin P, Bellisant E, Le Tilzo Y, et al. Effects of
and to define new therapeutic approaches. A critical eval- epinephrine compared with the combination of donutamine
uation of all that we call evidence-based medicine is and norepinephrine on gastric perfusion in septic shock. Clin
essential to delineate that therapy which truly offers new Pharmacol Ther. 2002;71:381.
11. Gilbert EM, Hershberger RE, Weichmann RJ, et al. Pharma-
advantages and positively affects outcome. Therapies that
cologic and hemodynamic effects of combined beta-agonist
are counterintuitive to the goal of normalizing physiology stimulation and phosphodiesterase inhibition in the failing
and minimizing tissue hypoperfusion must be suspect heart. Chest. 1995;108:1524.
and, if attempted, done so with careful assessment and 12. Parker JD, Landzberg JS, Bittl JA, et al. Effects of beta-
close observation for untoward effects. adrenergic stimulation with dobutamine on isovolumic
relaxation in the normal and failing human left ventricle.
Circulation. 1991;84:140.
13. Bellomo R, Chapman M, Finfer S, et al. Low-dose dopamine
in patients with early renal dysfunction: A placebo-controlled
KEY POINTS randomized trial. Lancet. 2000;356:2139.
1. A clear understanding of basic hemodynamic physi- 14. Schaer GL, Fink MP, Parillo JE. Norepinephrine alone versus
ology is an essential component of the management norepinephrine plus low-dose dopamine: Enhanced renal
of hypotension and shock. blood flow with combined pressor therapy. Crit Care Med.
2. A clear delineation of pathophysiology, using all 1985;13:492.
15. Richer M, Robert S, Lebel M. Renal hemodynamics during
diagnostic and monitoring capabilities, provides the
norepinephrine and low-dose dopamine infusions in man.
best approach to these patients. Crit Care Med. 1996;24:1150.
3. Understanding the different pharmacologic proper- 16. Rivers E, Nguyen B, Havsted S, et al. Early goal-directed
ties of the hemodynamic agents is necessary to therapy in the treatment of severe sepsis and septic shock.
provide a rational therapeutic approach. N Engl J Med. 2001;345:1368.
4. Constant evaluation of the response to therapy, and 17. Dellinger RP, Carlet GM, Masur H, et al. Surviving sepsis
the implications of the effects of the chosen options, campaign guidelines for management of severe sepsis and
is essential. septic shock. Crit Care Med. 2004;32:858.
5. Awareness of the evidence, or lack thereof, of chosen 18. Rozenfeld V, Cheng JW. The role of vasopressin in the
therapies and the most up-to-date treatments is an treatment of vasodilation in shock states. Ann Pharmacother.
2000;34:250.
important responsibility of the clinician.
19. Shoemaker WC, Appel MPA. Prospective trial of supranormal
values of survivors as therapeutic goals in high risk surgical
REFERENCES patients. Chest. 1988;94:1176.
20. Boyd O, Grounds M, Bennett ED. A randomized clinical trial
1. Chapman CB, Wasserman E. Translation of Otto of the effect of deliberate perioperative increase of oxygen
Franks paper Zur Dynamik des Herzmuskels,. Z Biol. delivery on mortality in high-risk surgical patients. JAMA.
1895;32:370447; Am J Heart. 1959;58(2):282317. 1993;270:2699.
2. Sagawa K, Maughan L, Suga H, et al. Cardiac contraction 21. Hayes MA, Timmins AC, Yao EHS, et al. Elevation of
and the pressure-volume relationship. New York: Oxford systemic oxygen delivery in the treatment of critically ill
University Press; 1988. patients. N Engl J Med. 1994;330:1717.
3. Patterson SW, Starling EH. On the mechanical factors which 22. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented
determine the output of the ventricles. J Physiol (London). hemodynamic therapy in critically ill patients. N Engl J Med.
1914;48:357. 1995;333:1025.
4. Starling EH. The Linacre lecture on the law of the heart. In: 23. Haqaekonen S, Scannon P, Mischak RO, et al. Phase 2 study
Chapman B, Mitchell JH, eds. Starling on the heart. London: of murine monoclonal antilipid A antibody in bacteremic
Dowsons; 1965:121147. and nonbacteremic patients. Antimicrob Agents Chemother.
5. Bone RC. Toward a theory regarding the pathogenesis of 1988;32:710.
systemic inflammatory response syndrome: What we do 24. Cohen J, Carlet J. Intersept: An international muticenter
and do not know about cytokine regulation. Crit Care Med. placebo-controlled trial of monoclonal antibody to human
1996;24:163. tumor necrosis factor-alpha in patients with sepsis. Crit Care
6. Silverman HJ, Penaranda R, Orens JB, et al. Impaired beta- Med. 1996;24:1431.
adrenergic stimulation of cyclic adenosine monophosphate 25. Opal SM, Fisher CJ, Dhainaut JFA. Confirmatory inter-
in human septic shock: Associated with myocardial hy- leukin-1 receptor antagonist trial in severe sepsis: A phase III,
poresponsiveness to catecholamines. Crit Care Med. 1993; randomized, double-blind, placebo-controlled, multicenter
21:31. trial. Crit Care Med. 1997;25:1115.
7. Levy B, Bollaert PE, Charpentier C, et al. Comparison 26. Gerard C, Bruyns C, Marchant A, et al. Interleukin-10
of norepinephrine and dobutamine to epinephrine for reduces the release of tumor necrosis factor and prevents
hemodynamics, lactate metabolism and gastric tonometric lethality in experimental endotoxemia. J Exp Med. 1993;177:
variables in septic shock: A prospective randomized study. S47.
Intensive Care Med. 1997;23:282. 27. Avontuur JAM, Nothenius RPT, van Bodegom JW, et al.
8. Meier-Hellmann A, Reinhart K, Bredle DL, et al. Epinephrine Prolonged inhibition of nitric oxide synthesis in severe septic
impairs splanchnic perfusion in septic shock. Crit Care Med. shock: A clinical study. Crit Care Med. 1998;26:660.
1997;25:399. 28. Mitaka C, Hirata Y, Yokoyama K, et al. A selective inhibitor
9. Moran JL, OFathartaigh MS, Peisach AR, et al. Epinephrine for inducible nitric oxide synthase improves hypotension
as an inotropic agent in septic shock: A dose-profile analysis. and lactic acidosis in canine endotoxic shock. Crit Care Med.
Crit Care Med. 1993;21:70. 2001;29:2156.
C H A P T E R 19 / H Y P O T E N S I O N A N D S H O C K S TAT E S 285
29. Schumer W. Steroids in the treatment of clinical septic 34. Korhonen R, Lahti A, Hamalainen M, et al. Dexamethasone
shock. Ann Surg. 1976;184:333. inhibits nitric-oxide synthase expression and nitric oxide
30. Bone RC, Fisher CJ, Clemmer TP, et al. A controlled clinical production by destabilizing mRNA in lipopolysaccharide-
trial of high-dose methylprednisolone in the treatment of treated macrophages. Mol Pharmacol. 2002;62:698.
severe sepsis and septic shock. N Engl J Med. 1987;317:653. 35. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety
31. The Veterans Administration Systemic Sepsis Cooperative of recombinant human activated protein C for severe sepsis.
Study Group. Effect of high dose glucocorticoid therapy on N Engl J Med. 2001;344:699.
mortality in patients with clinical signs of systemic sepsis. 36. Fourrier F. Recombinant human activated protein C in the
N Engl J Med. 1987;317:659. treatment of severe sepsis: An evidence-based review. Crit
32. Bollaert PE, Charpentier C, Levy B, et al. Reversal of late Care Med. 2004;32:S534.
septic shock with supraphysiologic doses of hydrocortisone. 37. Rice TW, Bernard GR. Therapeutic intervention and targets
Crit Care Med. 1998;26:645. of sepsis. Annual Rev Med. 2005;56:225.
33. Briegel J, Forst H, Haller M, et al. Stress doses of 38. Schrier RW, Wang W. Acute renal failure and sepsis. N Engl
hydrocortisone reverse hyperdynamic shock: A prospective, J Med. 2004;351:159.
randomized, double-blind single center study. Crit Care Med. 39. OBrien JM Jr, Abraham E. New approaches to the treatment
1999;27:723. of sepsis. Clin Chest Med. 2003;4:521.
CHAPTER PERIOPERATIVE HYPERTENSION
20 Philip G. Boysen
CASE SUMMARY 3 days, the patient was weaned from nitroglycerin, and
trandolapril, amiodipine, metoprolol, and hydrochlorthi-
41-year-old woman, weighing 135 kg and azide were given for BP control. Upper airway obstruction
A
a body mass index of 42 is referred for cleared, and she was discharged home on the fifth post-
surgery due to inflammatory bowel disease operative day without incident.
and chronic gastrointestinal bleeding, lead-
ing to anemia. She has a history of hyperten-
sion and type II diabetes. Her blood pressure
(BP) and glucose have been poorly controlled, and she INTRODUCTION
is referred for further evaluation and treatment before Hypertension has become a global health care issue, cross-
surgery. Her preoperative BP was 180/90 mm Hg despite ing all geographic boundaries.1 Economic globalization
therapy with fosinopril and verapamil. A -blocker was has led to an increasingly sedentary lifestyle, a high-
added to the regimen about a week before surgery. The pa- calorie and high-fat diet, and an increased incidence of
tient had a full cardiac evaluation within the last 6 months, comorbidities, such as obesity and diabetes mellitus.2
including coronary angiography, which revealed nonob- Even with such negative influences, longevity is increas-
structive coronary disease and a left ventricular ejection ing, and an older population also has a higher incidence
fraction of 55%. She denied orthopnea, and exercise was of hypertension. BP and BP patterns increase with age.
limited by body weight. Her preoperative hematocrit is
Systolic blood pressure (SBP) increases throughout a life-
24%. Her electrocardiogram (ECG) shows left ventricular
time. Diastolic blood pressure (DBP) increases with age,
hypertrophy with strain pattern, unchanged from pre-
but plateaus between the age of 50 and 60, and then be-
vious studies. On the morning of surgery, her BP was
gins to decline. Therefore at approximately the same age,
190/100 mm Hg in the sitting position, with no ortho-
the pulse pressure begins to increase. In patients older
static changes. Her heart rate is 80 bpm, and a fingerstick
than 50 years, isolated systolic hypertension is the most
shows a blood sugar of 120 mg per dL. Owing to poor pe-
common presentation of hypertension.3
ripheral intravenous access, a preoperative central venous
catheter is placed in the right internal jugular vein under
ultrasound guidance. Two units of packed red blood cells
were transfused, and metoprolol was administered and
titrated to achieve a heart rate under 70 bpm. Following a How Has the Definition and
successful induction of general anesthesia, a small bowel Parameters of Hypertension
resection and anastamosis proceeded without incident,
with the BP maintained at approximately 140/80 mm Hg.
Changed?
During emergence and following extubation, her BP rose
steadily to 200/100 mm Hg, and the patient complained of For many years, the major concern to avoid complications
pain. While in the postanesthesia care unit (PACU), she due to hypertension was focused on diastolic hyperten-
remained hypertensive despite the administration of in- sion.4 However, recent data from large observational
travenous narcotics and labetalol. The patient also began studies indicate a closer association between isolated sys-
to complain of obstruction to the upper airway, result- tolic hypertension and coronary artery disease and stroke.
ing in oxygen desaturation to 85%. Therapy with bi-level Often, isolated systolic hypertension is also associated
positive airway pressure (BiPAP) was instituted, and in- with an increased pulse pressure. Both systolic hyper-
travenous nitroglycerin was initiated for control of her tension and increased pulse pressure are now a focus of
hypertension. The patient was then transferred to the in- much more aggressive treatment.5 Whereas tight control
tensive care unit for further management. During the next of BP in the outpatient setting is thought to improve
286
C H A P T E R 2 0 / P E R I O P E R AT I V E H Y P E R T E N S I O N 287
TABLE 20.1 Clinical Classification of Hypertension phenomenon occurs when the SBP is >140 mm Hg, and
the DBP is <90 mm Hg.3 When the difference between the
SBP DBP SBP and the DBP is >90 mm Hg, there is an increased
Category (mm Hg) (mm Hg) risk of stroke, coronary artery disease, and preeclampsia.
The pulse pressure provides information concerning the
Optimal <120 <80
coupling between the left ventricle and the arterial tree.
Normal <130 <85
An elevation of the pulse pressure >65 mm Hg reflects the
Prehypertension 130139 8589
stiffness and loss of elasticity of the conduit vessels, that
Mild hypertension 140159 9099
is, the aorta, and the reflection of the energy wave prop-
Moderate hypertension 160179 100109
agated during ventricular ejection. With vessel stiffening
Severe hypertension >180 >110
and lack of compliance, failure of the vasculature to relax
Isolated systolic >140 <90
and accommodate ejected volume results in early reflec-
hypertension
tion of the pulse wave, which augments systole rather than
Pulse pressure >60
resulting in diastolic augmentation of blood flow to vital
hypertension
organs.16 In turn, the constant and increased stress im-
SBP, systolic blood pressure; DBP, diastolic blood pressure.
posed on the vascular system is exposed and breaks down
the elastic elements of the vascular wall, causing further
changes in time varying elastance of the vascular anatomy.
outcome, the need for tight control in the perioperative Other factors that contribute to loss of vessel distensibility
setting is unclear, making the challenge to the anesthesi- are age, glucose tolerance, coronary artery disease, hyper-
ologist even greater in terms of assessing and managing lipidemia, and inflammatory responses. This progressive
the hypertensive patient.6,7 deterioration is the final common pathway for multiple
In the general population, the association of hyper- risk factors. Increased pulse pressure is also significantly
tension and increased cardiovascular risk is well estab- associated with postoperative cardiac, renal, and cerebral
lished.8,9 Studies involving more than 1 million human events, whereas neither systolic nor diastolic hypertension
subjects indicate that death from ischemic heart disease showed a similar relation.1719
and stroke increases from an SBP as low as 115 mm Hg
and a DBP as low as 75 mm Hg. This has led epidemi-
ologists to examine the assertion that the upper limit of
a normal BP reading is 140/90 mm Hg. Moreover, it How Is the Diagnosis
has prompted the need to reevaluate the target BP levels
for treatment, because the risk of cardiovascular compli- of Hypertension Made?
cations increases at BP readings previously thought to
be normal.10 The current classification of hypertension
based on severity is listed in Table 20.1.
For the anesthesiologist, the problem is even more
MEASUREMENTS DURING
complex. First, there is an increased number of patients SLEEP
with diagnosed hypertension undergoing surgical pro-
cedures. Second, a large number of patients undergoing BP values tend to fluctuate during the waking hours, de-
preoperative evaluation lack the diagnosis of hypertension pending on activity and sympathetic state. There are also
and are clearly unaware, despite sustained BP readings major changes in BP and heart rate during physiologic
in excess of 140/90 mm Hg.11 Third, a subset of treated sleep.20 Light sleep, or REM sleep, is associated with in-
patients have abnormally high preoperative BP readings creased sympathetic tone, resulting in tachycardia and
either due to noncompliance or ineffective therapy.12 elevated BP. Deep sleep is characterized by parasympa-
thetic dominance, with low BP and heart rate. Although
these fluctuations in BP may be in the range of 40 to
50 mm Hg, they are not always innocuous. Myocardial
What Type of Decisions Face ischemia, infarction, and sudden death often occur at the
end of the 90-minute cycle leading to light sleep, and
the Anesthesiologist in the dreaming with rapid eye movement.
Operating Room?
In each of these cases, the anesthesiologist is faced with a MEASUREMENTS IN
decisioneither to proceed with the surgery and the anes- THE SITTING POSITION
thetic, or to cancel or delay the surgery while waiting for
better control of the BP through more aggressive therapy. For the reasons elucidated in the preceding text, the
Although there are no data to indicate that isolated hyper- standards for making the diagnosis of hypertension have
tension alters anesthetic risk, most patients have hyper- recently been closely examined.21,22 Sustained hyperten-
tension in combination with multiple comorbidities.1315 sion is diagnosed only after multiple readings on separate
The issue of increased pulse pressure hypertension occasions. The BP is to be measured in the sitting posi-
and perioperative outcome deserves special mention. This tion, with a cuff that is appropriately sized to the arm
288 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
circumference, and positioned at the level of the right that must also be addressed complicates the approach to
atrium. The cuff is inflated well in excess of the antici- an individual patient. Also, the perioperative period is a
pated SBP and slowly deflated (no more than a 10 mm Hg time of increased surgical stress and altered physiology.
decrease per three heart beats), and the examiner listens There is often an excessive release of catecholamines de-
carefully for the Korotkoff sounds, with the bell of the spite adequate management of pain and anxiety.23 Many
stethoscope positioned close to the cuff over the brachial surgical procedures and techniques result in ischemia-
artery. A minimum of two readings are taken per session reperfusion, with the release of mediators and tissue
and the results are averaged. If the white coat phe- injury.24 There can be an increased cellular and immune
nomenon is suspected (i.e., anxiety and elevated BP in response, platelet activation, and a compromise in mi-
the presence of a physician), paramedical personnel can crovasular blood flow.25 In the presence of hypertension
take the BP with the physician absent. This alternative, and other comorbidities, this constellation of physiologic
however, brings into question the training and monitor- changes can result in a higher risk for stroke, myocardial
ing of office personnel to ensure that the measurements infarction, or decompensated heart failure.2628
are appropriately made. With tight control of measuring Whereas some patients come to the anesthesiologist
devices and personnel, even a 2-mm Hg rise in the BP is for preoperative evaluation unaware that they are hyper-
deemed significant. tensive, many patients will be on chronic therapy for the
control of BP with varying degrees of success. The anesthe-
siologist must be prepared to manage these eventualities
MONITORING DEVICES and devise a safe and effective anesthetic prescription.
The preoperative assessment should include the search
The gold standard for BP measurement continues to be for evidence of end-organ damage, including brain, kid-
the mercury manometer. Aneroid sphygmomanometers neys,29 and heart.30 Knowledge of the agents employed to
are more typically used but are not as accurate. Given the treat hypertension is essential in formulating this plan.
concerns of maintaining devices that contain elemental
mercury, many facilities are replacing these measuring
devices with automated BP devices; most of these are
oscillometric devices that keep a constant pressure in ANTIHYPERTENSIVE AGENTS
the cuff to establish a mean BP, then search for the
SBP and the DBP. Because these are now the standard The site of action of antihypertensive drugs can be
of care for intraoperative readings, they are preferred in interpreted in terms of the physiology that generates
preanesthesia evaluation clinics. intravascular pressure (see Fig. 20.1). BP is the prod-
The use of devices to continuously measure BP for uct of cardiac output and systemic vascular resistance.
24 hours, much like the monitoring used to detect cardiac Antihypertensive agents that have a direct cardiac ef-
arrhythmias, are available but have not been widely fect and lower cardiac output include diuretics (usually
employed. However, there is every reason to recommend the first line of therapy), -blockers, and calcium chan-
that patients learn to monitor their own BP at home.22 nel blockers (CCBs). Commonly employed agents that
Wrist devices are easy to use and inexpensive, and can lower systemic vascular resistance include angiotensin-
provide valuable information as to variation in BP during converting enzyme (ACE) inhibitors, angiotensin-1 (AT-1)
wakefulness. blockers, -blockers, 2 agonists, CCBs, and diuretics.
Any vasodilator or sympatholytic drug will similarly lower
BP. -Blockers additionally have been shown to reduce
perioperative cardiac risk.7,31,32
CURRENT DEFINITIONS The molecular sites of action provide additional infor-
Whereas in the past, normal BP was considered mation and indicate why combinations of drugs are often
to be 120/80 mm Hg, and the cut-off for treatment used (see Fig. 20.2). Vascular tone is maintained or altered
140/90 mm Hg, current definitions include a diagnosis by a cascade beginning with the synthesis of angiotensino-
of prehypertension. This state is characterized by an gen, which, in the presence of renin, is converted to AT-1.
SBP between 120 and 139 mm Hg, and a DBP of 80 to In the presence of ACE, this is converted to AT-2, which in-
89 mm Hg. For this group of patients, the recommended teracts with the AT-1 receptor. This sequence of activation
treatment is lifestyle modification, emphasizing weight can be blocked at each of these distinct stepsby employ-
loss, diet low in sodium, and frequent aerobic exercise.21 ing renin inhibitors, ACE inhibitors,33 AT-2 antagonists,
and AT-1 receptor blockers.34
FIGURE 20.1 Antihypertensives: Therapeutic targets. Blood pressure, systemic vascular resistance,
cardiac output and the therapeutic targets for commonly used classes of antihypertensive drugs.
CO, cardiac output; CCB, calcium channel blockers; ACE, angiotensin-converting enzyme; ARB,
angiotensin receptor blocker.
choice of an ACE inhibitor (see Fig. 20.3). The therapeu- inhibitors have been shown in some instances to have
tic goal is to maintain the SBP <130 mm Hg and DBP severe, and often refractory, hypotension on anesthetic
<80 mm Hg. The second step in therapy is to add either a induction. However, it has also been shown that the risk
-blocker or a CCB, and the third step is to add both agents is ameliorated if the last dose of ACE inhibitor is 10 hours
to the ACE inhibitor. Although previously, it was recom- or more before anesthetic induction.36
mended to avoid a -blocker in patients with heart failure,
these agents can be used to great advantage in patients
already being treated with an ACE inhibitor (Fig. 20.3).
Such change in the recommended therapeutic regi- How Is Hypertension Managed
men is not without consequence to the anesthesiologist.
Patients receiving antihypertensive therapy with ACE Intraoperatively?
Anesthetic induction can be challenging in patients being
Angiotensinogen treated with antihypertensives, and more so because most
Renin inhibitor patients are given diuretics as part of their therapeutic
regimen. In addition to extracellular fluid depletion due to
Renin diuretic therapy, patients have been NPO, and often suffer
from further volume depletion due to bowel cleansing
regimens.
AT I
ACE Inhibitor
EXTRACELLULAR FLUID
AT Converting enzyme
VOLUME
The effect of extracellular fluid volume depletion can be
AT II graphically displayed by consideration of left ventricular
AT Antagonist
pressure-volume relations during the cardiac cycle (see
Fig. 20.4). With normal physiology, there is an increase in
AT-II Receptor left ventricular volume (abscissa), with little change in left
ventricular pressure (ordinate). With closure of the mitral
FIGURE 20.2 The molecular metabolic pathway with a direct valve at the beginning of left ventricular contraction, there
relation to systemic vascular tone, and the sites of action of is an isovolumic rise in pressure until the aortic valve
specific therapeutic agents. AT, angiotensin; ACE, opens and the chamber begins to eject blood into the aorta.
angiotensin-converting enzyme. Pressure continues to rise to a maximal or end-systolic
point (closely akin to the measured SBP), the aortic valve
290 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Blood pressure >15/10 mm Hg above goal Blood pressure >15/10 mm Hg above goal
SCr <1.8 mg/dL SCr 1.8 mg/dL
FIGURE 20.3 The flow chart represents recent recommendations for patients diagnosed with
diabetes mellitus and/or renal disease indicating more aggressive therapeutic endpoints, and
identifying ACE inhibitors as first line therapy. SCr, serum creatinine; BP, blood pressure; ACE,
angiotensin-converting enzyme; CCB, calcium channel blocker.
(From: Bakris GL, Williams M, Dworkin L, et al. Preserving renal function in adults with
hypertension and diabetes: A consensus approach. Am J Kid Dis. 2000;36:646.)
closes, and there is an isovolumic fall in chamber pressure dynamics in near-normal ranges. However, the result
until the mitral valve opens and left ventricular filling is a tall and narrow pressure-volume loop with a
begins again. The separation of the isovolumic volumes decreased stroke volume. Anesthetic induction with drugs
is the stroke volume, and the area within the pressure- that alter cardiac contractility or suddenly cause severe
volume loop defines left ventricular stroke work. vasodilatation can result in a precipitous fall in BP. Agents
The heart responds to an increase in impedance or should be selected to avoid this occurrence or induction
afterload in a variety of ways. With extreme elevations drugs can be given in divided doses, and replenishment of
in afterload, the heart may fail and be incapable of intravascular volume and blood flow returning to the heart
responding to the challenge. Left ventricular end-diastolic (leg elevation, intravenous fluids) should be accomplished.
volumes and corresponding pressures will rise, altering
the pressure-volume loop, with decreased capacity of the
left ventricle to perform work. Antihypertensive therapy ANESTHETIC MAINTENANCE
is meant to intervene before this occurs and at a time
when ventricular performance is preserved, such that a Anesthetic maintenance is usually provided by inhala-
decrease in afterload (vasodilatation) and a decrease in tional agents and augmented with other intravenous
extracellular fluid volume (diuretics) maintain ventricular agents. All the volatile anesthetics (except for nitrous
C H A P T E R 2 0 / P E R I O P E R AT I V E H Y P E R T E N S I O N 291
P P
A V V B
Plans must also be made to manage anesthetic emer- TABLE 20.2 Intravenous Antihypertensive Drugs
gence. Recurrence of hypertension is common when
inhalational agents have been discontinued. If manage- Drugs Dosage
ment of pain with opiates and other intravenous agents
Fenoldopam 0.1 g/kg/min; max
is adequate, and -blockers have been administered with
1.6 g/kg/min
good response, incremental doses of hydralazine (5 mg
Esmolol 500 g/kg over 1 min,
bolus doses up to 20 mg) can maintain the BP during
2550 g/kg over 1 min
emergence and recovery.
Metoprolol 515 mg IV, target heart rate
5080 bpm
Labetalol 20 mg in divided 5-mg doses
q5min
How Is a Patients Blood Hydralazine 5-mg increments, maximum
Pressure Maintained during 20 mg
Nitroprusside 0.5 g/kg/min; max
Postanesthesia Recovery? 2 g/kg/min
Nitroglycerin infusion 25100 g bolus; 2 g/kg/min
The major focus during the early stages of postoperative Amiodipine 50100 g/kg/h
recovery from anesthesia, during which time the patient Diltiazem 0.25 mg/kg bolus
is monitored in the PACU, is to maintain BP control, but Nicardipine 5 mg/h; maximum 15 mg/h
with the adjunctive effects of volatile anesthetic agents. Clonidine Can be administered orally
Although the same tenets as to rapid onset and metabolism 0.10.3 mg b.i.d;
apply, there is also the necessity to return the patient to a corresponding slow release
physiologic state that will allow discharge from the PACU. patch 2.5/5.0/7.5 mg.
If it is necessary to continue infusion therapy to maintain
BP control, arrangements must be made to transfer the
patient to a high-acuity unit for monitoring. It might also
marked natriuresis. The half-life of infused fenoldapam is
be prudent to continue invasive monitoring if begun in the
five minutes.
operating room. Otherwise, transition to a hospital bed
Clonidine is a central adrenergic 2 agonist that pro-
or discharge to home would be appropriate, depending
vides sustained vasodilation.45 Clonidine patches provide
on the nature of the surgical procedure and the course of
a sustained effect with dosing patches of 2.5, 5.0, 7.5 mg
anesthetic.
corresponding to 0.1, 0.2, 0.3 mg oral dosing. Cloni-
If moderate hypertension was observed and/or treated
dine should not be used with -blockers, because this
during anesthesia, and the patient had heretofore not
combination of therapeutic agents can result in severe
been diagnosed with hypertension, the patient should
bradycardia.
be advised to seek further evaluation and possible
The use of CCBs in the perioperative period presents
treatment from a primary care physician. Those treated
similar caveats.46 Dihydropyridine CCBs (amlodipine,
with chronic antihypertensive agents should resume their
nicardipine, nimodipine) have no effect on heart rate
medications as soon as possible. For some procedures
or cardiac and myocardial contractility. The nondihy-
requiring a prolonged surgical recovery and inability
drophyridines (verapamil, diltiazem) are potent coronary
to administer oral drugs, parenteral therapy may be
vasodilators, which suppress myocardial contractility and
required.
result in potent suppression of the SA and AV node.
Although it is true that hypertensive patients with
Some commonly used drugs for perioperative BP
end-organ damage are at greatest risk of developing
control are listed in Table 20.2.
complications, it is also true that these complications
involve the same end organs. Stroke, myocardial is-
chemia and/or infarction, and renal failure are unto-
ward consequences that can occur acutely during the SUMMARY AND
perioperative period. It is not uncommon for some post- CONCLUSION
operative complications to be noted several days after
the procedure, giving rise to the question whether the Hypertension is a global and worldwide health problem.
complication(s) bore a causal relation to the anesthetic. Public health and outcome studies suggest a more aggres-
Careful and complete recording of all physiologic events sive approach to treatment and have resulted in lowering
and therapeutic interventions make retrospective assess- the limits of what has been considered normal BP and
ments more feasible and improve patient care and patient changed the therapeutic target.47 Anesthesiologists are
safety. often presented with patients who are unaware of their
In terms of therapeutic choices, special circumstances disease or have been inadequately treated.
and drug interactions are pertinent to anesthesiologists. While absolute cut-offs for the SBP and the DBP have
The intravenous infusion of fenoldopam, a potent pe- not been established, the existence of comorbidities have
ripheral dopamine (D1) antagonist markedly improves an added impact on outcome. Also, the presence of a
renal perfusion, and vasodilates and lowers BP. Owing to wide pulse pressure correlates with postoperative renal
an abundance of D1 receptors in the nephron, there is damage, whereas the SBP and the DBP do not.
C H A P T E R 2 0 / P E R I O P E R AT I V E H Y P E R T E N S I O N 293
In addition to the effects of volatile anesthetic agents 2. Hanada S, Kawakami H, Goto T, et al. Hypertension and
in controlling BP, there are intravenous agents and infu- anesthesia. Curr Opin Anaesthesiol. 2006;19:315.
sion therapy that satisfy the need for strict BP control, 3. Aronson S, Fontes ML. Hypertension : A new look at an old
problem. Curr Opin Anaesthesiol. 2006;19:59.
rapid onset of action, and a short duration of effect.
4. Potyk D, Raudaskoski P. Preoperative cardiac evaluation for
Postanesthetic care includes the identification of the prob-
elective noncardiac surgery. Arch Fam Med. 1998;7:164.
lem to the patient, and the facilitation of referral for 5. Pickering TG, Hall JE, Appel LJ, et al. Circulation. 2005;
chronic care. 111:697.
6. Spahn DR, Priebe HJ. Preoperative hypertension: Remain
wary? yes-cancel surgery? no. Br J Anaesth. 2004;92:461.
7. Palmer J. Hypertension and perioperative risk. Br J Anaesth.
2004;93:305.
KEY POINTS 8. Fleisher LA. Preoperative evaluation of the patient with
1. Hypertension has become a global health issue, and hypertension. JAMA. 2002;287:2043.
the severity and duration are associated with target 9. Howell SJ, Sear JW, Foex P. Hypertension, hypertensive
heart disease and perioperative cardic risk. Br J Anaesth.
organ damage (brain, heart, kidneys) and increased
2004;92:570.
morbidity and mortality.
10. Chobanian AV, Bakris GL, Black HR, et al. The seventh report
2. Guidelines for the diagnosis of hypertension have of the joint national committee on prevention, detection,
recently been reexamined. Also, the categories of hy- evaluation, and treatment of high blood pressure. JAMA.
pertension have changed on the basis of longitudinal 2003;289:2560.
epidemiologic studies which suggest that SBP should 11. Hollenberg SM. Preoperative cardiac risk assessment. Chest.
be <130 mm Hg. and DBP should be <80 mm Hg. 1999;115:5157.
An increased pulse pressure, or pulse pressure hy- 12. Casadei B, Abuzeid H. Is there a strong rationale for
pertension, is associated with renal damage and deferring elective surgery in patients with poorly controlled
perioperative complications. hypertension? J Hypertens. 2005;23:19.
13. Schroder D. The preoperative period summary. Chest. 1999;
3. At the time of preoperative evaluation, approximately
115:44S46S.
one third of patients who are hypertensive will have
14. Cheung AT. Exploring an optimum intra/postoperative
no previous diagnosis of hypertension. Additionally management strategy for acute hypertension in the cardiac
another third diagnosed as hypertensive will be surgery patient. J Card Surg. 2006;21:S8.
receiving inadequate treatment. 15. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guide-
4. There is no definitive level of unacceptable BP line update for perioperative cardiovascular evaluation for
relating to perioperative complications in patients noncardiac surgeryexecutive summary a report of the
with isolated hypertension. However, risk is increased American College of Cardiology/American Heart Association
due to the frequent presence of comorbidities such as Task Force on Practice Guidelines. Circulation. 2002;105:
obesity and Diabetes Mellitus. 1257.
5. Current recommendations for outpatient therapy of 16. Herrington DM, Brown V, Mosca L, et al. Relationship be-
tween arterial stiffness and subclinical aortic atherosclerosis.
hypertension and diabetes and/or renal disease are
Circulation. 2004;110:432.
more aggressive, the target values for BP being 17. Dowanski M, Norman J, Wolz M, et al. Cardiovascular risk
<130/80 mm Hg. For these patients, ACE inhibitors assessment using pulse pressure in the 1st national health
are now suggested as first line therapythese agents and nutrition examination survey. Hypertension. 2001;38:
have been associated with profound and persistent 793.
hypotension with anesthetic induction if dosing has 18. Chertow GM, Lazarus JM, Christensen CL, et al. Preoperative
occurred within the previous 10 hours. renal risk qualification. Circulation. 1997;95:878.
6. Intraoperative management of the hypertension de- 19. Fontes ML, Aronson S, Weng WS, et al. Increased pulse pres-
pends on: Assessment of the need for increased and sure adds to the risk of stroke after surgery. Anesthesiology.
invasive monitoring; deepening the anesthetic with 2002;38:184.
20. Nieto FJ, Young TB, Lind BK, et al. Association of sleep-
volatile agents when appropriate (most agents are
disordered breathing and sleep apnea and hypertension in a
organ protective); and adding additional intravenous large community-based study. JAMA 2000;283:18291836.
agents as needed. 21. Smith L. New AHA recommendations for blood pressure
7. In the operating room and PACU, the usual choice measurement. Am Fam Physician. 2005;72:1391.
of agents includes those administered intravenously, 22. Moser M. Comments on the new AHA recommendations
with rapid onset and short duration of action, and for blood pressure measurement. J Clin Hypertens. 2005;
predictable metabolism. 7:71.
8. Appropriate referral for high-acuity monitoring in the 23. National Institutes of Health. The sixth report of the Joint
immediate postoperative period or for long term care National Committee on prevention, detection, evaluations,
is the responsibility of the anesthesiologist. and treatment of high blood pressure. Arch Int Med. 1997;
157; 2413.
24. Julia PL, Buckberg CD, Acar C, et al. Studies of controlled
REFERENCES reperfusion after ischemia XXI. Reperfusate composition:
Superiority of blood cardioplegia over crystalloid cardiople-
1. Asia Pacific Cohort Studies Collaboration. Blood pressure gia in limiting reperfusion damageimportance of oxygen
and cardiovascular diseases in the Asia-Pacific region. free radical scavengers in blood cells. J Thorac Cardiovasc
J Hypertens. 2003;21:707. Surg. 1991;101:303.
294 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
25. Huskowitz A, Mangano DT. Inflammatory cascade: A final 37. Kawaguchi M, Drummond J, Cole DJ, et al. Effect of
common pathway for perioperative injury. Anesthesiology. isoflurane on neuronal apoptosis in rats subjected to focal
1996;85:957. cerebral ischemia. Anesth Analg. 2004;98:798.
26. Toller WG, Metzler H. Acute perioperative heart failure. Curr 38. DeHert SG, Van der Linden P, Cromheecke S, et al. Car-
Opin Anaesthesiol. 2005;18:129. dioprotective properties of sevoflurane in patients under-
27. Hernandez AF, Whellan DJ, Stroud S, et al. Outcomes in going coronary surgery with cardiopulmonary bypass are
heart failure patients after major noncardiac surgery. J AM related to the modalities of its administration. Anesthesiol-
Coll Cardiol. 2004;44:1446. ogy. 2004;101:299.
28. Pastor PB, Bellows W, Leung JM. The prevalence of pre- 39. Piriou V, Chiari P, Lhuillier F, et al. Pharmacological
operative diastolic filling abnormalities in geriatric surgical preconditioning; comparison of desflurane, sevoflurane,
patients. Anesth Analg. 2003;97:1214. isoflurane and halothane in rabbit myocardium. Anesth
29. Dehne MG, Junger A, Hartmann B, et al. Serum creatinine Analg. 2001;92:SCA39.
and perioperative outcomesa matched-pairs approach 40. Weiskopf RB, Cahalan MK, Eger EI, et al. Cardiovascular
using computerized anaesthesia records. Eur J Anaesthesiol. actions of desflurane in normocarbic volunteers. Anesth
2005;22:89. Analg. 1991;73:143.
30. Slogoff S, Keats AS. Myocardial ischemia revisited. Anesthe- 41. Lee HT, Ota-Setlik A, Fu Y, et al. Differential protective
siology. 2006;105:214. effects of volatile anesthetics against renal ischemia
31. Lindenauer PK, Pekow P, Wang K, et al. Perioperative reperfusion inury in vivo. Anesthesiology. 2004;101:1313.
beta-blocker therapy and mortality after major noncardiac 42. Gan TJ, Sopitt A, Maroof M, et al. Goal-directed intraopera-
surgery. N Engl J Med. 2005;353:349. tive fluid administration reduces length of hospital stay after
32. Vandenbunder B, Mignon A. Perioperative beta-blocker major surgery. Anesthesiology. 2002;97:238.
therapy and mortality. N Engl J Med. 2005;35:2513. 43. Varon J, Marik PE. The diagnosis and management of
33. Comfere T, Spring J, Kumar MM, et al. Angiotensin system hypertensive crises. Chest. 2000;118:214.
inhibitors in a general surgical population. Anesth Analg. 44. Halpern JA, Goldberg M, Neely C, et al. Postoperative hyper-
2005;100:636. tension: A multicenter, prospective, randomized comparison
34. Strauss MH, Hall AS, Tsuyuki T, et al. Angiotensin recep- between intravenous incardipine and sodium nitroprusside.
tor blockers may increase risk of myocardial infarction: Crit Care Med. 1992;20:1637.
Unraveling the ARB-MI paradox. Circulation. 2006;114: 45. Wallace AW. Clonidine and modification of perioperative
838. outcome. Curr Opin Anaesthesiol. 2006;19:411.
35. Bakris GL, Williams M, Dworkin L, et al. Preserving 46. IV Nicardipine Study Group. Efficacy and safety of intra-
renal function with hypertension and diabetes: A consensus venous nicardipine in the control of postoperative hyperten-
approach. Am J Kidney Dis. 2000;36:647. sion. Chest. 1991;99:393.
36. Francis GS. ACE inhibition in cardiovascular disease. N Engl 47. Verdecchia PV, Angili F. Natural history of hypertension
J Med. 2000;342:201. sub-types. Circulation. 2005;111:1094.
CHAPTER CARDIAC ARREST
A
elective abdominal aortic aneurysm surgery. to allow chest rise, and inspiratory time of 1 second
Preoperative evaluations and cardiovascular should be provided.
risk stratification highlight a history of stable The airway rescuer should pay attention to avoid hy-
angina. A recent echocardiogram shows a perventilation while chest compressions are provided.
left ventricular ejection fraction of 50% with Defibrillations should be provided as soon as possible,
left ventricular hypertrophy. He is able to participate because VF is a shockable rhythm.
in moderate recreational activity, such as golf, and is Only one 200 J biphasic shock should be provided
now chest pain-free for the last 3 months on atenolol for each 2-minute cycle of cardiopulmonary resuscita-
and aspirin. He has a mild aortic stenosis but has been tion (CPR).
told by the family cardiologist that his valve disease is One mg of epinephrine every CPR cycle or vasopressin,
nonamendable to surgery at this time. 40 units to replace the first and second dose of
Following the induction of anesthesia, and utilizing epinephrine should be delivered immediately. In shock-
a balanced technique with opioids and isoflurane, the resistant CPR, IV amiodarone 300 mg should be
heart rate increases shortly to 130 bpm with 1.5 mm ST administered, and followed by repeated IV boluses of
segment depression noted in lead V5. A 60-mg bolus 150 mg alternated with a vasopressor.
of esmolol decreases the heart rate to 80 bpm, but Postoperative myocardial dysfunction is expected and
does not change the ST depression. Transesophageal should be treated with a positive inotrope infusion.
echocardiography (TEE) is performed. Findings show left Postoperative neurologic dysfunction (stupor or coma)
ventricular lateral wall hypokinesia. Further reduction is expected and should be treated with hemodynamic
of the heart rate with esmolol resolves the wall motion and general supportive care.
Post-cardiac arrest (CA) hypothermia should be consid-
abnormality. The decision is reached to proceed with
ered in selected cases.
surgery, which is uneventful.
The patient is transported to the intensive care KEY CONCEPTS
unit (ICU), still intubated and hemodynamically stable. 1. Is this an anesthesia-related cardiac arrest (ACA)? Yes,
Upon connecting the patient to the surgical intensive although the most common form of cardiac dysrhyth-
care unit (SICU) ECG leads, you notice that the arterial mia during arrest is severe bradycardia followed by
line goes flat and ventricular fibrillation (VF) is shown asystole (45%). VF is the cause of CA in up to 14% of
on the ECG rhythm strip. The pulse is assessed for the cases.
10 seconds, confirming a pulseless rhythm. The best 2. Is there a direct link between the anesthetic manage-
chance to restore spontaneous circulation in this patient ment and ventricular defibrillation? Unlikely. The most
is dependent on effective basic and advanced cardiac life important predictor of postoperative myocardial in-
support (ACLS): farction (MI) and its complications are the patients
specific risk and the surgical specific risk.
Chest compressions should be started immediately 3. What are the determinants for successful restoration
at a rate of 100 times per minute. A motivated of spontaneous circulation (ROSC) in these patients?
rescuer should achieve at least 1-inch depth with each Effective CPR, avoidance of hyperventilation, early
compression and allow full chest recoil after each defibrillation, and early pharmacologic therapy should
stroke. be provided to enhance coronary perfusion pressure
A defibrillator should be called in immediately. during chest compressions. Appropriate knowledge
295
296 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
emerged as an important factor in the causation of ACA, Inadequate ventilation leading to hypoxia
with most anesthesia mishaps occurring either at the Severe dysrhythmias, mostly bradycardia induced by
time of induction of anesthesia or during the recovery hypoxia or drug management, usually narcotics or
phase.10 Recent series from Australia11 and from Europe12 succinylcholine
confirmed these findings and, in addition, identified Unrecognized hypovolemia resulting in prolonged se-
ventilatory problems as the most common cause of ACA. vere hypotension
Less frequently, prolonged and untreated fluctuations
This grouping of patients according to the above
of blood pressuremore frequently hypotension and
seems to be valid up to 12 hours post anesthesia on
occasionally accelerated hypertensionwere identified to
the wards, in the recovery room and the ICU, and on the
be responsible for episodes of severe bradycardia and
wards.12
eventually asystole. These reports also documented the
Further studies on anesthesia-related mortality in-
average patient who experienced ACA to be (i) older,
clude the retrospective, single institution review of
(ii) of male gender, (iii) with a higher ASA physical status
250,000 anesthetic records,21 which shows the mortal-
score, (iv) undergoing emergency surgery, (v) having
ity of ACA to be relatively low, ranging from 0.55:10,000
a prolonged surgical time, and (vi) surgery performed
to 2.4:10,000 procedures. In a similar study conducted by
after 3 Pm. In addition, ACA was often preceded by the
the Mayo Clinic,22 ACA is defined as a condition requiring
administration of medications. Further analysis of these
either closed chest compression or open cardiac mas-
issues introduced obligatory quality control, and outcome
sage performed between the onset of anesthesia or during
reviews by the specialty of anesthesiology, which proved
transport to the ICU. The two outcome variables observed
extremely valuable.
in the survey were survival for at least 1 hour after re-
A 10-year retrospective review published in 2002
showed the overall ACA rate in a large academic institution suscitation and survival to discharge from the hospital.
in the United States to vary from 1.37 per 10,000 to 0.69 Probable causes of ACA were grouped into the following
per 10,000 anesthetics.13 Further trends were observed three categories:
overseas where, after a temporary increase in anesthesia- 1. Intraoperative hemorrhage
related deaths (associated with twofold increase of major 2. Preexisting cardiac disease and
cardiovascular and neurosurgical operations), the mortal- 3. Hypoxia, either at intubation or at extubation
ity from ACA sharply declined.14
Even after the results of these studies became Overall, 24 CAs were directly attributed to anesthesia.
generally known, the direct causeeffect relationship Despite the variety of practices, the rate of anesthesia-
between the choice of anesthesia and ACA was difficult to related mortality is not substantially different between
identify. The pieces of the puzzle, however, fell into place Europe12 and North America.22
at the end of the 1980s when confidentiality agreements
between investigators and government agencies allowed
the development of a massive database on ACAs.15
These surveys showed that most cardiac deaths were What Is the Pathophysiology
multifactorial and/or related to inefficient control of the
airway and asphyxia. Respiratory complications have also
of Cardiac Arrest?
been noted as important contributors to morbidity and
ACAs.16
About the same time, the American Society of Anes- HYPOXIA
thesiologists began reporting nationwide insurance claims
arising from anesthetic complications. Even with the Irreversible hypoxic or ischemic brain damage is a
limitations of voluntary reporting, these reports reliably devastating complication, which may occur when, at
confirmed that unrecognized airway obstructions were normal body temperatures, the brain is deprived of
the cause of ACA in approximately 25% of the cases. With its oxygen supply for more than 5 to 7 minutes.23
the introduction of pulse oximetry as a standard of care in Such a situation may develop in the context of a
1984, the number of ACAs caused by unrecognized airway cannot intubatecannot ventilate scenario. Such airway
obstruction decreased significantly. Additionally, such a management failure may be caused by misplacement
registry also documented the prevalence of spinal anes- of the endotracheal tube, airway obstruction, airway
thesia overdose as a common cause of ACA. This type of collapse, accidental extubation, or aspiration of gastric
survey became essential to provide insight17,18 and formu- contents. Laryngospasm induced by mechanical irritation
late policies aimed at improving the quality of anesthesia during inadequate depth of anesthesia or bronchospasm
care.17,19 This eventually resulted in the formulation of of anaphylactic or intrinsic origin may also cause severe
the American Society of Anesthesiologists (ASA) Practice episodes of hypoxia. Errors in oxygen supply seldom occur
Guidelines for Management of the Difficult Airway.20 A but, when they do, are devastating.24 The proportion of
byproduct of this process was the development of safety- ACAs caused by failure of adequate ventilation remained
first mentality of anesthesiologists, with a consequential relatively constant at approximately 35% in the 1980s.10
decrease in their professional liability premiums. These This increased somewhat in the 1990s25 when airway- and
registries listed the following several event categories that ventilation-related CA during intubation or extubation
were frequently associated with ACAs: amounted to approximately 45% of all ACAs, with the
298 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
cause of most of these mishaps being either lack of Although the number of maternal deaths due to
proper monitoring and/or underestimation of the level general anesthesia shows a substantial decrease, airway
of sedation. management failures in obstetric anesthesia still occur.
At present, the low reported incidence of ACA from This may be associated with displacement of the stom-
hypoxia or hypercarbia is in large part due to the ach by the gravid uterus and high risk of aspiration.
introduction of pulse oximetry and capnography into Other pregnancy related, physiologic changes may also
the daily practice of anesthesia. In fact, the ASA Closed contribute to adverse outcomes, including ACA. These
Claims Study reported that 57% of hypoxia-related deaths changes include diaphragmatic elevation and decreased
could have probably been avoided simply with the use functional residual capacity, both of which reduce oxy-
of pulse oximetry and capnography.26 Hypoxic brain gen lung reserves. The oxyhemoglobin dissociation curve
damage may also occur during prolonged hypotension. is shifted to the left, thereby resulting in less oxygen re-
In general, although older age and comorbidities have lease. Hemodilution decreases hemoglobin concentration,
been associated with a worse outcome, they did not seem and oxygen consumption is increased, factors that con-
to influence, per se, the occurrence of hypoxemic CA. tribute to the development of ACA approximately 10-fold.
Hyperventilation frequently results from the preven- Many of these patients are subjected to general anesthe-
tion and treatment of hypoxia, caused by occasional sia on an emergency basis, secondary to fetal distress.
inability to intubate and ventilate; however, if used in- A confidential review in the early 1980s attributed general
discriminately, it may be harmful. Anesthesiologists, in anesthesiarelated maternal mortality to difficult intu-
the presence of adequate lung compliance, have tradi- bation in 40% of cases, equipment failure in 18%, and
tionally learned to link the phenomenon of cyclic blood postoperative hypoventilation in 5%.32 Although the dan-
pressure variationwhen positive pressure ventilation ger of hypoxia in the pregnant woman still persists, recent
is appliedto hypovolemia or lung overinflation.27 Re- reports show that anesthesia for cesarean section is now
cent evidence supports the knowledge that inadvertent 30 times safer than it was 50 years ago. This is most
hyperventilation is common during CPR. Unnecessary hy- likely due to the widespread use of regional anesthesia
perventilation, that is, too many breaths or too large VTs and improved monitoring.33,34
given during CPR, is an inherent risk for death because it
may raise intrathoracic pressure to levels high enough to
impede venous return and decrease coronary and cerebral
perfusion, thereby compromising the success of CPR.28 ANESTHESIA-RELATED
The aforementioned clinical and laboratory obser- CARDIAC ARREST
vations led to an important change in the resuscitation
guidelines for CA for adult victims with an advanced air- Life-threatening dysrhythmias occur during anesthesia in
way device (ETT, LMA and Combitube) in placethat is, approximately 0.4% of the patient population.35 Their
to maintain a respiratory rate no >10 breaths per minute, occurrence may be related to the anesthetic technique,
with an inspiratory time of 1 second and a VT limited to which in turn impacts on the hemodynamic variables and
chest rise, (estimated 500 mL in the adult patient).29 may eventually cause CA.
Pharmaceuticals such as neuromuscular blocking The most common forms of cardiac dysrrhythmias
agents12,13 with the potential to decrease respiratory during anesthesia are bradycardia or asystole (45%), ven-
drive can also be associated with hypoxemic CA. Human tricular tachycardia or fibrillation (14%), and pulseless
and environmental factors may also contribute to the electrical activity (PEA) (7%). In the presence of dys-
occurrence of hypoxemic CA, especially by performing rrhythmias, a high index of suspicion for undetected
ineffective CPR. CPR is more likely ineffective if CA occurs hypoxia should be the rule, and resuscitation should be
after the typical hospital working hoursafter 5 Pm or performed keeping in mind the pathophysiology of lo-
during the weekendsprobably secondary to the reduced cal, general, and neuraxial anesthesia and their effect on
number of specialists present in the hospital after hours, resuscitation efforts. The physician or other health care
as well as the emergent nature of procedures. It is also no providers prior knowledge of the patients medical his-
surprise that the outcome of ACAs is the best in tertiary tory, their immediate awareness of the probable cause of
referral centers where personnel with airway skills are arrest, and the initiation of medical management within
available throughout the night.30 seconds also influence survival. Unfortunately, failure to
Pregnant women and children, especially neonates, increase the FiO2 to 1.0, forgetting to close the vapor-
are highly susceptible to hypoxemic CA. Both, respiratory izer with the inhalational anesthetic, and unnecessarily
and circulatory events are equally distributed in children delayed defibrillation or pharmacologic interventions still
and infants, occurring in 19 per 10,000 and 2.1 per occur in the operating room.
10,000 respectively.31 The adjusted ACA is approximately The cardiovascular effects of inhaled anesthetic
ninefold higher compared to adults. However, because agents may include myocardial depression, parasympa-
of the low incidence of comorbidities, as well as the thetic or sympathetic stimulation, increased myocardial
neuronal plasticity in the very young, the outcome of ACA excitability, and severe hypotension. The latter is most
is generally better in this population. Anesthetic mishaps likely to occur in patients with valvular heart disease, heart
causing hypoxic CA in infants are also possible during the block, constrictive pericarditis, or anaphylactic reaction.
maintenance of anesthesia. Relative hypovolemia from Inhalational anesthetics may also hinder atrioventricular
preoperative fasting may be a contributing event. conduction, and have a direct negative inotropic effect
C H A P T E R 21/ C A R D I A C A R R E S T 299
that can sensitize the myocardium to the arrhythmic ef- and decrease in cardiac output. Mixed CA, caused by
fects of catecholamines.36 In animal studies, overdose hypoxia and dysrhythmias, as well as metabolic-induced
with inhalational agents has been found to interfere with CA, may occur in special clinical situations. Massive hem-
coronary autoregulation and create transient episodes orrhage and cardiac diseases, such as cardiomyopathy,
of sympathetic hyperactivity, both of which may re- myocardial ischemia, and acute myocarditis may also
sult in myocardial ischemia. This usually resolves when lead to ACA by causing decreased systemic oxygen de-
the anesthetic is terminated; low ejection fraction may livery and coronary perfusion. Hypothermia during the
persist and contribute to postoperative cardiovascular in- course of surgery or intracardiac diagnostic procedures
stability. Previously unrecognized coronary artery disease may increase myocardial irritability and evoke physiologic
may also lead to fatal arrhythmias and to the failure of responses, leading to severe dysrhythmias. Interestingly,
resuscitation.37 hypoxic ACA actually has a better prognosis than ACAs
Intravenous drugs, such as etomidate,38 succinyl- from other causes. For example, a recent series showed
choline,39 and propofol,40,41 by their ability to increase that 16 out of 20 patients having suffered hypoxic ACA
vagal activity, may predispose to asystole. Dexmedetomi- survived to hospital discharge.30
dine, an -2 adrenergic receptor agonist with sedative-
analgesic and anxiolytic properties and a full agonist to
the -2 adrenergic receptor, may act synergistically with
general anesthesia to cause severe bradycardia and hemo-
dynamic instability. This occurs primarily by potentiation How Can Regional Anesthesia
of the effects of other negative chronotropic drugs, such Lead to Cardiac Arrest?
as digoxin and pyridostigmine, or with the effects of a
neuraxial block. The resulting asystole is usually brief and
responds well to parasympatholytic agents. Hypoxemia- Up to 50% of CAs occurring during local or regional
induced sympathetic stimulation may be followed by anesthesia may be avoided by timely recognition and
severe bradycardia and asystole.1 This may be facilitated correction of inadequate ventilation.45 In this respect,
by increased serum potassium, acute metabolic and respi- the database of the American Society of Anesthesiology
ratory acidosis, or by the cardiovascular depressant effect Closed Claim Study, a project of the ASA Committee
of the anesthetic itself. Hypercapnia from hypoventilation on Professional Liability,18 revealed surprising clinical
leads to an increase in circulating catecholamines. The trends. In each case, CA occurring from local or regional
combination of succinylcholine and dexmetomidine is anesthesia was unexpected, as the patients ASA status
commonly associated with initial bradycardia followed by was low; additionally, the outcome was, in general, poor.
asystole. This is more likely to occur with repeated admin- In 30% of the 14 cases reviewed, a spinal anesthetic was
istration.42 The mechanism is probably competition for applied in the course of an emergency procedure, and the
available cholinergic receptors by succinylcholine, direct use of tetracaine seemed to be the drug most commonly
stimulation of the carotid baroreceptors, and accumula- associated with CA. Most anesthesiologists involved in
tion of acetylcholine. Remifentanil, a short-acting, potent these cases were experienced.
narcotic has also been associated with severe cardiac Despite the obvious selection biases resultant from
depression.43 self-reporting, there were some special features of pa-
Surgical manipulation of different organs, such as the tients suffering CA in local or regional anesthesia (see
rectum, uterus, cervix, larynx, bronchial tree, bladder and Table 21.1). These findings indicate that, despite the pres-
urethra, mesentery, the carotid sinus, heart, dura, biliary ence or immediate availability of an anesthesiologist, the
tract, extraocular muscles, and testicles all could lead crisis situation was often appreciated too late, the treat-
to severe bradyarrhythmias by enhancing an unopposed ment ineffective, and the neurologic recovery limited;
vagal tone. indeed, only four patients regained consciousness. Even
Abnormalities of potassium and calcium metabolism those patients were left with various degrees of cogni-
are often seen in patients undergoing either elective tive dysfunction. One may conclude that hypoventilation
or emergent surgical interventions. Studies on the dys- induced by opioids, benzodiazepines, or hypnotics may
rhythmic effects of hypokalemia not only confirmed have enhanced a preexistent sympathetic blockade pro-
that hypokalemia may endanger patients with MI,44 but duced by a relatively high spinal anesthesia, and that
also conclusively linked rapid correction of chronic hy- the anesthesiologists level of awareness of this potential
pokalemia to ACA. An occasional side effect of succinyl- interaction was low.33,46
choline is the acute onset of hyperkalemia and consequent Studies on the mechanism of circulatory collapse
cardiovascular instability. This usually occurs in patients during central neuraxial blockade, that is, total spinal
with thermal injuries, upper or lower motor neuron dam- anesthesia, revealed that, in a situation where pregan-
age, or other critical illness resulting in immobilization. glionic efferent sympathetic nerve fibers are effectively
The manifestation is usually late, approximately 1 month blocked46 and the autonomic sympathetic fibers of the
after the initial injury, and is related to extrajunctional heart are denervated at the T1-T4 level, the release of
neuromuscular receptor upregulation. endogenous catecholamines is blunted by blockade of the
Air embolism, as well as pulmonary thromboem- efferent sympathetic adrenal medulla fibers from T5-L2.
bolism, may also induce bradycardia and asystole, the This leads to venous and arterial dilation, uncompen-
latter because of increase in right ventricular afterload sated blockade of the adrenal medulla, and predominance
300 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 21.1 Special Features of Patients Suffering doses of epinephrine during CPR, correlate with poor
Cardiac Arrest during Local or Regional Anesthesia outcomes.54
The use of high dose epinephrine also comes with
Combined use of intravenous opioids, benzodi- some risk. Animal studies show that epinephrine increases
azepines, and/or hypnotic agents to achieve a deep, myocardial oxygen consumption and causes ventricular
sleeplike state dysrhythmias, ventilationperfusion mismatching, and
CA usually occurred 525 minutes after the last drug postresuscitation myocardial dysfunction.55
administration, and was preceded by a few minutes The use of vasopressin during neuraxial anesthesia is
of unexplained and undertreated bradycardia and undefined. Recent experiments showed some advantage of
hypotension vasopressin over epinephrine when cerebral and coronary
Cyanosis was present in most cases, suggesting respi- blood flows were compared after induced VF.56 In
ratory depression that may exacerbate the effects of animal models of electrically induced VF during epidural
sympathetic blockade. The highest documented sen- neuraxial anesthesia, a single dose of vasopressin was
sory level was T4 T1 level. Arterial blood gases found comparable at 5 minutes to multiple doses of
during CA confirmed hypoxemia, which was immedi- epinephrine in achieving better perfusion of the heart
ately corrected by endotracheal intubation and brain. This implied that the advantage of vasopressin
Airway was secured in a timely fashion in most of over epinephrine has never been tested clinically.57
the patients; however, cardiopulmonary resuscitation An unrecognized high level of neuraxial block in the
was usually delayed for a few minutes after the arrest sedated patient, combined with delayed administration of
occurred direct catecholamines, has been identified by the Closed
Ephedrine, the most common vasopressor used to Claims Study as a potential cause of CA. Although all neu-
increase heart rate and raise blood pressure, yielded raxial anesthesia techniques have been known to cause CA,
minimal therapeutic success spinal anesthesia clearly has the worst track record.18,51
In most cases, epinephrine was administered, on
average, 5 min after the CA was recognized
near-toxic bupivacaine blood levels, there was no in- Therefore, one may conclude that formal anesthesia train-
creased incidence of arrhythmias.60 On this specific topic, ing contributes to safe practices, good clinical outcomes,
there is no information available in the 2005 American and high patient satisfaction.
Heart Association Resuscitation Guidelines. Some areas outside the operating room, but still
Lidocaine, which binds to the same ion channel, dis- within hospital grounds, are also at increased risk if
sociates more rapidly than bupivacaine. Another potential sedation is provided without trained anesthesia personnel.
benefit of lidocaine comes from the increased automatic- These areas include psychiatric, gastroenterology, and
ity of ectopic pacemakers that potentially jump start the interventional radiology suites. Ironically, the cardiology
heart in standstill. It has been stipulated that lidocaine catheterization laboratory has also shown poor outcomes
given in high concentrations may displace bupivacaine from CAs. In a 10-year retrospective study, the overall
from the sodium channel receptors, reducing its myocar- incidence of CA was 21.9 cases per 10,000 procedures.
dial toxicity. The negative inotropic effects of bupivacaine Although this rate of occurrence decreased from 33.9 per
are proven more complex than just the blockade of sodium 10,000 before 1995 to 13.1 per 10,000 after 1995, only
ion influx, and not all of its activity can be reliably reversed 56.1% of the patients who suffered ACA left the hospital
by lidocaine.61 It is possible, indeed, that although lido- alive.71
caine displaces the bupivacaine bound to plasma protein, The adverse effects of sedation in the pediatric pop-
it may also acutely potentiate its toxicity.62 The best re- ulation are particularly notable in the outpatient setting,
view of this topic is available in a report to the Food and where the margin of safety for both circulatory and
Drug Administration.63 Intralipid has also been reported respiratory complications is narrow. Inefficient moni-
to remarkably attenuate bupivacaine toxicity and increase toring, lack of appropriate medical evaluation, medi-
the efficacy of resuscitation.64 cation errors, and lack of suitable recovery procedures
seem to contribute to poor outcomes.72 A recent review
concluded that in this patient population, adverse out-
comes of ACAs are related mostly to failure to rescue.
The outcome was even worse if ACA occurred in office
How Does Cardiac Arrest environments.
The overall message from these reviews is clear:
Occur in the Ambulatory Proper personnel, planning, and monitoring are of
or Outpatient Setting? paramount importance if we are to lower the incidence
and improve the outcome of ACAs in areas remote from
the operating room. ASA monitoring standards, including
Anesthesia in ambulatory settings seems to be associated
pulse oximetry and supplemental oxygen, capnography
with increased risk of CA.65 Review of the literature has
when applicable, recording of blood pressure at least ev-
identified the following groups as prone to ACA: The mor-
ery 5 minutes, and continuous ECG monitoring in at least
bidly obese, premature infants younger than 60 weeks, pa-
one lead should be maintained at respective locations
tients with a hematocrit <30%, children with recent upper
whenever sedation is used.
respiratory infections, patients with history of malignant
hyperthermia (MH), sporadic cases of drug interactions,
and those given inadvertent administration of vasopres-
sors in the presence of monamine oxidase inhibitors.66
When complications occur in outpatient surgical of-
How Can Anesthesia-Related
fices unattended by anesthesiologists, mortality increases Cardiac Arrest Be Prevented?
to 9.2 cases per 100,000, a 10-fold rise, compared to a fully
staffed ambulatory surgical center.66 A disturbingly high The protection provided by the perioperative use of
number of ACAs happen in dental offices, where heavy -adrenergic antagonists in patients undergoing noncar-
sedation and nitrous oxide inhalation may be admin- diac surgery is widely documented.73 Studies indicate
istered in absence of the ASA-required monitoring67 by that perioperative blockade not only lowers the chances
nonanesthesiologist health professionals, or even by office of early cardiovascular complications in many patients,
personnel. In the early 1990s, a survey of dentists practic- but influences long-term mortality as well. Studies in
ing in the United States yielded 43 cases of ACAs, with a the 1980s have shown for the first time that periop-
total mortality of 81.4%.68 The use of general anesthesia erative mortality is related not only to the nature of
and inadequate monitoring were statistically associated the surgical procedures and to comorbid conditions,
with mortality, while preexisting diseases were not. Also, but also to preventable treatment and management er-
the greater the number of pharmacologic sedative agents rors,74 as well as to the experience of health providers.
used, the higher the risk of CA.69 Basic or ACLS was used The adjusted ratio of failure to rescue were greater by
in less than half of these cases. In contrast to dentists, oral a factor of 1.13 when care was delivered by noncerti-
maxillofacial surgeons, whose training included elements fied, mid-career anesthesiologists (p < 0.04); whether the
of anesthesiology as part of their residency curriculum, anesthesiologist graduated from a US or foreign medical
reported no instances of CAs during either local anes- school did not alter this score. This relation appeared also
thesia, general anesthesia, or conscious sedation. Their valid for anesthesia subspecialization. In a retrospective
overall patients satisfaction was also remarkably high.70 review of outcomes from ACAs in approximately 2,000
302 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
basis of initial test results, detailed in vitro and in vivo respiratory acidosis, hypoxemia, rhabdomyolysis with
analysis of immunocompetence should be performed, in- increased creatine kinase, myoglobinuria, hyperthermia
cluding skin tests with the suspected drugs in a dilution of (usually a late sign), and cardiac dysrhythmias.
1 per 1,000 to 1 per 10,000. Recent data indicate that aller- Awareness of the problem and the introduction of
gic reactions occur between 2.2 to 22.4 cases per 10,000 sodium dantrolene into the pharmacopeia of the anes-
anesthetics, with only 3% to 4% being life-threatening.90 thesiologist reduced the mortality of MH from 80% in
Anaphylactic shock has been identified as a coexisting or the 1960s to less than 10% today.94 If MH occurs in
major indeterminate factor for dysrhythmic ACA.10 the operating room, all volatile anesthetic agents and
succinylcholine administration should be immediately
Prolonged QT Interval Syndrome discontinued. The patient should be cooled and venti-
lated with 100% oxygen at high flow rate. Consideration
Arrhythmias characterized by prolongation of the QT seg- should also be given to the use of intravenous calcium as
ment occasionally are difficult to treat, and lead to ACAs.91 an add-on to the resuscitation protocol. A solution of 50%
A stimulated adrenergic state, intrinsic myocardial dis- dextrose and IV insulin may provide metabolic substrate
ease, electrolyte abnormalities, hypokalemia and hypocal- and promote the intracellular movement of potassium.
cemia, and antiarrhythmic drugs including quinidine, The suggested dose is one unit of insulin for each 10 g of
procainamide, imipramine, amiodarone, phenothiazines, glucose. Early administration of bicarbonate and hyper-
and serotonin antagonists can exacerbate acquired QT ventilation may be necessary to counter extravasation of
prolongation syndrome. If the diagnosis is suspected, ag- potassium. The main therapy, however, is the immediate
gressive preoperative correction of calcium, potassium, administration of sodium dantrolene, beginning with a
and magnesium levels, as well as mild blockade and bolus of 2.5 mg per kg and repeated every 5 minutes to a
perioperative cardiac pacing should be considered. total dose of 10 mg per kg. Detailed information should
be given to the patient and the family on how to take
Malignant Hyperthermia appropriate steps to protect themselves from an MH crisis
should they have to undergo general anesthesia again.95
The anesthesiologist rarely encounters a known case of
MH ACA.92 Although there are no certain clinical predic-
tors of MH, prodromal conditions such as succinylcholine-
induced masseter spasm are occasionally encountered. How Is Anesthesia-Related
A hypermetabolic state with increased lactate levels, CO2
production, muscle rigidity and massive rhabdomyoly-
Cardiac Arrest Managed?
sisdemonstrated by an increase of creatine phosphok-
inase in both urine and serumare usually part of the In 2005, the International Consensus on Cardiopulmonary
clinical picture. Resuscitation and Emergency Cardiovascular Care Sci-
The cause of CA is believed to be due to hyperkalemia ence published its recommendations.96 A single algorithm
secondary to massive rhabdomyolysis, potentiated by was published for the management of CA with two limbs:
sympathetic stimulation and severe metabolic acidosis. The first, to manage VF and pulseless ventricular tachy-
The pathophysiologic trigger usually present in MH is in- cardia; and the second, to manage PEA and asystole.
creased sensitivity to either halogenated anesthetic agents In VF and pulseless ventricular tachycardia, emphasis
such as halothane, isoflurane, sevoflurane and desflu- was placed on early electrical therapy with rapid, effective
rane, and/or depolarizing neuromuscular blocking drugs, compressions, allowing full chest recoil. Electrical therapy
such as succinylcholine. The culprit drug induces an im- administered every 2 minutes can be effective in restoring
balance of ionized calcium homeostasis in the skeletal regular rhythm fibrillation (VF). If VF or pulseless VT
muscle, which in turn increases intracellular calcium and persists after the first or second shock, 1 mg of epinephrine
exaggerates glycolysis, along with high demand for acetyl- can be given intravenously, intracardiac, or interosseous
triphosphase. The massive activity of the latter blocks (IO), and repeated every 3 to 5 minutes. A single (IV or IO)
myofilament relaxation and sequestrates calcium in both dose of vasopressin (40 U) may also be administered in lieu
the sarcoplasmic reticulum and in the sarcolemma itself. of the first or second dose of epinephrine. After the first
Fever, combined with elevated CO2 productionwhich dose of vasopressor, if VF or pulseless VT still persists,
may be measured by end-tidal CO2 are indicators of amiodarone in a 300 mg bolusor if this drug is not
a hypermetabolic response. More than half of the fam- immediately available, lidocaine 75 to 100 mg IVshould
ilies with history of MH have the ryanodine receptor be administered, followed by repeated electric shocks. In a
mutations, which is the calcium channel of the sarcoplas- well oxygenated patient, the success rate of the first shock
matic reticulum. This has an autosomal dominant trait is high, especially when biphasic waveforms are used.
with variable penetration. In families with known MH The historical Shock! Shock! Shock! with escalating
causative mutations, molecular genetic testing for MH energy sequence has now been replaced by a single shock
susceptibility is mandatory.93 Because of the heterogene- followed by immediate CPR. At a compression rate of
ity of the MH trait, the diagnosis of MH must also be 100 per min, and compressionventilation ratio of 20:1,
confirmed with muscle biopsy and halothanecaffeine five cycles are performed every 2 minutes. Either an initial
in vitro muscle contraction test. Clinical findings that raise biphasic electrical dose of 150 J to 200 J (truncated
suspicion of MH are family history, masseter rigidity, exponential waveform) or 120 J (rectilinear waveform)
C H A P T E R 21/ C A R D I A C A R R E S T 305
should be administered. If the biphasic waveform is led to the overall, single, most important modification
unknown, 200 J is recommended. The second dose should of past CPR guidelines: compression to ventilation ratio
be at least the same amount or preferably higher. If only (C:V) to a universal 30:2 for all patients except children
a monophasic defibrillator is available, the recommended (younger than 8 years) until an advanced airway device
dose is the highest (360 J). is inserted. For children younger than 8 years who de-
In the PEA and asystole algorithm, either 1 mg of velop CA secondary to asphyxia, a 15:2 compression to
epinephrine or 40 units of vasopressin are recommended ventilation ratio is recommended.
as immediate therapy. Cardiac pacing is no longer
recommended for asystole, with emphasis on assessing
VF and the differential diagnosis for the cause of arrest.
The rationale for the removal of pacing is not defined What Types of Medicolegal
but most likely is a result of simplification of the
algorithm and the lack of success of cardiac pacing
and Ethical Issues Are Relevant
in out-of-hospital environments. With early recognition to Cardiac Arrest?
and experience, cardiac pacing is an effective approach
to asystole management. Attempts at defibrillation may When CA results in a malpractice suit, the testimony of
be effective only in the very early stage of pulseless medical experts is pivotal. Unfortunately, the definition
arrhythmia or during the course of effective CPR. of an expert in anesthesia is still at the discretion of
the judge who may apply criteria that are too liberal. The
appropriateness of the testimony of some expert witness
is also open to criticism.99 To enhance the quality of this
How Long to Resuscitate? process, the ASA now provides professional standards for
expert testimony.
A clear and appropriate informed consent is an
In the absence of clear guidelines on the time limits al-
ethical duty and also the cornerstone of a successful
lowed to resuscitate a patient with ACA, the issue remains
legal defense. It should include goal and procedure-
a difficult one. The 2005 American Heart Association
directed options and define alternative approaches to the
Guidelines, while suggesting spans of time and level of
procedure proposed. Disagreements within the family in
efforts of resuscitation, very appropriately leaves plenty
general, but especially if a patient is unable to provide
of latitude for physicians on a case-by-case basis. In a his or her own informed consent, should be avoided. The
unique case of ACA, the patient had a full neurologic re- presence of a non-health care provider, such as a social
covery after 5 hours of CPR. In most cases, if there is no worker, and proper documentation are good strategies for
restoration of the circulation, CPR is seldom successful the physician who wants to avoid misunderstanding or
after 30 minutes. The presence or return of spontaneous misinterpretation. Institutional support for achieving an
respiratory within 7 minutes and the presence of cranial end-of-life care policy is also important. The ASA website
nerve reflexes shortly after CA correlate with favorable offers useful coverage of the entire spectrum of relevant
neurologic outcomes.97 issues (www.ASAhq.org).
The epidemiology of dysrhythmic ACAs is unique. In
fact, it is the only clinical situation where hypoxemia or
hypercarbia and/or dysrhythmias leading to CA frequently
coexist. These precipitating factors can occur during MAC, What Is Expected from a
regional, or general anesthesia. The most common forms
of cardiac dysrhythmias are bradycardia or asystole. Do-Not-Resuscitate Order?
The patients chance of survival is also influenced by
factors such as the physicians or health care providers Whereas patients may usually focus on their functional
knowledge of the patients medical history and the status, physicians usually center on diagnosis and life
probable cause of arrest, as well as initiating medical expectancy. For most anesthesiologists, the patients de-
management within seconds. To ensure the best quality of sire to receive surgical therapy appears to be inconsistent
the resuscitation, all anesthesia personnel must maintain with possible directives against resuscitation. Extensive
up-to-date knowledge of the current CPR guidelines. reviews on this topic100 have shown that young age, elec-
One of the striking findings of the 2005 International tive surgery, good functional status, the risk of iatrogenic
Consensus Conference on Cardiopulmonary Resuscita- events, lack of family support, and an inadequate intel-
tion was the awareness of the poor quality of chest lectual background can all influence the decision as to
compressions often provided.98 Good CPR applied si- institute a do-not-resuscitate order.100
multaneously with pharmacological interventions and Despite the widespread trend to perform complex
defibrillation should enhance the ROSC and remains a surgery on the older population, the issue of do-not-
prerequisite of adequate cerebral and coronary perfusion resuscitate order has been infrequently discussed with
in CA. Even during appropriate CPR, blood flow is only the patient before surgery. A recent survey in the United
approximately 30% of normal, so less ventilation than States on this subject indicates that most anesthesiol-
normalthat is, fewer breaths and smaller volumeis ogists simply try to avoid this issue;101 approximately
enough to match ventilation to perfusion. This realization two thirds of the respondents disregarded such requests
306 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
in the perioperative period, and only one half of them to improve performance skill and enhance the knowledge
discussed the issue with the patients guardian. If the of the science of resuscitation in both academic and
patient was in an unresponsive state, 33% to 76% of the nonacademic environments.105
anesthesiologists surveyed were of the opinion that the use
of positive pressure ventilation, pharmacologic support,
and defibrillation applied intraoperatively did not con-
stitute resuscitation. This behavior is probably based
CONCLUSIONS
on the belief that resuscitation in the odds ratio (OR) by The attitude that anesthetic death is due to ignorance,
skilled personnel and appropriate equipment immediately negligence, and only rarely to inadequate scientific knowl-
available is most likely successful, when compared to CA edge is flawed.24 Although the risks versus benefits of
occurring in other areas of the hospital. It has also been anesthetic agents and technique are being considered on a
suggested that the need for an anesthetic and surgical patient-by-patient basis, inadequate planning, faulty mon-
procedure may not allow faithful following of a patients itoring, and lack of basic resuscitation maneuvers are still
directives. For example, postoperatively, a 72-hour au- implicated in more than half of CAs.
tomatic delay for do-not-resuscitate orders should be Anesthesiologists should maintain a central role in
instituted.102 In our view, the routine use of general resuscitation in the operating room, continue their lead-
anesthesia techniques, such as endotracheal intubation, ership in resuscitation in other areas of the hospital, and
mechanical ventilation, pharmacologic manipulation of extend their activity beyond the institutions boundaries.
hemodynamics, and blood loss replacement should not be To be able to respond to this challenge, he or she should
regarded as excessive resuscitation efforts defined in the be knowledgeable enough to master and, if necessary,
do-not-resuscitate order. modify the available resuscitation guidelines and be able
Despite the common perception that the presence to adapt them to individual special work environments.
of a do-not-resuscitate order correlates with a general
reduction in the aggressiveness of care, such orders should
indeed be carried out whenever the patient either does not
wish to receive full CPR, or is too ill to benefit from it. KEY POINTS
It is clear, however, that adequate time must be spent
with the patient preoperatively to discuss these issues.103 1. The management of CA has undergone significant
Because the intrinsic nature of general anesthesia implies changes, and the American Heart Association has
aggressive intervention on the cardiac and respiratory provided tremendous momentum to the education of
system, at the time of this writing this issue remains still health care personnel and the public. The most recent
controversial. 2005 international consensus emphasizes evidence-
based guidelines and recommends algorithm simpli-
fication.
2. The special environment of the operating room and
the presence of an anesthesiologist provide a unique
How Has the Advent of Patient approach to the management of CA due to the
Simulation Helped the expertise in airway management and immediately
Physician? available interventional monitoring.
3. Anesthetic administration has its own risks of causing
a cardiac or respiratory arrest, with a specific
Since the 1940s, ACA has been attributed to a variety of and immediate differential diagnosis necessary to
causes including underlying disease, the surgical proce- maximize neurologic outcome after ROSC.
dure itself, and anesthetic mishap. Although the revolution 4. The basis of modern resuscitation from CA in and
in monitoring has greatly contributed to safety in anes- outside the operating room are the same and include
thesia, paradoxically it also led to a false sense of security the following:
and lowering the acuity of direct observation. Patient Early recognition of the event and differential
simulators, which introduced integrated monitoring and diagnosis, high quality of chest compression, and
bedside physical examination in realistic crisis scenar- adequate matching of ventilation to the low flow
ios, following the aviation safety model, became valuable state
tools of education.104 Such simulators are based on a Early defibrillation for shockable rhythm, effective
number of advantages by reproducing and having the po- pharmacologic approach and use of circulatory
tential to perform the following tasks: (i) Alter different adjuncts to enhance coronary perfusion pressure
crisis scenarios; (ii) analyze human behavioral patterns and prevent recurrence of malignant dysrhythmias
scientifically; and (iii) examine issues of teamwork, lead- Therapeutic hypothermia when indicated to limit
ership, and communication, all without any risk to the neurologic damage from low flow state
patient. Simulation can also be used to assess perfor-
mance, training, and update clinical skills, which allows
for the incorporation of new technology, standards, guide- REFERENCES
lines, and research to analyze human factors under stress. 1. Lagasse RS. Anesthesia safety: Model or myth? A review
With growing focus on behavior and performance, the of the published literature and analysis of current original
use of full scale anesthesia simulation is an excellent way data. Anesthesiology. 2002;97:1609.
C H A P T E R 21/ C A R D I A C A R R E S T 307
2. Silber JH, Kennedy SK, Even-Shoshan O, et al. Anesthe- 25. Currie M, Mackay P, Morgan C, et al. The wrong drug
siologist direction and patient outcomes. Anesthesiology. problem in anaesthesia: An analysis of 2000 incident
2000;93:152. reports. Anaesth Intensive Care. 1993;21:596.
3. Beecher HK: The first anesthesia death, with some remarks 26. Tinker JH, Dull DL, Caplan RA, et al. Role of monitoring
suggested by it on the fields of the laboratory and the clinic devices in prevention of anesthetic mishaps: a closed claims
in the appraisal of new anesthetic agents. Anesthesiology. analysis. Anesthesiology. 1989;71:541.
1941;2:443. 27. Lamia B, Chemla D, Richard C, et al. Clinical review:
4. Rosenberg J, Wahr J: Cardiac arrest during anesthesia. interpretation of arterial pressure wave in shock states.
In: Paradis N, Halperin H, Nowak R, eds. Cardiac Arrest: Crit Care. 2005;9:601.
The Science and Practice of Resuscitation Medicine. 1st ed. 28. Aufderheide TP, Sigurdsson G, Pirallo RG, et al.
Baltimore: Williams & Wilkins, 1996:783811. Hyperventilationinduced hypotension during cardiopul-
5. Macintosh RR: Deaths under anesthetics. Br J Anaesth. monary resuscitation. Circulation. 2004;109:1960.
1948;21:197. 29. Baskett P, Nolan JP, Parr M: idal volumes which are per-
6. Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia ceived to be adequate for resuscitation. Resuscitation. 1996;
in surgical mortality. JAMA. 1961;178:261. 31:231.
7. Jude JR, Bolooki H, Nagel E: Cardiac resuscitation in the 30. Dumot JA, Burval DJ, Sprung J, et al. Outcome of adult
operating room: Current status. Ann Surg. 1970;17:948. cardiopulmonary resuscitations at a tertiary referral center
8. Cliffton BS, Hoffman WIT: Deaths associated with anaes- including results of limited resuscitation. Arch Intern Med.
thesia. Br J Anaesth. 1963;35:250. 2001;161:1751.
9. Harrison GG: Death attributable to anaesthesia: A 10-year 31. Tiret L, Nivoche Y, Hatton F, et al. Complications related
survey (19671976). Br J Anaesth. 1978;50:1041. to anaesthesia in infants and children. A prospective survey
10. Kennan RL, Boyan CP: Cardiac arrest due to anesthesia. A of 40240 anaesthetics. Br J Anaesth. 1988;61:263.
study of incidence and causes. JAMA. 1985;253:2373. 32. Sachs BP, Oriol NE, Ostheimer GW, et al. Anesthetic-related
11. Morgan CA, Webb RK, Cockings J, et al. Cardiac arrest: maternal mortality, 1954 to 1985. J Clin Anesth. 1989;
An analysis of 2000 incident reports. Anaesth Intens Care. 1:333.
1993;21:626. 33. Ngan Kee WD: Confidential enquiries into maternal deaths:
12. Biboulet P, Aubas P, Dubourdieu J, et al. Fatal and non-
50 years of closing the loop. Br J Anaesth. 2005;94:413.
fatal cardiac arrests related to anesthesia. Can J Anesth.
34. Cooper GM, McClure JH: Maternal deaths from anaes-
2001;48:326.
thesia. An extract from Why Mothers Die 20002002,
13. Newland MC, Ellis SJ, Lydiatt CA, et al. Anesthetic-related
the confidential enquiries into maternal deaths in the
cardiac arrest and its mortality. A report covering 72,959
United Kingdom: Chapter 9: Anaesthesia. Br J Anaesth.
anesthetics over 10 years from a US teaching hospital.
2005;94:417.
Anesthesiology. 2002;97:108.
35. Fasting S, Gisvold SE: Serious intraoperative problemsA
14. Bodlander FM: Deaths associated with anaesthesia. Br J
five-year review of 83,844 anesthetics. Can J Anaesth.
Anaesth. 1975;47:36.
2002;4:545.
15. Lunn JN, Hunter AR, Scott DB: Anaesthesia-related surgical
36. Hunter JM: Synergism between halothane and labetalol.
mortality. Anaesthesia. 1983;38:1090.
Anaesthesia. 1979;34:257.
16. Kubota Y, Toyoda Y, Kubota H, et al. Frequency of
37. Wenzel V, Padosch SA, Voelckel WG, et al. Survey of
anesthetic cardiac arrest and death in the operating room
effects of anesthesia protocols on hemodynamic variables in
at a single general hospital over a 30-year period. J Clin
porcine cardiopulmonary resuscitation laboratory models
Anesth. 1994;6:227.
17. Caplan RA, Ward RJ, Posner KL, et al. Adverse respiratory before induction of cardiac arrest. Comp Med. 2000;50:644.
events in anesthesia: a closed claims analysis. Anesthesiol- 38. Van den Hurk AW, Teijen HJ: Cardiac complications during
ogy. 1990;72:828. use of etomidate. Anaesthesia. 1983;38:1183.
18. Caplan RA, Ward RJ, Posner KL, et al. Unexpected cardiac 39. Inoue K, Reichelt W: Asystole and bradycardia in adult
arrest during spinal anesthesia: A closed claims analysis of patients after a single dose of suxamethonium. Acta
predisposing factors. Anesthesiology. 1988;68:5. Anaesthesiol Scan. 1986;30:571.
19. Cheney FW: The American Society of Anesthesiologists 40. Guise PA: Asystole following propofol and fentanyl in an
Closed Claims Project. What have we learned: How has anxious patient. Anesth Intensive Care. 1991;19:116.
if affected practice, and how will it affect practice in the 41. Tramer MR, Moore RA, McQuay HJ: Propofol and
future? Anesthesiology. 1999;91:552. bradycardia: Causation, frequency, severity. Br J Anaesth.
20. American Society of Anesthesiologists Task Force on Guide- 1997;78:642.
lines for Management of the Difficult Airway: Practice 42. Martyn JA, Richtsfeld M: Succinylcholine-induced hy-
guidelines for management of the difficult airway. Anes- perkalemia in acquired pathologic states: Etiologic fac-
thesiology. 1993;78:597. tors and molecular mechanisms. Anesthesiology. 2006;104:
21. Olsson GL, Hallen B: Cardiac arrest during anaesthesia. 158.
A computer-aided study in 250,543 anaesthetics. Acta 43. Elliott P, OHare R, Bill KM, et al. Severe cardiovascular
Anaesthesiol Scand. 1988;32:653. depression with remifentanil. Anesth Analg. 2000;91:58.
22. Sprung J, Warner ME, Contreras MG, et al. Predictors 44. Glaser R: Chronic hypokalemia and intraoperative dys-
of survival following cardiac arrest in patients undergoing rhythmias. Anesthesiology. 1986;61:408.
noncardiac surgery: A study of 518,294 patients at a tertiary 45. Pollard JB: Cardiac arrest during spinal anesthesia: Com-
referral center. Anesthesiology. 2003;99:259. mon mechanisms and strategies for prevention. Anesth
23. Adams JH: Hypoxic brain damage. Br J Anaesth. 1975;47: Analg. 2001;92:252.
121. 46. Stienstra R: Mechanisms behind and treatment of sudden,
24. Foex P: Cardiac arrest during anesthesia. Am J Emerg Med. unexpected circulatory collapse during central neuraxis
1984;2:341. blockade. Acta Anaesthesiol Scand. 2000;44:965.
308 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
47. Carpenter RL, Caplan RA, Brown DL, et al. Incidence 69. Holden CG, Monaghan D, Cassidy M: Retention of car-
and risk factors for side effects of spinal anesthesia. diopulmonary resuscitation skills of dental nurses. SAAD
Anesthesiology. 1992;76:906. Dig. 1996;13:3.
48. Bainbridge FA: The influence of venous filling upon the rate 70. Perrott DH, Yuen JP, Andresen RV, et al. Office-based
of the heart. J Physiol (Lond). 1915;50:65. ambulatory anesthesia: Outcomes of clinical practice of oral
49. Hainsworth R: Reflexes from the heart. Phys Rev. and maxillofacial surgeons. J Oral Maxillofac Surg. 2003;
1991;71:617. 61:983.
50. Campagna JA, Carter C: Clinical relevance of the Bezold- 71. White PF, Roizen MF. Cardiac arrest in a day surgery
Jarisch Reflex. Anesthesiology. 2003;98:1250. patient. Anesthesiology. 1990;72:771.
51. Kopp SL, Horlocker TT, Warner ME, et al. Cardiac arrest 72. Cote CJ, Notterman DA, Karl HW, et al. Adverse seda-
during neuraxial anesthesia: Frequency and predisposing tion events in pediatrics: A critical incident analysis of
factors associated with survival. Anesth Analg. 2005;100: contributing factors. Pediatrics, 2000;105:805.
855. 73. Fleisher LA, Eagle KA. Lowering cardiac risk in noncardiac
52. Auroy Y, Benhamou D, Bargues L, et al. Major compli- surgery. N Engl J Med. 2001;345:1677.
cations of regional anesthesia in France. Anesthesiology. 74. Williamson JA, Webb RK, Sellen A, et al. The Australian
2002;97:1274. Incident Monitoring Study, Human failure: an analysis of
53. Rosenberg JM, Wahr JA, Sung CH, et al. Coronary perfusion 2000 incidence reports Anaesth Intensive Care. 1993;21:
pressure during cardiopulmonary resuscitation after spinal 678.
anesthesia in dogs. Anesth Analg. 1996;82:84. 75. Arbous MS, Meursing AE, van Kleef JW, et al. Impact of
54. Callaham M, Barton C: Prediction of outcome of car- anesthesia management characteristics on severe morbidity
diopulmonary resuscitation from end-tidal carbon dioxide and mortality. Anesthesiology. 2005;102:257.
concentration. Crit Care Med. 1990;18:358. 76. Needleman J, Buerhaus P, et al. Nurse-staffing levels and
55. Tang W, Weil MH, Sun S, et al. Epinephrine increases the quality of care in hospitals. N Engl J Med. 2002;346:
the severity of postresuscitation myocardial dysfunction. 1715.
Circulation. 1995;92:3089. 77. Aiken LH, Clarke SP, Cheung RB, et al. Educational levels
56. Wenzel V, Krismer AC, Arntz HR, et al. The European Re- of hospital nurses and surgical patient mortality. JAMA.
suscitation Council Vasopressor during Cardiopulmonary 2003;290:1617.
Resuscitation Study Group. A Comparison of Vasopressin 78. Rivers E, Nguyen B, Havstad S, et al. Early Goal-Directed
and Epinephrine for Out-of-Hospital Cardiopulmonary Re- Therapy Collaborative Group. Early goal-directed therapy
suscitation. N Engl J Med. 2004;350:105. in the treatment of severe sepsis and septic shock. N Engl J
57. Krismer AC, Hogan QH, Wenzel V, et al. The efficacy Med. 2001;345:1368.
of epinephrine or vasopressin for resuscitation during 79. Monk TG, Saini V, Weldon BC, et al. Anesthetic man-
epidural anesthesia. Anesth Analg. 2001;93:734. agement and one-year mortality after noncardiac surgery.
58. Johns RA, DiFazio CA, Longnecker DE. Lidocaine constricts Anesth Analg. 2005;100:4.
or dilates rat arterioles in a dose-dependent manner. 80. Russ MJ, Bailine SH: Asystole and bradycardia-related to
Anesthesiology. 1985;62:141. anesthetic induction during ECT: A case report. J of ECT.
59. Clarkson CW, Hondeghem LM. Evidence for a specific 2004;20:195.
receptor site for lidocaine, quinidine and bupivacaine 81. Van Brocklin MD, Hirons RR, Yolton RL: The oculocardiac
associated with cardiac sodium channels in guinea pig reflex: A review. J Am Optom Assoc. 1982;53:407.
ventricular myocardium. Circ Res. 1985;56:496. 82. Sprung J, West AJ, Bacon DR, et al. Myocardial stunning
60. Groban L, Deal DD, Vernon JC, et al. Cardiac Resuscitation during abdominal hysterectomy in a healthy woman. J Clin
after incremental overdosage with lidocaine, bupivacaine, Anesth. 2003;15:537.
levobupivacaine, and ropivacaine in anesthetized dogs. 83. Hohn L, Schweizer A, Morel DR, et al. Circulatory failure
Anesth Analg. 2001;92:37. after anesthesia induction in a patient with severe primary
61. Davis NL, de Jong RH. Successful resuscitation following pulmonary hypertension. Anesthesiology. 1991;91:1943.
massive bupivacaine overdose. Anesth Analg. 1982;61:62. 84. Jones DH, Schlatter MG, Cornelius AS, et al. A massive
62. Coyle DE, Sperelakis N. Bupivacaine and lidocaine block- pulmonary tumor embolism after surgical manipulation
ade of calcium-mediated slow action potentials in guinea and biopsy of a pelvic mass. Anesth Analg. 2000;90:322.
pig ventricular muscle. J Pharm Exper Ther. 1987;242: 85. Keane TK, Hennis PJ, Bink-Boelkens MTE. Non-drug
1001. related asystole associated with anaesthetic induction.
63. Albright GA. Clinical aspects of bupivacaine toxicity. Anaesthesia. 1991;46:38.
Presented to the Food and Drug Administration. Anesthetic 86. Shenkin HA, Cheney RH, Govons SR, et al. On the diagnosis
and Life Support Drugs Advisory Committee, October 4, of hemorrhage in manA study of volunteers bled large
1983. amounts. Am J Med Sci. 1944;208:421.
64. Weinberg G, Ripper R, Feinstein DL, et al. Lipid emulsion 87. Finfer SR. Pacemaker failure on induction of anaesthesia.
infusion rescues dogs from bupivacaine-induced cardiac Br J Anaesthesia. 1991;66:509.
toxicity. Reg Anesth Pain Med. 2003;28:198. 88. Ohm OJ, Bruland H, Pedersen OM, et al. Interference
65. Memery HN: Anesthesia mortality in private practice. A ten- effect of myopotentials on function of unipolar demand
year study. JAMA. 1965;194:1185. pacemakers. Br Heart J. 1974;36:77.
66. Bryson GL, Chung F, Cox RG, et al. Canadian Ambulatory 89. Domino KB, Smith TC. Electrocautery-induced reprogram-
Anesthesia Research Education Group. Patient selection in ming of a pacemaker using a precordial magnet. Anesth
ambulatory anesthesia - An evidence-based review: part II. Analg. 1983;62:609.
Can J Anaesth. 2004;51:782. 90. Watkins J. Investigation of allergic and hypersensitivity
67. Brahams D: Death in the dentists chair. Lancet. 1989;991. reactions to anaesthetic agents. Br J Anaesth. 1987;59:104.
68. Krippaehne JA, Montgomery MT. Morbidity and mortality 91. Wig J, Bali IM, Singh RG, et al. Prolonged Q-T inter-
from harmacosedation and general anesthesia in the dental val syndrome. Sudden cardiac arrest during anaesthesia.
office. J Oral Maxillofac Surg. 1992;50:691. Anaesthesia. 1979;34:37.
C H A P T E R 21/ C A R D I A C A R R E S T 309
92. Hopkins PM. Malignant hyperthermia: Advances in clinical 99. Liang BA, Walman AT. Who can be an expert in anesthesia
management and diagnosis. Br J Anaesth. 2000;85:118. malpractice suits? A case of general anesthesia, cardiopul-
93. Girard T, Treves S, Voronkov E, et al. Molecular genetic monary risk, and patient death. J Clin Anesth. 2003;15:
testing for malignant hyperthermia susceptibility. Anesthe- 395.
siology. 2004;100:1076. 100. Eliasson AH, Parker JM, Shorr AF, et al. Impediments to
94. Krause T, Gerbershagen MU, Fiege M, et al. DantroleneA writing do-not-resuscitate orders. Arch Intern Med. 1999;
review of its pharmacology, therapeutic use and new 159:2213.
developments. Anaesthesia. 2004;59:364. 101. Clemency MV, Thompson NJ. Do Not Resuscitate (DNR)
95. Christiansen LR, Collins KA. Pathologic findings in malig- Orders and the anesthesiologist: A survey. Anesth Analg.
nant hyperthermia: a case report and review of literature. 1993;76:394.
Am J Forensic Med Pathol. 2004;25:327. 102. Bernar JL, Gabowski EW. Suspending do-not-resuscitate
96. 2005 American Heart Association Guidelines for Cardiopul- orders during anesthesia and surgery. Surg Neurol.
monary Resuscitation and Emergency Cardiovascular Care. 1993;40:7.
Circulation. 2005;112[Suppl I]:1211. [Also available at: 103. Clemency MV, Thompson NJ. Do not resuscitate orders in
http://www.circ.ahajournals.org/content/vol112/24 suppl/ the perioperative period: patient perspectives. Anesth Analg.
#INDEX. Accessed March 20, 2007.] 1997;84:859.
97. Fabbri LP, Nucera M, Becucci A, et al. An exceptional case 104. Lindekaer AL, Jacobsen J, Andersen G, et al. Treat-
of complete neurologic recovery after more than five-hour ment of ventricular fibrillation during anaesthesia in an
cardiac arrest. Resuscitation. 2001;48:175. anaesthesia simulator. Acta Anaesthesiol Scand. 1997;41:
98. Abella BS, Alvarado JP, Myklebust H, et al. Quality of 1280.
cardiopulmonary resuscitation during in-hospital cardiac 105. Gaba DM. Anaesthesiology as a model for patient safety in
arrest. JAMA. 2005;293:363. health care. BMJ. 2000;320:785.
C . N E U R O LO G I C
CHAPTER INCREASED INTRACRANIAL
22
PRESSURE
Cheri A. Sulek
A
the emergency department with a history of the underlying pathophysiologic processes involved,
of confusion, intermittent slurred speech, the impact of secondary neuronal injury, and the informa-
difficulty with expressing and understand- tion obtained from cerebral monitoring available today.
ing speech, and progressive headaches. At It is important to recognize the impact of hemodynamic
the onset of his symptoms 3 months ago, manipulations, pharmacologic interventions, anesthetic
he was evaluated at an outside hospital and diagnosed agents, and ventilation on ICP and cerebral perfusion
with a left-sided stroke and possible seizure disorder. pressure (CPP). The goals of this chapter are to provide an
The etiology of his stroke was not determined, and he understanding of : (i) ICP and its determinants; (ii) the reg-
was discharged home on anticonvulsants. He improved ulation of ICP and its effects on CPP in normal and patho-
with physical therapy after discharge; however, his fam- logic states; (iii) the impact of anesthetic agents on ICP;
ily noted a progressive decline in language function and and (iv) the management of intracranial hypertension.
worsening headaches. His medical history was significant
for a myocardial infarction status post coronary artery by-
pass grafting, hypertension, smoking, and hyperlipidemia.
Magnetic resonance imaging (MRI) of the brain with con-
trast revealed a multilobulated increasing mass in the left What Are the Structures that
temporal and parietal lobes, measuring 5 5 cm, with Comprise the Intracranial
a large amount of peritumor edema. A significant mass
effect was noted on the left lateral ventricle and an 8-mm
Compartment?
left-to-right shift, with early transtentorial herniation. He
was started on dexamethasone, with significant improve- The intracranial compartment is defined by its con-
ment in his headaches, and remained on anticonvulsant tents: Brain tissue, cerebrospinal fluid (CSF), blood, and
therapy. He underwent a left craniotomy for biopsy and meninges. They are encased by the calvarium and commu-
tumor debulking under general anesthesia without com- nicate with the spinal axis through the foramen magnum.
plications. Pathology was consistent with glioblastoma In the absence of pathology, the intracranial volume re-
multiforme. mains constant within the neuraxis. The calvarium is a
bony, nondistensible, semiclosed container that strictly
limits both intracranial volume and any expansion by
INTRODUCTION acute or chronic pathologic processes.
Brain tissue alone accounts for 88% of the total
Maintenance of normal intracranial pressure (ICP) is de- intracranial volume. Eighty percent of the brain volume is
termined and closely regulated by well-defined structural water, and 20% of this is sequestered extracellularly.1 This
components of the intracranial compartment. The intro- extracellular environment is tightly regulated by an intact
duction of intracranial pathology will ultimately exhaust blood-brain barrier, which selectively permits diffusion
the mechanisms that maintain cerebral homeostasis and and active transport of limited substances.
310
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 311
70
Considered Abnormal?
50
lia e,
mp nc
e
40
is 5 to 15 mm Hg; normal values are 1.5 to 6 mm Hg D
low h e
30
Hi
supine position, pressures are considered equal within the Low elastance,
craniospinal axis. Lumbar subarachnoid pressures reflect 20 high compliance
ICP as long as the two compartments freely communicate
and no obstruction exists. It is now recognized that 10 C
ICP is not always uniform within the craniospinal axis A B
when pathology is present. Local differences in ICP 0
and cerebral blood flow (CBF) often exist in areas of Intracranial volume
pathology (i.e., brain tumors or traumatic brain injury
[TBI]), although, global ICP may be recorded as normal FIGURE 22.1 Intracranial pressure-volume curve. Along the
and is misleading.8,9 ICP is considered abnormal with horizontal axis (points AB), a change in intracranial volume is
levels of 15 mm Hg in infants, 18 mm Hg in children offset by a reciprocal decrease in another intracranial
younger than 8 years, and 20 mm Hg in older children and component, allowing for spatial compensation. Large changes in
adults.7 Neurologic outcome, especially in head-injured intracranial volume can occur without a change in intracranial
patients, appears to correlate with the degree and duration pressure. At point C, compensatory mechanisms are exhausted.
of intracranial hypertension.10 Sustained elevations in ICP Between points CD, small changes in intracranial volume
that exceed 25 to 30 mm Hg are frequently associated result in large changes in intracranial pressure.
with poor outcome or are fatal. The detrimental effects (Modified from: Fig. 44.1 in Kirby RR, Gravenstein N, Lobato
of intracranial hypertension are the result of cerebral EB, et al., eds. Clinical anesthesia practice, 2nd ed. WB
ischemia and the direct compressive effects on cerebral Saunders; 2002:836.)
structures.
312 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
INTRACRANIAL PRESSURE
MONITORS
An ICP monitor does not replace the neurologic examina-
tion but can guide treatment when the patient is comatose
and intracranial hypertension is suspected. The Glasgow
Coma Scale (GCS) is the gold standard utilized to deter-
mine the need for ICP monitoring in head-injured patients.
A GCS of 8 or less in head-injured patients warrants
the placement of a monitoring device.13 Any therapeutic
intervention that may elevate ICP in a patient with com-
promised intracranial elastance warrants consideration of
placement of an ICP monitor. The risks and benefits of
ICP monitoring must be carefully weighed in the presence
of coagulopathy. This issue has become even more rele-
vant with the increased number of liver transplantations
being performed, some of which will be done in patients
with encephalopathy and coexisting coagulopathy. Indi-
cations for ICP monitoring and the device utilized often
reflect the bias and experience of the neurosurgeon and
any institutional protocols that exist.
FIGURE 22.3 Magnetic resonance imaging of the brain There are several devices for clinical monitoring of
depicting a large brain tumor with significant mass effect and ICP. The intraventricular catheter is considered the gold
edema formation. standard of recording devices and is used to determine
the accuracy of new monitors. It has the dual advantage
of ICP measurement and therapeutic CSF drainage to
information (see Fig. 22.3). In patients with severe head lower ICP. Technical problems limit its use in patients
injury, the presence of midline shift and compressed ambi- with small or compressed ventricles such as pseudotumor
ent cisterns usually predict increased ICP or herniation.28 cerebri or a large brain tumor. Its placement requires
It is important to correlate neuroradiologic findings with more skill than with other monitoring devices and has
the neurologic examination because the clinical presenta- been associated with a higher incidence of infection, with
tion often reflects the chronicity of the underlying disease a greater potential to cause brain injury during placement.
process. The subarachnoid bolt or screw is associated with a
lower incidence of infection, less potential for brain injury,
and requires less skill to place. Mollman et al. determined
that the subarachnoid bolt was less accurate for ICP
monitoring when compared to either a subarachnoid
When Is Intracranial Pressure catheter or ventriculostomy technique.31 When ICP is
Monitoring Indicated and high, subarachnoid bolts are less reliable because of
obstruction of the lumen of the bolt by brain tissue and
Which Device? its propensity to produce falsely low readings.
Fiberoptic ICP monitoring systems are used fre-
The intraventricular catheter (or ventriculostomy) was in- quently and can be placed in the intraparenchymal,
troduced in 1960 by Lundberg and has remained the gold subdural, or intraventricular compartment. Fiberoptic
standard for ICP monitoring.29 Newer monitoring modal- monitors have proven to be reliable, accurate devices
ities have been developed; ICP can be recorded from an when compared with intraventricular monitors.32,33 The
epidural, subdural, subarachnoid, ventricular, or intra- devices have major drawbacks including significant drift,
parenchymal location, depending upon the device used. inability to recalibrate in vivo, expense and fragility of the
The monitoring devices presently in use include catheters, fiberoptic catheter. It is recommended that these devices
bolts, screws, and fiberoptic cables. ICP monitoring is be replaced every 5 days to minimize false readings and
most often achieved from a supratentorial location and infectious complications.
is less commonly employed in the posterior fossa due to
risk of cranial nerve or brainstem injury.30 It is pre-
sumed that once the monitor is in position, the ICP
recorded is reflective of pressure in the supratentorial What Is the Role
and infratentorial compartments in the absence of ob-
struction; however, differences in ICP have been recorded
of Transcranial Doppler?
not only between supratentorial and infratentorial com-
partments, but also within the same compartment (i.e., Transcranial doppler (TCD) imaging provides an inexpen-
supratentorial). sive, noninvasive bedside method to document severe
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 315
to sevoflurane precipitously increases CBF and impairs normal. The combination of nitrous oxide and isoflurane
autoregulation.48 The data is conflicting: Sevoflurane may or sevoflurane appears to increase CBF and should not be
precipitate seizure activity at high anesthetic concentra- used when ICP is elevated.
tions but should be a safe volatile agent when used in
clinically relevant concentrations.
Desflurane is a volatile anesthetic with a low blood- INTRAVENOUS AGENTS
gas partition coefficient that allows for rapid emergence
to evaluate the neurologic status of the patient. The Barbiturates have profound effects on the cerebral vas-
effects of desflurane on CBF and ICP are conflicting, culature and metabolism that are coupled. Thiopental is
but there is evidence that it has significant cerebral the most widely used barbiturate in anesthesia practice;
vasodilating properties.49,50 CBF increases in a dose- however, pentobarbital is commonly utilized in the inten-
dependent manner and is augmented by hemodynamic sive care unit (ICU) to lower ICP and establish barbiturate
support in animals.50,51 Reductions in CMRO2 following coma. With the exception of methohexital, which when
electroencephalogram (EEG) changes are similar to given in small doses may trigger seizure activity, the ef-
isoflurane anesthetic.50 The response of the cerebral fects of sodium thiopenthal on the brain reflect those of
vasculature to CO2 is intact in dogs and humans at all barbiturates utilized in anesthesia practice, and will be
clinically relevant concentrations of desflurane.52 discussed in the following text.
The effect of desflurane on ICP is controversial, par- Thiopental consistently lowers CBF and CMRO2 in
ticularly in humans. ICP elevation has been documented a dose-dependent fashion in animals and humans.54
in canine and porcine studies, and are similar to those Cerebral oxygen consumption is reduced by 55% when
reported under isoflurane anesthesia in the absence of the EEG becomes isoelectric with thiopental adminis-
pathology.50,52 It is speculated that ICP elevations are tration.54 Even with massive doses of thiopental, no
attributable to an elevated CBV reflecting desfluranes further reductions in CMRO2 are observed, and adenosine
potent vasodilating properties. Because animal and hu- triphosphate (ATP), lactate, pyruvate, and phosphocrea-
man studies are conflicting, desflurane should probably tine concentrations are normal. In addition, thiopental
be used cautiously in patients with abnormal intracranial suppresses neuronal synaptic transmission but preserves
elastance. the metabolic function required to maintain cellular in-
Nitrous oxide has been traditionally utilized as a tegrity and homeostasis.
supplemental anesthetic during neurosurgical procedures ICP is consistently and effectively lowered with
and was previously considered to have minimal effects thiopental in patients with intracranial hypertension.55
on CBF and ICP. CBF studies have often been difficult to Barbiturates have minimal effect on ICP in individu-
interpret because of species differences, CBF technique, als with normal elastance. Barbiturates are utilized to
or the use of supplemental anesthetics that also have treat ICP elevations, especially when intracranial hyper-
the potential to affect the cerebral vasculature and tension is intractable and conventional measures have
metabolism. There is now substantial evidence to suggest failed. When barbiturates are used to control intracranial
that nitrous oxide has powerful cerebral vasodilating hypertension, the drug is titrated to maintain 90% sup-
properties and should not be used in patients with elevated pression of total EEG activity. Systemic hypotension is
ICP. Nitrous oxide by itself can significantly elevate CBF; common, especially with high doses, and requires aggres-
however, it is speculated that increases may be due to sive vasopressor treatment to maintain adequate CPP.
light anesthesia and are not reflection of direct cerebral Barbiturates provide cerebral protection during focal
vasodilation. but not global ischemia, particularly when given before
The effects of nitrous oxide on CBF are caused by the ischemic event.56 The cerebral protective effects of
direct vasodilatory actions or indirectly by changes in barbiturates are not solely related to CMRO2 reduction,
CMRO2 and may depend upon the depth of anesthesia. It but likely involve the attenuation of secondary neuronal
is postulated that the vasodilatory activity of nitrous ox- injury (i.e., free radical scavenging, alteration of fatty acid
ide predominates at increased depths of anesthesia when metabolism).
CMRO2 is maximally suppressed by the volatile agent Etomidate lowers CMRO2 and CBF, but blood flow
in use. changes are independent of changes in metabolism.57 This
Recent studies have demonstrated that hyperventila- drug has been reported to preferentially suppress cortical
tion does not consistently prevent increases in CBF, even structures, but has the potential to induce seizure activity.
if established before nitrous oxide administration.53 The Its use consistently augments the amplitude of the cortical
administration of nitrous oxide is contraindicated in the somatosensory-evoked potential, perhaps by preferen-
presence of an existing pneumocephalus because its use tially affecting inhibitory postsynaptic potentials, thereby
in this scenario can elevate ICP and precipitate a life- leaving excitatory postsynaptic potentials unopposed.
threatening tension pneumocephalus. If nitrous oxide is Etomidate effectively lowers ICP, thereby reflecting
to be utilized during a craniotomy, the agent should be its powerful vasoconstrictive effects on CBF.58 Because
initiated at the onset of surgery and not added after dural CBF changes occur more rapidly than CMRO2 reduction,
closure. it is possible for ischemia to occur. Although etomidate
Nitrous oxide is an acceptable anesthetic for neuro- reduces CMRO2 and produces an isoelectric EEG, it
surgical procedures, as long as intracranial elastance is has not been proven to provide cerebral protection
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 317
during focal ischemia.56 Cerebrovascular response to CO2 dexmedetomidine has been shown to lower CBF, it has
remains intact with etomidate anesthesia.57 no effect on cerebral metabolism, suggesting metabolism-
Although etomidate is associated with greater hemo- flow uncoupling.71 It is currently used for sedation in the
dynamic stability than barbiturates, it has several unde- neurointensive care unit and during awake craniotomies
sirable side effects including myoclonic activity during to facilitate complex neurologic testing, including neu-
induction and, with prolonged use, adrenocortical sup- rocognitive function.
pression. Additionally it should be used cautiously in The cerebral effects of benzodiazepines are variable
patients with seizure disorders. depending on the animal model, background anesthetic,
Propofol decreases both CBF and CMRO2 in humans. and neurologic status of the subject. Benzodiazepines
Its ability to reduce CBF has been attributed to parallel can be regarded as metabolic depressants possessing
changes in metabolism.59 Propofol can suppress neuronal vasoconstrictive properties that are less potent than barbi-
function to produce a burst-suppressed or isoelectric turates. Intravenous diazepam in animals decreased CBF
EEG with clinically relevant doses much like barbiturates, without significant changes in CMRO2, 72 and reductions
etomidate, or isoflurane. in CMRO2 and CBF have been reported in humans.73
Propofol reduces or has minimal effects on ICP.47,60 Diazepam also preserves the reactivity of the cerebral
The largest reductions in ICP have been observed in vasculature to CO2. 73
patients with intracranial hypertension and, to a lesser The effects of midazolam on ICP are variable and
extent, when elastance is normal.59,61 Hypotension, even may depend on the neurologic status of the patient
in young normal patients, can occur with anesthetic and whether the drug is given as a large bolus or by
doses of propofol and must be treated aggressively. infusion.74 Significant reductions in MAP can occur with
Cerebral autoregulatory mechanisms may be reduced or bolus administration, resulting in a dangerously low CPP,
impaired in head-injured patients receiving high doses of which may elevate ICP in some patients.74
propofol.62 Cerebrovascular reactivity to CO2 is preserved Benzodiazepines are safe when administered to
with clinical doses of propofol.63 Patients receiving neurosurgical patients as long as hemodynamics are
propofol have a clearer sensorium postoperatively when supported but should be used cautiously in unventilated
compared with their counterparts receiving traditional patients because respiratory depression and hypercapnia
inhalational anesthetics. can occur alone or in combination with other anesthetics.
Lidocaine has a biphasic effect on CBF and CMRO2 .
At low doses sufficient to produce sedation and general
anesthesia, CMRO2 and probably CBF are reduced;
however, at higher doses, seizures are precipitated,
causing dramatic rises in CMRO2 and CBF. Lidocaine
What Are the Effects
has been reported to either decrease or have no effect of Narcotics on Intracranial
on ICP.64 Intratracheal lidocaine may be more effective
than intravenous lidocaine, in attenuating ICP elevations
Pressure?
during endotracheal suctioning.64
Ketamine has potent cerebral stimulant properties The effects of narcotics on CBF, CMRO2 , and ICP are
demonstrated by its propensity to induce hallucinations highly variable. Their cerebral effects depend on the fol-
and seizure activity.65 Ketamine causes dramatic in- lowing: (i) Animal model; (ii) background anesthetic;
creases in CBF and CMRO2 . Dawson et al. reported an (iii) use of supraclinical doses; (iv) mode of drug ad-
increase in CBF by 80% and CMRO2 by 16% after ke- ministration (bolus vs. infusion); (v) baseline intracranial
tamine administration in dogs.65 Similar CBF elevations elastance; (vi) presence or absence of neuromuscular
not accompained by alterations in CMRO2 have been blockade; and (vii) mode of ventilation (spontaneous vs.
observed in humans.66 In humans, ketamine selectively controlled). Narcotics, especially the synthetic opioids, are
increases CBF in the frontotemporal and parieto-occipital an important component of the anesthetic management
regions of the brain.67 of neurosurgical patients and contribute to hemodynamic
Ketamine consistently elevates ICP or CSF pressure, stability and a rapid, smooth emergence.
which is substantial when used as the primary anesthetic Morphine has been reported to produce significant re-
agent in patients with normal and abnormal elastance.68 duction in CMRO2 and CBF in normocapnic anesthetized
If hemodynamic instability is present in head-injured dogs, which can be readily reversed by nalorphine.75
patients, etomidate is preferred to ketamine for the High doses of morphine sulfate should be avoided in
induction of anesthesia. patients with abnormal elastance due to potential his-
Dexmedetomidine is an 2 -adrenergic agonist that tamine release and subsequent hypotension that may
has gained popularity in neuroanesthesia and neurosur- lower CPP.
gical procedures. It possesses sedative, anxiolytic, and Fentanyl may have either no effect or an increase in
analgesic properties but is devoid of respiratory de- ICP in humans.76,77 The mechanism of ICP elevation may
pression and anesthetic or opioid properties. Limited be secondary to increases in CBF and perhaps increased
data exists on the cerebral effects of dexmedetomi- resistance to reabsorption of CSF at high narcotic doses.78
dine. In human studies that are available, dexmedeto- Fentanyl, in high doses, can precipitate seizure activity
midine lowered or had no effect on ICP.69,70 Although with adverse effects on CBF, CMRO2 , and ICP.
318 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Sufentanil has been found to decrease CMRO2 and as long as large bolus doses are avoided and systemic
to either increase or decrease CBF in dogs.79,80 Similarly, hypotension is aggressively treated.
it has been observed to have either no effect or increase
CBF in healthy human subjects.81
Sufentanil has been reported to produce inconsistent
effects on ICP. This opioid has been used safely in neu- What Are the Effects
rosurgical patients with abnormal intracranial elastance
without associated elevations in ICP, brain retractor pres- of Neuromuscular Blockade
sure, or adverse brain conditions upon dural opening.77,82 Agents on Intracranial
Marx et al. were among the first to document CSF pres-
sure increases in neurosurgical patients.76 The numerical
Pressure?
changes in CSF pressure with sufentanil, albeit statisti-
cally significant, were relatively small (4.4 1.1 mm Hg). Succinylcholine causes dramatic elevations in CBF and
Sperry et al. also observed ICP elevations in severely head- ICP in both humans and animals, even in the absence of
injured patients after a sufentanil bolus despite aggressive fasciculations.89,90 It is speculated that succinylcholine-
ongoing measures to reduce ICP.83 Albanese et al. demon- induced increases in CBF and ICP reflect afferent muscle
strated similar results with sufentanil (1 g per kg and spindle activation with stimulation of portions of the
infusion of 0.005 g/kg/minute) in head-injured patients motor and somatosensory cortex and are attributed to
sedated with propofol and relaxed with vecuronium.84 The secondary increases in PACO2 from enhanced muscle oxy-
ICP elevations were transient, and were accompanied by gen consumption.89,90 Pretreatment with a defasciculating
marked reductions in MAP and CPP. The transient effects dose of nondepolarizing muscle relaxant may prevent fas-
on ICP did not appear to adversely affect outcome in the ciculations, but does not prevent CBF elevations unless
population studied. intubating doses of nondepolarizing agents are used as
Conflicting studies documenting the cerebral effects pretreatment.91 Succinylcholine causes significant ICP
of alfentanil also exist. In neurosurgical patients and elevations in patients with both normal and abnormal
non-neurosurgical patients anesthetized with isoflurane intracranial elastance.
and nitrous oxide, low or high doses of alfentanil de- Succinylcholine should be avoided, if possible, in pa-
creased or had no effect on Vmca by TCD.85 As with tients with intracranial hypertension or space-occupying
fentanyl and sufentanil, brain retractor pressure was unaf- lesions. If succinylcholine is used when intracranial elas-
fected by alfentanil in neurosurgical patients undergoing tance is abnormal, pretreatment with a nondepolarizing
craniotomies.82 CSF pressure elevations have been doc- muscle relaxant should be considered and use of lidocaine
umented with alfentanil administration in patients with may attenuate, not prevent, elevations in ICP. The use of
brain tumors.76 Marked reductions in MAP often result barbiturates for induction of anesthesia and rapid con-
from alfentanil and warrant aggressive treatment espe- trol of the airway with a brief period of hyperventilation
cially when intracranial elastance is abnormal. may also attenuate the ICP elevations that occur with suc-
Remifentanil is frequently used in neuroanesthesia cinylcholine. The increases in ICP with succinylcholine
practice. It is a potent opioid with rapid blood-brain are transient and less profound than those produced by
equilibration and fast metabolism and offset. In humans laryngoscopy. If a patient with intracranial hypertension
with intracranial pathology, the use of remifentanil during presents with a full stomach or a difficult airway, the use
craniotomies or as an analgesic in the ICU setting of succinylcholine is justified. Rocuronium may be a vi-
has not been associated with significant changes in able alternative to succinylcholine when rapid intubating
ICP.86,87 Remifentanil does not alter the rate of CSF conditions are required.
formation or resistance to reabsorption of CSF.88 It Vecuronium has no effect on ICP in neurosurgical pa-
is expected, as with the other synthetic opioids, that tients, and rocuronium has similar cerebral and systemic
the significant hypotension observed with remifentanil hemodynamics effects.92 Pancuronium (0.1 mg per kg)
could result in reduction in CPP and ICP elevation in caused no changes in CBF, CMRO2 , ICP, or EEG in
patients with abnormal elastance. Aggressive measures to dogs anesthetized with halothane.93 Whereas pancuro-
correct hypotension associated with remifentanil should nium can elevate heart rate and blood pressure, nar-
be instituted in patients with increased ICP. cotics are commonly used concurrently and appear to
In summary, opioids have a long-standing history of blunt the hemodynamic effects of the nondepolarizing
safety in neurosurgical patients; however, some studies agent.
contend that fentanyl, sufentanil, alfentanil, and likely Atracurium administered in clinically relevant doses
remifentanil may cause ICP elevations. The opioid- (0.5 mg per kg) produced no significant change in ICP
induced increase in ICP may be attributable to one of in patients with intracranial pathology.94 Atracurium-
the following: (i) Use of supraclinical or large bolus induced histamine release is related to high doses
doses; (ii) influence of background anesthetic; (iii) opioid- and rapid administration of the drug. In high doses
induced chest wall rigidity; or (iv) associated hypotension- administered to dogs, EEG arousal was seen but not
reducing CPP. Such increases, if they occur, appear to be accompanied by changes in CMRO2 , CBF, or ICP.93 The
transient and not associated with an adverse outcome. metabolite, laudanosine, can precipitate seizures and may
Narcotics are safe for surgery of space-occupying lesions be directly responsible for cerebral stimulation.
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 319
induction is warranted. Intubation should not be at- intraoperatively: (i) Induce additional hyperventilation, if
tempted until full relaxation occurs to avoid precipitous ischemia is not suspected; (ii) administer furosemide,
rises in ICP associated with coughing during endotracheal if the patient is not hypovolemic; (iii) verify head
tube placement. The duration of laryngoscopy should be position and absence of neck compression; (iv) increase
kept to <15 seconds to minimize the hemodynamic re- head elevation as long as the blood pressure is stable;
sponse to intubation. If a difficult airway is anticipated or (v) maintain CPP >60 mm Hg with intravenous fluids or
extensive facial or cervical injuries preclude the placement vasopressors; (vi) treat hypertension as long as it is not a
of an endotracheal tube under general anesthesia, awake protective mechanism; (vii) drain CSF if ventriculostomy
intubation should be the technique of choice to secure the in place; (viii) administer barbiturates; or (ix) administer
airway. Maneuvers that compromise venous return from hypertonic saline.
the cerebral circulation should be avoided, including final Intraoperative fluid management is directed toward
positioning with the head in extreme flexion or lateral ro- maintenance of euvolemia. Hypovolemia is not recom-
tation, or taping the endotracheal tube circumferentially mended because of the danger of associated hypotension,
around the neck. as hypotension below autoregulatory levels causes a com-
Maintenance of anesthesia in patients with abnormal pensatory vasodilation, which can lead to elevation of
elastance should be performed with the same goals as ICP, and ultimately worsening CPP. Isotonic fluids are
induction. The use of low concentrations of isoflurane the mainstay of fluid replacement. Hypertonic saline con-
or sevoflurane has minimal effects on CBF and is sistently reduces ICP and may be a useful adjunct to
safe in conjunction with simultaneous mild degrees decrease cerebral edema while maintaining hemodynamic
of hyperventilation. A continuous infusion of narcotics stability.101,102 Glucose-containing solutions are avoided,
is preferable during long duration cases to promote especially when a risk of cerebral ischemia exists.
hemodynamic stability, ensure steady-state plasma levels, At the conclusion of surgery, the PACO2 should be nor-
and provide a smooth emergence at the conclusion malized so as to minimize the degree of pneumocephalus,
of surgery. Neuromuscular blockade is maintained to which can be significant after CSF drainage and posterior
prevent movement in the pinion system, which can fossa procedures, and prevent reaccumulation of blood
result in scalp lacerations or cervical spine injuries. that can occur with subdural hematomas. In any case
Ventilation is controlled to maintain normocapnia or mild where ICP elevations are expected to persist or develop
hypocapnia. Hyperventilation has detrimental effects on postoperatively, controlled ventilation may be necessary.
CBF in patients with severe TBI and is contraindicated The emergence period often requires the use of antihyper-
within the first five days post injury and particularly within tensives to control blood pressure elevations during light
the first 24 hours.13 When hyperventilation is required, planes of anesthesia. Nitroglycerin and sodium nitroprus-
it should be initiated only for brief periods to control side are still not recommended after the mass lesion has
intracranial hypertension when all conventional measures been removed because edema may occur postoperatively.
have failed. Frequent arterial blood-gas determinations The anesthesiologist should tailor the anesthetic to pro-
are mandatory to evaluate the end-tidal to arterial gradient vide a rapid, smooth emergence to expedite performance
of CO2 . The addition of positive end-expiratory pressure of the neurologic examination. Extubation is planned at
(PEEP) to maintain oxygenation increases dead space the conclusion of surgery unless controlled ventilation is
ventilation, mandating frequent evaluation of PACO2 . warranted, worsening of head injury is expected, mental
When ICP monitoring is not available or utilized, status was impaired preoperatively, neurologic examina-
MAP should be maintained close to preoperative val- tion is significantly changed, or large doses of barbiturates
ues. Hypotension should always be aggressively treated; have been administered during the case.
marked hypertension is also treated unless it is suspected
that the elevation is a protective measure to maintain
CPP. Hypertension is often associated with placement
of the pinion system and during surgical opening until How Is Intracranial
the bone flap is removed. It can be attenuated or pre- Hypertension Managed
vented by the administration of supplemental narcotics,
barbiturates, inhalational agents, or -blockade agents Medically and Surgically?
(i.e., esmolol).
If an ICP monitor is not in place, the dura and brain The management of intracranial hypertension involves ag-
can be inspected for tenseness when the bone flap is gressive surgical intervention, if warranted, and medical
removed. Although ICP measurement may be zero after therapy in the ICU. Once the primary injury has occurred
the dura is open, the brain can still protrude through (i.e., closed head injury, hematoma), it is important to
the craniotomy site, resulting in ischemia or difficult minimize secondary neuronal injury that is often precipi-
closure. Mannitol, 20% 0.5 to 1.0 g per kg, can be tated by ischemic or TBI. Recent attention and treatment
administered before removal of the bone flap. If the has been directed toward prevention of secondary injury
response to mannitol is not adequate, furosemide may because it appears to play a significant role in the extent
be useful but should only be administered as long as of neurologic injury and ultimate neurologic outcome.
the patient is euvolemic and hemodynamically stable. Most of the therapy used to treat secondary injury is still
If ICP remains elevated despite hyperventilation and experimental, but may prove to be beneficial in improving
mannitol, the following measures should be considered outcome for patients in the future. The primary treatment
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 321
of ICP has evolved with improvements in cerebral moni- return from the cerebral circulation as central venous
toring and a better understanding of the pathophysiology pressures rises. Low levels of PEEP can be used safely as
of head injury. long as PACO2 levels are followed closely. If oxygenation
Cerebral ischemia (CBF <18 mL/100 gm/minute) worsens, requiring increasing PEEP levels, hypoxemia
occurs either globally or regionally in approximately 30% should be aggressively treated by increasing PEEP levels
of head-injured patients when CBF measurements are or pressure control inverse ratio ventilation, regardless of
made within 4 to 6 hours of injury.103 The incidence effects on ICP. It is well known that hypoxemia before
would be higher if CBF measurements were determined initial resuscitation and during hospitalization adversely
earlier after injury. It is speculated that ischemia may affects neurologic outcome.
result from vasospasm that has been demonstrated Mannitol is a hypertonic, hyperosmolar agent that
radiographically or by TCD in as many as 40% of effectively reduces brain water by establishing an osmotic
patients after head injury.103 Within 24 hours of injury, gradient that favors the movement of water from the
a relative or absolute hyperemia occurs in select patients brain interstitium into the vasculature. Mannitol is only
and is often uncoupled from CMRO2 (AVDO2 is low). effective in areas of intact blood-brain barrier, and onset
An uncoupling of CBF and CMRO2 can occur early is within 10 to 20 minutes of administration. Within
after injury; however, it is most pronounced within minutes of administration, ICP rises in response to direct
24 hours of head injury.22 Autoregulation is intact vasodilation, followed by a decrease in ICP and rise in CPP
in 60% to 70% of head-injured patients.104 Loss of that often lasts up to 90 minutes. Mannitol administerd
cerebral autoregulation or cerebrovascular reactivity to in bolus fashion is the mainstay of ICP treatment in head-
CO2 after head injury is associated with poor prognosis. injured patients and therapy is guided by serum osmolality
Because CBF and metabolic status cannot reliably be (315 mOsm per kg or less). The combination of mannitol
predicted after head injury, ICP and additional cerebral and furosemide has an additive effect on ICP reduction
monitoring are vital to guide therapy and perhaps predict and duration of diuresis.107 Mannitol may also have the
outcome. Monitors now available include brain tissue added benefit of free radical-scavenging activity.
O2 monitoring, TCD, CBF monitoring, brain chemistry Aggressive treatment of hypotension is of paramount
(microdialysis), and jugular venous O2 saturation with importance to maintain CPP. If autoregulation is intact,
AVDO2 . A jugular bulb venous oxygen saturation (SjvO2 ) hypotension lowers CPP, causing cerebral vasodilation
<50% for longer than 5 minutes is indicative of ischemia and exacerbation of intracranial hypertension, which in
and warrants treatment, and may be the result of elevated turn lowers CPP, thereby initiating a vicious cycle. In con-
ICP, hypotension, aggressive hyperventilation, hypoxia, or trast, CBF falls dramatically with systemic hypotension in
anemia. Significant reductions in brain tissue O2 tensions the absence of autoregulation. Current guidelines for pa-
indicate a CPP <50 mm Hg.13 tients with TBI are the maintenance of CPP at a minimum
Treatment is generally initiated when ICP is 20 mm Hg of 60 mm Hg using pressors or fluids as long as there is
or greater. At least 50% of patients with severe head in- no evidence of cerebral ischemia.13 If ischemia is present,
jury will develop intracranial hypertension at some point CPP may need to be maintained at a higher level. CPP
during the course of hospitalization.105 Historically, hy- guidelines were revised in 2003 due to the increased risk
perventilation was the mainstay of treatment for lowering of development of acute respiratory distress syndrome
ICP; however, the current trend is to maintain normo- with maintenance of CPP >70 mm Hg in head-injured
capnia or mild hypocapnia (PACO2 in low to mid 30s). patients.13
Hyperventilation acutely lowers CBF by vasoconstriction Head elevation, usually at 30 degrees in a neutral
of the end-arterioles and is effective until CSF pH normal- position has been a standard of care for lowering ICP and
izes within 1 or 2 days after initiation. It is common to improving cerebral venous drainage. Feldman determined
observe a rebound in ICP elevation after discontinuation that head elevation to 30 degrees decreased ICP without a
of hyperventilation that is related to CSF pH changes. significant reduction in CPP.108 ICP monitors are zeroed
Cerebrovascular reactivity to CO2 initially may be atten- at the level of the external auditory canal, and the same
uated after head injury but is restored within the first should be done with arterial pressure monitoring to
24 hours of injury. Hyperventilation can reduce CBF accurately reflect CPP.
even further when cerebral ischemia is present after head Sedation, pain control, and neuromuscular blockade
injury. Unless SjvO2 measurement or other cerebral mon- are utilized as part of the global treatment of ICP. Agitation
itoring devices are available, aggressive hyperventilation or pain may commonly contribute to elevations in ICP and
should be avoided and only instituted if there is no ev- can be treated with sedation or narcotics. Sedation is fre-
idence of ischemia and all other conventional measures quently accomplished with propofol or dexmedetomidine
have been attempted. to ensure rapid emergence for neurologic assessments. If
Concomitant pulmonary injuries (including aspira- neuromuscular blockade is instituted for agitation, the use
tion, contusions, and adult respiratory distress syndrome) of a twitch monitor is mandatory to minimize prolonged
often accompany head injury and result in hypoxemia, blockade and ensure reversibility. Neuromuscular block-
requiring the use of PEEP. When PEEP of 10 cm H2 O or ade should be accompanied by sedation; it should not be
less is utilized with frequent determinations of PACO2 , no instituted until other measures fail because the neurologic
significant elevations in ICP occur.106 When PEEP >10 cm examination, except for pupillary findings, can be lost.
H2 O is employed, ICP elevations result not only from in- The drainage of CSF can improve spatial compen-
creased deadspace ventilation, but obstruction of venous sation and can be highly effective in reducing ICP.
322 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
A ventriculostomy must be utilized to provide both Hyperthermia is deleterious and should be aggressively
ICP monitoring and CSF drainage. In the presence of lowered. Hypothermia can still be used as an ICP-lowering
a space-occupying lesion, lumbar CSF drainage is con- technique.
traindicated. Drainage of CSF is a vital component of ICP Early surgical intervention to evacuate mass lesions
treatment in patients with severe TBI. is mandatory to lower ICP, increase CPP, and improve
Maintenance of euvolemia is advocated for head- neurologic outcome. The presence of a mass lesion
injured patients. Historically, volume restriction was amenable to surgical intervention is the treatment of
employed to reduce brain water but often resulted in choice early in injury and should be done expeditiously
hypotension, especially in patients with concomitant in- to avoid worsening the outcome (i.e., rapidly expanding
juries and bleeding. In the intact brain, the movement of epidural hematoma with impending herniation). Many
fluid across the brain is dictated by plasma osmolality and patients sustaining head injury do not have operable
not plasma oncotic pressure.109 Isotonic crystalloid (i.e., lesions and require medical treatment to control ICP. If
normal saline) is safe for use in head-injured patients. Nor- ICP is uncontrollable with maximum medical therapy,
mal saline use did not promote increases in brain water a decompressive craniectomy may be considered to
in rats with cryogenic lesions even though plasma oncotic lower ICP. Historically, decompressive craniectomy was
pressure was significantly reduced.109 Hypotonic fluids considered an option for treatment only when other
(including lactated Ringer) exacerbate cerebral edema by measures have failed. There is currently ongoing research
lowering plasma osmolality and should not be used in pa- to evaluate the effects of early decompressive craniectomy
tients with intracranial hypertension. Hypertonic saline (perhaps within 24 hours of injury) to lower ICP and
has been shown to consistently lower ICP with reduc- improve outcome.13
tions similar to mannitol, as well as improve regional Barbiturates are often instituted when conventional
cerebral blood flow (rCBF) and cardiovascular status dur- measures to lower ICP have failed. Barbiturates are highly
ing early resuscitation without the use of large volumes effective in reducing ICP by lowering CMRO2 and CBF but
of fluid. Hypertonic saline ideally can be used in pa- can be associated with significant systemic hypotension.
tients with severe head injury and hemorrhage to reduce Pentobarbital is the most common barbiturate utilized
ICP and stabilize hemodynamics. Adverse effects from to induce barbiturate coma in the ICU. Barbiturates are
hypertonic saline include significant hypernatremia and titrated to 90% suppression of total EEG activity in an
hyperchloremic metabolic acidosis. The rapid elevation of attempt to control the amount of drug administered;
serum sodium from previously normal levels has not been CMRO2 reduction from 90% EEG suppression to iso-
associated with pontine myelinolysis. electricity is not clinically significant. Vasopressors are
Hyperglycemia should be aggressively treated when often needed to maintain adequate CPP. Once barbiturates
the risk of cerebral ischemia is present. Elevated are administered, neurologic assessments are impossible,
plasma glucose levels provide a substrate for anaerobic and monitoring modalities such as somatosensory-evoked
metabolism during periods of ischemia, which elevate potentials may be difficult to interpret. In focal cerebral is-
lactate levels and worsen intracellular acidosis leading to chemia, barbiturates convey cerebral protection by reduc-
neuronal injury. Lam et al. documented that head-injured ing CMRO2 and probably by attenuating secondary injury;
patients with a GCS of 8 or less had a worse neurologic and although they are effective in lowering ICP, they do
outcome if postoperative glucose levels exceeded 200 mg not appear to improve outcome in head-injured patients.
per dL.110 Higher glucose levels were also recorded on Steroids have a beneficial role in reducing edema
admission and postoperatively in patients who died or associated with brain tumors and are a mainstay of
remained in a persistent vegetative state. A better out- treatment. Steroids do not, however, improve outcome
come was reported for those patients with glucose levels or lower ICP in head-injured patients. Two independent
<150 mg per dL either on admission or postoperatively. studies have validated the failure of dexamethasone to al-
Serum glucose levels are now maintained at 120 mg per ter the neurologic outcome of head-injured patients.112,113
dL or less with an insulin infusion in the ICU setting. If Steroids are associated with significant complications in-
hypoglycemia occurs, it should be aggressively corrected cluding hyperglycemia, infections, and gastrointestinal
because it has even more detrimental effects on the brain bleeding; they should only be utilized to manage edema
than mild hyperglycemia. formation of a space-occupying lesion.
Treatment of head-injured patients with hypothermia Treatment of head-injured patients in the future
is controversial and currently not recommended. How- promises to focus on the cascade of events that contribute
ever, in a systematic review of 12 therapeutic hypothermia to secondary neuronal injury. Most agents that would
trials, there was a 19% reduction in risk of death and a be used to treat secondary neuronal injury are still
22% reduction in risk of poor neurologic outcome in head- experimental and await validation in human trials.
injured patients treated with mild-moderate hypothermia
compared with the normothermic patients.111 Hypother-
mia has the advantages of lowering CMRO2 , CBF, and OUTCOME
ICP, as well as inhibiting excitatory amino acid release
and providing membrane stability. From data that is The outcome of patients with intracranial hypertension
currently available, it is recommended that head-injured depends largely upon the underlying pathologic process
patients arriving hypothermic to the hospital should not and its chronicity. With operable hematomas, prognosis
be rewarmed unless they are profoundly hypothermic. may be favorable if evacuation is expedited. The prognosis
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 323
of patients with brain tumors depends upon the pathology 6. Nitrous oxide is a potent cerebral vasodilator.
of the lesion. The outcome of head-injured patients is 7. Desflurane may increase ICP to a greater extent than
predicted by the following: sevoflurane or isoflurane.
Age of the patient 8. CPP should be maintained at a minimum of
Intracranial pathology 60 mm Hg in patients with severe TBI as long as
Initial GCS there is no evidence of ischemia.
The number of episodes of hypoxemia and hypotension 9. Hyperventilation is contraindicated in head-injured
Degree and duration of ICP elevations patients unless required emergently to lower ICP.
CMRO2 10. Induction of hypothermia is currently not indicated
Cerebrovascular response to CO2 for neuroprotection in patients with head injury or
Presence of ischemia patients at risk for cerebral ischemia.
19. Lassen NA, Feinberg I, Lane MH. Bilateral studies of 40. Cucchiara RF, Theye RA, Michenfelder JD. The effects of
cerebral oxygen uptake in young and aged normal subjects isoflurane on canine cerebral metabolism and blood flow.
and in patients with organic dementia. J Clin Invest. Anesthesiology. 1974;40:571.
1960;39:491. 41. Lam AM, Mayberg TS, Eng CC, et al. Nitrous oxide-
20. Roland PE, Larsen B, Lassen NA, et al. Supplementary isoflurane anesthesia causes more cerebral vasodilation
motor area and other cortical areas in organization of than an equipotent dose of isoflurane in humans. Anesth
voluntary movements in man. J Neurophysiol. 1980;43:118. Analg. 1994;78:462.
21. Muizelaar JP, Schroder ML. Overview of monitoring of 42. Adams RW, Cucchiara RF, Gronert GA, et al. Isoflurane
cerebral blood flow and metabolism after severe head and cerebrospinal fluid pressure in neurosurgical patients.
injury. Can J Neurol Sci. 1994;21:S6. Anesthesiology. 1981;54:97.
22. Muizelaar JP, Marmarou A, DeSalles AAF, et al. Cerebral 43. Artru AA. Effects of enflurane and isoflurane on resistance
blood flow and metabolism in severely head-injured chil- to reabsorption of cerebrospinal fluid in dogs. Anesthesiol-
dren. J Neurosurg. 1989;71:63. ogy. 1984;61:529.
23. Risberg J, Ancri D, Ingvar DH. Correlation between cerebral 44. Hoffman WE, Edelman G, Kochs E, et al. Cerebral
blood volume and cerebral blood flow in the cat. Exp Brain autoregulation in awake versus isoflurane-anesthetized
Res. 1969;8:321. rats. Anesth Analg. 1991;73:753.
24. Grubb RL, Raichle ME, Eichling JO, et al. The effects of 45. Scheller MS, Tateishi A, Drummond JC, et al. The effects
changes in PaCO2 on cerebral blood volume, blood flow, of sevoflurane on cerebral blood flow, cerebral metabolic
and vascular mean transit time. Stroke. 1974;5:630. rate for oxygen, intracranial pressure, and the electroen-
25. Severinghaus JW, Chiodi H, Eger EI II, et al. Cerebral blood cephalogram are similar to those of isoflurane in the rabbit.
flow in man at high altitude. Circ Res. 1966;19:274. Anesthesiology. 1988;68:548.
26. Shapiro HM. Intracranial hypertension: Therapeutic and 46. Scheller MS, Nakakimura K, Fleischer JE, et al. Cerebral ef-
anesthetic considerations (Review). Anesthesiology. 1975; fects of sevoflurane in the dog: Comparison with isoflurane
43:445. and enflurane. Br J Anaesth. 1990;65:388.
27. Ropper AH. Lateral displacement of the brain and level of 47. Petersen KD, Landsfeldt U, Cold GE, et al. Intracranial
consciousness in patients with an acute hemispheral mass. pressure and cerebral hemodynamic in patients with cere-
N Engl J Med. 1986;314:953. bral tumors: A randomized prospective study of patients
28. Eisenberg HM, Gary HE, Aldrich EF, et al. Initial CT find- subjected to craniotomy in propofol-fentanyl, isoflurane-
ings in 753 patients with severe head injury. J Neurosurg. fentanyl, or sevoflurane-fentanyl anesthesia. Anesthesiology.
1990;73:688. 2003;98:329.
29. Lundberg N. Continuous recording and control of ventric- 48. Iacopino DG, Conti A, Battaglia C, et al. Transcranial
ular fluid pressure in neurosurgical practice. Acta Psychiatr doppler ultrasound study of the effects of nitrous oxide on
Neurol Scand Suppl. 1960;149:1. cerebral autoregulation during neurosurgical anesthesia:
30. Rosenwasser RH, Kleiner LI, Krzeminski JP, et al. Intracra- A randomized controlled trial. J Neurosurg. 2003;99:58.
nial pressure monitoring in the posterior fossa: A prelimi- 49. Muzzi DA, Losasso TJ, Dietz NM, et al. The effect of
nary report. J Neurosurg. 1989;71:503. desflurane and isoflurane on cerebrospinal fluid pressure in
31. Mollman HD, Rockswold GL, Ford SE. A clinical com- humans with supratentorial mass lesions. Anesthesiology.
parison of subarachnoid catheters to ventriculostomy and 1992;76:720.
subarachnoid bolts: A prospective study. J Neurosurg. 1988; 50. Lutz LJ, Milde JH, Milde LN. The cerebral functional,
68:737. metabolic, and hemodynamic effects of desflurane in dogs.
32. Chambers IR, Mendelow AD, Sinar EJ, et al. A clinical Anesthesiology. 1990;73:125.
evaluation of the camino subdural screw and ventricular 51. Holmstrom A, Rosen I, Akeson J. Desflurane results in
monitoring kits. Neurosurgery. 1990;26:421. higher cerebral blood flow than sevoflurane or isoflurane at
33. Brean A, Eide PK, Stubhaug A. Comparison of intracra- hypocapnia in pigs. Acta Anaesthesiol Scand. 2004;48:400.
nial pressure measured simultaneously within the brain 52. Lutz LJ, Milde JH, Milde LN. The response of the canine
parenchyma and cerebral ventricles. J Clin Monit Comput. cerebral circulation to hyperventilation during anesthesia
2006;20:411. with desflurane. Anesthesiology. 1991;74:504.
34. Feri M, Ralli L, Felici M, et al. Transcranial doppler 53. Algotsson L, Messetter K, Rosen I, et al. Effects of nitrous
and brain death diagnosis. Crit Care Med. 1994;22: oxide on cerebral haemodynamics and metabolism during
1120. isoflurane anaesthesia in man. Acta Anaesthesiol Scand.
35. Goodwin SR, Friedman WA, Bellefleur M. Is it time to use 1992;36:46.
evoked potentials to predict outcome in comatose children 54. Michenfelder JD. The interdependency of cerebral func-
and adults? Crit Care Med. 1991;19:518. tional and metabolic effects following massive doses of
36. Adams RW, Gronert GA, Sundt TM, et al. Halothane, thiopental in the dog. Anesthesiology. 1974;41:231.
hypocapnia, and cerebrospinal fluid pressure in neuro- 55. Shapiro HM, Galindo A, Wyte SR, et al. Rapid intraopera-
surgery. Anesthesiology. 1972;37:510. tive reduction of intracranial pressure with thiopentone. Br
37. Drummond JC, Todd MM, Toutant SM, et al. Brain surface J Anaesth. 1973;45:1057.
protrusion during enflurane, halothane, and isoflurane 56. Drummond JC, Cole DJ, Patel PM, et al. Focal cerebral
anesthesia in cats. Anesthesiology. 1983;59:2883. ischemia during anesthesia with etomidate, isoflurane, or
38. Lee JG, Hudetz AG, Smith JJ, et al. The effects of halothane thiopental: A comparison of the extent of cerebral injury.
and isoflurane on cerebrocortical microcirculation and Neurosurgery. 1995;37:742.
autoregulation as assessed by laser-doppler flowmetry. 57. Renou AM, Vernhiet J, Macrez P, et al. Cerebral blood flow
Anesth Analg. 1994;79:58. and metabolism during etomidate anaesthesia in man. Br J
39. Newberg LA, Milde JH, Michenfelder JD. The cerebral Anaesth. 1978;50:1047.
metabolic effects of isoflurane at and above concentrations 58. Prior JGL, Hinds CJ, Williams J, et al. The use of etomidate
that suppress cortical electrical activity. Anesthesiology. in the management of severe head injury. Intensive Care
1983;59:23. Med. 1983;9:313.
CHAPTER 22/INCREASED INTRACRANIAL PRESSURE 325
59. Pinaud M, Lelausque JN, Chetanneau A, et al. Effects of velocity, and metabolism in dogs. Anesth Analg. 1991;72:
propofol on cerebral hemodynamics and metabolism in 177.
patients with brain trauma. Anesthesiology. 1990;73:404. 80. Milde LN, Milde JH, Gallagher WJ. Effects of sufentanil on
60. Girard F, Moumdjian R, Boudreault D, et al. The effect of cerebral circulation and metabolism in dogs. Anesth Analg.
propofol sedation on the intracranial pressure of patients 1990;70:138.
with an intracranial space-occupying lesion. Anesth Analg. 81. Mayer N, Weinstabl C, Podreka I, et al. Sufentanil does not
2004;99:573. increase cerebral blood flow in healthy human volunteers.
61. Ravussin P, Thorin D, Guinard JP, et al. Effect of propofol Anesthesiology. 1990;73:240.
on cerebrospinal fluid pressure in patients with and without 82. Herrick IA, Gelb AW, Manninen PH, et al. Effects of
intracranial hypertension. Anesthesiology. 1989;71:A120. fentanyl, sufentanil, and alfentanil on brain retractor
62. Steiner LA, Johnston AJ, Chatfield DA, et al. The effects of pressure. Anesth Analg. 1991;72:359.
large-dose propofol on cerebrovascular pressure autoregu- 83. Sperry RJ, Bailey PL, Reichman MV, et al. Fentanyl and
lation in head-injured patients. Anesth Analg. 2003;97:572. sufentanil increase intracranial pressure in head trauma
63. Karsli C, Luginbuehl I, Bissonnette B. The cerebrovascular patients. Anesthesiology. 1992;77:416.
response to hypocapnia in children receiving propofol. 84. Albanese J, Durbec O, Viviand X, et al. Sufentanil increases
Anesth Analg. 2004;99:1049. intracranial pressure in patients with head trauma. Anes-
64. Yano M, Nishiyama H, Yokota H, et al. Effect of lidocaine thesiology. 1993;79:493493.
on ICP response to endotracheal suctioning. Anesthesiology. 85. Mayberg TS, Lam AM, Eng CC, et al. The effect of alfentanil
1986;64:651653. on cerebral blood flow velocity and intracranial pressure
65. Dawson B, Michenfelder JD, Theye RA. Effects of ketamine during isoflurane-nitrous oxide anesthesia in humans.
on canine cerebral blood flow and metabolism: Modifica- Anesthesiology. 1993;78:288.
tion by prior administration of thiopental. Anesth Analg. 86. Engelhard K, Reeker W, Kochs E, et al. Effect of remifen-
1971;50:443. tanil on intracranial pressure and cerebral blood flow
66. Takeshita H, Okuda Y, Sari A. The effects of ketamine on velocity in patients with head trauma. Acta Anaesthesiol
cerebral circulation and metabolism in man. Anesthesiology. Scand. 2004;48:396.
1972;36:69. 87. Guy J, Hindman BJ, Baker KZ, et al. Comparison of
67. Hougaard K, Hansen A, Brodersen P. The effect of ketamine remifentanil and fentanyl in patients undergoing cran-
on regional cerebral blood flow in man. Anesthesiology. iotomy for supratentorial space-occupying lesions. Anes-
1974;41:562. thesiology. 1997;86:514.
68. Sari A, Okuda Y, Takeshita H. The effect of ketamine on 88. Artru AA, Momota T. Rate of CSF formation and resistance
cerebrospinal fluid pressure. Anesth Analg. 1972;51:560. to reabsorption of csf during sevoflurane or remifentanil in
69. Aryan HE, Box KW, Ibrahim D, et al. Safety and efficacy rabbits. J Neurosurg Anesthesiol. 2000;12:37.
of dexmedetomidine in neurosurgical patients. Brain Inj. 89. Lanier WL, Iaizzo PA, Milde JH. Cerebral function and mus-
2006;20:791. cle afferent activity following intravenous succinylcholine
70. Talke P, Tong C, Lee HW, et al. Effect of dexmedetomidine in dogs anesthetized with halothane: The effects of pre-
on lumbar cerebrospinal fluid pressure in humans. Anesth treatment with a defasciculating dose of pancuronium.
Analg. 1997;85:358364. Anesthesiology. 1989;71:87.
71. Werner C. Effects of analgesia and sedation on cere- 90. Lanier WL, Milde JH, Michenfelder JD. Cerebral stimu-
brovascular circulation, cerebral blood volume, cerebral lation following succinylcholine in dogs. Anesthesiology.
metabolism and intracranial pressure. Anaesthetist. 1995; 1986;64:551.
44(Suppl 3):S566. 91. Minton MD, Grosslight K, Stirt JA, et al. Increases in
72. Carlsson C, Hagerdal M, Kaasik AE, et al. The effects of intracranial pressure from succinylcholine: Prevention by
diazepam on cerebral blood flow and oxygen consumption prior nondepolarizing blockade. Anesthesiology. 1986;65:
in rats and its synergistic interaction with nitrous oxide. 165.
Anesthesiology. 1976;45:319. 92. Stirt JA, Maggio W, Haworth C, et al. Vecuronium: Effect on
73. Vernhiet J, Renou AM, Orgogozo JM, et al. Effect of a intracranial pressure and hemodynamics in neurosurgical
diazepam- fentanyl mixture on cerebral blood flow and patients. Anesthesiology. 1987;67:570.
oxygen consumption in man. Br J Anaesth. 1978;50:165. 93. Lanier WL, Milde JH, Michenfelder JD. The cerebral effects
74. Papazian L, Albanese J, Thirion X, et al. Effect of bolus of pancuronium and atracurium in halothane-anesthetized
doses of midazolam on intracranial pressure and cerebral dogs. Anesthesiology. 1985;63:589.
perfusion pressure in patients with severe head injury. Br J 94. Rosa G, Orfei P, Sanfilippo M, et al. The effects of
Anaesth. 1993;71:267. atracurium besylate (tracrium) on intracranial pressure and
75. Takeshita H, Michenfelder JD, Theye RA. The effects cerebral perfusion pressure. Anesth Analg. 1986;65:381.
of morphine and n-allylnormorphine on canine cerebral 95. Cottrell JE, Gupta B, Rappaport H, et al. Intracranial
metabolism and circulation. Anesthesiology. 1972;37:605. pressure during nitroglycerin-induced hypotension. J Neu-
76. Marx W, Shah N, Long C, et al. Sufentanil, alfentanil, and rosurg. 1980;53:309.
fentanyl: Impact on cerebrospinal fluid pressure in patients 96. Rogers MC, Hamburger C, Owen K, et al. Intracranial pres-
with brain tumors. J Neurosurg Anesthesiol. 1989;1:3. sure in the cat during nitroglycerin-induced hypotension.
77. From RP, Warner DS, Todd MM, et al. Anesthesia for cran- Anesthesiology. 1979;51:227.
iotomy: A double-blind comparison of alfentanil, fentanyl, 97. Dohi S, Matsumoto M, Takahashi T. The effects of
and sufentanil. Anesthesiology. 1990;73:896. nitroglycerin on cerebrospinal fluid pressure in awake
78. Artru AA. Dose-related changes in the rate of CSF for- and anesthestized humans. Anesthesiology. 1981;54:511
mation and resistance to reabsorption of CSF during 511.
administration of fentanyl, sufentanil, or alfentanil in dogs. 98. Marsh ML, Shapiro HM, Smith RW, et al. Changes in
J Neurosurg Anesthesiol. 1991;3:283. neurologic status and intracranial pressure associated with
79. Werner C, Hoffman WE, Baughman VL, et al. Effects sodium nitroprusside administration. Anesthesiology. 1979;
of sufentanil on cerebral blood flow, cerebral blood flow 51:336.
326 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
99. Michenfelder JD, Milde JH. The interaction of sodium 107. Pollay M, Fullenwider C, Roberts PA, et al. Effect of
nitroprusside, hypotension, and isoflurane in determining mannitol and furosemide on blood-brain osmotic gradient
cerebral vasculature effects. Anesthesiology. 1988;69:870. and intracranial pressure. J Neurosurg. 1983;59:945.
100. Giffin JP, Cottrell JE, Hartung J, et al. Intracranial pressure 108. Feldman Z, Kanter MJ, Robertson CS, et al. Effect of
during nifedipine- induced hypotension. Anesth Analg. head elevation on intracranial pressure, cerebral perfusion
1983;62:1078. pressure, and cerebral blood flow in head-injured patients.
101. Prough DS, Whitley JM, Taylor CL, et al. Regional cerebral J Neurosurg. 1992;76:207.
blood flow following resuscitation from hemorrhagic shock 109. Zornow MH, Prough DS. Fluid management in patients
with hypertonic saline. Anesthesiology. 1991;75:319. with traumatic brain injury. New Horiz. 1995;3:488.
102. Freshman SP, Battistella FD, Matteucci M, et al. Hypertonic 110. Lam AM, Winn HR, Cullen BF, et al. Hyperglycemia
saline (7.5%) versus mannitol: A comparison for treatment and neurological outcome in patients with head injury.
of acute head injuries. J Trauma. 1993;35:344. J Neurosurg. 1991;75:545.
103. Bouma GJ, Muizelaar P, Choi SC, et al. Cerebral circulation 111. McIntyre LA, Fergusson DA, Hebert PC, et al. Prolonged
and metabolism after severe traumatic brain injury: The therapeutic hypothermia after traumatic brain injury in
elusive role of ischemia. J Neurosurg. 1991;75:685. adults. A systematic review. JAMA. 2003;289:2992.
104. Bouma GJ, Muizelaar JP, Bandoh K, et al. Blood pressure 112. Dearden NM, Gibson JS, McDowall DG, et al. Effect of
and intracranial pressure-volume dynamics in severe head high-dose dexamethasone on outcome from severe head
injury. Relationship with cerebral blood flow. J Neurosurg. injury. J Neurosurg. 1986;64:81.
1992;77:15. 113. Cooper PR, Moody S, Clark WK, et al. Dexamethasone
105. Chan KH, Dearden NM, Miller JD, et al. Multimodality mon- and severe head injury: A prospective double-blind study.
itoring as a guide to treatment of intracranial hypertension J Neurosurg. 1979;51:307.
after severe brain injury. Neurosurgery. 1993;32:547. 114. Aldrich EF, Eisenberg HM, Saydjari C, et al. Diffuse brain
106. Cooper KR, Boswell PA, Choi SC. Safe use of PEEP in swelling in severely head-injured children: A report from the
patients with severe head injury. J Neurosurg. 1985;63:552. NIH traumatic coma data bank. J Neurosurg. 1992;76:450.
CHAPTER SEIZURES
23 Stephen Luney
A
about unconscious in the street and brought intensity of the discharges. Normally, the clinical mani-
by an EMS team to the hospital emergency festation of these events starts suddenly and does not last
room. He is in a disheveled state, smells of long.1
alcohol, and has been incontinent of urine.
The staff recognize him as a frequent patient
of the department, a homeless person who has been seen
for mental health and alcohol-related issues. How Does a Seizure Present?
On arrival in the department, he has eye opening to
speech and is confused but obeys commands. The initial
observations show him to have a core temperature of 38 C, Seizures are usually classified first into partial and
heart rate 110 bpm, blood pressure 110/55 mm Hg, and generalized, and then each of these is classified further
oxygen saturation of 92%. While lying semirecumbent on a essentially on the basis of the clinical manifestation of the
gurney and before blood samples could be taken, he begins seizure (see Table 23.2).
having myoclonic twitching. Immediately thereafter, he A tonic-clonic, or grand mal seizure is the classic
proceeds to have a tonic-clonic, grand mal seizure lasting form of fit or convulsion that most people imagine when
a total of 10 minutes. During this time, the back of the they think of a seizure. Sometimes such a seizure can be
gurney was lowered and his head was cradled to protect preceded by a prodromal period during which the patient
him from injury. may anticipate its onset. It may be that there is an increase
Oxygen is administered through a face mask at 12 L in myoclonic jerking. If the patient gets a sensation that
per minute, an intravenous catheter is inserted, and an a seizure is imminenttermed an auratechnically, it
arterial blood gas sample is taken. Venous blood samples means that the seizure is generalized secondarily. The
are taken and further examination is performed, which patient then becomes unconscious and may cry out and
show him to have a core temperature of 38 C, heart rate fall if standing at the time. For a short period of time,
130 bpm, blood pressure 90/50 mm Hg, and an oxygen the patient will exhibit tonic flexion, followed by a more
saturation of 97%. Fluids and a slow intravenous injec- prolonged period where they become rigid, and extend
tion of 4 mg of lorazepam are initiated. Arrangements are axially, with their jaw clamped shut and their eyes rolled
made to admit him to the intensive care unit for further up. Their limbs become stiff and adopt a position of
management. adduction and extension, with their fists clenched. During
this period, they are apneic and often become cyanotic.
This phase lasts up to 30 seconds and leads into the
clonic phase. This usually affects all the limbs, jaw,
and facial muscles. There can be excessive salivation
What Is a Seizure? and partially obstructed respiration, possibly complicated
by bleeding from the tongue or lips if they have bitten
By definition, a seizure is the transient clinical mani- themselves.
festations that result from an episode of epileptic neu- As the seizure continues, the convulsive movements
ronal activity. The etiology of seizures is vast (see become less frequent and may settle down to around 4 Hz,
Table 23.1). One of the characteristic findings is abnormal but become greater in their excursions. Concurrent with
synchronization of this activity, with either inadequate these motor manifestations are autonomic phenomena,
327
328 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 23.1 Etiology of Seizures TABLE 23.2 International League Against Epilepsy (ILAE)
Classification of Seizures
1. Metabolic
a. Congenital 1. Generalized (convulsive and nonconvulsive)
b. Acquired a. Absence seizures
(1) Hypoglycemia (1) Absence seizures
(2) Hyperglycemia (2) Atypical absence seizures
(3) Hyponatremia b. Myoclonic seizures
(4) Hypoxia c. Clonic seizures
(5) Hypocalcemia d. Tonic seizures
(6) Uremia e. Tonic-clonic seizures
(7) Toxins f. Atonic seizures
(8) Drugs (either withdrawal or intoxication) 2. Partial (local, focal seizures)
2. Infection a. Simple partial seizures
a. Systemic (febrile convulsion/hyperthermia) (1) With motor signs
b. Intracranial (2) With somatosensory or special sensory signs
(1) Meningitis (3) With autonomic symptoms or signs
(2) Encephalitis (4) With psychic symptoms
(3) Abscess b. Complex partial seizures
3. Structural (1) Simple partial onset followed by impaired
a. Gliotic scarring consciousness
(1) Temporal lobe sclerosis (2) With impaired consciousness from the outset
(2) Posttraumatic c. Partial seizures evolving to secondary generalized
(3) Post infection seizures (tonic-clonic, tonic or clonic)
b. Congenital malformations (1) Simple partial seizures evolving to generalized
c. Vascular malformations seizures
d. Tumor (2) Complex partial seizures evolving to generalized
seizures
(3) Simple partial seizures evolving to complex
including changes in heart rate and blood pressure partial seizures then evolving to generalized
and cutaneous flushing which last 30 to 60 seconds. seizures
These manifestations conclude with a short-lived tonic 3. Unclassified epileptic seizures
contraction of all the muscle groups, during which the
patient may become incontinent. In the final phase (lasting Data taken from: Shorvon SD. Handbook of epilepsy treatment.
from a few minutes to half hour), there is generalized Oxford: Blackwell Publishing; 2005.
muscle flaccidity.
Consciousness usually returns gradually, although
the patient is invariably confused postictally. Oftentimes, First of all, clinical signs may be helpful. Evidence of
patients will complain of severe headache and usually feel fluctuations of pupillary size (hippus), nystagmus, or
dreadful and extremely fatigued. They may also go into stereotypic cyclic motor manifestations greatly aids in
a deep sleep (but not unconsciousness), from which they confirming the diagnosis, as these are clinical signs that
awake later with generalized muscle aches and pains, and cannot be manufactured. Conversely, clinical signs that
frequently report a persisting headache. should arouse suspicion of (but are not pathognomonic
of) pseudoseizure include the following:
The patient squeezing the eyes shut or resisting them
being opened
How Do I Know It Is Really Rolling the head
a Seizure? Arching the back
Thrusting the pelvis
Poorly coordinated thrashing
In the acute setting in the emergency room or the
postanesthesia care unit, one of the questions that runs This is by no means an exhaustive list, but includes
through ones mind is Is this really a seizure or is this some of the more common features of pseudoseizure.
guy faking it? This does not seem like an unreasonable Another reason to suspect pseudoseizure is if the patient
question when you consider that, in an audit of patients does not respond promptly to initial treatment of the
referred to a specialized neurology center as status epilep- seizure, as the vast majority with a seizure do respond.
ticus, around half were either in a drug-induced coma or Nonconvulsive seizure is a much more difficult diagnosis
in pseudostatus.2 to make, and in the presence of coma can only be diag-
There are a number of things that are helpful in dif- nosed by electroencephalogram (EEG), which should be
ferentiating between pseudoseizure and a valid seizure.3 sought relatively early in the investigation stage.
CHAPTER 23/SEIZURES 329
TABLE 23.4 Weinberg Dose Regimen for Use TABLE 23.5 Hyponatremia Scenarios
in Humans
Hypovolemic
1. Administer 1 mL/kg Intralipida 20% over 1 min Gastrointestinal losses
2. Repeat twice more at 35 min Vomiting
3. Convert to an infusion at a rate of 0.25 mL/kg/min Diarrhea
continuing until hemodynamic stability restored Skin losses
4. Increasing the dose beyond 8 mL/kg is unlikely to be Third space losses
useful Renal losses
5. In practice, when resuscitating a 70-kg adult: Diuretics
a. Administer 500 mL Intralipid 20% in a 50-mL syringe Renal damage
b. Draw 50 mL and give it stat; then draw another Urinary tract obstruction
syringe and give a further 20 mL Adrenal insufficiency
c. Repeat and give another 20 mL of Intralipid Euvolemic
d. Then attach the 500 mL Intralipid bag and run it Syndrome of inappropriate secretion of antidiuretic
intravenously over the following 15 min hormone
Renal failure
a Intralipid
(Fresenius Kabi AG, Bad Homburg v.d.H. Germany). Water intoxication
Data taken from: Weinberg G. Lipid emulsion infusion rescues Hypokalemia
dogs from bupivacaine-induced toxicity (letter). Reg Anesth Pain Dysfunctional osmostat
Med. 2004;29:74. Hypervolemic
Congestive heart failure
of cardiovascular toxicity, especially bupivacaine, the in- Nephrosis
travenous injection of lipid may act as a circulating lipid Liver dysfunction
sink and will draw local anesthetic out of plasma7,8 (see Water intoxication
Table 23.4). Although this proposition remains contro-
versial, it is tempting and worthwhile to consider in the
presence of cardiovascular collapse presumed secondary One very special circumstance in which seizures may
to local anesthetic toxicity. Some suggest keeping In- occur during neurosurgery is during awake craniotomy
tralipid around for just that possibility and application.7 for the resection of dominant hemisphere seizure foci. In
this procedure, electric stimulation of the cortex is per-
formed to map out both the motor strip and speech center
so they may be preserved. The repeated stimulation of the
What Type of Surgery Places cortex risks evoking a seizure, which may consequently
necessitate proceeding to general anesthesia.
the Patient at Risk for Seizures? Neurosurgery is, however, not the only type of surgery
in which seizures occur. Other types of surgery may cause
seizures by significantly altering the patients biochem-
NEUROSURGERY istry. Of these derangements, hyponatremia is the most
common. A wide range of surgical patients may be hypona-
The surgical setting most associated with postopera- tremic and at risk of seizures, but what complicates this
tive seizures is intracranial supratentorial neurosurgery. issue further is that the hyponatremia may be in a setting of
Setting aside neurosurgery for trauma (which we will ex- hypovolemia, hypervolemia, or euvolemia (see Table 23.5).
amine under the intensive care section), large studies have Urologic procedures, such as transurethral resection of the
shown an incidence of approximately 15% for postoper- prostate, can produce dilutional hyponatremia secondary
ative seizures.1 This varies widely depending on the type to absorption of the glycine used for bladder irrigation.
of surgery, and ranges from as high as 92% for patients Typically, such seizures occur if the serum sodium falls
having surgery to drain cerebral abscesses, to approxi- below 115 mmol per L. In this setting under regional
mately 20% for glioma surgery or meningioma removal, anesthesia, premonitory symptoms such as apprehension,
and down to <5% for stereotactic procedures or ventric- confusion, and headache can alert the clinicians to evolv-
ular drainage. However, if the patient has had seizures ing hyponatremia.
preoperatively, the risk for all groups is higher. In terms
of timing, the seizures are likely to occur in 37% within
the first postoperative week. Over the ensuing year, 77%
will develop seizures, with the figure rising to 92% by the CAROTID AND CARDIAC
end of the second year.1 If the seizures occur within the SURGERY
first week, the likelihood of recurrence is approximately
40%. A small note of encouragement can be sounded for Carotid endarterectomy is a particular surgery where
patients with meningiomas: The removal of their tumor patients are at increased risk of seizures for a variety
may result in cessation of seizures in 30% to 60% of cases.9 of reasons. They may have plaque or thrombi embolize
CHAPTER 23/SEIZURES 331
to their brain and cause them to seizure; they may The statistics worsen when we consider open head
have inadequate cerebral perfusion following either the injury. In these patients, if early seizures have been a
manipulation or clamping of their carotid artery; or they feature, the risk of subsequently developing epilepsy is
may have air enter their cerebral circulation. Similarly, approximately 25%, compared with only 3% if there were
patients undergoing cardiac surgery may also suffer no early seizures. If there is a combination of either de-
embolic phenomena resulting in postoperative seizures, pressed skull fracture with tear of the dura, an intracranial
be it from microbubbles from within the heart or hematoma, or a period of posttraumatic amnesia, the risk
air entering through the bypass circuit. Indeed with of seizures progressively rises, surpassing 50% if all three
the advent of an ever-growing range of interventional combine. Sadly, posttraumatic seizures are difficult to
endovascular procedures, there is now a huge variety treat, with some series suggesting that half of the patients
of procedures that all run the risk of the materials followed up still have seizures more than a decade later.10
(stents, coils, glue, drugs etc.) embolizing to the cerebral
vasculature and resulting in seizures.
Hypoglycemia and hyperglycemia to recreational and illicit drugs and toxins in patients who
Hypomagnesemia present to the emergency room.
As discussed in the preceding text, in performing
Similarly renal and hepatic failure and encephalo-
some of our local anesthesia techniques, we risk evoking
pathies may be associated with seizures, as can derange-
seizures due to CNS toxicity. In addition to those drugs, a
ments of thyroid function, including hypothyroidism and
wide range of other medications may also be responsible
Hashimoto thyroiditis. (see Table 23.6). It may be that the drugs themselves are,
in fact, proconvulsant: They may interfere with existing
antiepileptic medications or they may elicit seizures in
DRUGS AND TOXINS the presence of renal or hepatic failure. Alternatively, they
may produce seizures due to their withdrawal, or patients
It is helpful to consider the fourth category in the
may exhibit direct cerebral toxicity when they take these
context of prescription drugs, but also later in relation drugs in excessive amounts (overdose). Of particular im-
portance is that most psychotropic medications are at risk
TABLE 23.6 Drugs That May Cause Seizuresa of causing seizures, not the least by overdose. Most an-
tiseizure medications can produce seizures after sudden
Drug Comment withdrawal.
Phenothiazines
Tricyclics
Selective serotonin
reuptake inhibitors
Monoamine oxidase
Are There Other Situations in
inhibitors an Anesthesia Context Where
Meperidine Especially in the presence
of renal impairment or
I Might Encounter Seizures?
monoamine oxidase
inhibitor In the obstetric setting, seizures most commonly develop
Narcotics In withdrawal in the presence of preeclampsia. (see Chapter 49).
-lactam antibiotics Owing to -aminobutyric
acid (GABA) antagonism
Isoniazid By antagonizing pyridoxal EMERGENCY ROOM
phosphate
In addition to patients who are known to have epilepsy,
Aminoglycosides,
along with the possibilities discussed in the previ-
metronidazole,
ous sections, there are a number of important likely
quinolones
scenarios:
Zidovudine
IV contrast media Alcohol-related seizures
Chemotherapeutic agents Recreational drugs
Theophylline Reflex epilepsies
-Blockers Cardiovascular events
Nonsteroidal
anti-inflammatory drugs
Antiarrhythmic agents ALCOHOL
Cimetidine
Local anesthetics Perhaps 15% of patients in the United States of America
Alcohol with epilepsy have alcoholism, and the greater the
Cocaine alcohol consumption, the greater the risk of seizures.
Phencyclidine Seizures may be seen at several different points in alcohol
Insulin Hypoglycemia consumption, and can take place both in withdrawal from
Levodopa alcohol (typically within 2448 hours) and in association
Thiazides with binge drinking. The etiology of seizures in association
Benzodiazepines In withdrawal or following with binge drinking is complex and may also involve
antagonism with hyponatremia, hypomagnesemia, or hypoglycemia. There
flumazenil is also the likelihood that other comorbidities such as
Salicylates liver failure may contribute to seizures. An important
Carbamazepine In withdrawal factor to keep in mind when treating such patients is
Barbiturates In withdrawal to include the administration of thiamine when treating
their seizures to avoid the development of Wernicke
a Note: this list is indicative, not exhaustive. encephalopathy.
CHAPTER 23/SEIZURES 333
TABLE 23.7 Drug Therapy in Status Epilepticus passage of a weakly acidic drug into the brain, as brain
pH will be higher than blood pH. Also, the permeability
1. Early status (030 min) of the bloodbrain barrier increases during convulsive
a. Lorazepam 4 mg IV or seizures, particularly at seizure foci where blood flow
b. Diazepam 1020 mg (IV or rectally) increases, thus increasing the concentration of drug at
2. Established status (3060/90 min) these sites.
a. Phenytoin 15 mg/kg IV (at maximum of 50 mg/min)
or
b. Fosphenytoin 15 mg PE/kg IV (at maximum of
100 mg PE/min) or Why Is It Important for Me to
c. Valproate 25 mg/kg IV (at 36 mg/min) Treat the Seizure? Will It Not
3. Refractory status (>60/90 min)
a. Induction dose of propofol followed by infusion Just Stop?
designed to effect burst suppression on EEG or
b. Dose of thiopental sufficient to effect seizure If seizures continue to occur despite the initial manage-
control followed by infusion designed to effect burst ment of early status, the stage of established status may
suppression on EEG or then ensue (see Table 23.7) at which point it may be
c. Midazolam 0.10.4 mg/kg IV, followed by infusion necessary to consider securing the patients airway for
sufficient to effect burst suppression on EEG a variety of reasons. The drugs that are given for the
seizure may cause either respiratory depression or loss
PE, phenytoin equivalents; EEG, electroencephalogram, IV, intra- of airway reflexes, and therefore it becomes important to
venously. both secure the airway and support ventilation. Once this
stage is reached, plans should be made to arrange transfer
of the patient to an intensive care unit (if not already
1. If the seizures recur rapidly admitted) for further supportive management. If the stage
2. If the seizure lasts more than 10 minutes of refractory status is reached, further measures may be
3. If the seizure lasts longer than customary for that required in terms of intravenous fluid therapy, inotropic
patient (if such is known) support, EEG monitoring, and so on. This is due in part
If any of these conditions are present, drug therapy to the fact that as the patients status changes from estab-
should be initiated as shown in Table 23.7 (under Early lished to refractory, so do the physiologic changes from
Status). phase I or compensated to phase II or decompensated.
The potency of benzodiazepines decreases as seizures at this point, as this will most likely prevent the
continue because of the acute down-regulation of seizures from developing into status epilepticus. His initial
-aminobutyric acid (GABA-A) receptors.12 Non- management involved protection from injury during the
GABAergic drugs such as phenytoin become even less seizure and supportive measures such as oxygen therapy
effective in the later stages of status epilepticus. For status and intravenous fluids. Given the need for ongoing
epilepticus, the prognosis is not good, with a mortal- support, his transfer to the intensive care unit was
ity of approximately 20%. Worse still, the mortality of appropriate. His further investigations would include
refractory status epilepticus is high (48%), with only blood sampling for glucose, sodium, magnesium, urea
29% of patients returning to their premorbid functional and potassium. His mental health history would suggest
baseline.13 the need to take blood samples for levels of psychotropic
medications. Culture of blood and urine would also be
important, along with consideration of CT scanning of
the brain to exclude intracranial hematoma, tuberculosis,
My Patient Is No Longer or stroke in particular.
SeizingWhat Do I Do? (Is My
Patient an Epileptic Now?)
KEY POINTS
Epilepsy is a disorder of the brain characterized by an 1. If emergency treatment for patients with repetitive
ongoing liability of recurrent epileptic seizures. As a seizures is started early, generally the patient can be
definition, it is not very helpful if your patient has just prevented from proceeding to status epilepticus.
had a couple of seizures due to sepsis or hyponatremia. 2. Status epilepticus can cause devastating cardiovascu-
Being pragmatic in the clinical setting, we can state that a lar and neurologic deterioration, with potential for
liability to further attacks can be considered if two or more extremely poor outcome.
spontaneous attacks have occurred, as it is reasonable to 3. A range of situations may precipitate seizures, not the
assume that more attacks are likely. However, there are least of which are a variety of anesthesia interventions
epileptic seizures that are not considered to warrant a and surgical procedures.
diagnosis of epilepsy. Both childhood febrile seizures and 4. There is a relatively discreet range of blood tests
provoked seizures (i.e., acute symptomatic seizures) are that can rapidly guide diagnosis and direct corrective
such that they do not indicate an ongoing liability to treatment of seizures
recur. Epidemiologists divide epilepsy into the following 5. The management of ongoing seizures requires ad-
four categories: mission to an intensive care unit because both
the condition itself and the therapeutic interven-
1. Acute symptomatic epilepsy (i.e., provoked, where
tions undertaken to treat it may both require res-
there is a cause)
piratory and cardiovascular support in the form
2. Remote symptomatic (a cause has been present for at
of airway protection, artificial ventilation, and
least 3 months)
inotropes.
3. Congenital (the cause existed at birth)
4. Idiopathic (the cause is unknown)
REFERENCES
1. Shorvon SD. Handbook of epilepsy treatment. Oxford: Black-
CASE STUDY INTERPRETATION well Publishing; 2005.
2. Walker MC, Howard RS, Smith SJ, et al. Diagnosis and
The 73-year-old man in the case summary at the start treatment of status epilepticus on a neurological intensive
of the chapter illustrates many of the issues involved. In care unit. QJM. 1996;89:913.
his history, he is known to have issues regarding alcohol 3. Walker M. Status epilepticus: An evidence based guide. BMJ.
2005;331:673.
consumption. He is homeless; this fact, combined with
4. Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison
his alcohol use, calls into question his nutritional state. of lorazepam, diazepam and placebo for the treatment of
Further, it places him at higher risk of tuberculosis. out-of-hospital status epilepticus. N Engl J Med. 2001;345:
His age points to cardiovascular risk factors, and his 631.
mental health history suggests consideration of Alzheimer 5. Morishima HO, Pedersen H, Finster M, et al. Bupivacaine
disease or the use of psychotropic medications. Of note toxicity in pregnant and nonpregnant ewes. Anesthesiology.
in his examination is the presence of fever as a possible 1985;63:134.
marker of infection. The presence of incontinence may 6. Picard J, Meek T. Lipid emulsion for local anaesthetic
indicate a variety of possibilities, including a urinary toxicity. Anaesthesia. 2006;61:107.
7. Moore N, Kirton CB, Ane J. Lipid emulsion to treat overdose
tract infection or simply ictal incontinence. The reported
of local anaesthetic. Anaesthesia. 2006;61:607.
history of writhing about is strongly suggestive of
8. Weinberg G. Lipid emulsion infusion rescues dogs from
seizure, given the seizure that is witnessed in the bupivacaine-induced toxicity (letter). Reg Anesth Pain Med.
department. Consequently, therapy is best instituted 2004;29:74.
336 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
9. Chozick BS, Reinert SE, Greenblatt SH. Incidence of seizures 12. Kapur J, Macdonald RL. Rapid seizure-induced reduction
after surgery for supratentorial meningiomas: A modern of benzodiazepine and Zn2+ sensitivity of hippocampal
analysis. J Neurosurg. 1996;84:382. dentate granule cell GABA-A receptors. J Neurosci. 1997;17:
10. Frey C. Epidemiology of post traumatic epilepsy: A critical 7532.
review. Epilepsia. 2003;44(Suppl 10):11. 13. Kendall JL. Intranasal midazolam in patients with status
11. Brodie MJ, Kwan P. Epilepsy in elderly people. BMJ. 2005; epilepticus. Ann Emerg Med. 1997;29:415.
331:1317.
CHAPTER PERIOPERATIVE STROKE
CASE SUMMARY
What Is a Stroke and a
n 82-year-old man presents for a right
Transient Ischemic Attack, and
A
carotid endarterectomy. During the last
6 weeks, he had two episodes of transient Why Is It Important to Know?
weakness of the left upper extremity that
lasted for approximately 2 and 14 hours, Stroke refers to the sudden onset of a neurologic deficit
respectively. A carotid ultrasound revealed a due to a focal disruption of the cerebral circulation. Stroke
90% stenosis of the right internal carotid artery and an occurs in the setting of either brain ischemia (88% of all
80% stenosis of the left internal carotid artery. Computer strokes) or hemorrhage (12%, see Fig. 24.1). Approxi-
tomographic angiography confirmed these findings and mately 700,000 Americans have a stroke per yearthat is,
did not show significant intracranial or vertebral vascular one person every 45 seconds. In adults, stroke is the third
disease. leading cause of death, behind heart disease and cancer,
His past medical history shows long-standing hyper- and a leading cause of serious, long-term disability.1
tension that is reasonably controlled by medications, A transient ischemic attack (TIA) is a brief episode of
with current blood pressures of 155/92 mm Hg and neurologic dysfunction lasting <24 hours that is caused
163/85 mm Hg in the right and left arms, respectively. by reversible ischemia in a vascular territory. A TIA is a
The patient had an inferior myocardial infarction 7 years medical emergency, because 10% of patients with a TIA
before this presentation, which is evident from electro- will go on to have a stroke within 30 days of the TIA,
cardiogram. He denies current anginal symptoms, but with over half of these strokes occurring within 48 hours.2
his exercise tolerance is limited to approximately four Typical TIAs can present as sudden but transient blindness
metabolic equivalents (METs) because of degenerative in one eye (amaurosis fugax), or as other transient deficits
joint disease. He has a 70-pack-year history of smoking, clearly confined to an arterial territory. As TIAs have often
but quit smoking at the time of his myocardial infarc- resolved by the time the patient presents to the emergency
tion. Previous general anesthetics for an appendectomy room, leaving no neurologic deficits on examination, a
and bilateral total hip arthroplasties more than 10 years detailed clinical history is critical to recognizing this sign
ago were uneventful. Current medical therapy includes as- of impending stroke. Because a TIA is considered as an
pirin (acetylsalicylic acid [ASA] 81 mg), metoprolol, and impending stroke, the pathologic mechanisms, workup
a diuretic. and treatment are the same as for ischemic stroke.
The patient undergoes right carotid endarterectomy
under general anesthesia with shunt placement during
the period of carotid cross-clamping. His intraoperative
course is remarkable for a labile blood pressure, requiring What Are the Known Risk
the frequent administration of vasopressors and anti-
hypertensives to control hypotension and hypertension, Factors for Ischemic Stroke?
respectively. He awakens from anesthesia with a dense left
hemiplegia, which resolves completely during the course The risk factors for stroke are presented in Table 24.1.
of the next 36 hours. The most important risk factors are those that also
337
338 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Stroke Hemorrhagic
12% stroke
88% - Intraparenchymal
- Subarachnoid
Ischemic stroke
underlie atherosclerotic vascular disease in general, such age, male gender, smoking, hypertension, and evidence
as age and hypertension,1 and a history of a previous for generalized atherosclerosis, not unusual for stroke
TIA or stroke (see Fig. 24.2).3 The patient, from the case patients in general and for those undergoing carotid
description, presents with a combination of advanced endarterectomy in particular. Less well established, but
possible, risk factors for stroke are increased daytime
TABLE 24.1 Risk Factors for Stroke sleepiness and obstructive sleep apnea.4
FIGURE 24.2 Annual risk of stroke among patients in various high-risk subgroups. TIA, transient
ischemic attack.
(Adapted from: Wilterdink JL, Easton JD. Vascular event rates in patients with atherosclerotic
cerebrovascular disease. Arch Neurol. 1992;49:857.)
b = 1000 P ADC
FIGURE 24.4 Acute embolic right middle cerbral artery stroke What Are Watershed
with cortical involvement, which can be seen as an abnormal
signal on DWI (diffusion-weighted imaging)-MRI (bright on
Infarctions?
B1000 [left] and dark on ADC [right]). This patient suffered a
cardioembolic stroke during an acute myocardial infarction. Watershed infarctions are a subgroup of atherosclerotic
MRI, magnetic resonance imaging. ADC, apparent diffusion strokes and are caused by low-flow states in patients
coefficient. with severely stenosed or occluded arteries. They occur
in the distal collateralization areas, the borderzones of
C H A P T E R 2 4 / P E R I O P E R AT I V E S T R O K E 341
FIGURE 24.6 Middle cerebral artery/anterior cerebral artery borderzone (watershed) stroke after
massive upper GI bleed with severe hypotension for 20 minutes or longer. FLAIR (fluid attenuated
inversion recovery) image on left, DWI (diffusion-weighted imaging) with B1000 middle and DWI
with ADC on right). GI, gastrointestinal.
major cerebral arteries (ACA/MCA or MCA/PCA territory, Rare genetic causes (CADASIL [cerebral autosomal
see Fig. 24.6). A typical scenario is prolonged systemic dominant arteriopathy with subcortical infarcts and
hypotension during cardiac arrest or surgery with labile leukoencephalopathy] or MELAS [mitochondrial en-
blood pressures, during which an adequate CPP cannot cephalopathy with lactacidosis and strokes]) and others
be maintained to perfuse the most distal endzones of the
If no etiology can be found, strokes are called
large arteries. Clinical symptoms include blood-pressure
cryptogenic and compose approximately 30% to 40% of
sensitive fluctuations of the neurologic examination,
all strokes.14 Secondary stroke prevention strategies are
causing waxing and waning as stereotypic symptoms.
difficult to determine because of the heterogeneity of this
On imaging studies, a symmetric cortical distribution
subgroup of strokes. The recent Warfarin versus Aspirin
is sometimes seen. The location of watershed infarctions
in Recurrent Stroke Study15 showed that warfarin was not
also depends on the collateral supply of the cortex. Sources superior to aspirin in the prevention of recurrent ischemic
of collateral flow are derived from three areas: stroke in patients with a prior noncardioembolic stroke,
1. Collateral flow through the circle of Willis. Of note, but provided some suggestion of potential benefit in favor
more than one third of patients do not have a complete of oral anticoagulants in the selected subpopulation of
circle of Willis and are unable to draw on the collateral cryptogenic stroke. The small number of patients with this
flow between the left and right hemispheres through subtype prevented this finding from reaching statistical
the anterior communicating artery or between the significance, and therefore further studies of this subgroup
anterior and posterior circulation through posterior are needed.
communicating arteries.
2. Leptomeningeal collaterals that provide some over-
lap between endarteries within each hemisphere are
another source. What Is the Pathophysiology
3. The third source is extracranial to intracranial collat- and Mechanism of Acute
erals such as communications between the external
carotid system and internal carotid system through the Stroke?
facial arteries.
Stroke and TIA are due to thrombus formation and occlu-
sion of cerebral vessels. A thrombus (clot) tends to develop
in a very limited number of locations (see Fig. 24.7).
What Happens When No Stroke A clot may occlude a blood vessel locally at the site of
its formation, or it may break off and embolize, thereby
Etiology Is Found leading to occlusion of a more distal vessel, typically at a
(Cryptogenic Strokes)? bifurcation. This is the case particularly for thrombi origi-
nating from the heart, aorta, internal carotid, or vertebral
arteries.
If the workup for the more common etiologies of stroke
Once the thrombus has wedged and occluded a
remains unrevealing, more unusual etiologies, among
vessel, cerebral blood flow and oxygen delivery to the
others, need to be considered, as follows:
downstream territory are severely decreased. If cere-
Arterial dissections bral blood flow decreases below the critical limit of
Hypercoagulable states 18 mL/100 g/minute, irreversible neuronal damage begins
Hyperviscosity to take place. The extent of neuronal damage depends on
Central nervous system (CNS) angiitis: how far the blood flow decreases below this critical limit,
342 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
n
mbra
rctio
Collateral flow
Infa
Penu
Intracranial Penetrating
atherosclerosis artery
disease
s
bu
Flow reducing
m
Carotid plaque with
ro
carotid stenosis
Th
arteriogenic emboli
Cardiogenic emboli
Left
ventricular pressure. A reduction in systemic blood pressure can,
thrombi therefore, increase the size of an infarction. Speedy
restoration of blood flow can save brain tissue and
crucially limit the extent of a stroke. New imaging
techniques such as perfusion-computer tomography or
FIGURE 24.7 Common sites of thrombus formation and perfusion-magnetic resonance imaging are now utilized
atheromatous disease as possible etiologies of stroke. Note that to image this brain at risk and may help predict stroke
cardiogenic emboli may have different origins. outcome.16,17
(Reproduced from: Albers GW, Amarenco P, Easton JD, et al.
Antithrombotic and thrombolytic therapy for ischemic stroke:
the Seventh ACCP Conference on Antithrombotic and
Thrombolytic Therapy. Chest 2004;126:483S.) What Happens at the Molecular
Level in Acute Stroke?
and how long it remains decreased. In addition to the Neuronal ischemia activates several pathologic mecha-
period of time, factors such as cerebral autoregulation, nisms that, if left unchecked, not only individually lead
CPP, the extent of cerebral collateral flow, hemoglobin to cell death but, in concert, augment and accelerate
concentration, and arterial oxygen saturation play cru- cell damage.18 The major pathogenic mechanisms are
cial roles in determining the size of the infarct. CPP, excitotoxicity, peri-infarct depolarizations, inflammation,
hemoglobin concentration, and oxygen saturation may and programmed cell death (apoptosis) (see Fig. 24.9).19
change significantly in the perioperative period, and Their interplay and severity are dependent on the de-
therefore offer opportunities for either mitigating or ag- gree and duration of ischemia, and develop over time
gravating secondary injury, particularly in patients with a (see Fig. 24.10).
preoperative or intraoperative stroke. Impairment of cerebral blood flow with cessation of
The tolerance for ischemia differs for different areas oxygen and glucose delivery leads to energy depletion
of the brain. It is usually greater for the brain stem than and loss of membrane potentials, because ionic gradients
for the cortex, whereas the pyramidal neurons of the hip- can no longer be maintained. The depolarization activates
pocampus are most vulnerable to ischemia. Tolerance for presynaptic voltage-dependent Ca2+ channels, causing the
ischemia also depends on the extent of previous damage. massive release of glutamate and other excitatory amino
Patients with previous strokes may have a decreased re- acids. At the same time, energy-dependent reuptake of
serve for repeated ischemia. Regions in which cerebral glutamate is impeded. There is increasing evidence that
blood flow is severely decreased will undergo infarction. the ensuing excitotoxic damage is mediated through the
These regions correspond to the core zone (center zone) of activation of different glutamate receptors in the gray and
a stroke. The surrounding area of tissue around the core white matter; activation of N-methyl-D-aspartate (NMDA)
is referred to as the ischemic penumbra. Here, cerebral receptors predominates in damage at the neuronal level
blood flow is in the range of 15 to 23 mL/100 g/minute. (gray matter), whereas activation of AMPA/kainate re-
Brain tissue in the penumbra is at serious risk for injury ceptors is central to damage at the axonal level (white
but potentially salvageable (see Fig. 24.8). matter).20 Both lead to Na+ and Ca2+ influx, and K+ efflux,
In the penumbra, because of cerebral autoregulation, accelerating the loss of physiologic ionic gradients. Wa-
resistance vessels are maximally dilated, and cerebral ter passively follows the ionic influx. The resulting brain
blood flow becomes a linear function of perfusion edema is one of the earliest markers of stroke on imaging
C H A P T E R 2 4 / P E R I O P E R AT I V E S T R O K E 343
lu
lu
G
lu
G
G
lu
G
lu Na Depolarization K
G
lu
G
lu
G
Ca2 Na
Edema K
Ca2
Mitochondrial
damage
Enzyme
Induction/Activation
Apoptosis
Necrosis
Free radicals
Membrane NO
degradation DNA damage
Proinflammatory
mediators
Leukocyte
Microglial infiltration
activation Inflammation
FIGURE 24.9 Simplified overview of the molecular pathophysiology in brain ischemia (see text for
details).
(Adapted from: Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an
integrated view. Trends Neurosci. 1999;22:391.)
Excitotoxicity
Peri-infarct
depolarizations Inflammation
Impact
Apoptosis
FIGURE 24.10 Cascade of damaging events in focal cerebral ischemia. The x-axis reflects the
evolution of cascades over time, the y-axis shows the impact of each element of the cascade on final
outcome, the schematic picture of the brain shows the extent of core and penumbra as it would be
seen histologically or on imaging studies. Note the time-dependent changes in the relative
importance of pathologic mechanisms at play and the time-dependent changes in the size of core
and penumbra. For details, see text.
(Adapted from: Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of ischaemic stroke: an
integrated view. Trends Neurosci 1999;22:391.)
344 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
studies, and its extent is one of the major determinants of In the penumbra, where ischemia is milder than in the
survival beyond the first few hours of stroke. core, cell death occurs predominantly through apoptosis.
The influx of Ca2+ triggers a number of nuclear and This cellular suicide is mediated by caspases,29 protein-
cytoplasmatic events, such as the activation of prote- cleaving enzymes that modify homeostasis and repair
olytic enzymes and extracellular matrix proteins that proteins which, in turn, kill cells. Caspase-1 and caspase-3
lead to the destruction of cell structures. Ca2+ also seem to play a major role in ischemia-mediated apoptosis.
activates phospholipase A2 and cyclooxygenase (COX), One important way to activate caspases is to release
leading to free radical production, with lipid peroxida- cytochrome C from the mitochondria.23,30 After mild
tion and membrane damage. Oxygen free radicals, in (30 minutes), reversible MCA occlusion, cytochrome C
turn, trigger inflammation and apoptosis. Nitric oxide release is detected at 6 hours and caspase processing at
(NO), which is produced by the Ca2+ -dependent enzyme 9 hours following ischemia. Cell death becomes prominent
neuronal nitric oxide synthase (NOS), reacts with su- between 24 and 72 hours.31 Caspase inhibitors have the
peroxide to form the highly reactive peroxinitrite, which ability to attenuate the volume of dead tissue and decrease
promotes tissue damage.21 Free radicals impair mito- the neurologic deficit.32
chondrial function. The mitochondria cease to produce Reperfusion injury occurs after the blood flow to the
adenosine triphosphate (ATP), swell, and release more ischemic brain has been restored. It involves paradoxical
oxygen free-radicals22 as well as cytochrome C.23 Cy- tissue injury with selective neuronal necrosis. Restoration
tochrome C provides an important trigger for apoptosis. of energy metabolism and an abrupt increase in free
The neurons therefore find themselves with damage to radicals hasten the destructive processes set in motion
membranes, as well as structural and functional pro- during the ischemic interval.18 Reperfusion can manifest
teins. This damage is further aggravated by a severely as regional breakdown of the blood-brain barrier and
diminished ability to synthesize new protein for repair or ischemic injury to microvessels leading to brain edema
survival.18 and hemorrhagic conversion respectively. These latter
The metabolic and ionic changes as described in the risks limit the window of time for reperfusing strategies
preceding text affect the core and penumbra of the stroke in the treatment of ischemic stroke.
differently. Necrosis of the core develops minutes after
the onset of ischemia. Cells die rapidly through lipol-
ysis, proteolysis, and breakdown of ionic homeostasis.
The penumbra, however, can either progress to infarc- How Are These Concepts
tion more slowly by the same mechanisms as the core, or Applied to Clinical Care of
exhibit secondary phenomena, such as spreading depolar-
ization, postischemic inflammation, reperfusion injury, Acute Ischemic Stroke?
and apoptosis. Peri-infarct depolarizations occur in areas
where enough energy and perfusion are present to re- The most important aspect of acute stroke management
polarize previously depolarized cells; this repolarization is determining the time of onset (Time is brain)33,34
occurs at the expense of further energy consumption. In so that the appropriate treatment can be targeted. For
response to increased concentrations of glutamate and this purpose, the time (last seen normal) should be
K+ , the same cells depolarize again. These repetitive de- used. Acute stroke treatments, such as recombinant tissue-
polarizations have been demonstrated in animal models, plasminogen activator (rt-PA), are employed to establish
occurring at a rate of several events per hour, and can early recanalization and salvage the ischemic penum-
be recorded for at least 6 to 8 hours after the onset of bra.35,36 However, rt-PA, may only be administered within
ischemia. As the number of depolarizations increase, the 3 hours of the onset of stroke symptoms. The risk of
infarct grows larger.24 hemorrhage, most notably intracranial hemorrhage, in-
The activation of second messenger systems, increase creases not only steadily with the time, but exponentially,
in oxygen free radicals, and hypoxia itself trigger a number once 3 hours from symptom onset have elapsed.36 Fur-
of proinflammatory genes by inducing transcription fac- ther analysis of the same study suggests that patients
tors.19 Injured brain cells, in turn, produce mediators of treated 0 to 90 minutes from stroke onset have better odds
inflammation, such as platelet-activating factor, TNF-, of improvement at 24 hours and favorable outcome at
and interleukin 1-,19 5 to 7 days after ischemia neu- 3 months than patients treated later than 90 minutes.37
trophils cross the vascular wall, followed by macrophages By adhering strictly to validated patient selection criteria
and monocytes. Glia cells also undergo changes. Astro- (see Table 24.2), the risk of an intracranial hemorrhage
cytes become hypertrophic 4 to 6 hours after ischemia, with thrombolytic therapy is <7%.36,38
while microglia become activated. Postischemic inflam- In the perioperative period, this approach to stroke
mation contributes to ischemic damage by several mech- treatment needs to be modified, because thrombolysis by
anisms. Neutrophils may not only worsen ischemia by intravenous rt-PA can risk fatal bleeding. One benefit of
obstructing the microvasculature, but also produce toxic the perioperative period is that the time of stroke onset,
mediators. Examples of such mediators are NO derived in and thereby its cause, is frequently easier to determine
toxic amounts from inducible NOS (iNOS);25 superoxide than for patients in the emergency room. This offers
and toxic prostanoids derived from COX 2 (2)26,27 and the option of targeted interventions, such as hyperbaric
TNF-, a cytokine that can exacerbate ischemic injury oxygen therapy for the treatment of a stroke caused
and trigger apoptosis.28 by an air embolus, or neuroradiologic interventions
C H A P T E R 2 4 / P E R I O P E R AT I V E S T R O K E 345
TABLE 24.2 Criteria for Intravenous Thrombolysis with Recombinant Tissue-Plasminogen Activator in Acute Stroke
Inclusion Criteria Absolute Exclusion Criteria Relative Contraindications
Focal neurologic deficit Bacterial endocarditis Excessive bleeding Trauma within 3 mo
with significant risk of Hemorrhage or well- risk CPR with chest compression within
long-term disability established infarct on 10 d
No intracranial computed tomography Stroke within 3 mo
hemorrhage on Surgery within 14 d
computed tomography Gastrointestinal, urologic or respiratory
hemorrhage within 21 d
Known bleeding diathesis (includes
renal and hepatic insufficiency and
dialysis)
PTT >40 s, INR >1.7, platelets
<100,000/L
Severe hypertension despite treatment
Severe neurologic deficit, age >75,
early edema or mass effect on
computed tomography
Last seen normal <3 h Brain lesion with high Diminished benefit Life expectancy from other causes
ago likelihood of <1 yr
hemorrhage (tumor, Rapidly improving symptoms
AVM, aneurysm, Seizure at stroke onset To rule out Todd paralysis
contusion) Glucose <50 or May cause focal neurologic deficit in
>400 mg/dL the absence of stroke
CPR, cardiopulmonary resuscitation; PTT, partial thromboplastin time; INR, international normalized ratio; AVM, arteriovenous malformation.
such as balloon angioplasty for arterial vasospasm after perturbations such as hypercoagulability4951 and post-
subarachnoid hemorrhage or mechanical clot disruption operative complications such as hypotension and atrial
for thromboembolic stroke.39,40 fibrillation.47
to 5%, 2.8% to 8.4%, and 6.5% to 9.2%, respectively.5961 1. Strategies to protect the brain and
Embolic strokes outnumber hemodynamic strokes by ap- 2. Strategies to prevent the aggravation of a potential or
proximately 3:1.59,60 Research efforts to minimize embolic real cerebral ischemic injury
events have been focused on avoiding extracorporeal cir- The latter, although not glamorous, are supported by
culation,62,63 minimizing manipulation of the aorta,64 both experimental and clinical studies, whereas protective
especially if it is diseased, and preventing postoperative strategies remain largely unproven in the clinical realm.
atrial fibrillation.65 Strategies to prevent the aggravation of cerebral
ischemic injury are aimed at maintaining cerebral perfu-
sion, normothermia, and normoglycemia. Normal cere-
bral perfusion requires an adequate blood pressure,
Are There Perioperative because blood vessels in the penumbra are maximally
dilated, and blood flow is pressure-dependent. Note that
Interventions That Can CPP, generally calculated as mean arterial pressure (MAP)
Minimize the Patients minus the intracranial pressure (ICP), is influenced by
body position and should either be corrected, if the head
Neurologic Risk? of the patient is significantly above the heart, or preferably
be based on a blood-pressure measurement referenced
The primary prevention of perioperative stroke is cen- to the external auditory meatus. In a patient such as
tered on the preoperative modification of risk factors the one presented in the case summary, an adequate
(Table 24.1). These include the control of hypertension blood pressure in the face of his postoperative neurologic
and the modification of risk factors for cardiovascular dis- deficit may be in the hypertensive range, because chronic
ease. Consideration should be given to -blockade, which hypertension causes a shift of the cerebral autoregula-
has been shown to reduce myocardial infarction,66,67 atrial tory curve toward higher pressures.79 Specifically, given
fibrillation,6870 and possibly stroke rate,71 as well as the preoperative blood pressure of 160/90 mm Hg in the
continuation of statin therapy, which is thought to sta- patient presented in the case summary, a reasonable goal
bilize atherosclerotic plaque.7274 Patients receiving an- for CPP may be 90 to 110 mm Hg. In stroke patients,
ticoagulation or antiplatelet medication for documented there is evidence that antihypertensive drugs decrease
vascular disease or arterioembolic events require individ- blood flow in the penumbra80 and that decreased blood
ualized assessment of perioperative anticoagulationthat pressure may worsen outcome after a stroke.81,82 Severe
is, weighing the risk of intraoperative bleeding against the hypertension, on the other hand, should be treated to
risk of thromboembolism.7577 For the patient described decrease the attendant risk of hemorrhagic conversion
in the case summary, at least the continuation of as- and cerebral edema.83 In addition to adequate blood pres-
pirin throughout the perioperative period of his carotid sure, euvolemia is an important component of ensuring
endarterectomy would be recommended.78 normal cerebral perfusion, particularly under anesthesia.
Intraoperative and postoperative interventions aimed Excessive fluid administration, as in hypertensive hyper-
at minimizing a patients neurologic risk fall into the volemic hemodilution, patterned after the treatment of
following two categories: vasospasm in patients with subarachnoid hemorrhage,
C H A P T E R 2 4 / P E R I O P E R AT I V E S T R O K E 347
shows no proven beneficial effect in patients with, or at flow and its regulation may still be abnormal 2 weeks
risk for, stroke.83,84 Management of ventilation should after a stroke,102,103 whereas damage to the blood-brain
aim for normoxia and normocapnia. Although hypocap- barrier and evidence for inflammation may persist for
nia decreases ICP, evidence from traumatic brain injury more than 4 weeks.104 106 On the other hand, carotid
shows that hyperventilation may be associated with in- endarterectomies have been performed safely as early as
creased oxygen extraction, anaerobic metabolism/lactate 2 weeks after a nondisabling stroke.107,108 Carotid en-
release, and worsened neurologic outcome.85,86 Body tem- darterectomies, however, remove a source of emboli and
perature has a potent effect on the tolerance of the CNS stroke, and may promote healing by improving blood
for ischemia; hyperthermia unequivocally worsens neu- flow in the penumbra of an existing stroke. In addition,
rologic outcome after a stroke.8789 These data provide these studies excluded patients with severe strokes, point-
a compelling rationale for avoiding the excessive intra- ing to the importance of considering size, location, and
operative warming of patients and aggressively treating functional impact of a stroke in planning a surgical pro-
fever in patients with a stroke. Hypothermia, on the other cedure. Careful consideration should therefore be given
hand, currently has no well-defined place in the treatment to delaying elective surgery for 4 to 6 weeks after a stable
of perioperative stroke;84 it prolongs ischemic tolerance, neurologic status is achieved in patients in whom a large
but does not provide neuroprotection in patients preemp- volume of tissue is affected by the stroke, because orga-
tively cooled during surgery for intracranial aneurysms.90 nizing the infarcted area into a glial scar will take longer.
Its therapeutic use in stroke patients is a current subject The same consideration applies to patients with marked
of investigation.91,92 functional impairment from a stroke, because any periop-
Management of blood glucose may impact neurologic erative aggravation of the ischemic damage may further
outcome in patients with, or at risk for, perioperative diminish their quality of life. A small stroke in a nonelo-
stroke. Several studies have associated hyperglycemia quent area of the brain may, in contrast, require less of
with poor neurologic outcome in stroke.93,94 It is impor- a delay. The ideal interval between a stroke and a car-
tant to note that these studies neither establish causation diac procedure has also not been studied. Again, for the
nor provide guidance on the degree of glycemic con- reasons mentioned in the preceding text, it seems pru-
trol. Relevant to the perioperative period, normoglycemia dent to delay elective operations for 4 to 6 weeks after
has been shown to improve survival in surgical inten- the neurologic status has stabilized. Cardiac procedures
sive care95,96 and non-neurologic outcomes in cardiac that eliminate the source of a stroke may be undertaken
surgery.97,98 Insulin therapy needs to be tempered by earlier, especially in cases of small nondisabling strokes.
the knowledge that hypoglycemia can not only mimic
stroke by causing focal neurologic symptoms but can also
damage neurons directly.83 KEY POINTS
Despite the fact that most anesthetics, with the ex-
ception of etomidate,99 nitrous oxide,100 and ketamine, 1. A recent TIA is an impending stroke! Workup and
have been shown to cause a robust decrease in infarct treatment are same as for a stroke!
size in animal models of focal ischemia,101 no specific 2. The stroke type determines the outcome, risk of
cerebroprotective interventions can improve the outcome recurrent stroke, and means of secondary stroke
of a perioperative stroke in humans.19 One reason may be prevention.
that most perioperative strokes occur in the postoperative 3. Cortical neurologic signs point to an embolic cause
period, when the effects of an anesthetic have worn off. of stroke.
Another reason is that, while anesthetics may decrease the 4. Lacunar strokes typically indicate a small volume of
brains metabolic requirement, and may favorably impact injured tissue.
early pathogenic mechanisms in cerebral ischemia, the 5. In stroke treatment: TIME is BRAIN!
pathologic sequence activated by a perioperative stroke is 6. The penumbra is: BRAIN at RISK. Prevention of
sufficiently redundant to compensate for the interference secondary injury and improvement of perfusion to
with any single mechanism. In this regard, anesthetics the penumbra are the cornerstones of stroke therapy.
share the fate of many diverse drugs that were launched 7. A patient with an acute ischemic stroke should have
for neuroprotection based on compelling animal experi- a normal or high-normal CPP. A CPP target of 90 to
ments, only to falter in the setting of a clinical stroke.19,101 130 mm Hg will be applicable to most patients.
8. The size of a stroke and the pathologic mechanisms
driving the secondary damage change over time until
a glial scar is formed.
How Does the Presence of a 9. The perioperative period increases the risk for
stroke.
Stroke Affect the Timing of a 10. Elective surgery should be delayed for 4 to 6 weeks
Surgical Procedure? after a stroke.
The ideal time interval between a stroke and a surgical pro- REFERENCES
cedure has not been investigated. The pathophysiologic 1. American Heart Association. Heart Disease and Stroke
processes of cerebral ischemia remain active well beyond Statistics2005 Update. Dallas, Texas: American Heart
the time of the initial insult (Fig. 24.10). Cerebral blood Association; 2005. http://www.americanheart.org
348 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
/downloadable/heart/1105390918119HDSStats2005Update 25. Forster C, Clark HB, Ross ME, et al. Inducible nitric
.pdf. oxide synthase expression in human cerebral infarcts. Acta
2. Johnston SC, Gress DR, Browner WS, et al. Short-term Neuropathol (Berl). 1999;97:215.
prognosis after emergency department diagnosis of TIA. 26. Iadecola C, Forster C, Nogawa S, et al. Cyclooxygenase-2
JAMA. 2000;284:2901. immunoreactivity in the human brain following cerebral
3. Wilterdink JL, Easton JD. Vascular event rates in patients ischemia. Acta Neuropathol (Berl). 1999;98:9.
with atherosclerotic cerebrovascular disease. Arch Neurol. 27. Nogawa S, Zhang F, Ross ME, et al. Cyclo-oxygenase-2
1992;49:857. gene expression in neurons contributes to ischemic brain
4. Davies DP, Rodgers H, Walshaw D, et al. Snoring, daytime damage. J Neurosci. 1997;17:2746.
sleepiness and stroke: a case-control study of first-ever 28. Barone FC, Arvin B, White RF, et al. Tumor necrosis factor-
stroke. J Sleep Res. 2003;12:313. alpha. A mediator of focal ischemic brain injury. Stroke.
5. Ay H, Furie KL, Singhal A, et al. An evidence-based 1997;28:1233.
causative classification system for acute ischemic stroke. 29. Endres M, Kaps M, Moskowitz MA. Apoptosis and ischemic
Ann Neurol. 2005;58:688. infarct. Nervenarzt. 1998;69:459.
6. Fischer CM. Lacunes: small, deep cerebral infarcts. Neurol- 30. Green DR, Reed JC. Mitochondria and apoptosis. Science.
ogy. 1965;15:774784. 1998;281:1309.
7. Fischer CM, Cole M. Homolateral ataxia and crural 31. Endres M, Namura S, Shimizu-Sasamata M, et al. Attenu-
paresis: a vascular syndrome. J Neurol Neurosurg Psychiatry. ation of delayed neuronal death after mild focal ischemia
1965;28:48. in mice by inhibition of the caspase family. J Cereb Blood
8. Fischer CM. A lacunar stroke. The dysarthria-clumsy hand Flow Metab. 1998;18:238.
syndrome. Neurology. 1967;17:614. 32. Hara H, Friedlander RM, Gagliardini V, et al. Inhibition
9. Fischer CM, Curry HB. Pure motor hemiplegia of vascular of interleukin 1beta converting enzyme family proteases
origin. Arch Neurol. 1965;13:304. reduces ischemic and excitotoxic neuronal damage. Proc
10. Mohr JP, Caplan LR, Melski JW, et al. The Harvard co- Natl Acad Sci USA. 1997;94:2007.
operative stroke registry: a prospective registry. Neurology. 33. Gomez C. Time is brain. J Stroke Cerebrovasc Dis. 1993;3:1.
1978;28:754. 34. Saver JL. Time is brainquantified. Stroke. 2006;37:263.
11. Stroke Prevention in Atrial Fibrillation Study. Final results. 35. Albers GW, Amarenco P, Easton JD, et al. Antithrombotic
Circulation. 1991;84:527. and thrombolytic therapy for ischemic stroke: the seventh
12. Pruitt AA, Rubin RH, Karchmer AW, et al. Neurologic com- ACCP conference on antithrombotic and thrombolytic
plications of bacterial endocarditis. Medicine (Baltimore). therapy. Chest. 2004;126:483S.
1978;57:329. 36. The National Institute of Neurological Disorders and
13. Roeltgen DP, Weimer GR, Patterson LF. Delayed neurologic Stroke rt-PA Stroke Study Group. Tissue plasminogen
complications of left atrial myxoma. Neurology. 1981;31:8. activator for acute ischemic stroke. N Engl J Med. 1995;333:
14. Sacco RL, Ellenberg JH, Mohr JP, et al. Infarcts of 1581.
undetermined cause: the NINCDS Stroke Data Bank. Ann 37. Marler JR, Tilley BC, Lu M, et al. Early stroke treatment
Neurol. 1989;25:382. associated with better outcome: the NINDS rt-PA stroke
15. Mohr JP, Thompson JL, Lazar RM, et al. A comparison study. Neurology. 2000;55:1649.
of warfarin and aspirin for the prevention of recurrent 38. Tanne D, Kasner SE, Demchuk AM, et al. Markers of in-
ischemic stroke. N Engl J Med. 2001;345:1444. creased risk of intracerebral hemorrhage after intravenous
16. Parsons MW, Pepper EM, Chan V, et al. Perfusion computed recombinant tissue plasminogen activator therapy for acute
tomography: prediction of final infarct extent and stroke ischemic stroke in clinical practice: the multicenter rt-PA
outcome. Ann Neurol. 2005;58:672. stroke survey. Circulation. 2002;105:1679.
17. Sims J, Schwamm LH. The evolving role of acute stroke 39. Qureshi AI, Siddiqui AM, Suri MF, et al. Aggressive
imaging in intravenous thrombolytic therapy: patient selec- mechanical clot disruption and low-dose intra-arterial
tion and outcomes assessment. Neuroimaging Clin N Am. third-generation thrombolytic agent for ischemic stroke:
2005;15:421, xii. a prospective study. Neurosurgery. 2002;51:1319.
18. White BC, Sullivan JM, DeGracia DJ, et al. Brain ischemia 40. Pelz D, Lylyk P, Negoro M. Interventional neuroradiology.
and reperfusion: molecular mechanisms of neuronal injury. Stroke. 2004;35:381.
J Neurol Sci. 2000;179:1. 41. Knapp RB, Topkins MJ, Artusio JF Jr. The cerebrovascular
19. Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of accident and coronary occlusion in anesthesia. JAMA. 1962;
ischaemic stroke: an integrated view. Trends Neurosci. 182:332.
1999;22:391. 42. Larsen SF, Zaric D, Boysen G. Postoperative cerebrovascu-
20. Phan TG, Wright PM, Markus R, et al. Salvaging the lar accidents in general surgery. Acta Anaesthesiol Scand.
ischaemic penumbra: more than just reperfusion? Clin Exp 1988;32:698.
Pharmacol Physiol. 2002;29:1. 43. Parikh S, Cohen JR. Perioperative stroke after general
21. Beckman JS, Koppenol WH. Nitric oxide, superoxide, and surgical procedures. N Y State J Med. 1993;93:162.
peroxynitrite: the good, the bad, and ugly. Am J Physiol. 44. Evans BA, Wijdicks EF. High-grade carotid stenosis de-
1996;271:C1424. tected before general surgery: is endarterectomy indicated?
22. Dugan LL, Choi DW. Excitotoxicity, free radicals, and cell Neurology. 2001;57:1328.
membrane changes. Ann Neurol. 1994;35(Suppl):S17. 45. Ropper AH, Wechsler LR, Wilson LS. Carotid bruit and
23. Fujimura M, Morita-Fujimura Y, Murakami K, et al. the risk of stroke in elective surgery. N Engl J Med. 1982;
Cytosolic redistribution of cytochrome c after transient 307:1388.
focal cerebral ischemia in rats. J Cereb Blood Flow Metab. 46. Wong GY, Warner DO, Schroeder DR, et al. Risk of
1998;18:1239. surgery and anesthesia for ischemic stroke. Anesthesiology.
24. Mies G, Iijima T, Hossmann KA. Correlation between peri- 2000;92:425432.
infarct DC shifts and ischaemic neuronal damage in rat. 47. Hart R, Hindman B. Mechanisms of perioperative cerebral
Neuroreport. 1993;4:709. infarction. Stroke. 1982;13:766.
C H A P T E R 2 4 / P E R I O P E R AT I V E S T R O K E 349
48. Kam PC, Calcroft RM. Peri-operative stroke in general 67. Poldermans D, Boersma E, Bax JJ, et al. Dutch Echocardio-
surgical patients. Anaesthesia. 1997;52:879. graphic Cardiac Risk Evaluation Applying Stress Echocar-
49. Wilson D, Cooke EA, McNally MA, et al. Changes in diography Study Group. The effect of bisoprolol on
coagulability as measured by thrombelastography following perioperative mortality and myocardial infarction in high-
surgery for proximal femoral fracture. Injury. 2001;32:765. risk patients undergoing vascular surgery. N Engl J Med.
50. Fujii Y, Tanaka R, Takeuchi S, et al. Serial changes in 1999;341:1789.
hemostasis after intracranial surgery. Neurosurgery. 1994; 68. White HD, Antman EM, Glynn MA, et al. Efficacy and safety
35:26. of timolol for prevention of supraventricular tachyarrhyth-
51. Caprini JA, Arcelus JI, Laubach M, et al. Postoperative mias after coronary artery bypass surgery. Circulation.
hypercoagulability and deep-vein thrombosis after laparo- 1984;70:479.
scopic cholecystectomy. Surg Endosc. 1995;9:304. 69. Daudon P, Corcos T, Gandjbakhch I, et al. Prevention of
52. Bond R, Rerkasem K, Shearman CP, et al. Time trends in atrial fibrillation or flutter by acebutolol after coronary
the published risks of stroke and death due to endarterec- bypass grafting. Am J Cardiol. 1986;58:933.
tomy for symptomatic carotid stenosis. Cerebrovasc Dis. 70. Lamb RK, Prabhakar G, Thorpe JA, et al. The use of atenolol
2004;18:37. in the prevention of supraventricular arrhythmias following
53. Brott TG, Brown RD Jr, Meyer FB, et al. Carotid revascu- coronary artery surgery. Eur Heart J. 1988;9:32.
larization for prevention of stroke: carotid endarterectomy 71. Amory DW, Grigore A, Amory JK, et al. Neuroprotection
and carotid artery stenting. Mayo Clin Proc. 2004;79:1197. is associated with beta-adrenergic receptor antagonists
54. Krul JM, van Gijn J, Ackerstaff RG, et al. Site and patho- during cardiac surgery: evidence from 2,575 patients.
genesis of infarcts associated with carotid endarterectomy. J Cardiothorac Vasc Anesth. 2002;16:270.
Stroke. 1989;20:324. 72. Poldermans D, Bax JJ, Kertai MD, et al. Statins are
55. Riles TS, Imparato AM, Jacobowitz GR, et al. The cause associated with a reduced incidence of perioperative
of perioperative stroke after carotid endarterectomy. J Vasc mortality in patients undergoing major noncardiac vascular
Surg. 1994;19:206. surgery. Circulation. 2003;107:1848.
56. Spencer MP. Transcranial Doppler monitoring and causes 73. Libby P, Ridker PM, Maseri A. Inflammation and
of stroke from carotid endarterectomy. Stroke. 1997;28:685. atherosclerosis. Circulation. 2002;105:1135.
57. Payne DA, Jones CI, Hayes PD, et al. Beneficial effects of 74. Kennedy J, Quan H, Buchan AM, et al. Statins are associated
clopidogrel combined with aspirin in reducing cerebral with better outcomes after carotid endarterectomy in
emboli in patients undergoing carotid endarterectomy. symptomatic patients. Stroke. 2005;36:2072.
Circulation. 2004;109:1476. 75. Harder S, Klinkhardt U, Alvarez JM. Avoidance of bleeding
58. Naylor AR. There is more to preventing stroke after carotid during surgery in patients receiving anticoagulant and/or
surgery than shunt and patch debates. Eur J Vasc Endovasc antiplatelet therapy: pharmacokinetic and pharmacody-
Surg. 2005;29:329. namic considerations. Clin Pharmacokinet. 2004;43:963.
59. McKhann GM, Grega MA, Borowicz LM Jr, et al. Stroke and 76. Jafri SM. Periprocedural thromboprophylaxis in patients
encephalopathy after cardiac surgery: an update. Stroke. receiving chronic anticoagulation therapy. Am Heart J.
2006;37:562. 2004;147:3.
60. Salazar JD, Wityk RJ, Grega MA, et al. Stroke after cardiac 77. Kearon C, Hirsh J. Management of anticoagulation before
surgery: short-and long-term outcomes. Ann Thorac Surg. and after elective surgery. N Engl J Med. 1997;336:1506.
2001;72:1195. 78. Taylor DW, Barnett HJ, Haynes RB, et al. Low-dose
61. Bucerius J, Gummert JF, Borger MA, et al. Stroke after and high-dose acetylsalicylic acid for patients undergoing
cardiac surgery: a risk factor analysis of 16,184 consecutive carotid endarterectomy: a randomised controlled trial. ASA
adult patients. Ann Thorac Surg. 2003;75:472. and Carotid Endarterectomy (ACE) Trial Collaborators.
62. Sellke FW, DiMaio JM, Caplan LR, et al. Comparing Lancet. 1999;353:2179.
on-pump and off-pump coronary artery bypass grafting: nu- 79. Paulson OB, Strandgaard S, Edvinsson L. Cerebral autoreg-
merous studies but few conclusions: a scientific statement ulation. Cerebrovasc Brain Metab Rev. 1990;2:161.
from the American Heart Association council on cardio- 80. Lisk DR, Grotta JC, Lamki LM, et al. Should hypertension
vascular surgery and anesthesia in collaboration with the be treated after acute stroke? A randomized controlled trial
interdisciplinary working group on quality of care and using single photon emission computed tomography. Arch
outcomes research. Circulation. 2005;111:2858. Neurol. 1993;50:855.
63. Cheng DC, Bainbridge D, Martin JE, et al. Does off-pump 81. Brott T, Lu M, Kothari R, et al. Hypertension and its
coronary artery bypass reduce mortality, morbidity, and treatment in the NINDS rt-PA stroke trial. Stroke. 1998;29:
resource utilization when compared with conventional 1504.
coronary artery bypass? A meta-analysis of randomized 82. Gold JP, Charlson ME, Williams-Russo P, et al. Improve-
trials. Anesthesiology. 2005;102:188. ment of outcomes after coronary artery bypass. A random-
64. Kapetanakis EI, Stamou SC, Dullum MK, et al. The ized trial comparing intraoperative high versus low mean
impact of aortic manipulation on neurologic outcomes after arterial pressure. J Thorac Cardiovasc Surg. 1995;110:1302.
coronary artery bypass surgery: a risk-adjusted study. Ann 83. Adams HP, Jr., Adams RJ, Brott T, et al. Guidelines for
Thorac Surg. 2004;78:1564. the early management of patients with ischemic stroke:
65. Zimmer J, Pezzullo J, Choucair W, et al. Meta-analysis of A scientific statement from the Stroke Council of the
antiarrhythmic therapy in the prevention of postoperative American Stroke Association. Stroke. 2003;34:1056.
atrial fibrillation and the effect on hospital length of stay, 84. Adams H, Adams R, Del Zoppo G, et al. Guidelines for the
costs, cerebrovascular accidents, and mortality in patients early management of patients with ischemic stroke: 2005
undergoing cardiac surgery. Am J Cardiol. 2003;91:1137. guidelines update a scientific statement from the Stroke
66. Mangano DT, Layug EL, Wallace A, et al. Multicenter Council of the American Heart Association/American
Study of Perioperative Ischemia Research Group. Effect Stroke Association. Stroke. 2005;36:916.
of atenolol on mortality and cardiovascular morbidity after 85. Muizelaar JP, Marmarou A, Ward JD, et al. Adverse
noncardiac surgery. N Engl J Med. 1996;335:1713. effects of prolonged hyperventilation in patients with severe
350 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
head injury: a randomized clinical trial. J Neurosurg. hospital outcome after cardiac surgery in diabetic patients.
1991;75:731. Anesthesiology. 2005;103:687.
86. Bouma GJ, Muizelaar JP, Choi SC, et al. Cerebral cir- 98. Lazar HL, Chipkin SR, Fitzgerald CA, et al. Tight glycemic
culation and metabolism after severe traumatic brain in- control in diabetic coronary artery bypass graft patients
jury: the elusive role of ischemia. J Neurosurg. 1991;75: improves perioperative outcomes and decreases recurrent
685. ischemic events. Circulation. 2004;109:1497.
87. Azzimondi G, Bassein L, Nonino F, et al. Fever in acute 99. Drummond JC, Cole DJ, Patel PM, et al. Focal cerebral
stroke worsens prognosis. A prospective study. Stroke. 1995; ischemia during anesthesia with etomidate, isoflurane, or
26:2040. thiopental: a comparison of the extent of cerebral injury.
88. Reith J, Jorgensen HS, Pedersen PM, et al. Body tempera- Neurosurgery. 1995;37:742.
ture in acute stroke: relation to stroke severity, infarct size, 100. Soonthon-Brant V, Patel PM, Drummond JC, et al. Fentanyl
mortality, and outcome. Lancet. 1996;347:422. does not increase brain injury after focal cerebral ischemia
89. Wang Y, Lim LL, Levi C, et al. Influence of admission body in rats. Anesth Analg. 1999;88:49.
temperature on stroke mortality. Stroke. 2000;31:404. 101. Patel PM, Drummond JC. Cerebral physiology and the
90. Todd MM, Hindman BJ, Clarke WR, et al. Mild intraopera- effects of anesthetics and techniques. In: Miller RD, ed.
tive hypothermia during surgery for intracranial aneurysm. Anesthesia. Philadelphia: Elsevier Science; 2005:813.
N Engl J Med. 2005;352:135. 102. Paulson OB. Regional cerebral blood flow in apoplexy
91. Hammer MD, Krieger DW. Hypothermia for acute is- due to occlusion of the middle cerebral artery. Neurology.
chemic stroke: not just another neuroprotectant. Neurol- 1970;20:63.
ogist. 2003;9:280. 103. Agnoli A, Fieschi C, Bozzao L, et al. Autoregulation of cere-
92. Krieger DW, De Georgia MA, Abou-Chebl A, et al. Cooling bral blood flow. Studies during drug-induced hypertension
for acute ischemic brain damage (cool aid): an open pilot in normal subjects and in patients with cerebral vascular
study of induced hypothermia in acute ischemic stroke. diseases. Circulation. 1968;38:800.
Stroke. 2001;32:1847. 104. Olsen TS. Regional cerebral blood flow after occlusion of
93. Bruno A, Biller J, Adams HP Jr, et al. Acute blood glucose the middle cerebral artery. Acta Neurol Scand. 1986;73:321.
level and outcome from ischemic stroke. Trial of ORG 105. Schwamm LH, Koroshetz WJ, Sorensen AG, et al. Time
10172 in Acute Stroke Treatment (TOAST) Investigators. course of lesion development in patients with acute stroke:
Neurology. 1999;52:280. serial diffusion- and hemodynamic-weighted magnetic
94. Davalos A, Castillio J. Potential mechanisms of worsening. resonance imaging. Stroke. 1998;29:2268.
Cerebrovasc Dis. 1997;7(Suppl 5):19. 106. Albers GW. Diffusion-weighted MRI for evaluation of acute
95. Van den BG, Wilmer A, Hermans G, et al. Intensive insulin stroke. Neurology. 1998;51:S47.
therapy in the medical ICU. N Engl J Med. 2006;354: 107. Ballotta E, Da Giau G, Baracchini C, et al. Early versus
449. delayed carotid endarterectomy after a nondisabling is-
96. Van den BG, Wouters P, Weekers F, et al. Intensive chemic stroke: a prospective randomized study. Surgery.
insulin therapy in the critically ill patients. N Engl J Med. 2002;131:287.
2001;345:1359. 108. Ricco JB, Illuminati G, Bouin-Pineau MH, et al. Early
97. Ouattara A, Lecomte P, Le Manach Y, et al. Poor intraop- carotid endarterectomy after a nondisabling stroke:
erative blood glucose control is associated with a worsened a prospective study. Ann Vasc Surg. 2000;14:89.
CHAPTER COMA AND DELAYED EMERGENCE
25 Roger S. Mecca
Y
has been in the postanesthesia care unit thetics, and the use of depth of anesthesia indicators,
(PACU) for 60 minutes following a total hip such as end-expired volatile agents and processed elec-
replacement under spinal anesthesia with in- troencephalogram (EEG) monitoring. In spite of these
trathecal bupivacaine and morphine, along advances, some patients still exhibit a significantly de-
with intravenous sedation with midazolam. pressed state of consciousness after surgery. For most
During placement of the prosthesis, the patient became of these individuals, persistent sedation resolves during
very agitated and complained of discomfort. This led the early recovery interval without intervention; however,
his intraoperative anesthesia provider to induce a gen- occasionally a patient will not regain consciousness as
eral endotracheal anesthetic that incorporated fentanyl, expected. Evaluation and treatment of prolonged uncon-
isoflurane, and vecuronium. Placement of the prosthe- sciousness after anesthesia is one of the most perplexing
sis was extremely difficult, and the amount of blood lost and anxiety-provoking challenges presented to a practi-
throughout the case was significant. tioner during postoperative care.
Upon review of the patients chart, a history of dia- To quickly complete a thorough and accurate analysis
betes, hypertension, atrial fibrillation, and stable angina of prolonged unconsciousness, an organized and consis-
is noted. Physical examination reveals an unconscious, tent approach should be utilized that addresses potential
spontaneously ventilating patient with small, poorly re- problems in order of their likelihood and their impact
active pupils, a carotid bruit on the right, and a subtle, on the patients well-being. These measures will expedite
holosystolic murmur at the left sternal border. His blood the identification and resolution of the etiology of de-
pressure is 20% below his chronic baseline, and his arte- layed emergence and minimize the chances of adverse
rial oxygen saturation has varied between 97% and 93% on outcomes.
supplemental oxygen in the PACU. He has a central venous Prolonged unconsciousness after anesthesia can be
catheter in his right internal jugular vein. The patient does caused or exacerbated by a large number of adverse phys-
not respond at all to verbal stimulation and only twitches ical conditions. Unfortunately, most of these conditions
his arms feebly in response to a firm tactile stimulus. are progressive and generate time-sensitive morbidity,
The PACU nurse asks you how much longer it will and can be life-threatening if not addressed promptly.13
take for the patient to regain consciousness. The orthope- Many of these conditions are also easily reversible without
dic surgeon wants you to assure the family that, in fact, he permanent injury if discovered and treated early in their
will emerge from anesthesia. How would you assess the clinical course.
etiology of his state of unconsciousness to answer these
questions?
RESPIRATORY SYSTEM
A state of unconsciousness poses a variety of secondary
What Is the Impact risks that are independent of the underlying etiology.
of a Prolonged State For example, factors that depress the central nervous
system (CNS) sufficiently to produce unconsciousness
of Unconsciousness? usually also depress, to some degree, the centers that
are responsible for regulating ventilation. An unconscious
In current practice, emergence from anesthesia is more patients awareness of ventilation and volition to breathe
rapid and predictable than ever before, owing to such are undoubtedly impaired. Central monitoring of changes
351
352 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
in lung volume is also blunted by a reduced level of or nausea, nor can disorientation secondary to cerebral
consciousness. The medullary respiratory center near the hypoperfusion be assessed.
fourth ventricle primarily regulates the partial pressure of
carbon dioxide in arterial blood (PACO2 ) by monitoring
the pH concentration in the cerebrospinal fluid. Any
factor that interferes with the sensitivity of this center
AUTONOMIC
encourages the acceptance of a higher than usual AND SYMPATHETIC
PACO2 , thereby leading to hypoventilation and respiratory
acidemia. Even a moderate level of somnolence, as
NERVOUS SYSTEMS
occurs during normal sleep, will result in some degree Factors that depress the CNS sufficiently to prolong
of hypercarbia and a reduction of serum pH. unconsciousness will also depress centers that regulate
Deeper levels of unconsciousness exert a proportion- the autonomic nervous system; therefore, sympathetic
ally greater degree of ventilatory depression and respira- nervous system responses are blunted to pain, noxious
tory acidemia. A patients hypoxic ventilatory drive, which sensations, and changes in systemic blood pressure,
responds to changes in partial pressure of oxygen in arte- arterial pH, and other physiologic variables. Attenuation
rial blood (PaO2 ), is usually more sensitive to depression of sweating and tachycardia caused by hypoglycemia in
from drugs and other factors than the PACO2 drive. There- an anesthetized patient is a well acknowledged example
fore, hypoventilation secondary to unconsciousness can of this phenomenon.
also lead to hypoxemia. In severe cases of respiratory
center depression, hypoventilation or complete apnea can
quickly progress to severe hypoxemia or end-stage respi-
ratory acidemia, leading to death. EXPENDITURE OF MEDICAL
In unconscious patients, decreased tone and coor- RESOURCES
dination of the pharyngeal and laryngeal musculature
promotes the increase of upper airway resistance to gas Other aspects of delayed emergence impact medical risk
flow; such a reduction in upper airway patency increases and expenditure of resources. Prolonged unconsciousness
the chances that hypoxemia and/or hypercarbia will de- lengthens the time that the patient spends in a postanes-
velop in spontaneously ventilating patients. The likelihood thesia care setting, thereby increasing staff costs due to the
of obstructive, negative-pressure pulmonary edema is also amount of care and observation that the patient requires.
increased. In extreme cases, augmented resistance can While attending to an unconscious patient, the health care
progress to complete upper airway obstruction. team may provide less coverage to other postanesthesia
Beyond the impact on airway patency, a depressed patients, and therefore, additional staff may be required,
consciousness also impairs the protective reflexes of the further adding to hospital costs. In addition to expenditure
airway. The probability of postoperative regurgitation of resources, these modifications diminish the efficiency
and aspiration is increased in an unconscious patient, and overall safety of postanesthesia care. Workplace safety
especially if the patient is left in a supine position with for postanesthesia staff is also an issue. Because uncon-
the head in the midline. scious patients cannot participate in self-care, they must
be positioned and moved by staff members, which in-
creases the exposure of staff members to injury during
lifting or fall prevention.
IMPAIRED REFLEXES
A patient in an unconscious state is also prone to inci-
dental injuries in the postoperative care environment. For
example, an unconscious patient is unable to acknowl- How Can A State of Prolonged
edge and respond to pain from entrapment of skin folds
or contact of body parts with rigid surfaces; this condition
Unconsciousness Be Generally
can lead to pressure necrosis in tissues, compartment syn- Assessed?
dromes, peripheral nerve damage, or myoglobinuria with
renal impairment. The possibility of positioning-related Two general issues need to be addressed when deciding
and fall-related injuries varies inversely with the level of whether a postoperative patient is exhibiting prolonged
consciousness in postoperative patients. For example, the unconsciousness after anesthesia. The first involves differ-
shifting of oxygen equipment placed on the face of an un- entiating whether the patient is truly unconscious, merely
conscious patient can cause corneal abrasion. Prolonged asleep, or exhibiting a disordered sensorium.
unconsciousness after anesthesia and surgery can mask
the symptoms of an evolving surgical complication or a
threatening unrelated condition and, consequently, can
delay diagnosis and treatment. For example, a patient VERBAL AND TACTILE
in an unconscious state cannot acknowledge abdominal STIMULATION
pain from a ruptured viscus or lower extremity paralysis
from an evolving epidural hematoma. Neither will such For purposes of this discussion, unconsciousness can
a patient exhibit agitation caused by metabolic acidemia be defined as the absence of any meaningful, directed
C H A P T E R 2 5 / C O M A A N D D E L AY E D E M E R G E N C E 353
response to ordinary levels of verbal or non-noxious TABLE 25.1 Medical History Pertinent to the Differential
tactile stimulation.4 If a patient can be aroused with gentle Diagnosis of Prolonged Unconsciousness
tactile stimulation, mumbles a vague verbal response to
a question, and/or purposefully withdraws an extremity Preexisting abnormalities in level of consciousness
from a mildly uncomfortable physical stimulus, he or Stroke or transient ischemic attacks
she is not unconsciousness but is merely somnolent. Intracranial pathology
Similarly, if a patient is combative or exhibits confusion Epilepsy or other seizure disorder
or disorientation, he or she is not unconscious but likely Chronic hepatic dysfunction
exhibiting an unusual emergence reaction. However, if no Atrial fibrillation or flutter
response to verbal interaction is forthcoming, and only a Congenital heart disease, septal defects, heart murmurs
deep reflex response to a painful tactile stimulus is elicited, Chronic metabolic or electrolyte abnormalities
a patient should be considered unconscious. Inborn errors of metabolism
Medication history
Severe malnutrition
Substance abuse
RATE OF EMERGENCE Deafness
The second issue involves estimating the reasonable pe- Unrecognized head trauma
riod in which the patient is expected to regain conscious- Possibility of unrecognized asphyxia
ness after the anesthetic, taking into consideration the Exposure to environmental toxins
specific conditions of the situation at hand. Obviously, Carbon monoxide exposure
there is an acceptable interval after the termination of Ingestion of poisons
general or regional anesthesia with deep sedation during
which decreased responsiveness is expected. The amount
of time that a patient requires to regain consciousness practitioner performing the postoperative examination
after anesthesia is terminated depends on the following was not involved in the intraoperative anesthetic and is
factors: not familiar with the patients condition before induction
Type and duration of anesthesia and surgery of anesthesia.
Type and dosage of agents employed A review of the admission history and physical
Timing of final doses and discontinuation of agents examination, the preanesthesia evaluation, and other
The occurrence and severity of unexpected events sources, such as inpatient progress notes, nursing notes,
or referral forms can yield invaluable insight. During this
Many factors that are intrinsic to each patient also review, assess whether preexisting factors are contributing
affect the rate of emergence; therefore, establishing the to the patients state of unconsciousness such as level
acceptable duration of unconsciousness post anesthesia of mental dysfunction or any history of epilepsy or
is a highly individualized judgment that depends on the trauma-related seizure disorder. Take special note of
clinical circumstances. Nevertheless, it is possible to pre- preexisting medical conditions that can impact CNS
dict an interval within which a vast majority of patients function, such as cerebral vascular disease, transient
will regain consciousness after a reasonably conducted ischemic attacks, stroke, intracranial tumor, cerebral
anesthetic. Most patients emerging from general anesthe- aneurysm, or previous head trauma. The presence of
sia should regain consciousness and react purposefully supraventricular dysrhythmias such as atrial fibrillation
to verbal and tactile stimuli within 15 minutes of admis- or flutter should lead one to consider the possibility
sion to the PACU. (For practical purposes, ignore the 5 of cerebral thromboembolism secondary to migration
to 10 minutes that usually elapse between the discon- of atrial clots. A history of congenital heart disease,
tinuation of a general anesthetic and admission to the septal defect, endocarditis, or heart murmur may point
PACU. Doing so expands the actual interval to 20 to 25 toward paradoxical cerebral embolization with thrombus,
minutes after the cessation of anesthetic delivery.) Even vegetations, air, or fat. Cirrhosis, chronic hepatitis, or
a patient who is highly sensitive to the residual sedative other disorders of liver function may indicate an element
effects of anesthetic agents should respond to verbal or of hepatic encephalopathy. Chronic metabolic conditions,
tactile stimuli within 30 minutes of admission to the PACU electrolyte disorders, or inborn errors of metabolism can
(35 to 40 minutes after cessation of anesthesia). A state of certainly affect the level of consciousness, particularly if
unconsciousness that persists for more than 30 minutes exacerbated by intraoperative conditions.
after admission to the PACU is considered prolonged and A patient may be using medications on a chronic
should be aggressively evaluated. basis that can depress the level of consciousness or lead to
unusual cross-reactions with agents administered during
surgery. For example, baclofen taken preoperatively or
MEDICAL HISTORY given during a procedure can significantly impair the
postoperative level of consciousness.5,6 Postanesthesia
An initial assessment of a patients medical history can arousal can also be affected by the use of herbal
help elucidate the cause of prolonged unconsciousness medications, such as St. Johns wort.7 If possible, also
(see Table 25.1). This is particularly important when the assess the nutritional status and look for alcohol or
354 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
other substance abuse. A history of deafness occasionally TABLE 25.2 Intraoperative History Pertinent to the
explains an emerging patients lack of response to verbal Differential Diagnosis of Prolonged Unconsciousness
stimuli. Finally, scrutiny of events that occurred in
the immediate interval before surgery is important. In Level of responsiveness before induction
trauma patients or those requiring emergency surgery, Level of responsiveness during emergence
the possibility of unrecognized head injury, asphyxia, or Duration and concentration of inhalational anesthetics
exposure to carbon monoxide, environmental toxins, or Time and dosage of premedications
ingested poisons should be evaluated. Timing and dosage of opioids
Timing and dosage of sedatives and antiemetics
Episodes of airway obstruction
PERIOPERATIVE EVENTS Episodes of arterial oxygen desaturation
Episodes of hypotension
With respect to the surgical procedure, reviewing docu- Episodes of significant hyperventilation
mentation that clarifies the patients peri-induction state Amount of blood loss
of responsiveness and behavior helps elucidate whether Intraoperative rhythm changes
unrecognized acute intoxication with drugs or alcohol is Extreme or unusual intraoperative positioning
a contributing factor. The time and amount of sedative or Interventions near the cerebral circulation
hypnotic drugs administered for premedication should be Placement of central vascular catheter or pacing device
noted, being especially vigilant for longer-acting sedatives
given orally or rectally, because these exhibit a delayed
peak effect and a much longer duration of action than
parenterally administered, shorter-acting sedatives. to help establish the level of autonomic nervous system
A review of the actual anesthesia record is critical: depression. Evaluating the rate, depth, and character of
spontaneous ventilation helps in assessing the degree of
Check for any documentation describing the patients cerebral depression from medications, particularly if rel-
mental status just before the induction of anesthesia. atively large dosages of opioids were incorporated into
Assess the amount and timing of medications admin- the general anesthetic. Examining pupillary size and
istered during surgery, such as sedatives and opioids, responses may not yield any conclusive information in de-
particularly those causing significant CNS depression. termining the diagnosis of postoperative unconsciousness
Evaluate the duration, concentration, and discontinu- because medications, autonomic nervous system tone,
ation time of inhalational anesthetics, especially when and even eye surgery can all affect pupillary reactions
one of the more soluble agents is utilized in high concen- during emergence.
trations for a long period or if it is continued through the
end of surgery as a strategy for extubation or transport
under a deep level of anesthesia. Response to External Stimulus
Note any unusual intraoperative events such as tran-
sient airway obstruction, periods of low arterial oxygen A provocative test that can be very useful in determining
saturation, prolonged decreases in systemic blood pres- the source of unconsciousness is assessing the patients
sure, dysrhythmias, or blood loss. response to external stimuli. Verbal input should be sharp
and loud enough to slightly startle. Although obvious, it is
Discussing the patients intraoperative course (see important to use the patients correct name as part of the
Table 25.2) with the anesthesia provider often helps verbal query, as a somnolent patient may ignore a verbal
elucidate a perspective that may not be evident from stimulus that he does not realize is directed toward him.
the anesthesia record alone. Using the first name or a moniker may be more effective
Knowing the patients mental status in the operat- in eliciting a response.
ing room at the end of surgery and upon admission to If no response is elicited by verbal input, include
the PACU will distinguish whether unconsciousness was a tactile stimulus, such as a light skin pinch, trapezius
present in the operating room or appeared during or after squeeze, or a sternal rub. Tactile stimulation seems to
transfer to the PACU. A review of the PACU admission re- provoke a greater level of arousal than verbal stimulation,
port can be very useful in making a differential diagnosis.
TABLE 25.4 General Causes of Prolonged of postoperative sedation when opioids are administered
Unconsciousness After Anesthesia intraoperatively is related to the timing, route, and to-
tal dosage of agents administered. When long- and
Residual sedation from opioids intermediate-acting opioids are used, the resolution of se-
Residual sedation from inhalational aesthetics dation is slower than that caused by residual inhalational
Residual sedation from premedications or antiemetics anesthetics because opioids require hepatic metabolism
Hypercarbia or hypocarbia and/or renal excretion for clearance. Prolonged depression
Hypoxemia is especially common after the intraoperative administra-
Hypothermia tion of longer-acting opioids such as morphine, meperi-
Cerebral hypoperfusion dine, or hydromorphone. However, other opioids such
Hypoglycemia or hyperglycemia as fentanyl that usually have a shorter duration of clin-
Hyperosmolar or hypoosmolar states ical action, secondary to redistribution, can also exhibit
Carbon monoxide poisoning a long tail of sedation when high dosages or continu-
Coexisting medical illness ous infusions are administered.8 It is less likely that the
Central neurologic events shortest-acting opioids, such as alfentanil, sufentanil, and
Spurious unconsciousness particularly remifentanil, will significantly contribute to
postoperative unconscious states unless very high dosages
are given over extensive periods.
Intramuscular opioid administration leads to slower
perhaps because the sensory input is amplified through uptake and prolonged action, especially in surgical pa-
the reticular activating system. The degree of tactile tients who are hypothermic or hypovolemic. The admin-
stimulation should be reasonable and should cause no istration of intrathecal or epidural opioids can result
risk of physical injury; the use of needles or other sharp in the rostral spread of opioid into the cerebral ventri-
devices to generate pinprick sensation is unnecessary. As cles, thereby resulting in unconsciousness and ventilatory
a rule of thumb, any maneuver that you would allow depression.
a colleague to apply to you at the bedside should be Several interesting aspects of opioid pharmacology
appropriate to apply to a patient. At this early juncture, can increase their impact on prolonged unconsciousness.
assessment of other neurologic signs such as deep tendon Some opioids are metabolized to active metabolites that
reflexes or cranial nerve responses yields little value. prolong and add to central depression. Sedation induced
by opioids is usually accompanied by a decrease in spon-
taneous minute ventilation that slows the washout of
residual inhalational anesthetics, thereby prolonging se-
How Do You Assess the dation. Opioids exert a synergistic effect on the depressant
properties of sedatives, leading to a greater degree of se-
Different Causes of Prolonged dation than the sedative would have caused by itself. Also,
Unconsciousness? when compared to other agents, the intense analgesic
influence of opioids minimizes the arousal generated by
postoperative pain; this effect also blunts the response to
The general causes of a prolonged state of unconscious- tactile stimuli and accentuates the depressant effects of
ness after anesthesia are listed in Table 25.4. sedatives or antiemetics.
The administration of additional opioids in the PACU
adds to the residual depression from medications that
were provided intraoperatively. While determining the
RESIDUAL EFFECTS reason for prolonged unresponsiveness, it is appropriate
OF ANESTHETICS to assume that the patient does not perceive significant
levels of pain, and therefore, the initiation of analgesic and
Residual sedation from anesthetic agents often con-
sedative regimens, such as loading for patient-controlled
tributes to prolonged unconsciousness after surgery. Gen-
analgesia, should be delayed until the source of uncon-
erally, an unconscious state related to residual anesthesia
sciousness is determined. Assessing pain levels should be
is time-limited and characterized by a rapid and progres- avoided at this time, as well as administering analgesic
sive lessening of depth. The rate of emergence varies with medications based solely on signs indicating increased
the type of anesthetics used and the specific characteristics sympathetic nervous system activity. Generally, a patient
of the individual patient. Also, prolonged unconsciousness who generates significant tachycardia and hypertension in
from residual anesthesia almost always reflects the com- response to postoperative pain will also exhibit some de-
bined effects of several agents, each of which exhibits a gree of consciousness. In an unresponsive patient, these
different rate of resolution. physical signs can reflect critical abnormalities of oxy-
genation, ventilation, systemic perfusion, or intracranial
Opioids pressure. Administering opioid analgesics under these cir-
cumstances could result in the patients death.
Opioids are frequently implicated in producing a pro- To assess whether prolonged unconsciousness is re-
longed state of unconsciousness. The degree and duration lated to residual opioids, small, incremental, titrated
356 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
doses of intravenous naloxone (40 g increments) can anesthetics, also wane during this interval. Therefore,
be administered. Careful titration can reverse both venti- the likelihood of re-sedation after flumazenil reversal
latory depression and sedation without precipitating the is insignificant, unless a benzodiazepine overdose has
dangerous reversal of analgesia and excess sympathetic occurred.
nervous system activity that can result. Unless a patient
has received a massive opioid overdose, the ventilatory Reversal Agents
rate and level of consciousness will increase with a total
dosage of 200 g or less of intravenous naloxone; if un- There are no specific reversal agents available to coun-
consciousness persists, it is most likely not related to the teract the depressant effects of barbiturates, propofol,
depressant effects of residual opioids on the CNS. phenothiazines, and butyrophenones. The administration
of intravenous physostigmine (1.25 mg) generates a de-
Sedatives and Antiemetics gree of central arousal that can counteract, but not reverse,
depression from sedatives, antiemetics, and other depres-
The administration of sedative premedication to achieve sant medications such as baclofen.6,10,11 Application of
anxiolysis or amnesia can contribute to prolonged un- this modality is usually not warranted unless the etiol-
consciousness, particularly if long-acting sedatives (e.g., ogy of sedation is unclear and immediate resolution is
pentobarbital, hydroxyzine, promethazine, lorazepam) important.
are administered orally, rectally, or intramuscularly. The
possibility of unacknowledged self-premedication by Inhalational Anesthetics
patients with long-acting oral sedatives or other psy-
chotropic medications should also always be considered. High alveolar partial pressures of residual volatile anes-
Even the judicious use of intravenous midazolam before thetic agents can sometimes leave a patient deeply sedated
induction can, in some patients, affect the level of con- early in the postoperative course. This phenomenon oc-
sciousness in the PACU. The administration of sedatives curs predominantly after extended exposure to high con-
or antiemetics as part of the anesthetic regimen like- centrations of a more soluble agent such as isoflurane.12
wise adds even more profound depression in the PACU, During long surgical procedures, significant amounts of
especially if given toward the end of surgery. soluble anesthetic agents build up in tissues that have
Parenteral medications such as propofol, short- lower perfusion levels, consequently leading to a more
acting barbiturates, or etomidate by frequent bolus or gradual washout after discontinuation. Obese patients
continuous infusion can generate high circulating serum may be at particular risk of prolonged sedation after long
levels and resultant redistribution of high concentrations procedures, given their relatively high proportion of body
of drug into the tissues. The delayed excretion of fat. Also, if high inspired concentrations are continued
medication can cause or accentuate delayed awakening through the end of surgery to maintain bronchodilation
after discontinuation.9 or to facilitate a deep extubation, alveolar partial pres-
Antiemetics, such as droperidol, prochlorperazine, or sures and level of sedation will naturally be higher during
scopolamine, have sedative side effects that can augment the initial recovery period.
the residual sedation from anesthetics. Other antiemetic It is unlikely that low solubility agents, such as
agents such as dexamethasone and serotonin-blocking sevoflurane and desflurane, are the primary cause of per-
agents (e.g., ondansetron, dolasetron) do not exhibit sistent unconsciousness because they are eliminated very
significant sedative side effects, and therefore do not rapidly, soon after their discontinuation in the operating
contribute to postoperative unconsciousness. room. However, these agents may contribute to seda-
Evaluating the contribution of sedatives or antiemet- tion when combined with other, longer-acting parenteral
ics to prolonged unconsciousness is not quite as straight- medications such as opioids. Nitrous oxide is seldom im-
forward as evaluating the influence of opioids. If a patient plicated, because of its low solubility and relatively weak
has received benzodiazepines, intravenous flumazenil, a anesthetic properties.
competitive benzodiazepine antagonist, can be given in Considering the inevitable washout of anesthetic
titrated, incremental dosages of 0.1 mg every 2 minutes. In vapor during normal breathing, it would be unusual
the perioperative setting, <1 mg is typically needed to re- that the residual effects from any volatile inhalational
verse residual benzodiazepine effect. Flumazenil directly anesthetics would be the primary cause of an unconscious
reverses the sedation caused by midazolam, diazepam, state that lasts over 30 minutes after discontinuation of
lorazepam, and other benzodiazepines, although its dura- inhalational anesthesia. However, if the residual volatile
tion of action is relatively short. If unconsciousness due agent should, indeed, significantly contribute to prolonged
to benzodiazepines is reversed by intravenous flumazenil, unconsciousness, then the culprit can be easily detected
it is theoretically possible that benzodiazepines duration from breath odor or through a quantitative analysis of
could exceed that of flumazenil reversal, leading to the expired gas using standard intraoperative monitors. There
return of unconsciousness an hour or so after reversal. are no specific agents available that will reverse residual
However, the dosages of benzodiazepines used in con- sedation from volatile anesthetics.
temporary anesthesia care are low enough that the serum If it is essential to assess whether residual inhalational
concentration decreases significantly during the effective anesthesia is causing prolonged unconsciousness, the
duration of flumazenil. In addition, the sedative effects administration of intravenous physostigmine can be
of other medications, such as opioids and inhalational tried to counteract the sedative effects. However, simply
C H A P T E R 2 5 / C O M A A N D D E L AY E D E M E R G E N C E 357
allowing adequate time for the inhalational agent to wash regional technique, an unconscious state can reflect either
out through alveolar ventilation will provide sufficient persistent local anesthetic effects on the CNS or postictal
differentiation. If the administration of appropriate doses depression secondary to seizure activity.13 Adjunctive
of naloxone, flumazenil, and physostigmine does not agents, such as clonidine, are sometimes included in
elicit a response, unconsciousness is not likely related the local anesthetic solution and can independently exert
to sedation from residual anesthetic medications. central depression and augment unconsciousness. The ad-
ministration of naloxone has been shown to be efficacious
in reversing central depression related to clonidine.14 The
Neuromuscular Relaxants intra-arterial injection of relatively small volumes of local
Neuromuscular agents do not have any significant seda- anesthetic solution during blocks located near the head
tive or analgesic properties and therefore would not and neck can produce profound obtundation.15
exacerbate postoperative unconsciousness. Rarely, pro- The inadvertent subarachnoid injection of local anes-
found residual neuromuscular blockade can mimic un- thetic during placement of an epidural or other block can
consciousness during recovery by completely eliminating generate a total spinal that incorporates relatively high
any voluntary motor response to verbal or tactile stimuli concentrations of local anesthetic and adjunctive medica-
in a conscious but completely paralyzed patient. Although tions directly into the intracranial cerebrospinal fluid.16
unlikely, this degree of neuromuscular blockade can occur This can produce a dense, but reversible, postoperative
after gross overdose with neuromuscular blocking agents coma with physical signs similar to those seen after se-
or if reversal agents are omitted. Similarly, complete vere cerebral anoxia.17,18 Again, relatively small volumes
postoperative paralysis is conceivable in patients with of anesthetic solution can cause profound central depres-
unrecognized neuromuscular disease or in those exhibit- sion if injected into the cervical or thoracic subarachnoid
ing phase II blockade caused by excessive succinylcholine space. The rostral spread of intrathecally administered
administration or pseudocholinesterase deficiency. opioids can also prolong an unconscious state, and is usu-
Observation of spontaneous ventilation, deep tendon ally accompanied by profound depression of spontaneous
reflex activity, or any purposeful motion categorically ventilation.19
eliminates residual paralysis as an explanation for lack of The risk of physical injury to the intracranial struc-
responsiveness. If neuromuscular function is sufficiently tures after a spinal or epidural anesthetic is small, but
recovered to allow these manifestations of skeletal neu- real. The loss of cerebrospinal fluid with dural punc-
romuscular function, a conscious patient will be able to ture can cause displacement of cranial contents, tearing
generate a motor response to stimuli, no matter how weak. of small bridging veins, and formation of a subdural
Moreover, if a patient is conscious but completely para- hematoma.20,21 Dural puncture in a patient with unrec-
lyzed, he or she would likely exhibit panic and signs of ognized increased intracranial pressure and obstructive
sympathetic nervous system hyperactivity such as sweat- hydrocephalus can rapidly lead to herniation of the ten-
ing, tearing, tachycardia, hypertension, and dysrhythmias. torium and brain death. Both increased intracranial pres-
If any doubt exists concerning the level of residual neu- sure and obstructive hydrocephalus could well present as
romuscular blockade, a simple check with a peripheral inexplicable unconsciousness after regional anesthesia.
nerve stimulator will reveal whether a motor response is
possible or not. Again, any motor response to peripheral
nerve stimulation that is not produced by direct stimu-
lation of the muscle eliminates residual neuromuscular What Are Other Causes
blockade from the differential diagnosis of unrespon-
siveness.
of Prolonged Unconsciousness?
Once it has been determined that the residual effects of
Regional Anesthesia and Monitored medications are not related to prolonged unconscious-
Anesthesia Care ness, other less common causes need to be considered.
Serious inadequacy of ventilation or oxygenation, enough
After surgery with regional anesthesia or monitored anes- to cause deviations of the PACO2 and/or the PaO2 from
thesia care, a prolonged unconscious state more likely normal physiologic ranges, can primarily produce or con-
reflects a serious, ongoing physiologic problem or a sig- tribute to a state of unconsciousness during recovery from
nificant intraoperative misadventure. Delayed emergence anesthesia.
rarely occurs as a consequence of these anesthetic tech-
niques themselves. Using deep intraoperative sedation
and/or opioid analgesia to allay anxiety or to supplement
a regional technique can result in prolonged unconscious- HYPERCARBIA AND
ness during recovery. Dense residual analgesia after a RESPIRATORY ACIDEMIA
successful regional technique can eliminate postoperative
discomfort that would otherwise offset this sedation and The impact of acute hypercarbia and respiratory acidemia
maintain postoperative arousal. on the brain is somewhat more pronounced than
After a patient exhibits a toxic reaction from an in- that of acute metabolic acidemia because carbon diox-
travascular injection or uptake of local anesthetic during a ide molecules diffuse across the bloodbrain barrier
358 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
oxygen applied by facemask. These interventions entail hypoperfusion. These conditions can produce a prolonged
minimal risk and can be life-saving. state of unconsciousness after surgery.32
As with hypoxemia, prolonged unconsciousness re-
lated to hypotension or cerebral hypoperfusion constitutes
a medical emergency that requires the cerebral perfusion
HYPEROXIA pressure be immediately raised into a more appropriate
In routine clinical circumstances, hyperoxia has no range. Short-term, aggressive, intravenous hydration, and
significant impact on the level of consciousness. Prolonged the judicious use of pressor agents such as ephedrine
ventilation with 100% oxygen can produce early signs or neosynephrine, are useful for establishing this as are
of pulmonary oxygen toxicity accompanied by vague atropine or glycopyrrolate if hypotension is caused by
agitation or slight changes in mentation, but impact bradycardia. Addressing the primary cause underlying
on level of consciousness is minimal. Ventilation with the hypotension is critical.
100% oxygen under hyperbaric conditions generates very
high cerebral oxygen partial pressures and elicits seizure
activity that can mimic unconsciousness or cause postictal HYPOGLYCEMIA
depression. However, this phenomenon does not occur at
ambient atmospheric pressure. Acute hypoglycemia during recovery generally occurs in
patients with preexisting diabetes or glucose intolerance,
and is a result of excessive endogenous insulin secre-
tion or exogenous insulin administration in conjunction
HYPOTENSION AND with inadequate glucose infusion. Severe hypoglycemia
CEREBRAL HYPOPERFUSION produces unconsciousness by depriving the CNS of essen-
tial energy substrates necessary for neuronal function.33
An adequate PaO2 , hemoglobin content, and oxygen A realistic suspicion that unresponsiveness may reflect
saturation reading does not necessarily guarantee that hypoglycemia should trigger an immediate empirical trial
cerebral perfusion pressure and blood flow are sufficient of intravenous 50% dextrose. Under these conditions, it
to provide enough oxygen to meet CNS requirements. is inappropriate to delay the administration of glucose
Any condition that interferes with cerebral perfusion can until hypoglycemia is corroborated by a serum glucose
reduce the delivery of oxygen and energy substrates to determination. Caution should be used when making a
the CNS, consequently generating cerebral anoxic injury provisional diagnosis of hypoglycemia as a cause of ob-
similar to that produced by severe hypoxemia. The most tundation in an unconscious patient; if unconsciousness
common cause of acute cerebral hypoperfusion during is secondary to an acute anoxic or ischemic event, the ad-
or after surgery is a severe reduction in systemic blood ministration of emergency glucose can produce iatrogenic
pressure, often secondary to hypovolemia, decreased hyperglycemia that can accentuate the degree of central
peripheral vascular resistance, dysrhythmia, or an acute neurologic injury and worsen the prognosis for recovery.34
myocardial event.
The systemic blood pressure at which critical cerebral
hypoperfusion occurs varies among individuals, and can HYPERGLYCEMIA
be relatively high in patients with cerebrovascular dis-
ease, preexisting hypertension, or increased intracranial Acute hyperglycemia can interfere with the level of con-
pressure.29,30 It is, therefore, important to assess base- sciousness by increasing serum osmolarity compared
line blood pressure as part of the differential diagnosis of to that in cerebral tissues. The resulting osmolar gra-
prolonged unconsciousness. In the absence of these condi- dient pulls fluid from the intracellular and interstitial
tions, a blood pressure sufficient to generate a detectable spaces to the intravascular space of the CNS, essentially
peripheral pulse and a measurable pulse oximeter reading producing acute cerebral cellular dehydration. Acutely
will usually provide enough cerebral perfusion to main- high serum glucose concentrations can produce a hyper-
tain some level of consciousness, although mentation may glycemic, hyperosmolar coma.35 Concurrent ketoacidosis
be clouded. and metabolic acidemia can worsen the impact on the
Cerebral perfusion can sometimes be globally com- state of consciousness. Determination of serum glucose
promised in spite of adequate aortic perfusion pressures. and serum osmolarity will quickly reveal whether severe
The risk of cerebral hypoperfusion is increased when hyperglycemia is contributing to a prolonged unconscious
surgery is performed in the sitting position, especially state. Acute perioperative hyperglycemia is generally
with extreme flexion of the neck.31 Compression of the treated with hydration and intravenous insulin, often in
carotid arteries from external contact or a hematoma conjunction with a potassium infusion.
in the neck can also impede cerebral perfusion, partic-
ularly in patients with severe cerebrovascular disease.
Intraoperative interference with cerebral venous return HYPONATREMIA
instigated by external compression of the jugular veins,
high intrathoracic pressures, jugular venous cannulation, The level of consciousness can also be impacted by an
or extreme head and neck positioning can lead to cere- acute hypoosmolar state (<260 mOsm per L). The re-
bral edema, increased intracranial pressure, and cerebral sulting osmolar gradients cause fluids to shift from the
360 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
intravascular compartment to the interstitial and intracel- a decreased body temperature contributing to an uncon-
lular spaces of the CNS, in effect producing acute cerebral scious state is greater in traumatized or burned patients,
edema.36 In surgical patients, the uptake of hypotonic patients who exhibit significant preoperative hypother-
irrigating solutions during hysteroscopy, transurethral mia, poorly rewarmed patients after cardiopulmonary
prostatic surgery, joint irrigations, or similar procedures bypass, or in patients whose intraoperative tempera-
can produce acute hyponatremia (Na <125 mEq per L), ture maintenance is suboptimal. A differential diagnosis
thereby markedly reducing the level of consciousness.37,38 simply requires an accurate determination of core body
Impairment of consciousness in these cases is likely ex- temperature.
acerbated by glycine and ammonia toxicity as well.39,40
Rarely will inappropriate antidiuretic hormone secretion
cause significant dilutional hyponatremia in postoperative
patients with head trauma or in those who have under- How Do Coexisting Diseases
gone craniotomy or transsphenoidal pituitary surgery.41
Preoperative polydipsia, water intoxication, or fresh wa- Affect Delayed Awakening?
ter drowning can lead to similar dilutional states.42
Severe, symptomatic hyponatremia seldom appears as A wide variety of coexisting diseases can be implicated in
an idiopathic finding in otherwise healthy postoperative delayed awakening. Patients with porphyrias can exhibit
patients.43 The excessive intravenous administration of unconsciousness after exposure to barbiturates, propofol,
free water is an iatrogenic cause of dilutional hypona- and other classes of medications.49,50 Patients with
tremia.44 Another cause of cerebral impairment from Hunter syndrome and other mucopolysaccharide storage
hypo-osmolarity is the postdialysis disequilibrium syn- diseases are also prone to prolonged unconsciousness
drome. The potential contribution of hyponatremia and after anesthesia.51 Other causes that predispose to delayed
hypo-osmolarity to prolonged postoperative unconscious- emergence or persistent unconsciousness are related to a
ness can be easily assessed with serum electrolyte and large number of inborn errors of metabolism.52
osmolarity determinations. If severe hyponatremia is im- Many critical systemic illnesses affect cerebral func-
plicated, care should be taken to restore the serum sodium tion and can therefore prolong unconsciousness after
gradually; too rapid a correction of hyponatremia can pre- surgery.44,53 Clinically hypothyroid patients can be espe-
dispose to the development of central pontine myelinosis, cially slow to emerge from general anesthesia, whereas
which will compound the neurologic problem.45 patients with obstructive sleep apnea can be particu-
larly sensitive to the depressant effects of opioids or
inhalational anesthetics. Patients suffering from kidney
HYPERNATREMIA insufficiency with unresolved uremic encephalopathy are
at high risk for delayed emergence after surgery and
Hypernatremia can produce a hyperosmolar state similar anesthesia, as are those with liver failure who exhibit
in clinical presentation to that caused by hyperglycemia. hepatic encephalopathy.54 Persistent coma can occur af-
This degree of hypernatremia often occurs after a period ter liver transplantation, accompanied by cerebral edema
of marked dehydration or in patients suffering from dia- or intracerebral hemorrhage and increased intracranial
betes insipidus characterized by insufficient secretion of pressure.55,56 Neurologic complications and coma are
antidiuretic hormone and loss of free water through the also relatively common after cardiac transplantation.57,58
kidneys.46 Unconsciousness due to a hypernatremic, hy- Systemic sepsis significantly impairs mental status, both
perosmolar state is rare in postoperative patients, given directly and secondarily through hypotension.
the amount of intravenous isotonic hydration that is usu-
ally administered in the perioperative period. However,
pathologic conditions such as hydatid disease can result in
persistent unconsciousness related to hyperosmolarity.47 How Is Spurious
Unconsciousness Evaluated?
HYPOTHERMIA
Under certain circumstances, patients can exhibit mini-
Reduction of core body temperature depresses the level mal levels of arousal after surgery and anesthesia in the
of arousal and increases the impact of depressant med- absence of any abnormal physiology. Children who were
ications on the level of consciousness.48 Hypothermia exhausted before surgery are sometimes difficult to arouse
below 33 C significantly impairs consciousness, whereas after anesthesia, especially if sleep patterns are disrupted
a core temperature below 30 C produces coma, with fixed by emergency surgery at night. Other patients who are
pupillary dilation and areflexia.42 accustomed to chronically using CNS stimulants (such as
Hypothermia is rarely a primary cause of postop- coffee) may be difficult to arouse after surgery.
erative unconsciousness. Nonetheless, a moderately de- Occasionally, during emergence from anesthesia, a
creased core body temperature can certainly augment the patient will feign unresponsiveness for secondary gain or
depression of unconsciousness from other causes in pa- will suffer a hysterical conversion reaction that presents
tients recovering from surgery and impair the clearance of as a state of unconsciousness.59 Factitious disorder
medications that obtund consciousness. The possibility of is another rare cause of spurious unconsciousness.60
C H A P T E R 2 5 / C O M A A N D D E L AY E D E M E R G E N C E 361
Differentiating between an actual and spurious uncon- an underlying seizure disorder, metabolic abnormalities,
scious state is a clinical challenge. In a supine patient who side effects of anesthetic medications, or eclampsia.79,80
is feigning unconsciousness, dropping the patients hand Physical manifestations of the seizure can be missed be-
toward the face will often result in the arm falling to the cause of the subtle presentation, neuromuscular paralysis,
side rather than toward the nose as gravity would normally or its misdiagnosis as agitation or tremor. Seizure activity
direct it. Of course, care should be taken to guard the pa- is particularly difficult to diagnose in neonates.81 Uncon-
tients face should the hand actually fall onto the nose in sciousness related to seizure activity could reflect ongoing
a truly unconscious patient. Another useful ploy is to ver- chaotic electrical activity in the brain, postictal depres-
bally announce a plausible bedside diagnostic test that is sion, or cerebral ischemia secondary to hypoventilation
allegedly pathognomonic of coma. The test should relate and/or high cerebral metabolic rates.
a discreet sensory stimulus (e.g., stroking the forehead) The possibility of unrecognized head trauma, in-
to an unrelated reflex motor response such as bilateral tracranial hemorrhage, vasospasm, or cerebral edema
clenching of the fists. If light pressure is applied to the must be considered in trauma patients or those recover-
forehead and the fists clench, the patient obviously heard ing from intracranial surgery.8284 After long intracranial
and processed the verbal misinformation, and is actually surgical procedures, patients sometimes awaken very
conscious. A more definitive differentiation of feigned un- slowly in the absence of complicating factors.85 De-
consciousness can be gleaned through assessment of a layed emergence correlates with intracranial tumor size
bispectral index or a formal EEG analysis. and extent of the surgical intervention.86 Any condition
that leads to increased intracranial pressure can incite
a prolonged unconscious state after anesthesia.32 The
possibility of central neurologic problems should also
When Should a Neurologic be considered in patients who exhibit significant intra-
operative or postoperative hypertensive episodes and in
Examination Be Performed? preeclamptic parturients.
A relation between delayed awakening and child
abuse has also been described.87 Postoperative tension
PRIMARY NEUROLOGIC pneumocephalus is another rare cause of prolonged un-
CAUSES consciousness in neurosurgical patients.88 Patients who
suffer direct mechanical trauma to the brain during surgi-
The entire process of making a differential diagnosis cal interventions can manifest prolonged unconscious-
outlined previously can be completed in a matter of ness. Intracerebral structures can be damaged during
minutes. If a diagnosis remains elusive after all the more sphenoid sinus procedures or middle ear procedures. In
common etiologies of postoperative unconsciousness have patients with facial fractures or those who have under-
been ruled out, a neurologic evaluation should be obtained gone transsphenoidal surgery, the inadvertent passage of
as soon as possible. Neurologic consultation should not nasogastric or nasotracheal tubes through the cribiform
be delayed to secure noninvasive CNS imaging studies. plate into the intracranial cavity can obviously produce
Many primary neurologic events that produce per- severe brain injury intraoperatively.89,90
sistent unconsciousness in postoperative patients are Postoperative cerebrovascular accidents are relatively
embolic in nature. The risk of cerebral thromboembolism rare and often occur later in the postoperative course,
and related neurologic dysfunction is highest after prox- although the possibility that prolonged unconsciousness
imal major vascular, carotid, cardiac, or invasive neck could reflect a new stroke should be considered in any
surgery.6165 Patients who have undergone internal jugu- patient.9193 The probability that cerebral ischemia or
lar or subclavian cannulation are also at risk, as are those stroke can be caused by excessive hyperventilation during
with atrial dysrhythmias,66 septal aneurysms,67 carotid mechanical ventilation is real. Undiagnosed cerebral
bruits, or hypercoaguable states. Paradoxical fat, throm- anoxia from intraoperative hypotension, dysrhythmias, or
bus, or air embolism through a patent foramen ovale68 hypoxemia must always be evaluated as a potential cause.
or right-to-left intracardiac or pulmonary shunts can also
migrate into the cerebral circulation and cause profound
disruption of CNS function.6971 Even in the absence of
an intracardiac defect, venous fat embolization that can What Should Be Done When a
occur after trauma or during orthopedic surgical pro-
cedures can produce profound neurologic side effects Previously Responsive Patient
and present as failure to awaken from anesthesia.7274 Develops the Acute Onset of
Similarly, patients who suffer blunt chest trauma, baro-
trauma during positive-pressure ventilation, or airway Unconsciousness?
trauma during instrumentation can exhibit prolonged un-
consciousness from air embolism into the cerebral arterial If a previously responsive patient lapses into uncon-
circulation.75,76 sciousness after emergence from anesthesia, the need
Occasionally, postoperative unresponsiveness reflects for determining an immediate differential diagnosis is
subclinical grand mal seizure activity.77,78 Perioperative much more urgent. It is very unlikely that a previously
seizures can be incited by delirium tremens, emergence of responsive patient will suddenly lose consciousness due
362 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
to the residual effects of intraoperative anesthetic medica- 2. Zelcer J, Wells DG. Anaesthetic-related recovery room
tions, considering that they steadily decrease throughout complications. Anaesth Intensive Care. 1996;15:168.
the recovery period. There are no realistic scenarios in 3. Zelcer J, Wells DG. Anaesthetic-related recovery room
which a phenomena such as renarcotization, bi-phasic complications. Anaesth Intensive Care. 1987;15:168.
4. Sebel PS, Bowdle TA, Ghoneim MM, et al. The incidence
responses to opioids, or recurarization will generate a
of awareness during anesthesia: A multicenter United States
clinical picture of sudden unconsciousness. study. Anesth Analg. 2004;99:883.
Medication-induced loss of consciousness can occur 5. Lyew MA, Mondy C, Eagle S, et al. Hemodynamic instability
in a postoperative patient if a large dose of parenteral and delayed emergence from general anesthesia associated
opioid or sedative medication was administered just with inadvertent intrathecal baclofen overdose. Anesthesiol-
before a PACU admission or early in the postoperative ogy. 2003;98:265.
course, or if it was left in the IV tubing deadspace and 6. Gomar C, Carrero EJ. Delayed arousal after general anesthe-
inadvertently flushed in after arrival to the PACU. Also, sia associated with baclofen. Anesthesiology. 1994;81:1306.
the acute withdrawal of a highly noxious stimulus in the 7. Crowe S, McKeating K. Delayed emergence and St. Johnss
PACU (e.g., tracheal extubation) could allow underlying wort. Anesthesiology. 2002;96:1025.
8. Mutch WAC, Ringaert KRA, Ewert FJ, et al. Continuous
residual depression from intraoperative medications to
opioid infusions for neurosurgical procedures: A double-
emerge and reduce the level of consciousness. However, blind comparison of alfentanil and fentanyl. Can J Anaesth.
these circumstances are rare. Loss of consciousness in a 1991;38:710.
previously responsive patient most likely reflects a life- 9. Apfelbaum JL, Grasela TH, Hug CC Jr, et al. The initial clini-
threatening physical change that is acutely affecting the cal experience of 1819 physicians in maintaining anesthesia
patients CNS function. Such an occurrence should be with propofol: Characteristics associated with prolonged
considered a medical emergency until proved otherwise. time to awakening. Anesth Analg. 1993;77:S10.
10. Artru AA, Hui GS. Physostigmine reversal of general anesthe-
sia for intraoperative neurological testing: Associated EEG
changes. Anesth Analg. 1986;65:1059.
11. Mauret P, Backman S, Bonhomme V, et al. Physostigmine
KEY POINTS reverses propofol-induced unconsciousness and attenuation
1. If a patient does not generate a meaningful, directed of the auditory steady state response and bispectral index in
response to ordinary levels of verbal or tactile stimuli human volunteers. Anesthesiology. 2000;93:708.
within 30 minutes of PACU admission, he or she is 12. Todd MM, Warner DS, Sokoll MD, et al. A prospective, com-
parative trial of three anesthetics for elective supratentorial
exhibiting prolonged unconsciousness that requires a
craniotomy. Anesthesiology. 1993;78:1005.
differential diagnosis.
13. Auroy Y, Narchi P, Messiah A, et al. Serious complications
2. A state of unconsciousness increases the risk of related to regional anesthesia. Anesthesiology. 1997;87:479.
aspiration, hypoventilation, airway obstruction, and 14. Botero M, Enneking FK. Reversal of prolonged unconscious-
incidental injury. ness by naloxone after an intravascular injection of a local
3. Most instances of prolonged unconsciousness are pro- anesthetic and clonidine. Anesth Analg. 1999;88:1185.
duced by opioids, although sedatives, antiemetics, 15. Ross S, Scarborough CD. Total spinal anesthesia following
and residual inhalational anesthetics can all con- brachial plexus block. Anesthesiology. 1973;39:458.
tribute. 16. Palkar NV, Boudreaux RC, Mankad AV. Accidental total
4. Selective titrated reversal of sedation from opioids spinal block: A complication of an epidural test dose. Can J
with naloxone or from benzodiazepines with flumaze- Anaesth. 1992;39:1058.
17. Park PC, Berry PD, Larson MD. Total spinal anesthesia
nil helps in determining both the differential diagno-
following epidural saline injection after prolonged epidural
sis and therapy of prolonged unconsciousness. anesthesia. Anesthesiology. 1998;89:1267.
5. Residual neuromuscular blockade seldom, if ever, 18. Lubenow T, Keh-Wong E, Kristof K, et al. Inadvertent
contributes to postoperative unresponsiveness. subdural injection: A complication of an epidural block.
6. Once sedation from anesthetic medications is ruled Anesth Analg. 1988;67:175.
out, consider hypercarbia, hypoxemia, hypotension, 19. Aromaa U, Lahdensuu M, Cozanitis DA. Severe complica-
hypoglycemia, hyponatremia, hyperosmolarity, or tions associated with epidural and spinal anaesthesias in
spurious unconsciousness as potential causes of a Finland 19871993. Acta Anaesthesiol Scand. 1997;41:445.
prolonged unconscious state. 20. Vaughan DJA, Stirrup CA, Robinson PN. Cranial subdural
7. If a firm etiology is not discovered once the likely haematoma associated with dural puncture in labour. Br J
Anaesth. 2000;84:518.
causes of unconsciousness have been evaluated, con-
21. Mercieri M, Paolini A, Lupoi D, et al. Postpartum cerebral
sult a neurologist to evaluate for a primary CNS event. ischaemia after accidental dural puncture and epidural blood
8. Consider any acute loss of consciousness that occurs patch. Br J Anaesth. 2003;90:98.
after admission to the PACU as a medical emergency 22. Rafferty GF, Saisch SGM, Gardner WN. Relation of hypocap-
that requires immediate intervention until proven nic symptoms to the rate of fall of end-tidal CO2 in normal
otherwise. subjects. Respir Med. 1992;86:335.
23. Skippen P, Seear M, Poskitt K, et al. The effect of
hyperventilation on regional blood flow in head-injured
REFERENCES
children. Crit Care Med. 1997;25:1402.
1. Hines R, Barash PG, Watrous G, et al. Complications 24. Fu ES, Downs JB, Schweiger J, et al. Supplemental oxygen
occurring in the postanesthesia care unit: A survey. Anesth impairs detection of hypoventilation by pulse oximetry.
Analg. 1992;74:503. Chest. 2004;126:1552.
C H A P T E R 2 5 / C O M A A N D D E L AY E D E M E R G E N C E 363
25. Downs JB. Has oxygen administration delayed appropriate 48. Sessler DI. Complications and treatment of mild hypother-
respiratory care? Fallacies regarding oxygen therapy. Respir mia. Anesthesiology. 2001;95:534.
Care. 2003;48:611. 49. Parikh SS, Chung F. Prolonged loss of consciousness
26. Lieberman JA, Weiskopf RB, Kelley SD, et al. Critical and elevated porphyrins following propofol administration.
oxygen delivery in conscious humans is less than 7.3 ml Anesthesiology. 1998;89:1029.
O2 kg1 min1 . Anesthesiology. 2000;92:407. 50. Suarez JI, Cohen ML, Larkin J, et al. Acute intermit-
27. Berry PD, Sessler DI, Larson MD. Severe carbon monox- tent porphyria: Clinicopathologic correlation. Report of
ide poisoning during desflurane anesthesia. Anesthesiology. a case and review of the literature. Neurology. 1997;48:
1999;90:613. 1678.
28. Woehlck HJ, Dunning M III, Raza T, et al. Physical factors 51. Kreidstein A, Boorin MR, Crespi P, et al. Delayed awakening
affecting the production of carbon monoxide from anesthetic from general anaesthesia in a patient with Hunter syndrome.
breakdown. Anesthesiology 2001;94:453. Can J Anaesth. 1994;41(5 Pt 1):423.
29. Jin F, Chung F. Minimizing perioperative adverse events in 52. Neuvonen PT, van den Berg AA. Postoperative coma in a
the elderly. Br J Anaesth. 2001;87:608. child with carnitine palmitoyltransferase I deficiency. Anesth
30. Pietropaoli JA, Rogers FB, Shackford SR, et al. The Analg. 2002;92:646.
deleterious effects of intraoperative hypotension on outcome 53. Bleck TP, Smith MC, Pierre-Louis SJ, et al. Neurological
in patients with severe head injuries. J Trauma. 1992;33: complications of critical medical illnesses. Crit Care Med.
403. 1993;21:98.
31. Porter JM, Pidgeon C, Cunningham AJ. The sitting position 54. Wijdicks EFM, Plevak DJ, Rakela J, et al. Clinical and
in neurosurgery: A critical appraisal. Br J Anaesth. 1999;82: radiologic features of cerebral edema in fulminant hepatic
117. failure. Mayo Clin Proc. 1995;70:119.
32. Valero R, Faabregas N, Laopez A, et al. Neuroendoscopic 55. Wiesner RH, Wijdicks EFM, Krom RAF. Neurotoxicity in
intracranial pressure levels over 30 mmHg are associated liver transplant recipients with cyclosporine immunosup-
with high postoperative morbidity. Br J Anaesth. 1999; pression. Neurology. 1995;45:1962.
82:87. 56. Wijdicks EFM, de Groen PC, Wiesner RH, et al. Intracerebral
33. Malouf R, Brust JC. Hypoglycemia: Causes, neurological hemorrhage in liver transplant recipients. Mayo Clin Proc.
manifestations, and outcome. Ann Neurol. 1985;17:421. 1995;70:443.
34. Wass CT, Lanier WL. Glucose modulation of ischemic brain 57. McManus RP, OHair DP, Schweiger J, et al. Cyclosporine-
injury: Review and clinical recommendations. Mayo Clin associated central neurotoxicity after heart transplantation.
Proc. 1996;71:801. Ann Thorac Surg. 1992;53:326.
35. Brenner WI, Lansky Z, Engelman RM, et al. Hyperosmolar 58. Adair JC, Call GK, OConnell JB, et al. Cerebrovascular
coma in surgical patients: An iatrogenic disease of increasing syndromes following cardiac transplantation. Neurology.
incidence. Ann Surg. 1973;178:651. 1992;42:819.
36. Oh MS, Carroll HJ. Disorders of sodium metabolism: Hyper- 59. Adams AP, Goroszeniuk T. Hysteria: A cause of failure to
natremia and hyponatremia. Crit Care Med. 1992;20:94. recover after anaesthesia. Anaesthesia. 1991;46:932.
37. Van Boven MJ, Singelyn F, Donnez J, et al. Dilutional 60. Albrecht RF, Wagner SR, Leicht CH, et al. Factitious disorder
hyponatremia associated with intrauterine endoscopic laser as a cause of failure to awaken after general anesthesia.
surgery. Anesthesiology. 1989;71:449. Anesthesiology. 1995;83:201.
38. Gehring H, Nahm W, Baerwald J, et al. Irrigation fluid 61. Ballotta E, Da Giau G, Saladini M, et al. Carotid en-
absorption during transurethral resection of the prostate: darterectomy in symptomatic and asymptomatic patients
Spinal vs. general anaesthesia. Acta Anaesthesiol Scand. 1999; aged 75 years or more: perioperative mortality and stroke
43:458. risk rates. Ann Vasc Surg. 1999;13:158.
39. Ichai C, Ciais JF, Roussel LJ, et al. Intravascular ab- 62. Roach GW, Kanchuger M, Mora Mangano C, et al. Adverse
sorption of glycine irrigating solution during shoulder cerebral outcomes after coronary artery bypass surgery.
arthroscopy: A case report and follow-up study. Anesthe- N Engl J Med. 1996;335:1857.
siology. 1996;85:1481. 63. Arrowsmith JE, Grocott HP, Reves J, et al. Central nervous
40. Kirwan PH, Ludlow J, Makepeace P, et al. Hyperammon- system complications of cardiac surgery. Br J Anaesth. 2000;
aemia after transcervical resection of the endometrium. Br J 84:378.
Obstet Gynaecol. 1993;100:603. 64. Hogue CW, Sundt T, Barzilai B, et al. Cardiac and neuro-
41. Soroker D, Ezri T, Lurie S, et al. Symptomatic hypona- logical complications identify risks for mortality for both
traemia due to inappropriate antidiuretic hormone secretion men and women undergoing coronary artery bypass graft
following minor surgery. Can J Anaesth. 1991;38:225. surgery. Anesthesiology. 2001;95:1074.
42. Golden FS, Tipton MJ, Scott RC. Immersion, near-drowning 65. Hogue CW, De Wet CJ, Schechtman KB, et al. The impor-
and drowning. Br J Anaesth. 1997;79:214. tance of prior stroke for the adjusted risk of neurological
43. Arieff AI. Hyponatremia, convulsions, respiratory arrest, and injury after cardiac surgery for women and men. Anesthesi-
permanent brain damage after elective surgery in healthy ology. 2003;98:823.
women. N Engl J Med. 1986;314:1529. 66. Hanson EW, Lowson SM. Atrial fibrillation and thromboem-
44. Wijdicks E. Neurological complications in critically ill bolism. Anesth Analg. 1998;87:217.
patients. Anesth Analg. 1996;83:411419. 67. Gallet B, Malergue MC, Adam C, et al. Atrial septal
45. Laureno R. Central pontine myelinolysis following rapid aneurysm: A potential cause of systemic embolism. Br
correction of hyponatremia. Ann Neurol. 1983;13:232. Heart J. 1985;53:292.
46. Daugirdas JT, Kronfol NO, Tzamaloukas AH, et al. Hyper- 68. Fisher DC, Fisher EA, Budd JH, et al. The incidence
osmolar coma: Cellular dehydration and the serum sodium of patent foramen ovale in 1,000 consecutive patients:
concentration. Ann Intern Med. 1989;110:855. A contrast transesophageal echocardiography study. Chest.
47. Rakic M, Vegan B, Sprung J, et al. Acute hyperosmolar 1995;107:1504.
coma complicating anesthesia for hydatid disease surgery. 69. Dive AM, Dubois PE, Ide C, et al. Paradoxical cerebral fat
Anesthesiology. 1994;80:1175. embolism, an unusual cause of persistent unconsciousness
364 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
after orthopedic surgery. Anesthesiology. 2002;96: 83. Solenski NJ, Haley EC, Kassell NF, et al. Medical complica-
1029. tions of aneurysmal subarachnoid hemorrhage: A report of
70. Liu S, Holley HS, Stulberg SD, et al. Failure to awaken the multicenter, cooperative aneurysm study. Crit Care Med.
after general anaesthesia secondary to paradoxical venous 1995;23:1007.
embolus. Can J Anaesth. 1991;39:351. 84. Lee JH, Martin NA, Alsina G, et al. Hemodynamically
71. Weiss SJ, Cheung AT, Stecker MM, et al. Fatal paradoxical significant cerebral vasospasm and outcome after head
cerebral embolization during bilateral knee arthroplasty. injury: A prospective study. J Neurosurg. 1997;87:221.
Anesthesiology. 1996;84:721. 85. Manninen PH, Raman SK, Boyle K, et al. Early postoperative
72. Byrick RJ. Fat embolism and neurological dysfunction. complications following neurosurgical procedures. Can J
Anesth Analg. 1999;88:1427. Anaesth. 1999;46:7.
73. Ozelsel TJP, Hein T, Marcel RJ, et al. Delayed neurological 86. Schubert A, Mascha EJ, Bloomfield EL, et al. Effect of cranial
deficit after total hip arthroplasty. Anesth Analg. 1998; surgery and brain tumor size on emergence from anesthesia.
87:1209. Anesthesiology. 1996;85:513.
74. Edmonds CR, Barbut D, Hager D, et al. Intraoperative 87. Ott VV, Pollack MM, Riddick L, et al. Delayed awakening
cerebral arterial embolization during total hip arthroplasty. from anesthesia and child abuse. Anesthesiology. 1983;58:
Anesthesiology. 2000;93:315. 178.
75. Ho A, Ling E. Systemic air embolism after lung trauma. 88. Toung T, Donham RT, Lehner A, et al. Tension pneumo-
Anesthesiology. 1999;90:564. cephalus after posterior fossa craniotomy: Report of four ad-
76. Saada M, Goarin JP, Riou B, et al. Systemic gas embolism ditional cases and review of postoperative pneumocephalus.
complicating pulmonary contusion-diagnosis and manage- Neurosurgery. 1983;12:164.
ment using transesophageal echocardiography. Am J Respir 89. Paul M, Dueck M, Kampe S, et al. Intracranial placement of a
Crit Care Med. 1995;152:812. nasotracheal tube after transnasal trans-sphenoidal surgery.
77. Manner JM, Wills A. Post-operative convulsions: A review Br J Anaesth. 2003;91:601.
based on a case report. Anaesthesia. 1971;26:66. 90. Marlow TJ, Goltra DD Jr, Schabel SI. Intracranial place-
78. Modica PA, Templehoff R. Seizures during emergence from ment of a nasotracheal tube after facial fracture: A rare
anesthesia. Anesthesiology. 1989;71:296. complication. J Emerg Med. 1997;15:187.
79. Sutherland MF, Burt P. Propofol and seizures. Anaesth 91. Wong GY, Warner DO, Schroeder DR, et al. Risk of
Intensive Care. 1994;22:733. surgery and anesthesia for ischemic stroke. Anesthesiology.
80. Crider BA, Hansen-Grant S. Nonconvulsive status epilepticus 2000;92:425.
as a cause for delayed emergence after electroconvulsive 92. Limburg M, Wijdicks EF, Li H. Ischemic stroke after surgical
therapy. Anesthesiology. 1995;82:591. procedures: Clinical features, neuroimaging, and risk factors.
81. Setzer N. Perioperative presentation of seizures in neonates. Neurology. 1998;50:895.
Anesth Analg. 1996;82:875. 93. Mangano DT, Mangano CT. Perioperative stroke en-
82. Kalfas IH, Little JR. Postoperative hemorrhage: A survey of cephalopathy and CNS dysfunction. J Intensive Care Med.
4992 intracranial procedures. Neurosurgery. 1988;23:343. 1997;12:148.
CHAPTER INTRAOPERATIVE AWARENESS
26
AND RECALL
A
pital for sterilization surgery under general
anesthesia. After a smooth, uneventful in-
duction, she awakes and notices she cannot DEFINITIONS
move. She hears the gynecologist, who is The subject of awareness is quite controversial. For in-
late, come in and a quarrel unfold between stance, the American Society of Anesthesiologists (ASA)
him and the anesthesiologist who rants, Where the hell practice advisory that is discussed at the end of this
were you, your patient has been ready for an hour! She chapter fails to acknowledge the subtleties involved and
later feels a series of knife-like stabbing sensations in her presented here. Part of the controversy stems from the
abdomen. She panics and becomes frightened of what else notion that awareness refers to consciousness. Indeed,
is going to happen, fearing more pain. loss of consciousness is a common clinical term and im-
In the recovery room, the patient is restless. The staff portant intraoperative endpoint. But what exactly is lost
tells her that restlessness is a common side effect of anes- when patients lose consciousness? The biologic under-
thesia and that she should calm down. Only later does pinnings of this mental state, or its counterpartto be
she remember the things that happened during surgery, conscioushave yet to be determined despite persistent
and she decides to inquire further. However, the nurses scientific interest and pursuit.1 For reasons of practi-
do not take her seriously, leaving her feeling helpless cality, one of several definitions may be adopted. Not
and ignored. She then gets upset, angry, and decides to uncommonly, awareness denotes a conscious subjective
confront the anesthesiologist. He, too, initially denies her experience (I heard a man talk).2 In clinical anesthesia,
experience, noting that her vital signs were completely however, the term has a distinctly different meaning and
normal. Only when she repeats his exact words during the refers to patients with memories or recall of the surgical
quarrel with the gynecologist, does his attitude change. procedure (I heard the surgeon talk). Therefore, when
After listening and explaining what may have happened, discussing awareness during anesthesia, we not only refer
the anesthesiologist apologizes and notifies the gynecolo- to subjective experiences, but also to memory. The sig-
gist. After repeated discussion and two psychotherapeutic nificance of this distinctionbetween the literal meaning
sessions, the patient considers a formal complaint rather of awareness and its use in clinical anesthesiawill be-
than legal action. come apparent as we discuss monitoring techniques (see
Section, Are There Warning Signs During the Anes-
thetic That Tell Us Something Is Wrong?). Adequate
monitoring demands proper delineation of the subject
What Are We Talking About? matter at hand and raises the question: Is it intraoperative
awareness that we are concerned with or postoperative
Awareness is considered an undesirable outcome, a com- memory? How the two are related but not the same will
plication, of anesthetic management, and there are good be elucidated in this chapter.
reasons for this contention. Patients with vivid memories For now, let us distinguish between awareness
of their surgical procedure, especially when they felt the without recall and awareness with recall. Although the
pain, tend to suffer long-term sequelae. The experience of- consequences of awareness without recall are unknown,
ten marks an adverse, debilitating event in their life. This we have a considerable understanding of the occurrence
chapter reviews the occurrence and meaning for both the and phenomenology of awareness with concurrent recall.
365
366 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
incidence of 1 to 2 cases per 1,000, irrespective of geo- TABLE 26.2 Posttraumatic Stress Disorder
graphic location and potential differences in anesthetics
and techniques. With approximately 20 million general Persistent reexperiencing of the event (intrusive recol-
anesthetics administered in the United States annually, lections, nightmares, intense distress, anxiety)
approximately 26,000 cases of awareness with recall can Persistent avoidance of associated stimuli (doctors,
be expected to occur each year, or 100 per work day. hospitals, check-ups)
Persistent increased arousal
Numbing of general responsiveness
Duration of symptoms for over 1 month
Why Should I Care?
PATIENT CONCERN persist 1 month and profoundly affect the way in which
people feel, behave, and function, a posttraumatic stress
In contrast to those of us who consider the number disorder (PTSD) may have developed (see Table 26.2).
presented above sufficiently high to warrant continued Depending on the awareness experience and reaction of
efforts to prevent awareness, others feel that postoperative hospital staff, 15% of those with recall require some form
recall is relatively rare and not worth the trouble. If you of psychotherapy, whereas 10% develop PTSD.14,15 Even
feel the same way, keep in mind that many patients are years after the event, 50% of those affected may still suf-
concerned about this potentially adverse outcome. Before fer from the experience and be severely disabled because
their anesthetic, up to 54% worry about the possibility of psychiatric sequelae.16 Therefore, in spite of its rela-
of pain, paralysis, and mental distress during surgery.13 tively infrequent occurrence, recall of awareness can be a
Furthermore, when suffered, awareness is a major source life-changing event for the patient. In the section, What
of patient dissatisfaction.10 This is not surprising given To Do? suggestions are offered for what you can do,
the subjective feelings and experiences of individuals who intraoperatively and postoperatively, when you suspect
were aware during their surgery. Although not all accounts awareness.
of intraoperative awareness are necessarily horrific, the It is important to note that adverse sequelae are pri-
following review of patient recollections represents a marily associated with the use of neuromuscular blocking
typical frame of reference. drugs. Of the 14 patients in a Swedish study who experi-
enced awareness and had been given a muscle relaxant,
11 reported unpleasant aftereffects.9 In comparison, none
PATIENTS MEMORIES of the four nonparalyzed patients that reported awareness
experienced postoperative sequelae.
Patients with recall of awareness are most likely to
remember sounds and conversations (30% to 90%),
as opposed to seeing, feeling, and smelling things.14
A significant number (up to 40%) may remember being
in pain, an experience that mediates adverse aftereffects.
Who Is at Risk?
Interviews suggest that it is not necessarily the awakening
itself that distresses patients most, but rather the inability
to move or communicate (awake paralysis). Even when CLINICAL FACTORS
pain is not experienced, the complete lack of control gives
rise to feelings that worse is yet to come. Most patients who In the American incidence study, awareness with
wake up paralyzed (70% to 90%) panic and experience recall was associated with a surgical procedure
anxiety, whereas half feel helpless and powerless. This (abdominal/thoracic, cardiac, and ophthalmology vs. oth-
situation is compared to being buried alive, with mental ers) and sicker patients (American Society for Anesthe-
trauma mounting when pain is added. A minor proportion siologists Physical Status Classification III to V [ASA
(15%) may experience suffocation, impending death, or IIIV]).8 The latter finding probably reflects the use of
believe they are in a coma and will not emerge from the lower anesthetic doses in such patients, a common cause
anesthetic. Approximately two thirds of patients report a of awareness with recall. A higher incidence of awareness
change in their attitude toward anesthesia after having has been reported for cesarean section, and cardiac and
experienced awareness. trauma surgery, ranging from 1% to 4%.1719
Up to 70% of patients with recall of awareness expe- The use of neuromuscular blocking agents clearly
rience unpleasant sequelae, including sleep disturbances, increases the occurrence of recall. In the Swedish inci-
recurrent nightmares, flashbacks, daytime anxiety, and dence study, twice as many patients developed memories
distress. This group is also most likely to change its of awareness episodes when paralysis was maintained
opinion about anesthesia and become more afraid and ap- throughout the surgical procedure, as opposed to general
prehensive about anesthetic procedures. Patients describe anesthesia without relaxants.9 When no neuromuscular
staying away from hospitals and doctors to avoid being blocking agents are used, patients can signal awareness
reminded of the traumatic event, a situation that compro- through movement, and the anesthetic may in turn be
mises treatment compliance and success. When symptoms deepened. On the basis of research findings such as these,
368 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
vigilance and the judicious use of muscle relaxants is TABLE 26.3 Causes of Recall
advised. These relaxants should be administered only
when absolutely necessary, for instance with endotra- Inadequate doses of anesthetics
cheal intubation and during ventilation. However, avoid- Equipment failure, leaks (empty or disconnected va-
ing neuromuscular blockade does not necessarily prevent porizers)
awareness. For instance, some patients will not attempt Opioid-based anesthesia (opioids are not general
to move despite their awareness of being awake. anesthetics)
Whereas the use of neuromuscular blocking drugs Difficult intubation (remember to re-dose intravenous
affects the incidence of awareness with recall, benzodi- anesthetics)
azepines do not. Because of their sedative properties, these Hypotension (requiring discontinuation of anesthetic
drugs are valuable for relieving preoperative anxiety, and agents)
therefore are commonly used. Because of their amnestic Justified risk (inadequate dosage may be necessary to
properties, they are also often used to prevent patients preserve life)
from having unpleasant memories. Benzodiazepines have
a strictly anterograde effect, meaning that the drugs
specifically interfere with the creation of new memo-
ries rather than disrupting old ones.20 An individual on and, accordingly, settlement payments and jury award
benzodiazepines may accurately perceive and process in- were often granted (78%). The report did not address
formation, and may respond appropriately to commands sequelae of awake paralysis.
or questions, but very little if anything of the experience Recall in comparison was mostly likely to involve
will be stored in memory. This anterograde amnesia for the maintenance phase of anesthesia (80% to 85%),
events after the drug is given may be desired when an and a number of contributing factors were identified
adverse event (e.g., pain) is anticipated. Indeed, the pro- (see Table 26.3). Here, the anesthetic care was judged
phylactic use of benzodiazepines is not uncommon, and substandard in fewer cases (43%), and payments were
only few ponder about the ethical issue it raises: Where granted approximately half the time. Eighty-four percent
did the adverse event go if not into memory?21 Contrary of recall plaintiffs sustained temporary emotional distress,
to its established effect, however, benzodiazepines are whereas 10% had developed PTSD.
often given after an adverse event to ameliorate unpleas- The closed-claims project further showed that, inde-
ant memories. Retrograde targeting of memory is useless pendent of standard of care, anesthetic techniques that
given the drugs pharmacokinetics and may explain why rely on intraoperative opioids, muscle relaxants, and no
there is no evidence in the awareness literature to sug- or low volatile anesthetic concentrations substantially in-
gest that benzodiazepines or scopolamine are any more crease the frequency of (legal) claims for recall after
successful than ordinary anesthetics in inducing amnesia. general anesthesia.15 Other variables such as age, ASA
Similar to ordinary anesthetics, benzodiazepines may im- status, anesthesia personnel, use of benzodiazepines, dif-
pair memory by inducing sedation, but no studies suggest ferent induction technique, and inhalational agents or
that either class of drugs directly interferes with memory intravenous anesthesia, generally do not affect the occur-
that is already formed. Even when given before or upon rence of recall. Female gender was also associated with an
induction of anesthesia, benzodiazepines do not reduce increased frequency of recall claims, although it is unclear
the incidence of awareness with recall.9,22 whether this reflects an intrinsic higher risk that is possi-
The ASA evaluated more than 4,000 nondental claims bly associated with sex differences in drug metabolism23
from a number of American insurance companies for or a greater propensity for women to file suit. Failure
adverse anesthetic outcomes that occurred between 1961 to demonstrate gender differences in large prospective
and 1995.15 Seventy percent of those claims were filed awareness trials8,9 supports the latter.
between 1980 and 1990. Awareness accounted for a small
proportion (2%) of claims, which were split into the
following two categories: INDIVIDUAL DIFFERENCES
1. Awake paralysis (18 claims) and
Anesthetic requirements vary substantially from one in-
2. Recall (61 claims)
dividual to the next and arguably affect the likelihood
Most claims for awake paralysis were found to be of awareness. The question of whether some people are
related to intravenous infusion errors (56%) or syringe more prone to awareness than others is rarely addressed.
swaps (44%): Bags or syringes containing neuromuscular Case reports generally do not reveal common patterns
blocking drugs were unlabeled, mislabeled, or properly la- in patients experiencing postoperative recall, other than
beled but not checked before administration. The periods those factors already discussed that pertain to contex-
of highest vulnerability were the preinduction and induc- tual rather than individual characteristics. Among the few
tion periods when a muscle relaxant was given instead exceptions are a history of major depression11 and a his-
of a hypnotic or sedative agent. The analysis confirmed tory of awareness,18 both of which may subject people
that some practitioners injected a benzodiazepine after to a greater risk of awareness. The influence of preoper-
the muscle relaxant in an unsuccessful attempt to achieve ative anxiety or distress11 on intraoperative awareness is
retrograde amnesia. Expert reviewers judged most of the controversial,14 although stress does create a strong neu-
awake paralysis cases (94%) to represent substandard care romodulatory influence that regulates the consolidation of
C H A P T E R 2 6 / I N T R A O P E R AT I V E A W A R E N E S S A N D R E C A L L 369
several forms of memory. Such regulatory effects may not them. The feedback also helped patients understand that
yet have been properly characterized because of the focus their situation was noticed and addressed. This is impor-
on self-reporting rather than physiologic stress measures. tant because patients may be unsure as to whether they
Finally, our own research suggests that patients with good succeed in moving despite an intention to move. Feedback
preoperative memory are more likely to develop memories from the outside world helps overcome this distortion in
during anesthesia.24 The scope of this effect, which was proprioception during sedation anesthesia, and possibly
found with implicit rather than explicit memory, remains other states of altered consciousness. We were pleased to
to be seen. Other traits, such as speed of information learn that by talking and providing explanations when-
processing, were unsuccessful predictors of memory. ever there was reason to suspect awareness in one of our
patients, all felt at ease upon recovery from anesthesia
and hospital discharge. The section on purposeful move-
ment (see Section Are There Warning Signs During the
What to Do? Anesthetic That Tell Us Something Is Wrong?) covers
the details of communicating with your patient during
anesthesia by using hand movement.
PREOPERATIVELY
Patients are rarely informed that they will be paralyzed POSTOPERATIVELY
by the anesthetics, which adds to the confusion when
When awareness has not surfaced preoperatively or in-
they accidentally awaken during the anesthetic. Only at
traoperatively, consider asking your patient five standard
that point do they find out that they cannot move, speak,
questions (Table 26.1) after general anesthesia. Be advised
or breathe on their own. One can imagine the startling that recall of awareness may be delayed or seemingly
nature of such an experience. This raises the question of absent because of residual sedation and drug-induced
whether awareness should be discussed preoperatively, amnesia. Ideally, question or arrange for the patient to
especially in high-risk cases and when the continuous use be questioned in both the PACU and before discharge.
of neuromuscular blockade is planned. The prevalence This ensures a reliable assessment of memory. When con-
of patient concern over pain and paralysis before their fronted with a possible awareness case, it is important to
anesthetic is another reason why the subject of awareness treat the patient appropriately and sympathetically (see
may be addressed during a preoperative consult, although Table 26.4). An empathic reaction, together with an expla-
the anesthesiologist should avoid provoking unnecessary nation, helps patients understand what happened and that
worry. You can explain the unlikeliness of postoperative errors sometimes occur. On the other hand, ignoring pa-
memories as well as the experience of pain. When tients and denying the experience or the very existence of
neuromuscular blockade is not anticipated, patients can awareness episodes fuels anger and upset. Ask additional
be told they will be able to move if desired. By talking questions (see Table 26.5) to explore experiences, or ar-
to your patient before surgery, you can acknowledge a range for someone to do so if time constraints are an issue.
common concern and reduce the apprehension. Do not ignore the patient, and be sure to follow through.
In addition to interviewing patients before they leave
the PACU or hospital, it is worthwhile to offer repeated
INTRAOPERATIVELY discussion and explanation as needed. Such a follow-up
procedure allows patients with recall of awareness to
When awareness or paralysis is not discussed preopera- come to terms relatively quickly (within weeks) with what
tively, be prepared and willing to talk to your patients
when awareness occurs intraoperatively. Feedback from
the outside world helps the patient cope with a startling TABLE 26.4 Managing the Impact
situation, as one of our recent studies suggested.25 In
a group of deeply sedated patients scheduled for elec- Keep conversation and procedures in the OR respectful
tive ambulatory surgery, we monitored the occurrence to the patient
of awareness by regularly asking patients to squeeze our Interview patients after the procedure
hand. Whenever they did so reliably, we provided com- Take a detailed account of recollections, if present
forting feedback and told patients what was occurring If awareness occurred, apologize to the patient
and what to expect. Patients were also offered more anes- Sympathize with unpleasant experiences or events
thetic, which some, but not all, desired. Postoperatively, Assure the patient of the credibility of his/her account
those who remembered being awake stressed the impor- Explain what (might have) happened, and why
tance of our feedback during the episode. Although the Offer repeated discussion and psychologic referral
possibility of awareness had been discussed with all study Notify the surgeon and other key personnel
patients preoperatively as part of the informed consent Include a copy of the detailed account in chart
procedure, our feedback helped ease the mind of those Follow up or arrange for a follow-up (a week or month)
who experienced it. One patient had felt closed in, and two after discharge
others feared pain or the start of surgery. Hearing what
was going on and what to expect had been a great relief to OR, operating room.
370 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 26.5 Additional Questions for Patients with actually present in a fair amount (67%) of awareness cases,
Postoperative Recall but also in the case controls (21%). Research studies
consistently suggest that clinical signs are neither very
What did you notice (pain, paralysis, sounds, vision)? sensitive nor specific measures of heightened levels of
Did you feel something in your mouth or throat? consciousness.
Were you able to move?
What went through your mind?
Did you think you were dreaming? END-TIDAL GAS
How long did it last?
Did you try to alert anyone? CONCENTRATIONS
Did you inform the anesthesiologist/hospital staff? Similar incidents of recall have been observed in large
Have there been any consequences? groups of patients receiving neuromuscular blocking
Do you sleep alright at night? agents, irrespective of end-tidal anesthetic gas concen-
tration monitoring.9 Taking individual case studies and
large prospective trials into account, reliance on expired
happened and not seek or need further (legal) assistance.9 gas concentrations is not recommended as a method of
When symptoms do persist, PTSD (Table 26.2) may be detecting awareness.
developing, and it becomes important to obtain referral
to a psychiatrist or psychotherapist experienced in its
treatment. PURPOSEFUL MOVEMENT
A note in the patient file will prepare future caregivers
to better deal with the phenomenon. Also, if conversation A useful method is to assess hand movement in response
among operating staff remains professional and respectful to command while patients undergo general or sedation
to the patient, the memories that potentially develop will anesthesia (see Fig. 26.1). Repeated and consistent move-
not necessarily be upsetting. ment may very well indicate your patient is awake. Do
not resort to giving additional muscle relaxants but in-
stead assess the patients state of mind. Your efforts and
feedback are likely to be appreciated when the patient is
Are There Warning Signs awake (see Section What To Do?).
Take the patients hand, preferably the one used to
During the Anesthetic That Tell write, in yours as if you are to shake hands. Start your
Us Something Is Wrong? assessment, each time by saying the patients first name,
so he or she understands that they are being addressed.
Without their name, patients may actually hear you but
not respond because they think you are talking to someone
CLINICAL SIGNS else. After calling their name, ask the patient to squeeze
Contrary to what many believe, awareness with recall is your hand if they can hear you, and wait approximately
often not accompanied by intraoperative hypertension, 10 seconds. If no response occurs, repeat the command to
tachycardia, or other clinical signs we are taught to ensure the patient is nonresponsive. Alternatively, if the
look for. In the aforementioned analysis of legal claims, patient does squeeze your hand, ask him/her to squeeze
clinical signs of light anesthesia were absent in most your hand twice. Failure to comply with this command
cases: Hypertension was noted in 15% of recall cases and may be considered an inadequate (equivocal) response
tachycardia in only 7%.15 As most patients had received reflecting emerging wakefulness (patient responds but
muscle relaxants, movement occurred only rarely. In the
absence of relaxants, bear in mind that movement does
Squeeze my hand once
not imply a patient is regaining consciousness; it may just
represent an automatic bodily reflex.
Investigators studied the possibility that awareness No squeeze One squeeze
cases can be identified through inspection of anesthetic
records. They asked three experienced anesthesiologists Nonresponse Squeeze my hand twice
to rate the possibility of awareness and recall based on Asleep
the records of patients who had experienced it.14 For each
awareness case, two similar cases (matched controls) were No squeeze Two squeezes
selected in which no memory had been reported. Raters
initially judged all records in random order, and went on
Equivocal response Unequivocal response
to assess sets of three records, one of which belonged to Wakeful? Aware
the awareness case (forced-choice situation). Not only
did raters agree very poorly on instances of possible FIGURE 26.1 Assessing intraoperative awareness using verbal
awareness, but also failed to identify the true awareness command.
cases. In this study, hypertension and tachycardia were
C H A P T E R 2 6 / I N T R A O P E R AT I V E A W A R E N E S S A N D R E C A L L 371
not appropriately), or it may simply be a reflex to auditory TABLE 26.6 Preventive Measures
stimulation. In these instances, we recommend you repeat
the command to squeeze twice to see what happens. If the Preoperative visit, mention possible awareness, es-
response remains, repeat the procedure shortly thereafter pecially in high-risk cases: Cesarean section, trauma,
(within 5 to 10 minutes). If the patient is regaining cardiac surgery, obesity, alcohol- or drug-abuse, previ-
consciousness, he or she may respond more accurately ous awareness experience
within 5 or 10 minutes. Check anesthesia delivery machines before start of case
It is generally agreed that squeezing twice to com- Consider minimizing the use of muscle relaxants
mand unequivocally establishes awareness. Such a re- Assess patient responsiveness to verbal command
sponse requires intentional, higher level processes (e.g., (Fig. 26.1)
counting), and is hard to accommodate in a purely re- Assess anesthetic depth using EEG measures
flexive response model. Therefore, when patients squeeze
twice to command (Fig. 26.1), assert they are awake and EEG, electroencephalogram.
continue to explore how they feel. Ask them to squeeze
twice again (or thrice!) to see if they are alright, and
to stretch their fingers if not. We assessed the patients they are an invaluable source of information and reliably
feelings while they received deeply sedative infusions.25 identify patients who were aware. Moreover, an interview
Similarly, you may explore their desire to receive more gives these patients an opportunity to ease their mind
anesthesia or discern whether they are in pain. Do not be amidst mental turmoil and confusion while clinicians get
surprised to find that some patients are fine as they are: a second chance to make up for whatever was, or was
Partially awake. We have come across cases that consis- not, lost.
tently and unequivocally responded to command but who Although intraoperative awareness and postoperative
did not desire more anesthesia. recall are imperfectly related, there is good evidence to
Two final notes on movement. First, be sure to ac- suggest that postoperative memories arise from brief pe-
knowledge movement when you see itbe it an unequiv- riods of wakefulness. In our study of deep sedation, for
ocal, equivocal, or a random response. Because patients instance, patients with no postoperative recall (n = 47)
may lose the capacity to monitor the effects of their in- responded on average to 10% of the commands given
tentions to move under the influence of hypnotic/sedative during anesthesia, whereas patients with recall (n = 9)
drugs (see Section, What To Do? subsection Intraoper- responded to approximately 30% of commands.25 This
atively), acknowledging movement clarifies to them that notion offers great potential for intraoperative aware-
they succeeded. Second, the procedure of probing for ness monitoring and leads scientists to contend that the
purposeful movement as described here may be used in anesthetized patient must be monitored to detect and ad-
the presence of overall neuromuscular paralysis. This re- dress awareness effectively (see Table 26.6 for preventive
quires the inflation (to 250 mm Hg) of a tourniquet around measures).
the lower arm before muscle relaxants are administered,
and is referred to as the isolated forearm technique.26,27 It
excludes one hand from paralysis, which in turn can be ELECTROENCEPHALOGRAM
used to signal awareness, either spontaneously or upon
request. The technique may also be used in the face of long Much effort has been directed in recent years toward de-
surgeries, provided that the cuff is repeatedly deflated, or veloping a reliable monitor of anesthetic adequacy to pre-
else ischemia sets in after 25 to 30 minutes. When muscle vent intraoperative awareness. With the brain being the
relaxants are again to be administered, the tourniquet is target effect site of anesthetic agents and demonstrating
simply reinflated beforehand. definitive changes with increasing drug concentrations, ef-
Studies using response to command during anesthe- forts have concentrated on electroencephalogram (EEG)
sia consistently show that more patients are awake at activity. With some exceptions (e.g., rapid eye movement
some point during the anesthetic than postoperative in- [REM] sleep), an awake brain typically generates an array
terviews suggest. In one of our recent studies during deep of random, rapidly oscillating (high frequency, Hz), low-
sedation, 66% of patients responded to command, but amplitude (voltage, V) electrical signals. EEG monitors
only one in four remembered doing so when interviewed of the hypnotic state rely on the fact that sedative drugs
later.25 Similarly, patients deliberately awakened during markedly slow down the signal as it gains power in am-
certain neurosurgical procedures tend to remember very plitude. EEG characteristics carry useful information on
little of what happened.28 Observations such as these have hypnotic state.
made it clear that awareness is not necessarily reflected Any complex time-varying signal can be decomposed
in a memory, and illustrate the notion that intraopera- into sine-wave elements and converted into a represen-
tive awareness and postoperative recall are imperfectly tation of frequency versus amplitude, also known as a
related. Postoperative interviews tend to underestimate Fourier transformation. The resultant power spectrum is
the incidence of awareness, which calls for a shift in a series of discrete values, the frequency components,
attention toward intraoperative monitoring. The limited and their associated power. In the power spectrum, fre-
usefulness of postoperative interviews in awareness moni- quency bands ( , , , ) can be discriminated, as well
toring and detection does not warrant the conclusion that as the frequency below which 50% or 95% of the power
interviews should be omitted altogether. On the contrary, in the EEG resides (median frequency [MF] and spectral
372 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
edge frequency [SEF], respectively). All have been related responses to command.25 Other studies, using a variety
to psychophysiologic states, such as levels of conscious- of agents, support the superiority of BIS as an awareness
ness or sleep, with moderate success. Power variables do monitor.3135 These findings agree with the notion that
not necessarily display a uniform response to different BIS incorporates more EEG information than power vari-
drugs, which has hampered their clinical applicability. ables such as SEF and MF and, consequently, BIS may be
They may show a biphasic response that is not easily expected to coincide more accurately with the presence or
translated into a single number denoting one particular absence of consciousness.
state. Anesthesia-induced burst suppression poses another Deriving from a database, BIS remains a probabilistic
problem for power variables, referring to alternating pe- measure with imperfect prediction probabilities. This
riods of high versus low voltage isoelectric activity, and means that intraindividual and interindividual differences
spectral analysis cannot quantify the amount of phase in brain monitor output are observed at loss and return of
coupling in the EEG. consciousness. Although brain function monitors perform
Phase coupling refers to the synchronization of fre- well given the state of the art, puzzling observations do
quency components and is an important characteristic of occur (some attributable to artifacts or lack of appropriate
nonlinear systems such as the brain.29 The physiologic interpretation of available signs).36 Perfection will be
meaning of phase relationships is not completely un- difficult, if not impossible, to attain in the absence of
derstood, but synchronization of brain activity has been a gold (biologic) standard for consciousness and, to a
implicated in various accounts of the anesthetic mech- lesser extent, awareness. Odd readings will therefore occur
anism.30 Frequency relationships may be examined by and underscore the importance of continued vigilance on
bispectral analysis of the EEG and, for clinical purposes, the part of the anesthesia caregiver. Rather than rely
a linear indexthe bispectral index (BIS)was developed on a single parameter, it is fruitful to acknowledge and
that ranges from 0 (isoelectric brain) to 100 (fully awake). recognize its added value. Although BIS enhances the
The BIS incorporates both power and phase information monitoring of sedation, research also clearly highlights
of the EEG and, similar to other processed parameters, is its limitations when it comes to predicting memory. In
a mathematical abstraction. It was empirically derived by our study of trauma patients, although BIS was the sole
estimating the EEG parameters that best discriminated significant predictor of performance on a sophisticated
sedation status in a large database of subjects receiving and sensitive postoperative memory test, it covaried
hypnotics and opioids. The first monitoring unit and elec- only weakly with memory scores4 that is, although we
trode sensor received Food and Drug Association (FDA) demonstrated a clear, nonrandom relationship between
approval in the mid-1990s and, in 2003, a clearance for depth of hypnosis as measured by BIS and postoperative
marketing for an awareness indication was approved. memory, a large proportion of variance in memory scores
To evaluate monitoring technology for awareness de- was left unexplained. This shows that many factors
tection, a reference criterion must be established that other than depth of hypnosis contribute to memory,
determines whether a patient is aware. Experimentally, highlighting the complex and multifaceted nature of
this can only be achieved in the unstimulated patient, memory. Similarly, we have observed comparable BIS
and therefore many research studies into the usefulness readings in patients who developed recall during a deep
of EEG parameters have relied on volunteer subjects or sedation study and those without recall.25 Designed as
surgical patients in the preincision periods. As a reference a monitor of consciousness, BIS is suitable to signal
criterion, most used the response to verbal command intraoperativebut not postoperativeawareness. Given
(see Section Are There Warning Signs During the the observed relation between intraoperative awareness
Anesthetic That Tell Us Something Is Wrong? sub- and postoperative recall, it appears that brain function
section Purposeful Movement) or tactile stimulation. monitoring will indirectly affect the incidence of recall by
Some included noxious stimulation to explore the entire lowering the occurrence of intraoperative awareness if a
range of sedative effect. By including response to ver- steady state of unconsciousness is maintained.
bal command, however, investigators acknowledged that
consciousness is more than movement to painful stimula-
tion (e.g., intubation, incision). Typically, drug doses were
related or titrated to sedation states, which were associ-
OTHER
ated with continuously measured SEF, MF, and/or BIS ELECTROENCEPHALOGRAM
values. A primary finding across these studies of monitor- MONITORS
ing technologies is the superiority of EEG measures over
heart rate and blood pressure in predicting loss or return A most promising alternative to the processed parameters
of consciousness. This further confirmed that autonomic described in the preceding text is the auditory evoked re-
signs poorly reflect the state of mind, which makes perfect sponse (AER), another electrical signal derived from the
sense, given the nervous system they represent. Clini- brain. Unlike the spontaneous EEG that lies at the heart
cally, it urges anesthesia caregivers to rely on centrally of (bi)spectral analysis, the AER is induced by sweeps
generated physiologic changes. of auditory clicks. On this particular type of stimulation,
When the different EEG measures are compared, BIS the EEG shows a characteristic response that represents
tends to outperform SEF and MF. In a direct comparison the passage of neuronal signals from the cochlea to the
of the three, we found only BIS to discriminate reliably be- auditory cortex. When sufficient numbers of clicks are
tween a large number of nonresponses versus unequivocal administered, signal averaging yields the AER waveform
C H A P T E R 2 6 / I N T R A O P E R AT I V E A W A R E N E S S A N D R E C A L L 373
that typically consists of a series of peaks and troughs measured by BIS <45, to be a prime, independent risk
of different latencies (expressed in milliseconds) and am- factor. Every hour of deep anesthesia, rather than total
plitudes (V) while awake. The midlatency potential that duration of anesthesia, established a 24% increased risk
arises after 10 to 100 ms shows graded changes with of dying 1 year after surgery. A smaller, yet significant
anesthetic concentration, and therefore is most suitable risk was noted with intraoperative systolic hypotension.
to monitor depth of anesthesia.37 A major advantage of However, comorbidity still presents the biggest risk fac-
AER technology is that it derives from the individual tor. If these studies suggesting a possible cause and effect
brain rather than a database of brain states, thereby relationship between the depth of sedation and mortal-
potentially offering greater prediction accuracy. On the ity are confirmed, a monitor of anesthetic depth could
other hand, real-time signal quantification was impossi- prove critical to patient management, and therefore this
ble until the development of fast-tracking technology38 observation/question warrants further research.
and precluded the technology from being implemented
in operating room (OR) monitoring equipment. The cur-
rently available indices (A-line [Danmeter A/S, Odense,
Denmark] commercially, and A-line ARX Index [AAI] or
LOWER INCIDENCE
Auto Regressive Model with Exogenous Input [ARX] al- The benchmark test of monitoring efficacy is whether
gorithms for academic purposes) perform comparably to the technology has a notable effect on the incidence
BIS, which is often used as a reference in recent feasibility of awarenessthat is, whether the technology reduces
studies. Both monitors predict sedation states with good the occurrence of postoperative recall in a randomized
accuracy (85% to 95%) and correlate well with calculated controlled trial. Because recall after anesthesia is rare,
effect-site drug concentrations.3941 large numbers of subjects are required to demonstrate
There are two other monitors that are not essentially differences between a group that was monitored using
different from the EEG-based technologies described so the technology and a group that was monitored by other
far. Spectral entropy derives from EEG power anal- measures. At present, a randomized controlled trial has
yses and shows the limitations described earlier (see only been performed for BIS.18 In the United States
Section Electroencephalogram).42,43 Narcotrend clas- awareness study that left BIS monitoring at the discretion
sifies EEG traces into different sedation stages (A to F)44,45 of the attending anesthesiologist, no association between
with limited success.46,47 the use of BIS and the incidence of awareness was found.8
In the randomized controlled trial, which was executed in
Australia and included adults at high risk of awareness,
close to 2,500 patients were allocated to BIS-guided
Epiphenomena: What Is In anesthesia where the delivery of anesthesia was adjusted
a Monitor? to maintain a BIS of 40 to 60 between laryngoscopy and
wound closure, or to routine care, in which case an EEG
sensor was applied but the BIS monitoring unit was simply
not turned on. Observers blinded to group allocation
SHORT-TERM AND undertook the awareness follow-up using the structured
LONG-TERM OUTCOME interview mentioned earlier8,9 (Table 26.1). As before, the
primary outcome measure of interest was the incidence of
Hypnotic state monitoring using the latest (EEG) tech- confirmed awareness recollections. BIS-guided anesthesia
nology may not only be beneficial during anesthesia, but reduced the incidence by 82% (from 11 to 2 cases),
also has other implications beyond the surgical theatre. a number that closely resembled the findings of a
Compared to the standard practice of not using brain mon- Swedish nonrandomized, historic control study in low-
itors, titration to an optimal level of hypnosis (sufficiently risk, noncardiac patients.53 The Australian randomized
deep to prevent awareness but not unnecessarily deep) re- controlled trial, the first of its kind, demonstrates the clear
duces overall drug requirements and generally speeds up potential to reduce postoperative recall using cerebral
early recoveryfor instance, PACU discharge.48,49 Most rather than traditional monitoring techniques. In the BIS-
of the available evidence comes from studies using BIS, monitored group, awareness occurred at values above 55,
but similar secondary effects are now noted for AER and which aligns with observations of unequivocal responses
spectral-based measures. to command at values between 55 and 60,25,54 close to the
In the long run, these early benefits may take a dra- manufacturers guideline to maintain values below 60 to
matic turn. Anesthetic management during surgery has avoid intraoperative awareness.
been tentatively implicated in mortality rates 1 year out.
After an initial retrospective report,50 two prospective tri-
als in Europe51 and the United States52 recently addressed
the issue more rigorously. Both included significant num- How Does Awareness Affect
bers (more than 1,000 and 4,000) of noncardiac patients
and observed approximately a 5.5% mortality rate 1 year Children?
after surgery. Among the many recorded factors that could
potentially be associated with death, both groups identi- Awareness in children is even more controversial than its
fied duration (cumulative time) of deep anesthesia, as occurrence and prevention in adults. Perhaps because the
374 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
extraction of information is more complicated, awareness 8. Sebel PS, Bowdle TA, Ghoneim MM, et al. The incidence
studies in children are rare. One recent prospective of awareness during anesthesia: A multicenter United States
trial yielded a 0.8% incidence, suggesting that children study. Anesth Analg. 2004;99:833.
are four to eight times more likely than adults to be 9. Sandin RH, Enlund G, Samuelsson P, et al. Awareness during
anaesthesia: A prospective case study. Lancet. 2000;355:707.
aware.55 At present, it is unclear whether the higher
10. Myles PS, Williams DL, Hendrata M, et al. Patient satisfac-
incidence is something unique to childrensuch as tion after anaesthesia and surgery: Results of a prospective
increased anesthetic requirementsor whether it reflects survey of 10 811 patients. Br J Anaesth. 2000;84:6.
an overestimation because of specific study practices and 11. Ranta SO, Laurila R, Saario J, et al. Awareness with recall
assessment procedures.56 during general anesthesia: Incidence and risk factors. Anesth
The intraoperative monitoring of pediatric awareness Analg. 1998;86:1084.
also poses new challenges, not in the least because nor- 12. Liu WH, Thorp TAS, Graham SG, et al. Incidence of aware-
mal pediatric EEG differs markedly from adult EEG, ness with recall during general anaesthesia. Anaesthesia.
displaying more variation. This renders processed EEG 1991;46:435.
parameters such as BIS potentially less reliable in chil- 13. Klafta JM, Roizen MF. Current understanding of patients
attitudes toward and preparation for anesthesia: A review.
dren (but see recent studies57 ). In general, the changes in
Anesth Analg. 1996;83:1314.
electrical activity of the brain during growth and devel- 14. Moerman N, Bonke B, Oosting J. Awareness and recall
opment call for age-specific considerations and cautious during general anesthesia. Facts and feelings. Anesthesiology.
interpretation of EEG data in children. 1993;79:454.
15. Domino KB, Posner KL, Caplan RA, et al. Awareness
during anesthesia: A closed claims analysis. Anesthesiology.
1999;90:1053.
KEY POINTS 16. Lennmarken C, Bildfors K, Enlund G, et al. Victims of
awareness. Acta Anaesth Scand. 2002;46:229.
1. The latest ASA practice advisory for intraoperative 17. Phillips AA, McLean RF, Devitt JH, et al. Recall of intraoper-
awareness by the ASA58 is centered around postop- ative events after general anaesthesia and cardiopulmonary
erative recall. As can be concluded at the end of bypass. Can J Anaesth. 1993;40:922.
this chapter, intraoperative awareness and postoper- 18. Myles PS, Leslie K, McNeil J, et al. Bispectral index
ative recall are not mutually exclusive phenomena, monitoring to prevent awareness during anaesthesia: The
allowing clinicians and investigators to use one as a B-Aware randomised controlled trial. Lancet. 2004;363:1757.
19. Lyons G, Macdonald R. Awareness during caesarean section.
substitute for the other.
Anaesthesia. 1991;46:62.
2. Recall typically underestimates the incidence of 20. Buffett-Jerrott SE, Stewart SH. Cognitive and sedative effects
intraoperative awareness and represents only the tip of benzodiazepine use. Curr Pharm Des. 2002;8:45.
of the iceberg. 21. Hope MD. Pain and forgetting. JAMA. 2003;289:617.
3. Brain function monitors do not predict recall very 22. Myles PS, Leslie K, Forbes A, et al. A large randomized trial
well but are much better than traditional autonomic of BIS monitoring to prevent awareness in high risk patients:
parameters in signaling the loss and return of The B-Aware trial. Anesthesiology. 2003;99:A320.
consciousness. 23. Ciccone GK, Holdcroft A. Drugs and sex differences: A review
4. Brain function monitoring represents a major devel- of drugs relating to anaesthesia. Br J Anaesth. 1999;82:255.
opment in anesthesia practice management. Having 24. Kerssens C, Lubke GH, Klein J, et al. Memory function
during propofol and alfentanil anesthesia: Predictive value
the ability to signal intraoperative awareness and
of individual differences. Anesthesiology. 2002;97:382.
prevent it by maintaining sufficiently deep levels of 25. Kerssens C, Klein J, Bonke B. Awareness: Monitoring versus
hypnosis, in turn, offers great potential for the pre- remembering what happened. Anesthesiology. 2003;99:570.
vention of postoperative recall. 26. Russell IF, Wang M. Absence of memory for intra-operative
information during surgery with total intravenous anaesthe-
REFERENCES sia. Br J Anaesth. 2001;86:196.
27. Tunstall ME. Detecting wakefulness during general anaes-
1. Crick F, Koch C. A framework for consciousness. Nat thesia for caesarean section. Br Med J. 1977;1:1321.
Neurosci. 2003;6:119. 28. Nordstrom O, Sandin R. Recall during intermittent propofol
2. Tassi P, Muzet A. Defining the states of consciousness. anaesthesia. Br J Anaesth. 1996;76:699.
Neurosci Biobehav Rev. 2001;25:175. 29. Rampil IJ. A primer for EEG signal processing in anesthesia.
3. Bogetz MS, Katz JA. Recall of surgery for major trauma. Anesthesiology. 1998;89:980.
Anesthesiology. 1984;61:6. 30. Mashour GA. Consciousness unbound: Toward a paradigm
4. Lubke GH, Kerssens C, Phaf RH, et al. Dependence of explicit of general anesthesia. Anesthesiology. 2004;100:428.
and implicit memory on hypnotic state in trauma patients. 31. Katoh T, Suzuki A, Ikeda K. Electroencephalographic
Anesthesiology. 1999;90:670. derivatives as a tool for predicting the depth of sedation and
5. Crystal JD, Maxwell KW, Hohmann AG. Cannabinoid anesthesia induced by sevoflurane. Anesthesiology. 1998;88:
modulation of sensitivity to time. Behav Brain Res. 2003; 642.
144:57. 32. Liu J, Singh H, White PF. Electroencephalographic bis-
6. Andrade J. Learning during anaesthesia: A review. Br J pectral analysis predicts the depth of midazolam induced
Psychol. 1995;86:479. sedation. Anesthesiology. 1996;84:64.
7. Kerssens C, Sebel PS. BIS and memory during anesthesia. 33. Liu J, Singh H, White PF. Electroencephalographic bispec-
In: Ghoneim MM, ed. Awareness during anesthesia. Woburn: tral index correlates with intraoperative recall and depth of
Butterworth-Heinemann; 2001:103. propofol-induced sedation. Anesth Analg. 1997;84:185.
C H A P T E R 2 6 / I N T R A O P E R AT I V E A W A R E N E S S A N D R E C A L L 375
34. Iselin-Chaves IA, Flaishon R, Sebel PS, et al. The effect of the electroencephalographic effects of isoflurane with
of the interaction of propofol and alfentanil on recall, loss and without burst suppression. Anesthesiology. 2004;101:
of consciousness and the bispectral index. Anesth Analg. 847.
1998;87:949. 46. Russell IF. The Narcotrend depth of anaesthesia monitor
35. Glass PS, Bloom M, Kearse L, et al. Bispectral analysis mea- cannot reliably detect consciousness during general anaes-
sures sedation and memory effects of propofol, midazolam, thesia: An investigation using the isolated forearm technique.
isoflurane, and alfentanil in healthy volunteers. Anesthesiol- Br J Anaesth. 2006;96:346.
ogy. 1997;86:836. 47. Schneider G, Kochs EF, Horn B, et al. Narcotrend does
36. Rampil I, Mychaskiw GI, Horowitz M. False negative BIS? not adequately detect the transition between awareness and
Maybe, maybe not! Anesth Analg. 2001;93:798. unconsciousness in surgical patients. Anesthesiology. 2004;
37. Thornton C, Sharpe RM. Evoked responses in anaesthesia. 101:1105.
Br J Anaesth. 1998;81:771. 48. Johansen JW, Sebel PS, Sigl JC. Clinical impact of hypnotic-
38. Jensen EW, Nygaard M, Henneberg SW. On-line analysis titration guidelines based on EEG bispectral index (BIS)
of middle latency auditory evoked potentials (MLAEP) for monitoring during routine anesthetic care. J Clin Anesth.
monitoring depth of anaesthesia in laboratory rats. Med Eng 2000;12:433.
Phys. 1998;20:722. 49. Johansen JW, Sebel PS. Development and clinical appli-
39. Struys MM, Jensen EW, Smith W, et al. Performance of the cation of electroencephalographic bispectrum monitoring.
ARX-derived auditory evoked potential index as an indicator Anesthesiology. 2000;93:1336.
of anesthetic depth. Anesthesiology. 2002;96:803. 50. Weldon BG, Mahla ME, Van der Aa MT, et al. Advancing
40. Struys MM, Vereecke H, Moerman A, et al. Ability of age and deeper intraoperative anesthetic levels are associ-
the bispectral index, autoregressive modelling with exoge- ated with increased first year death rates. Anesthesiology.
nous input-derived auditory evoked potentials, and predicted 2002;96:A1097.
propofol concentrations to measure patient responsiveness 51. Lennmarken C, Lindholm M, Greenwald SD, et al. Con-
during anesthesia with propofol and remifentanil. Anesthe- firmation that low intraoperative BIS levels predict in-
siology. 2003;99:802. creased risk of post-operative mortality. Anesthesiology. 2003;
41. Vereecke HE, Vasquez PM, Jensen EW, et al. New composite 99:A303.
index based on midlatency auditory evoked potential and 52. Monk TG, Saini V, Weldon BC, et al. Anesthetic management
electroencephalographic parameters to optimize correlation and one-year mortality after noncardiac surgery. Anesth
with propofol effect site concentration: Comparison with Analg. 2005;100:4.
bispectral index and solitary used fast extracting auditory 53. Ekman A, Lindholm ML, Lennmarken C, et al. Reduction
evoked potential index. Anesthesiology. 2005;103:500. in the incidence of awareness using BIS monitoring. Acta
42. Vanluchene AL, Struys MM, Heyse BE, et al. Spectral entropy Anaesthesiol Scand. 2004;48:20.
measurement of patient responsiveness during propofol and 54. Flaishon R, Windsor A, Sigl J, et al. Recovery of con-
remifentanil. A comparison with the bispectral index. Br J sciousness after thiopental or propofol: Bispectral index
Anaesth. 2004;93:645. and the isolated forearm technique. Anesthesiology. 1997;86:
43. Vanluchene AL, Vereecke H, Thas O, et al. Spectral en- 613.
tropy as an electroencephalographic measure of anesthetic 55. Davidson AJ, Huang GH, Czarnecki C, et al. Awareness
drug effect: A comparison with bispectral index and pro- during anesthesia in children: A prospective cohort study.
cessed midlatency auditory evoked response. Anesthesiology. Anesth Analg. 2005;100:653.
2004;101:34. 56. Davis PJ. Goldilocks: The pediatric anesthesiologists
44. Kreuer S, Biedler A, Larsen R, et al. The Narcotrenda dilemma. Anesth Analg. 2005;100:650.
new EEG monitor designed to measure the depth of 57. Kerssens C, Sebel PS. To BIS or not to BIS? That is the
anaesthesia: A comparison with bispectral index monitor- question. Anesth Analg. 2006;102:380.
ing during propofol-remifentanil-anaesthesia. Anaesthesist. 58. Practice advisory for intraoperative awareness and brain
2001;50:921. function monitoring: A report by the American Society of
45. Kreuer S, Bruhn J, Larsen R, et al. Application of bis- Anesthesiologists task force on intraoperative awareness.
pectral index and narcotrend index to the measurement Anesthesiology. 104:847, 2006.
CHAPTER POSTOPERATIVE COGNITIVE
27
DYSFUNCTION
W
some patients will experience disturbances An insurance company executive had a peripheral
of cognition, and even delirium. Clinical ex- vascular surgery. Several months postoperatively, he
perience and numerous studies suggest that found it impossible to carry out required statistical
these changes will resolve spontaneously as calculations.
the patients pain subsides; as active drugs An 80-year-old woman had knee replacement surgery
in the central nervous system (CNS) undergo biotransfor- and a normal recovery until approximately 3 weeks
mation and elimination; and as hormone levels return to postoperatively when she awoke one morning wonder-
baseline. Soon after discharge from the hospital, physi- ing why her leg was sore and painful. She had lost all
cians usually lose contact with patients; occasionally, memory of her surgery event and rehabilitation. She
weeks or months after a surgery, we encounter a pa- continues to have no recollection of any these events.
tient who claims to have been plagued by subtle changes
in brain function. It is not unusual for patients to blame Concerns about cognitive or neurologic change af-
anesthesia because of its profound CNS effects. ter surgery or anesthesia have continued to grow since a
The following observed stories exemplify what the publication by Bedford in 1955 claiming that anesthesia
clinician may hear from a patient with possible postoper- induced mental changes in the elderly.1 Despite an in-
ative cognitive dysfunction (POCD): crease in publications related to the topic and improved
understanding about possible risk factors associated with
An attending anesthesiologist in his thirties underwent postoperative cognitive changes, we still do not know the
an exploratory thoracotomy and resection of a small specifics of how anesthesia influences cognitive change,
pulmonary nodule. He had an uneventful anesthetic what brain systems may be involved, or how best to treat
procedure and smooth recovery from anesthesia. He symptoms when they do occur. In this chapter we aim to
returned to work soon after but complained for several provide fodder for thought (and hopefully some helpful
months that he could not solve crossword puzzles tools) for clinicians who encounter postoperative cogni-
because he found it difficult to concentrate. tive dysfunction, as well as for researchers studying the
A 54-year-old woman, an avid and apparently compe- phenomenon.
tent bridge player, had to undergo repeated surgeries
for lysis of intra-abdominal adhesions. After each of
her surgeries, she was soon able to attend to her daily
chores as a housewife but for approximately 6 months, What Is Postoperative Cognitive
she found it difficult to play bridge because she could
not remember the cards that had been played. Dysfunction?
An 82-year-old woman was seen for the removal of a
small tumor near her chin. She was accompanied by two In research articles, the term, postoperative cognitive dys-
of her children. When the option of general anesthesia function, or POCD, denotes postoperative memory and/or
was mentioned, the children strongly objected, saying thinking problems that have been corroborated by neu-
that their mother had had a surgery 8 weeks ago and ropsychological testing. Although not yet a recognized
she had not been the same since. She was very forgetful diagnostic code in the International Classification of Dis-
and sometimes confused. eases (ICD), clinicians are beginning to use the term POCD
An executive secretary reported, approximately as a general description of postoperative patients who
3 months after a surgery under general anesthesia, complain of memory and thinking problems. Typically,
376
C H A P T E R 2 7 / P O S T O P E R AT I V E C O G N I T I V E D Y S F U N C T I O N 377
these complaints include difficulty staying focused on a pressure, congestive heart failure, repeat coronary artery
task, difficulty completing more than one task at the same bypass grafting, anemia, carotid arteriosclerosis, diabetes,
time, problems finding words, and difficulty recalling in- left ventricular ejection fraction lower than 45%, chronic
formation recently heard or read. Postoperative cognitive pulmonary disease, and unstable anginaall conditions
dysfunction should not be confused with delirium, which, that have been shown to affect postoperative morbidity.
in contrast to POCD, involves fluctuating orientation and Importantly, both clinicians and researchers have
confusion. a tendency to ignore the role of mental comorbidities
When POCD needs to be listed as a diagnosis, the (e.g., depression, presurgical cognitive impairment) as
Diagnostic and Statistical Manuals (DSM-IV)2 definition possible risk factors for postoperative complications. The
of Cognitive Disorder Not Otherwise Specified (ICD importance of mental health influences is alluded to by
Code Number: 296.5) may be appropriate. This requires a retrospective study by Berstein and Offenbartl.8 These
the clinician to document cognitive changes that cannot be authors report that out of 975 general surgery patients, 57
attributed to another etiology such as delirium, dementia, had postoperative fatal and nonfatal complications; most
or cognitive dysfunction due to a specific cause. of those who had a fatal complication had symptoms
of dementia before surgery (25 of 32 who died had
dementia). These findings suggest that a patient with a
presurgical mental disorder, specifically a dementia, may
Are Certain Patients at Risk be particularly sensitive to surgical events.
for Postoperative Cognitive
Dysfunction?
What Type of Assessments Can
As demonstrated by the examples in the preceding text,
an individual of almost any age or education level can
Be Useful for Identifying
experience POCD. The symptoms also vary. Clinicians, Postoperative Cognitive
therefore, need to remain attentive to all patients and
listen carefully to their postoperative complaints.
Dysfunction?
A number of studies demonstrate that, at least during
the first 3 postoperative months, elderly patients have As discussed in the preceding text, despite evidence point-
higher rates of POCD than young and middle-aged ing to population risk factors, predicting who is at risk
adults. In large studies of patients undergoing noncardiac for POCD remains problematic. Therefore, direct obser-
surgery, the prevalence of POCD among adults 60 years vations and documentation of patients cognitive abilities
of age and older was between 20% and 30% at 1 week before surgery and again after surgery may be helpful.
and 10% to 15% at 3 months.3,4 We know little about
POCD occurring beyond 1 year. One study reports that
only 1% of elderly adults experience the persistent effects
of cognitive decline after surgery,5 which may indicate
BRIEF COGNITIVE
that POCD is reversible even among the elderly. MEASURES
Apart from age, another strong risk factor for POCD
appears to be years of formal education. A low educa- In addition to the customary preoperative and postopera-
tional level has been shown to predict cognitive decline, tive clinical evaluations, a clinician can consider using a
whereas high educational levels seem to protect against few basic tests to assess general cognition. Two examples
cognitive decline following surgical procedures involv- include the Mini Mental State Examination (MMSE)9 and
ing cardiopulmonary bypass.6 One proposed rationale for the Clock Drawing test.10 These measures are simple to
this relationship involves the concept of cerebral cogni- administer and provide basic information about patients
tive reserve: Extensive education or high occupational orientation, memory, visual construction (drawing), and
attainment has been associated with the ability to re- planning abilities.
cruit neuronal structures to replace damaged processing
pathways.7 Other explanations for the relationship be- Mini Mental State Examination
tween POCD and educational level include test-taking
skills, enhanced social support, and excellent postopera- The MMSE is a frequently used measure of general
tive medical care. Of course, there are always exceptions cognitive status that examines orientation, learning and
to these findings, as shown from the case descriptions memory, attention, language functions, and visual con-
at the beginning of this chapter, and therefore pre- struction. The MMSE requires approximately 10 minutes
dicting who is at great risk for POCD will be at best to administer (guidelines available in the study9 ). A score
tentative. between 28 and 30 indicates intact cognitive functions.
Surprisingly, as compared to age and level of educa- A comparatively low postoperative score relative to a
tion, comorbidity has been largely ignored in research. preoperative score (e.g., presurgical score = 25, 1 day
Intuitively, we would expect increased susceptibility to postoperative score = 19) would be cause for concern
POCD in patients suffering debilitating conditions known and may require enhanced postanesthesia follow-up and
to affect postoperative morbidity, such as high blood neurologic and/or neuropsychological consultation.
378 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 27.3 Magnetic resonance imaging (MRI) changes in the brain before surgery and 3 days
post surgery. Presurgical and postsurgical axial diffusion-weighted MRI slices from a 50-year-old
man who underwent aortic valve and root replacement. At 3 days post surgery, multiple new
infarctions were found in the right subinsular region and extending to the right corona radiata and
centrum semiovale of the frontal lobe, all considered to be in the distribution of the right
leptomeningeal medial cerebral artery (MCA) and right MCA perforators.
(Picture and caption are complements of Thomas F, Floyd MD, Department of Anesthesiology,
University of Pennsylvania, Philadelphia, PA.)
MRI within the first 72 hours of surgery is particularly and external capsules) because they are highly susceptible
useful for identifying new frank and silent strokes. to numerous neurotransmitter and small vessel changes.
The importance of the basal ganglia, in particular, on
ones ability to attend and inhibit distracting information
has been demonstrated mostly by elegant animal studies19
Are Certain Brain Regions and also by studies on neurodegenerative illnesses such as
Parkinson disease.20 Additionally, the thalamus is a pri-
Associated with Postoperative mary subcortical structure necessary for both attention
Cognitive Dysfunction? and memory. It is, unfortunately, also the most common
site for ischemic changes and dysfunction, particularly
To date, no research has directly examined which brain among older adults.
regions are most often associated with the primary The subcortical nuclei are in very close proximity to
symptoms of POCD (e.g., mild to moderate attention the intricate vessels of the Circle of Willis and smaller
disturbance, word finding problems, difficulty learning intracranial arterioles, such as the lenticulostriate arteries
new information). We can speculate, however, which (see Fig. 27.4), which nourish important subcortical
anatomic zones may be most susceptible. We base this
speculation not only on a general understanding of
neuropsychology but also on recent research suggesting
that there may be subtypes of POCD.18 Some patients
experience primary disturbances in attention, whereas
others experience only learning and memory difficulties.
nuclei (e.g., caudate) and associated white matter tracts. The medial temporal lobe is well established as
Intracranial arterioles such as the lenticulostriate arteries being important for memory processes. Neuropsychology
are particularly vulnerable to hemodynamic changes.21 case studies first demonstrated the importance of medial
These changes may result in reduced arteriole wall temporal structures on learning and memory function.22
integrity thereby leading to reduced nourishment or In particular, the CA1 region in the hippocampus is an
possible protein leakage into the white matter or gray important site for memory consolidation and for directing
nuclei. This would lead to the development of white matter memory to storage. When this region is damaged,
changes (Fig. 27.2) and lacunae within the thalamus and individuals can immediately repeat information, but
basal ganglia.21 Such damage is hypothesized to result are unable to consolidate the information for later
in a disconnection to the frontal lobe thereby leading recall. For individuals who experience difficulty with
to attention disturbances. This disconnection can be at learning new information post surgery, hypoxic damage
least partially explained by neurotransmitter disruption, to hippocampal structures should be flagged as a possible
for the interplay between subcortical nuclei and white mechanism.
matter is largely regulated by inhibitory (GABAergic Also implicated in disruptions of new learning is
[ -aminobutyric acid]) and excitatory (glutaminergic) the basal forebrain that innervates the hippocampus
neurotransmitters.19 with essential cholinergic neurons. This structure is be-
lieved to be uniquely involved in a cholinergic system
of memory. Animal studies have demonstrated impaired
learning when the neuronal transmission of the basal
What Anatomic Systems Can forebrain is disrupted following the administration of
cholinergic (muscarinic) receptor antagonists.23 Unfortu-
Underlie Postoperative Memory nately, we do not know how anesthesia affects the basal
Difficulties? forebrain. Future research needs to address whether pa-
tients with reduced basal forebrain function (e.g., early
stage Alzheimer disease) are particularly sensitive to
The basis for impaired learning and memory following
general anesthesia and antimuscarinic drugs that cross
surgery is unclear, but is most likely multifactorial.
the bloodbrain barrier. Scopolamine provides a good
The possible mechanisms for learning and memory
example; the elderly are particularly sensitive to its CNS
disruption involve pathologic changes in the neural
effects.
regions that support memory functions. In humans,
At the subcortical gray matter level, the thalami,
amnestic disturbances have been classically associated
specifically the dorsomedial and anterior nuclei, have
with dysfunction of three neuroanatomic regions within
been shown to be important in the retrieval processes
the brain (and the pathways that interconnect them):
of memory. Damage to the thalami can result in recall
(a) The medial temporal lobe (hippocampus, entorhinal
impairment of recently learned information. This is
cortex); (b) the thalamus (dorsomedial, anterior nuclei);
markedly different from a lesion within the hippocampus
and (c) the basal forebrain. Some of these regions are
or basal forebrain disruption; the thalamus provides
shown in Figure 27.5.
assistance with recalling information that has already
been consolidated into long-term memory (i.e., tip of the
tongue phenomena). A range of perioperative factors that
alter vascular, neurotransmitter, and protein functions
can influence the function of the thalamus.
d
Does General Anesthesia Cause
Postoperative Cognitive
c Dysfunction?
b
a Most recent investigations using objective cognitive mea-
sures report equivalent rates of POCD among general and
regional anesthesia groups.24,25 At most, general anesthe-
sia has been associated with greater incidence rates of
cognitive decline in the very early postoperative hours.
FIGURE 27.5 Magnetic resonance imaging (MRI) coronal The effects of general anesthesia, over and above that
image depicting the (a) entorhinal cortex, (b) hippocampus, of regional anesthesia, dissipate by 1 week post surgery;
(c) thalamus, and (d) caudate. therefore, other factors must be at work causing long-term
postoperative cognitive dysfunction.
C H A P T E R 2 7 / P O S T O P E R AT I V E C O G N I T I V E D Y S F U N C T I O N 381
and reported having to do things slowly to complete activity of the cobalamin-dependent enzyme, methion-
them correctly. These symptoms reportedly disappeared ine synthase (or 5-methyltetrahydrofolatehomocysteine
2 weeks later.33 S-methyltransferase; Enzyme Commission code EC
Although both studies show that even after a pro- 2.1.1.13). Methionine is critically involved in the assembly
tracted exposure to halothane CNS effects had dissipated of the myelin sheath and DNA synthesis in rapidly prolif-
in <2 weeks, they also show that lipid-soluble anesthetics erating tissues. That the vast majority of patients tolerate
can reside for many days in nerve tissue. We do not know nitrous oxide so well must be attributed to the genetic
the consequences of such lengthy exposure. good fortunes of most of us.
Far more common than the rare genetic defects were
the anesthetic disasters related to the fact that nitrous
oxide, being a weak anesthetic, had in the past often
What Studies Link been given without sufficient oxygen. In a current classic
paper calling attention to the problem and attributing
Postoperative Cognitive it to hypoxemia, Batten and Courville vividly describe a
Dysfunction to the Lingering series of patients suffering often severe and lasting mental
disturbances after nitrous oxide administration.36 We do
Effects of Anesthesia? not know whether hypoxemia with nitrous oxide causes
more or less damage than hypoxemia in the absence of
nitrous oxide. The point is moot; cerebral hypoxemia
ETHER AND CHLOROFORM can undoubtedly impair the CNS tissue indefinitely.
Nonetheless, it is perplexing that some patients can
In studies dating back to the nineteenth century, refer- tolerate low-oxygen partial pressures better than others.
ence is made to patients who became mentally deranged While Batten and Courville pointed to nitrous oxide,
after anesthesia. For example, in a report, Insanity fol- cerebral hypoxemia can arise during use of any anesthetic
lowing the use of anaesthetics in operations,34 the author and, often enough, postoperatively under the influence of
made two points, which he proceeded to illustrate with opiate-induced respiratory depression.
vivid case histories: (i) Patients with a family and per-
sonal history of weak mindedness and who come of
insane stock are at risk of lasting mental derange- HYPOCARBIA
ment after anesthesia; and (ii) Any cause which will
give rise to delirium may set up a more chronic form In the past, hyperventilation during anesthesia was quite
of mental disorder quite apart from any febrile distur- commonly practiced in conjunction with nitrous oxide
bance. He quoted as examples alcohol and anesthesia: and neuromuscular blocking drugs, as it was recognized
Insanity frequently follows alcoholism . . . . Beside these that hypocapnia would lessen the required anesthetic
cases occurring in chronic drinkers, there are others doses.37 That this would have deleterious effects by
in which single bout of drinking or moderate drinking severely decreasing brain blood flow and oxygen uptake
associated with some shock or some cause of vital de- was soon recognized.38 Wollman and Orkin39 reported
pression is followed by a similar development of acute that in 20 patients (all below 60 years of age) whose PaCo2
delirious mania, and it is to these cases I specially call was below 24 mm Hg throughout the procedure there was
notice in relationship to the attack of insanity follow- postoperative prolongation of reaction time which lasted
ing the use of some anaesthetics. In this historically 3 to 6 days. Seventeen patients whose PaCo2 was greater
interesting paper, alcoholism and anesthesia are related, than 24 mm Hg did not demonstrate this prolongation.
and both are assumed to be capable of causing long- Later, Hovorka demonstrated that hypocarbia (28 mm Hg
term changes in susceptible (from insane stock) pa- for patients aged 60 years and older and 22 mm Hg for
tients. patients aged 46 and younger) during anesthesia caused a
deterioration in recovery tests that was noticeable up to
48 hours postoperatively.40
NITROUS OXIDE
A handful of case reports have pointed to the rare but
potentially devastating effects of nitrous oxide. A case What Systems Other than the
carefully studied by Erbe and Salis provides a stark
example.35 The authors describe the death of a 3-month- Central Nervous System Can
old infant who, within a week, had two uneventful ad- Incur Lingering Damages After
ministrations of anesthetics including nitrous oxidefirst
for a biopsy and then for the excision of a soft tissue Anesthesia?
tumor. The infant recovered from both anesthetics with-
out complications and each time was discharged home, As long as we do not understand how anesthetics may
apparently in good health. However, within 2 weeks, the affect CNS function in grossly leading to protracted or per-
infant developed severe neurologic deficits and died. The manent weak mindedness or resulting in mild POCD, we
authors explain that nitrous oxide irreversibly oxidizes need to examine all mechanisms by which anesthetics pro-
the cobalt atom of vitamin B12 , thereby inhibiting the duce organic damage outlasting their sojourn in the body.
C H A P T E R 2 7 / P O S T O P E R AT I V E C O G N I T I V E D Y S F U N C T I O N 383
Methoxyflurane and halothane provide two well-known surgical team to any concerns that may require additional
examples. attention (e.g., anterograde memory impairment) and also
The biotransformation of methoxyflurane by the provides recommendations to the staff caring for the pa-
hepatic and renal, inducible P-450 systems generates tient post surgery (inpatient as well as rehabilitation).
fluoride ions that have been shown to cause renal Postoperatively, the neuropsychologist assists with cogni-
damage.41 In a very few patients, halothane anesthesia was tive monitoring. When postoperative cognitive difficulties
followed by fatal hepatitis. Only years after the first case arise and the patient is stable, the opportunity for a cog-
reports appeared in the literature was it possible to form a nitive training program, focusing on problem areas, is
picture of how halothane, a small halogenated molecule, provided.
could induce, in rare instances, the characteristics of an The cognitive intervention program is based largely
immune-mediated reaction. We now understand that the on research indicating that older adults have the ability
likely culprit is not halothane, but a metabolite inducing to improve their memory, reasoning, and speed of
an antitrifluoroacetyl protein antibody. The fact that the processing functions.45 The intervention program is in
vast majority of patients suffer no such complications, collaboration with the Department of Clinical and Health
even after repeated exposure to the drug, points to a Psychology, particularly the cognitive aging specialty.
genetic susceptibility.42 Presently, we subdivide training based on the cognitive
domains affected (i.e., memory, attention, reasoning).
For example, memory training consists of learning new
strategies for encoding and ways to assist with retrieval.
What Lasting Effects Does For clinicians in the private sector, such a cognitive
intervention program will call for extraordinary efforts.
Surgery Have on Postoperative However, a clinician can recommend a formal neuropsy-
Cognitive Dysfunction? chological evaluation to patients or inform them about
other available cognitive treatment options. These rec-
Certain surgical procedures are associated with embolism ommendations will be more efficiently addressed if the
(e.g., air, clots, tumor, fat). Changes in clotting function anesthesiologist has the results on cognitive function from
intraoperatively and postoperatively may produce emboli the presurgical assessment (e.g., interview, MMSE, Clock
in the postoperative period also. The comforting image drawing).
that these venous emboli will be filtered by the lung and With regard to pharmacologic treatment approaches,
therefore will not gain access to the cerebral circulation we regret to state that, like cognitive interventions, we
needs to be abandoned. On one hand, up to 30% of patients are unaware of any formal studies. The value of neu-
can have a potentially patent foramen ovale that allows roprotective agents and acetylcholinesterase inhibitors
material from the right atrium to slip over to the left administered before surgery has been questioned, but
atrium when atrial pressure on the right exceeds that on randomized clinical studies are yet to be published. Such
the left.43 On the other hand, the lung is not as competent studies may one day alter surgical planning, particularly
a filter as has been thought in the past. Sulek et al. have for elective surgeries involving at-risk patients.
observed emboli to pass through the lungs and into the Overall, POCD and its treatment need to be formally
brain, even in patients with an intact atrial septum.44 addressed both clinically and in research. This issue will be
We do not know whether observed or other nonobserved particularly important in the coming years as the number
embolic events cause or contribute to POCD. of aged adults increase (the largest risk population)
and the number of elective surgeries in that age-group
increases.
to learn test procedures and improve test-taking strategies. This information alone is clinically meaningful and, from
These effects significantly influence an investigators a neuropsychological standpoint, may suggest subtle im-
interpretation of an individuals or groups performance pairments in cognitive function. Individual change scores
on cognitive measures over time. Practice effects are more allow researchers to examine changes in frequency rates
likely to be observed on tests that have a large speed based on a cut-off point. Researchers can choose to exam-
component,50 involve learning,51,52 or involve a solution ine POCD based on a decline by 1.0 or 1.5 or, for the most
that can be easily conceptualized once it is obtained.53 conservative approach, 2.0 standard deviations.46,49
Unfortunately, alternate forms help reduce, but do not Furthermore, researchers need to decide whether
completely eliminate, repeated test-taking effects. to define POCD by the score decline of a group or
Control groups are essential for POCD research. individuals using one test, two or more tests, or a
An age-matched, nonsurgical group that is tested with composite test score. Rasmussen et al.48 suggest a decline
the same measures at the same time intervals will at of 2 standard deviations on at least two out of five
the very least: (i) Provide information regarding alter- different tests that provide a total of seven different
nate testretest reliabilities; (ii) allow investigators to scores. Other researchers, however, may use more or
calculate nonsurgical practice effects for group com- fewer tests. The more tests used, the greater the likelihood
parison purposes; and (iii) provide normative refer- of identifying a change and, hence, producing a type I
ence scores for calculating z-based individual composite error. Composite scores, specifically domain composite
change indices (z-score = [(raw scorereference popu- test scores, may serve as a solution to this dilemma.
lation mean)/reference population standard deviation]). Composite scores are summed or averaged scores based
Furthermore, an age- and disease-matched control group on the group of discrete tests. Hence, composite scores
can help parcel out the effects of health changes from that typically improve reliability (the more scores, the more
of surgery. reliable the sum55 ), simplify the decision process as
Ceiling effects and floor effects influence the inter- to how many tests should be used to define POCD,
pretation of POCD severity and frequency as well. Ceiling and, consequently, reduce the number of statistical tests
effects may identify high baseline performers as impaired needed for group or individual comparison purposes.
at postoperative testing. Floor effects, on the other hand Importantly, the creation of domain composite scores
may protect poor baseline performers from demonstrating requires an advanced understanding of neurocognitive
further deterioration. These differences are especially pro- substrates and psychometrics. Some researchers may
nounced when dealing with changes in cognitive scores.54 attempt to use a factor analysis approach to create domain
These effects are attributable to the psychometric proper- composite scores. Although this approach is encouraged,
ties of cognitive measures (e.g., range of possible scores, caution is vital if the factors are to provide any valuable
normal distribution) and test the appropriateness for the information regarding domain differences. If a factor
population of interest (i.e., you would not administer a analysis is not hypothesis driven or conceptually based
16-word verbal learning test to patients with moderate on an understanding of neurocognitive substrates and
cognitive impairment, for this would most likely result associated test measures, the resulting factors are then less
in poor performance and a floor effect). Other important likely to explain the significant amount of the variance in
issues involve whether change scores should actually be scores and will often fail to provide meaningful composite
used to assess POCD and, if used, how the scores should domains for interpretation.
be calculated.48
Fortunately, most recent publications on POCD show a 14. Floyd TF, Pallav NS, Price CC, et al. Clinically silent cerebral
growing awareness of these issues. ischemic events after cardiac surgery: Their incidence,
regional vascular occurrence, and procedural dependence.
Ann Thorac Surg. 2006;81:2160.
15. Goto TT, Baba K, Matsuyama K, et al. Aortic atheroscle-
rosis and postoperative neurological dysfunction in elderly
KEY POINTS coronary surgical patients. Ann Thorac Surg. 2003;75:1912.
1. POCD is real. 16. Caplan LR, Hennerici M. Impaired clearance of emboli
2. More common in older patients. (washout) is an important link between hypoperfusion,
embolism, and ischemic stroke. Arch Neurol. 1998;55:1475.
3. Most patients with POCD recover within 3 months.
17. Roman GC, Erkinjuntti T, Wallin A, et al. Subcortical
4. Brief pre- and postoperative cognitive tests may ischaemic vascular dementia. Lancet Neurol. 2002;1:426.
provide a useful baseline and postoperative reference. 18. Price CC, Garvan CW, Monk T. Neurocognitive performance of
5. POCD rates are equivalent for general versus regional older adults with post operative cognitive dysfunction (POCD).
anesthesia. San Francisco: American Society of Anesthesiologists; 2003.
6. Procedures with potential for intravascular embolism 19. Mink J. The basal ganglia: Focused selection and inhibition
of any type appear to have a higher incidence of of competing motor programs. Prog Neurobiol. 1996;50:
POCD. 381.
7. Emboli can get to the brain through the pulmonary 20. Chudasama Y, Robbins TW. Functions of frontostriatal
circulation. systems in cognition: Comparative neuropharmacological
studies in rats, monkeys and humans. Biol Psychiatry.
8. Research needs to address the usefulness of cognitive
2006;73(1):19.
intervention for patients with POCD. 21. Pantoni L, Garcia JH. Cognitive impairment and cellular/
9. POCD research is best done in a collaborative setting vascular changes in the cerebral white matter. Ann N Y Acad
and with rigorous attention to methodological issues. Sci. 1997;826:92.
22. Scoville WB, Milner B. Loss of recent memory after
REFERENCES bilateral hippocampal lesions. J Neurol Neurosurg Psychiatry.
1957;20:11.
1. Bedford PD. Adverse cerebral effects of anaesthesia on old 23. Andrews JS, Jansen JH, Linders S, et al. Effects of disrupting
people. Lancet. 1955;269(6884):259. the cholinergic system on short-term spatial memory in rats.
2. American Psychiatric Association. Diagnostic and statistical Psychopharmacology (Berl). 1994;115:485.
manual of mental disorders (DSM-IV), 4th ed. Washington 24. Rasmussen LS, Johnson T, Kuipers HM, et al. Does anaesthe-
DC: American Psychiatric Association; 1994. sia cause postoperative cognitive dysfunction? A randomised
3. Moller JT, Cluitmans P, Rasmussen LS, et al. International study of regional versus general anaesthesia in 438 elderly
Study of Post-Operative Cognitive Dysfunction. Long-term patients. Acta Anaesthesiol Scand. 2003;47:260.
postoperative cognitive dysfunction in the elderly ISPOCD1 25. Williams-Russo P, Sharrock NE, Mattis S, et al. Cognitive
study. ISPOCD investigators. Lancet. 1998;351(9106):857. effects after epidural vs general anesthesia in older adults.
4. Monk T, Garvan CW, Dede DE. Predictors of postoperative A randomized trial. JAMA. 1995;274:44.
cognitive dysfunction following major surgery. Anesthesiol- 26. Jackson J. On a peculiar disease resulting from the use of
ogy. In Press. ardent spirits. N Engl J Med Surg. 1822;2:352.
5. Abildstrom H, Rasmussen LS, Rentowl P, et al. International 27. Korsakoff SS. Uber Erinnerungsstorungen (Pseudorem-
Study of Post-Operative Cognitive Dysfunction. Cognitive iniszenzen) bei polyneuritischen psychosen. Allgemeine
dysfunction 12 years after non-cardiac surgery in the el- Zeitschrift fur Psychiatrie; 1891:47.
derly. ISPOCD group. Acta Anaesthesiol Scand. 2000;44:1246. 28. Blass JP, Gibson GE. Abnormality of a thiamine-requiring
6. Newman MF, Croughwell ND, Blumenthal JA, et al. Predic- enzyme in patients with Wernicke-korsakoff syndrome.
tors of cognitive decline after cardiac operation. Ann Thorac N Engl J Med. 1977;297:1367.
Surg. 1995;59:1326. 29. Lassen HCA, Henriksen E, Neukirch F, et al. Treatment
7. Satz P. Brain reserve capacity on symptom onset after brain of tetanus: Severe bone-marrow depression after prolonged
injury: A formulation and review of evidence for threshold nitrous-oxide anesthesia. Lancet 1956;1:527530.
theory. Neuropsychology. 1993;7:273. 30. Eastwood DW, Green CD, Lambdin MA, et al. Effect of
8. Bernstein GM, Offenbartl SK. Adverse surgical outcomes nitrous oxide on the white-cell count in leukemia. N Engl J
among patients with cognitive impairments. Am Surg. 1991; Med. 1963;268:297.
57:682. 31. Layzer RB. Myeloneuropathy after prolonged exposure to
9. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. nitrous oxide. Lancet. 1978;2(8102):1227.
A practical method for grading the cognitive state of patients 32. Bruchiel KJ, Stockard JJ, Calverley RK, et al. Electroen-
for the clinician. J Psychiatr Res. 1975;12:189. cephalographic abnormalities following halothane anesthe-
10. Libon DJ, Malamut BL, Swenson MR, et al. Further analysis sia. Anesth Analg. 1978;57:244.
of clock drawings among demented and non-demented 33. Stark AH, Calverley RK, Smith NT. Psychological function-
subjects. Arch Clin Neuropsychol. 1996;11:193. ing after halothane or enflurane anesthesia. Anesth Analg.
11. Brandt J. The Hopkins verbal learning test: Development of 1980;59:245.
a new verbal memory test with six equivalent forms. Clin 34. Savage GH. Insanity following the use of anaesthetics in
Neuropsychol. 1991;5:125. operations. Br Med J. 1887;3:1199.
12. Wechsler D. Wechsler memory scale, 3rd ed. San Antonio: 35. Erbe RW, Salis RJ. Severe methylenetetrahydrofolate
The Psychological Corporation; 1997. reductase deficiency, methionine synthase, and nitrous
13. Price CC, Bowers D, Schmalfuss IM, et al. Pre-operative white oxideA cautionary tale. N Engl J Med. 2003;349:5.
matter alterations predicting post-operative cognitive decline 36. Batten CT, Courville CB. Mental disturbances following
in healthy older adults. J Int Neuropsychol Soc. 2006;12:265. nitrous oxide anesthesia. Anesthesiology. 1940;1:261.
386 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
37. Robinson JS, Gray TC. Observations on the cerebral effect of 47. Murkin JM, Stump DA, Blumenthal JA, et al. Defining
passive hyperventilation. Br J Anaesth. 1961;33:62. dysfunction: Group means versus incidence analysisa
38. Pierce ECJ, Lambertsen CJ, Deutsch S, et al. Cerebral statement of consensus. Ann Thorac Surg. 1997;64:904.
circulation and metabolism during thiopental anesthesia and 48. Rasmussen LS, Larsen K, Houx P, et al. The assessment of
hyperventilation in man. J Clin Invest. 1962;41:1664. postoperative cognitive function. Acta Anaesthesiol Scand.
39. Wollman SB, Orkin LR. Postoperative human reaction time 2001;45:275.
and hypocarbia during anaesthesia. Br J Anaesth. 1968; 49. Rasmussen LS. Defining postoperative cognitive dysfunc-
40:9209. tion. Eur J Anaesthesiol. 1998;15:761.
40. Hovorka J. Carbon dioxide homeostasis and recovery af- 50. Salinsky MC, Storzbach D, Dodrill CB, et al. Test-retest
ter general anaesthesia. Acta Anaesthesiol Scand. 1982;26: bias, reliability, and regression equations for neuropsycho-
498. logical measures repeated over a 1216-week period. J Int
41. Kharasch ED, Hankins DC, Thummel KE. Human kidney Neuropsychol Soc. 2001;7:597.
methoxyflurane and sevoflurane metabolism: Intrarenal flu- 51. Uchiyama CL, DElia LF, Dellinger LM, et al. Alternate forms
oride production as a possible mechanism of methoxyflurane of the auditory-verbal learning test: Issues of test compara-
nephrotoxicity. Anesthesiology. 1995;82:689. bility, longitudinal reliability, and moderating demographic
42. Brown BBJ, Gandolfi AJ. Adverse effects of volatile anaes- variables. Arch Clin Neuropsychol. 1995;10:133.
thetics. Br J Anaesth. 1987;59:14. 52. Duff K, Westervelt HJ, McCaffrey RJ, et al. Practice effects,
43. Hagen PT, Scholz DG, Edwards WD. Incidence and size of test-retest stability, and dual baseline assessments with the
patent foramen ovale during the first ten decades: A necropsy California verbal learning test in an HIV sample. Arch Clin
study of 965 normal hearts. Mayo Clin Proc. 1987;59:17. Neuropsychol. 2001;16:461.
44. Sulek CA, Davies LK, Enneking FK, et al. Cerebral microem- 53. Lowe C, Rabbitt P. International Study of Post-Operative
bolism diagnosed by transcranial Doppler during total knee Cognitive Dysfunction. Test/re-test reliability of the CANTAB
arthroplasty: Correlation with transesophageal echocardiog- and ISPOCD neuropsychological batteries: Theoretical and
raphy. Anesthesiology. 1999;91:672. practical issues. Cambridge neuropsychological test auto-
45. Ball K, Berch DB, Helmers KF, et al. Effects of cognitive mated battery. Neuropsychologia. 1998;36:915.
training interventions with older adults: A randomized 54. Browne SM, Halligan PW, Wade DT, et al. Cognitive per-
controlled trial. JAMA. 2002;288:2271. formance after cardiac operation: Implications of regression
46. Murkin JM, Newman SP, Stump DA, et al. Statement of toward the mean. J Thorac Cardiovasc Surg. 1999;117:481.
consensus on assessment of neurobehavioral outcomes after 55. Lezak M. Neuropsychological assessment, 3rd ed. New York:
cardiac surgery. Ann Thorac Surg. 1995;59:1289. Oxford Press; 1995.
CHAPTER SPINAL CORD INJURY
A
from the third floor balcony of his apartment cervical spine mobility. Upper cervical spine injuries
house while trying to retrieve his neighbors and the problem of a severe ligamentous injury without
keys. When the Emergency Medical Services radiographic abnormalities (SCIWORA) are more likely in
personnel respond to his residence, he is young children.1 The incidence of SCI in patients who are
found to be awake and alert, but unable to older than 60 years is rising, and falls are the most likely
move his arms and legs. Oxygen by face mask is applied, mechanisms of injury in this population. Nontraumatic
and he is carefully placed on a spine board with a cervi- SCI may result from degenerative spine disease, ischemia,
cal collar and transported via ambulance. In the emergency demyelination, inflammation, and extrinsic neoplastic,
department, he is noted to have complete motor and sen- hemorrhagic or pyogenic masses.
sory loss below C5, a blood pressure of 85/50 mm Hg,
heart rate of 46 beats per minute, respiratory rate of
22 breaths per minute, and O2 saturation reading of 95% Pathophysiology
by pulse oximetry. A 30 mg per kg intravenous bolus of
methylprednisolone is initiated, followed by an infusion Most experimental studies of SCI have been performed
of 5.4 mg/kg/hour for 23 hours. Cervical spine radiographs in animal models subjected to spinal cord trauma be-
reveal subluxation of C5 on C6 with ligamentous instabil- cause this is the most common mechanism of injury in
ity. The patient is placed in 10 lb of traction that reduces the the human population. In most cases, traumatic injury
misalignment of the spinal elements. He is admitted to the to the spinal cord is limited to stretching or shearing
intensive care unit (ICU) and receives supplemental oxy- of the neural elements and/or compression manifested
gen. Forced vital capacity measurements are performed by damage to spinal intramedullary blood vessels and
three times daily, and he is scheduled in the operating hemorrhage into the central gray matter of the cord.2
room for cervical spine stabilization the following day. In The critical contributions of secondary mechanisms of
the operating room, he is cooperative and receives careful injury to the ultimate outcome following acute SCI have
sedation and topicalization for an awake fiberoptic intu- been recognized, and therapy is targeted to limit sec-
bation. An arterial line is placed in his radial artery, and ondary injury. At the time of injury, the initial disruption
the intraoperative mean arterial pressure is maintained at of spinal cord blood vessels causes a loss of autoregula-
85 to 90 mm Hg. The patient is extubated following the tion and altered autonomic tone. Subsequent hypotension
procedure, and returns to the ICU for additional monitor- and probable neuronal hypoxia contribute to the release
ing. He is transferred to a rehabilitation center 5 days after of calcium within the neurons, which causes an un-
his initial injury. coupling of oxidative phosphorylation, vasospasm, and
activation of membrane phospholipases. These chemi-
cals, in turn, produce a breakdown of the cell membrane
and the release of free radicals. Inflammatory processes
What Baseline Knowledge Is are also important acutely and over the next several days,
as demyelination has been linked to the attraction of
Relevant? neutrophils and macrophages to the traumatized area.
Excitatory amino acids (EAAs) such as glutamate are
released and bind to N-methyl-D-aspartate receptors, caus-
TRAUMATIC SPINAL CORD ing receptor-gated channels to open and a further increase
INJURY in calcium flux. Nitric oxide synthetase is activated. De-
layed neuronal cell death, also referred to as apoptosis,
The average age of a patient with an acute spinal cord may appear remote from the immediate time of injury, al-
injury (SCI) is 32.3 years, and the typical victim is a though this process is likely to have been initiated through
387
388 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Cell death
apoptosis
What Are the Important
FIGURE 28.1 Schematic of spinal cord injury. Considerations for Acute Care
of the Spinal Injury Patient?
RNA and protein synthesis within minutes of injury
(see Fig. 28.1).
In 1990, results from the second National Acute CRITICAL CARE
Spinal Cord Injury Study (NASCIS-2) demonstrated that
patients achieve improvement in motor function, as well Spinal shock is characterized by flaccid paralysis of ex-
as pin and touch sensation at 6 weeks, 6 months, and tremities below the level of the lesion, and absent tendon
a year following treatment with methylprednisolone.3 and patellar reflexes. The hemodynamic consequences
Steroids have several theoretical beneficial effects, includ- of spinal shock have been more recently characterized
ing suppression of vasogenic edema by restoration of the as neurogenic shock.9 Systemic vascular resistance is
blood: Central nervous system (CNS) barrier, enhance- decreased and venous capacitance is increased. If the
ment of spinal cord blood flow, stabilization of lysosomal spinal cord lesion is above T5, interruption of sympa-
membranes, inhibition of pituitary endorphin release, al- thetic outflow to the heart (T1-T5) causes unopposed
teration of electrolyte concentrations in injured tissue, parasympathetic tone. Very often, patients manifest per-
and attenuation of the inflammatory response. Methyl- sistent bradycardia for several weeks following a severe
prednisolone is administered as an initial bolus of 30 mg cervical cord injury. Transient external stimuli such as
per kg intravenously, followed by an infusion of 5.4 mg per tracheal suctioning or passive movement in patients may
kg for 23 hours. It is crucial that the methylprednisolone further exacerbate bradycardia, which is responsive to
be initiated within 8 hours of injury. The NASCIS-3 further therapy with atropine.
refined the time period of treatment so that if a patient In the patient with a cervical spine injury, the res-
receives methylprednisolone treatment within 3 hours of piratory system becomes a focus of intensive care. The
injury, they should be maintained on the treatment proto- diaphragm, the principal muscle responsible for inspira-
col for 24 hours.4 Patients initiating steroid therapy 3 to tion, accounts for more than 60% of the air exchange
8 hours after injury should have the maintenance dose during quiet breathing. Innervation is through elements
extended for 48 hours.5 of the cervical spinal cord from C3 to C5. Therefore, if
In recent years, the administration of high dose no other injuries are involved, patients with an intact C5
methylprednisolone for acute SCI has become contro- innervation may not require ventilatory support acutely,
versial. Concerns have been voiced over the side effects but have the potential for deterioration at any time in
of the drug, including an increased incidence of pneu- the early phases of injury. Loss of expiratory function is
monia, pressure sores, wound infection, delayed healing, to be expected in those patients who have lost thoracic
gastrointestinal bleeding, deep venous thrombosis, acute input from the spinal cord, and gradual deterioration of-
adrenal insufficiency, and the small beneficial neurologic ten occurs up to 5 days after injury in the absence of
effects that have been reported utilizing posthoc data aspiration or pulmonary edema. The development of a
analysis in the clinical trials.6 Indeed, the Guidelines left-sided pneumonia is also common, as it is difficult
CHAPTER 28/SPINAL CORD INJURY 389
to clear secretions.10 Elective intubation to temporarily Often, these procedures are planned while the patient is
support ventilatory function is indicated, as many acutely still within the neurogenic shock phase of injury.13
quadriplegic patients will increase vital capacity over a Many factors related to surgery and anesthesia such
5-week period to at least double that measured on admis- as positioning, positive-pressure ventilation, relative hy-
sion. This improvement often correlates with the onset of povolemia, and myocardial depression or vasodilatation
the spastic phase of SCI when chest muscles become con- caused by anesthetic agents have the potential to exacer-
tracted and better able to support ventilatory mechanics. bate the neurogenic shock syndrome and cause deteriora-
With neurogenic shock, peripheral vascular resis- tion of spinal cord perfusion. Additionally, hypothermia
tance is decreased, and mean arterial blood pressure may aggravate circulatory instability. In those patients
may require support. Certain centers have been extremely who manifest circulatory instability, intraoperative moni-
aggressive in support of cardiac output, such that the toring should include an intra-arterial catheter and a cen-
therapeutic aim has been a mean arterial pressure of tral venous, pulmonary artery catheter or transesophageal
85 mm Hg or above, achieved with combinations of in- echocardiograph to provide ongoing assessments to max-
travenous fluids, colloids, vasopressors, and monitoring imize systemic and spinal cord perfusion.
utilizing a pulmonary artery catheter. Retrospective re- Electrophysiologic monitoring, such as somatosen-
views have demonstrated results that, when compared to sory evoked potential (SSEP) monitoring, or transcranial
historical controls, are encouraging; however, no random- motor evoked potentials (tMEP) may be requested in-
ized prospective trials with hemodynamic controls have traoperatively. Injury to the cord may disrupt the SSEP
been initiated to date. The Guidelines Committee of the signals and cause signal derangements in the case of a par-
AANS and CoNS Joint Section on Disorders of the Spine tial SCI, making interpretation difficult. Before surgery,
and Peripheral Nerves state that Class III evidence from obtaining an adequate baseline tracing may be impossi-
the literature suggests that maintenance of mean arterial ble. If SSEP monitoring is planned, a total intravenous
pressure at 85 to 90 mm Hg after acute SCI for 7 days anesthetic technique utilizing continuous infusions of nar-
is safe and may improve spinal cord perfusion, and ul- cotics and propofol improves the ability to obtain reliable
timately, neurologic outcome.11 Fluid titration utilizing SSEP signals.
changes in cardiac output and filling pressures is recom- Goals for optimal anesthetic management of the pa-
mended because of concern over the disruption of the tient should include minimal spine manipulation, mainte-
pulmonary-capillary endothelium and noncardiac edema, nance of adequate spinal cord perfusion and normocarbia,
which has been reported even in relatively young patients facilitation of intraoperative monitoring, and a goal of
following cervical SCI. Efforts to increase cardiac out- neurologic assessment of the patient as rapidly as possible
put and systemic oxygen delivery are based in theory on at the conclusion of the operative procedure. Assuming
experimental data which show that shortly after spinal that the patient is reasonably cooperative, a controlled,
cord trauma, there is a significant diminution of spinal awake fiberoptic technique to secure the airway in the
cord blood flow, and vasospasm in the central gray matter, patient with a cervical cord injury would be most of-
provoking further ischemic insults. If a vasopressor is nec- ten utilized, while succinylcholine is avoided following
essary, an agent with both -adrenergic and -adrenergic the first 24 hours after injury due to the proliferation of
action, such as dopamine or epinephrine, should be uti- extrajunctional myoneural receptors and associated hy-
lized to counter the loss of sympathetic tone and provide perkalemia following succinylcholine depolarization of
chronotropic support to the heart. these receptors. If clinical signs indicate that extubation
The state of neurogenic shock generally persists from is not desirable immediately following surgery because
1 to 3 weeks after injury and resolves with recovery of of instability in the cardiovascular or respiratory status
spinal reflexes. Recovery from spinal shock indicates of the patient, it is extremely useful for the patient to
that nonsynaptic neurotransmission is occurring and, be managed such that repeated neurologic examinations
in this phase, some patients may be hypersensitive can be performed. Recurrent compression may occur ow-
when neuroactive substances such as vasopressors are ing to vertebral malalignment, hematoma, swelling, or
administered. Receptors and synapses on the surface of delayed abscess formation. Rapid identification of such
partially denervated spinal cord cells may be upregulated compression and operative interventions are necessary.
through a denervation supersensitivity mechanism.12
injury. Complete spinal cord transection at a high cervical Hoffman et al. embarked upon the National Emergency
level (C1) results in pentaplegia: Paralysis of the lower X-Radiography Utilization Study (NEXUS) in an attempt
cranial nerves, the diaphragm, and loss of sensation and to better define those patients requiring radiographic
motor function of all four extremities. Quadriplegia refers cervical spine evaluation.15 More than 34,000 patients
to injuries at C3-C5, which results in the loss of sensation were enrolled at 21 medical centers across the United
and motor function of all four extremities, as well as States.16 The NEXUS low-risk criteria (see Table 28.2)
impairment of the diaphragm, while the cranial nerves were found to be 99.6% sensitive and 12.9% specific for
and sensation to the face and neck and accessory muscles cervical spine injury. In addition, Stiell et al. formulated
of the neck are spared. Tetraplegia refers to injuries the Canadian Cervical Spine Rule (see Fig. 28.2) which
involving C5 and C6, in which diaphragmatic function is encompasses criteria similar to the NEXUS low risk
retained as well as some movement of the upper extremity,
but there is a total loss of function of the lower extremities.
Paraplegia refers to injuries below T1, resulting in loss of Any high-risk factor that mandates
function of the lower extremities.14 radiography?
Incomplete SCI syndromes are listed in Table 28.1. Age 65 yr or dangerous mechanism
Of these, the central cord syndrome is the most common. or paresthesias in extremities
These patients frequently require surgery, either for
spinal stability or an associated injury, presenting the No
anesthesiologist with the quandary of managing a patient
Any low-risk factor that allows safe Yes
with a potential cervical spine injury. Because of the
potential for catastrophic neurologic injury, nearly every assessment of range of motion?
blunt trauma victim is routinely subjected to a plain film Simple rear-end motor vehicle collision
or sitting position in the emergency
radiographic cervical spine series. A complete series must No Radiography
department or ambulatory at any time
minimally include a lateral cross-table cervical spine, an or delayed (not immediate) onset of
anteriorposterior view, and an open-mouth odontoid neck pain or absence of midline
view. The occipitocervical junction and all seven cervical cervical-spine tenderness
Unable
vertebrae, including the C7-T1 junction, must be viewed.
To better delineate those emergency department Yes
patients who may be at low risk for cervical spine injury,
Able to rotate neck actively?
TABLE 28.2 National Emergency X-Radiography 45 degrees left and right
Number of Patients
with Finding (%) (N = 27) MRI Finding
25 (93%) Spinal cord edema or
contusion
13 (48%) Central or paramedian
disc herniation
11 (41%) Spinal stenosis
Data from Hendey GW, Wolfson AB, Mower WR, et al. Spinal cord Head of
injury without radiographic abnormality: Results of the National humerus
Emergency X-Radiography Utilization Study in blunt cervical trauma.
J Trauma. 2002;53:1. (Table 2, page 2) Scapulla
c6
criteria and additionally requires the patient to rotate c7
Clavicle
his head.17 This large, multicenter study of 8,283 patients
yielded a sensitivity of 99.4% and specificity of 45.1%.
Although a contentious debate as to the superiority of one
criterion versus the other continues, both studies suggest
that many patients are needlessly radiographed, and that
many cervical spines may be cleared in the emergency
department based on clinical factors.
Unfortunately, radiographic examination of the spine FIGURE 28.3 C6-7 subluxation (plain radiograph).
may not detect an uncommon SCI. Hendey et al. evaluated
the data provided by the NEXUS study to better determine
the incidence of SCIWORA.18 SCIWORA is defined as
SCI demonstrated by MRI when complete and adequate the radiographs of a trauma patient who sustained a C6-7
cervical spine plain radiographs revealed no injury. A subluxation, resulting in quadriplegia. Although the plain
total of 34,069 patients were entered into the NEXUS radiograph clearly demonstrates the C6-7 subluxation, the
database; 818 (2.4%) had cervical spine injury. Twenty- computed tomography (CT) reconstruction presents fine
seven (0.08%) were patients with SCIWORA with a variety detail of the bony architecture.
of MRI findings (see Table 28.3). Of note, all 27 patients
were found to have at least one documented clinical
finding which did not meet the low-risk NEXUS criteria,
and all were evaluated with plain radiograph films upon
initial presentation.
The NEXUS study revealed a typical distribution of
cervical fractures (C2 23.9%, C6 20.25%, C7 19.08%, C5
14.98%) and subluxations (C5-6 25.11%, C6-7 23.77%,
C4-5 16.96%) (15). Of most importance is the recognition
of those fractures that result in an unstable neck (see
Table 28.4), as those patients are at greatest risk of
suffering permanent neurologic injury. In summary,
trauma patients, particularly those who do not meet the
NEXUS or Canadian low-risk criteria, should be treated
as if they have an unstable spine. Figures 28.3 and 28.4 are
with an unstable neck, awake fiberoptic intubation should a potential wake-up test if requested should signals be
be given primary consideration. Modern approaches to lost.
airway anesthesia and facility with the fiberoptic bron- The descending motor tracts can be monitored with
choscope have made this a rapid and reliable technique. the newer transcranial motor evoked potentials. An
If an awake fiberoptic intubation is not feasible in electrical or magnetic stimulus is passed through the
the patient with an unstable cervical spine, other tech- cranium to the motor cortex and descends through the
niques may be employed which limit the movement of cerebral peduncles and pons to the medullary pyramids.
the cervical spine. Lennarson et al. studied cadavers with Most fibers cross in the pyramids to form the lateral
fresh ligamentous disruptions at C4-C5 and found man- corticospinal tract, although some remain ipsilateral,
ual in-line stabilization to be effective in reducing cervical creating the anterior corticospinal tract. The signal is
motion.26 Wahlen et al. found the Bullard laryngoscope recorded on extremities by electromyelography when the
(Circon, ACMI, Stamford, CT), a rigid optical stylet, and muscles respond to the descending impulse. As a result,
the intubating laryngeal mask airway cause less cervi- neuromuscular-blocking drugs are allowed to dissipate
cal motion than the Macintosh laryngoscope (MVL, Karl before signal acquisition. Inhalational anesthetics and
Storz Gmb & Co. KG, Tuttlingen, Germany) in patients benzodiazepines are avoided in favor of opioids and
with no known cervical pathology.27 However, no sta- propofol to optimize the capture of the initial stimulus
bilization maneuvers were undertaken in their study. at the motor cortex.30
Turkstra et al. found the Lightwand to be superior to direct
laryngoscopy with a Macintosh blade and manual in-line
stabilization.28 The GlideScope (Saturn Biomedical Sys-
tems Inc., Burnaby, B.C., Canada) reduced movement Have Can the Spinal Cord Be
50% in one of four cervical motion segments when com- Injured During Vascular
pared to direct laryngoscopy with the Macintosh blade,
but took 62% longer. Sahin et al. however, confirmed that Surgery?
fiberoptic intubation resulted in less motion than direct
laryngoscopy and the intubating laryngeal mask airway.29 Most injuries to the spinal cord involve trauma and
During intraoperative management of the patient damage to the surrounding ligamentous and skeletal
with degenerative spine disease, preventing further injury framework. It is often the pathologic movement of these
to the spinal cord is the primary goal. Unfortunately, structures, impinging upon the structural integrity of
few data exist to guide the anesthesiologist in achieving the spinal cord, which produces the damage. Another
optimal perfusion to the spinal cord. Maintenance of mechanism of injury is ischemia, which may occur as a
mean arterial pressure (zero-referenced to the head) at consequence of aortic surgery. Paraplegia and paraparesis
baseline or slightly elevated levels has been suggested are the most devastating complications in survivors of
as a means of preventing secondary SCI. Intraoperative thoracoabdominal aortic aneurysm (TAAA) repair. Not
monitoring may be utilized to evaluate the integrity of the only is the quality of life decreased, but there is also
spinal cord in the presence of optimal perfusion. SSEPs significant shortening of lifespan. In one large study
are frequently utilized and consist of peripheral electrical of 1,509 patients undergoing TAAA repair, the 5-year
stimuli and the proximal recording of those impulses as survival rate decreased from 62% for those without SCI,
they travel proximally to the somatosensory cortex. The to 44% for those who incurred an SCI.31 For the surgery
nerves chosen to stimulate are usually the median or itself, reported mortality is between 3% to 23%, with
ulnar nerve for the upper extremity and the posterior an additional 3% to 18% suffering SCI.32 The risk of
tibial nerve for the lower extremity. The impulse ascends SCI following a TAAA repair depends on many factors,
the spinal cord in the dorsal columns and synapses in including the type of aneurysm and the duration of aortic
the brainstem nuclei before crossing over to synapse in crossclamp time with its ensuing ischemia-producing
the thalamus. The third and final part of this pathway cytotoxic and reperfusion injury.
is its projection from the thalamus to the cortex. The TAAA is primarily caused by dissection (17%) or
impulse is usually recorded at several locations: Over degenerative atherosclerosis (80%). Of the two main types,
the peripheral nerve proximal to the stimulus, over the degenerative aneurysms tend to be more difficult to repair,
cervical cord, and over the contralateral somatosensory and are likely to have an increased risk of ischemic
cortex. The latency is recorded as the time it takes for a complications due to atheromatous emboli. The Crawford
signal to arrive from its stimulus to its recording electrode. classification system was devised by the vascular surgeon
Ischemia will result in the slowing of signal propagation E.S. Crawford, and stratifies patients according to the
and an increase in the measured latency. The amplitude aortic involvement.33 Types I, II, and III, are considered
of the signals are measured at the recording electrode thoracoabdominal aortic aneurysms. A Crawford type IV
in microvolts. Significant ischemia results in decreases aneurysm is not considered a TAAA, because it involves
of signal amplitude >50%. Both volatile anesthetics and subdiaphragmatic aorta only.33 Owing to their size and
nitrous oxide cause dose-dependent decreases in signal vascular involvement, type II TAAA repairs portend the
amplitude. Therefore, anesthesia for procedures utilizing greatest risk of spinal cord ischemia (see Table 28.5).
SSEPs has evolved to an opioid-based approach with The reason to fear intraoperative ischemia is the
reduced inhalation anesthetic doses to improve signal unique and precarious blood supply to the spinal cord
quality, provide for a rapid assessment, and prepare for which derives from three main arteries. The vertebral
394 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
arteries, or posterior inferior cerebellar arteries, give What about Spinal Cord
rise to the paired posterior spinal arteries which supply Protection?
the posterior one third of the spinal cord, including
dorsal columns and much of the spinothalamic tracts.
The anterior spinal artery, also arising from branches Multiple modalities are used to provide spinal cord
of the vertebral arteries, is a single vessel that supplies protection during a TAAA repair. These can be broadly
the anterior two thirds of the spinal cord. The descending divided into two main categories: Surgical-based and
motor pathways and spinal reflexes are dependent on anesthesia-based procedures. Often, the operative team
perfusion from the anterior spinal artery. In addition to will combine modalities and utilize several techniques
the posterior spinal arteries and the anterior spinal artery, during the repair.
multiple medullary and segmental radicular arteries Surgical-based modalities attempt to minimize spinal
from the deep cervical, intercostal, and lumbar arteries cord ischemia during aortic crossclamping and include
augment spinal cord blood flow. multiple segmental repairs with attendant multiple aortic
The middle thoracic region of the spinal cord tends to crossclamps, augmenting blood flow to the distal aorta
be less well supplied than other cord regions and is subject with partial left heart bypass and distal perfusion, reim-
to watershed areas. The largest radiculomedullary artery, plantation of intercostal and/or lumbar vessels and, in
arteria radicularis magna, or the artery of Adamkiewicz, some cases, deep hypothermic cardiopulmonary bypass
usually enters the spinal cord blood somewhere between (CPB) with circulatory arrest. With most of these tech-
T9 and L5. The contribution of this artery is variable, niques, neurophysiologic monitoring can help guide the
however, as Biglioli et al. found in his study of human surgeon with the repair.
cadavers; specifically, the arteria radicularis magna was There are two major types of neurophysiologic
found below T12 in 70% of cases and between L1 and monitoring used in aortic surgery: Transcranial motor
L3 in 64.7% of cases. Interestingly, the arteria radicularis evoked potentials and SSEP. Changes in latency or
magna was found below L2 in 23.5%, and was located on amplitude of the signals conducted following clamping of
the left side in most cases, almost 63%.34 the aorta, or loss of flow from critical intercostal arteries,
During surgical dissection between T5 and L5, 26 alerts the surgeon to the possible need to reimplant the
intercostal and lumbar arteries can be encountered; how- affected arteries. The anesthesiologist may also respond by
ever, in a series of 184 TAAA repairs, Jacob et al. discov- aggressively managing hemodynamics to optimize spinal
ered on average only five patent arteries in this area. In cord perfusion.
fact, in 8% of the patients, the critical blood supply for Of the two types of evoked potentials available,
the lower half of the anterior spinal artery came from the transcranial motor evoked potentials have superior
the pelvic circulation. During surgical repair, intercostal sensitivity to ischemia in the anterior spinal cord, which
vessels with orifices occluded by aortic plaque showed is the region predominantly affected by TAAA repair.
back-bleeding after endarterectomy, indicating that the Transcranial motor evoked potentials assess the integrity
vessels were still part of a collateral circulation. Jacobs of the anterolateral descending pathways supplied by the
goes further, stating that the artery of Adamkiewicz is anterior spinal artery and do not require signal averaging.
probably irrelevant and does not even exist in its assumed A single stimulation can provide real-time information.
function in most patients with TAAA.35 Still, many sur- Further, after a change or loss of the signal due to spinal
geons attempt to find the arteria radicularis magna with cord ischemia, the signal can regenerate within a short
presurgical arteriograms, in an attempt to delineate which period if perfusion is reestablished.
intercostals or lumbar vessels may be critical for cord SSEPs, on the other hand, monitor the activity of the
perfusion to spare or reimplant them during surgery.36 ascending posterolateral columns of the spinal cord, an
Because of the altered and collateralized blood supply area supplied by the posterior spinal arteries. The blood
in patients with TAAA, it becomes very difficult to deter- supply to the posterior spinal arteries is significantly less
mine how best to preserve spinal cord perfusion pressure, affected by aortic crossclamping, making SSEPs relatively
especially in the anterior spinal artery. Safe ischemic insensitive to this ischemic insult. Also, in comparison to
time during aortic crossclamp has been estimated to be the transcranial motor evoked potentials, loss of potentials
CHAPTER 28/SPINAL CORD INJURY 395
with SSEPs is much more gradual, with a delayed used, the infusion of iced saline through an epidural
restoration when perfusion is restored.38,39 catheter placed in the low thoracic region doubles the
The technical and procedural difficulties inherent CSF pressure. The investigators felt that the increased
to performing transcranial electric (stimulation) motor CSF pressure was innocuous as long as the spinal cord
evoked potentials (teMEP) include the need to avoid in- was appropriately cooled to a target range of 26 C.
halational anesthetics to optimize cortical stimulation, In this study, the mean CSF pressure during aortic
minimize neuromuscular blockade to record electromyo- crossclamp was 35.1 mm Hg, while an arbitrary 30 to
graphy (EMG) peripherally, and the understanding that 40 mm Hg gradient between mean CSF and arterial
signal loss peripherally could reflect either spinal cord is- pressures was maintained. After the aortic crossclamp was
chemia or distal limb ischemia from the aortic crossclamp. removed, CSF was allowed to drain to a mean pressure
Within the vascular surgery community, no consen- of 12 mm Hg. A paraplegia/paraparesis rate of 7% was
sus exists on the optimal method to minimize or avoid reported by these investigators. Interestingly, most (66%)
spinal cord ischemia during TAAA repair. Hypothermic neurologic deficits were not apparent immediately after
CPB with circulatory arrest is at one end of the spec- surgery and did not appear until 48 hours or longer after
trum and usually allows a prolonged aortic crossclamp the operation. Although the etiology of delayed injury
time with an extended duration of spinal cord ischemic is unknown, speculation suggests postoperative spinal
timeas long as 138 minutes.40 Although Kouchoukos cord swelling or hypotension leading to decreased cord
et al. had excellent success with reasonable morbidity and perfusion.44
mortality in a series of 161 patients, other surgeons shy A variety of pharmaceutical products have been
away from hypothermic CPB with circulatory arrest due to investigated experimentally and clinically to assess their
the serious complications of coagulopathy and bleeding, ability to decrease spinal cord damage in the face of
pulmonary cold injury, and the inflammatory response to ischemia. Because the results of laboratory animal studies
CPB.41 often cannot be applied to human trials, clinical studies
For some surgeons, a clamp and sew technique is a will be stressed.
viable option, although this requires an extremely efficient Papaverine, an opium alkaloid, has been placed in the
surgeon and short spinal cord ischemic times to prevent intrathecal space to increase spinal cord blood flow. In a
adverse outcomes.42 If a surgeon prefers clamp and sew, study of 33 randomized, high-risk Crawford type I and II
or employs limited bypass, anesthesia-based modalities TAAA repairs, study patients received intrathecal papaver-
are often used concurrently. These include cerebral spinal ine (IP), as a 3-mL bolus of 1% preservative-free solution,
fluid (CSF) drainage, epidural cooling, and intrathecal 20 minutes before aortic crossclamp, and then had CSF
administration of drugs, along with hemodynamic man- drainage to a CSF pressure of 10 cm H2 O at crossclamp.
agement to optimize spinal cord perfusion. After 1992, patients also underwent arteriofemoral bypass
CSF drainage is a widely used adjunct for TAAA with active cooling to 31 C. Neurologic injury occurred
repair. The rationale for drainage is simpleCSF pressure in 2/17 (11.8%) of the CSF drainage + IP group, and 7/16
increases with aortic crossclamping, and drainage allows (43.8%) of the control group, well above reported averages,
increased spinal cord perfusion pressure. Preoperatively, although in the end, only two patients had permanent neu-
a subarachnoid catheter is placed in the lumbar region, rologic damage that prevented ambulation. Although the
and CSF pressure recorded during the case and into papaverine was thought to be protective, the lack of CSF
the early postoperative period. During this time, CSF is drainage in the control group may have been the more
allowed to drain with gravity. The transducer is placed at important determinant.45
the level of the patients spine and, depending on the A second drug that has been used clinically is
local institutional protocol, CSF pressure is generally naloxone hydrochloride (naloxone), an opioid receptor
maintained at less than 15 mm Hg. By allowing the antagonist. Acher et al. infused naloxone at 1 g/kg/hour
catheter to remain postoperatively, CSF can be drained during TAAA repair. The surgeons reported a paraplegia
if delayed neurologic deficits develop, sometimes with rate of 3.4%, utilizing an evolving protocol (which also
improvement of symptoms. includes moderate hypothermia and CSF drainage to a
In a prospective trial by Coselli et al, patients who CSF pressure of 10 mm Hg), of which naloxone is felt to
underwent CSF draninage during TAAA repair had a be a core component.46,47
greatly reduced risk of paraplegia and paraparesis. The Naloxone attenuates the production of the EAAs. In
trial was halted mid-study when an 80% reduction in a study by Kunihara et al., naloxone was intravenously
neurologic risk was demonstrated.43 administered at 1 g/kg/hour for the duration of the
Epidural cooling utilizes an established neurologic TAAA repair, and continued for 72 hours postoperatively.
protective methodhypothermiaand focuses it on the Patients who suffered spinal cord injuries had significantly
thoracic region of the spinal cord, the area most at risk higher levels of EAAs in the CSF fluid. Although the levels
for poor perfusion. Cambria et al. described a large study of the EAAs were notably lower in those patients who
of 170 patients undergoing TAAA repair using epidural received naloxone, the clinical results were disappointing.
cooling (EC) for neurologic protection. Although the Neurologic injury rates were 3/16 (18.7%) in the naloxone
clamp and sew technique was used for the most repairs, group, and 2/11 (18.1%) in the control group.48 Still, the
cooling allowed for longer aortic crossclamp times. As concept is intriguing, and investigation continues.
opposed to CSF drainage, (in which the goal is to keep In summary, it is an operative team effort to prevent
spinal fluid pressure below 20 mm Hg), when EC is spinal cord ischemia and devastating neurologic injuries
396 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
during vascular surgery. Currently, multiple modalities the peripartum period.50 However, autonomic dysreflexia
are used, respecting surgical technical challenges and may still develop postpartum.
evolving concepts of neurologic protection.
PAIN
What Are the Concerns With Pain can be a very frequent and disabling sequel to SCI.
Chronic Spinal Cord Injury? Chronic pain can be labeled as sympathetically mediated
pain, central pain, phantom pain, dysesthetic pain, neu-
As longevity increases for patients with spinal cord in- ropathic pain, chronic nerve or deafferentation pain, or
juries, operative procedures are often necessary to further central dysesthetic syndrome.51 The development of pain
rehabilitation, control pain, evaluate urologic dysfunc- has been theorized to relate to loss of downstream inhibi-
tion, maintain skin integrity, and reverse additional tion, realignment of structural and synaptic connections,
neurologic degeneration. Because most patients sustain release of excitatory pathways, regrowth or sprouting
SCI during their reproductive years, pregnancies are of neurons in the area of injury, and activation of sec-
not unusual in this group. General considerations for ondary nociceptive pathways. Pain may be characterized
anesthetic management of chronic SCI patients include as lancinating or paroxysmal, and superimposed upon a
the avoidance of autonomic hyperreflexia, awareness of background of constant aching or dysesthetic pain. Treat-
hyperkalemia related to succinylcholine, and possible ment alternatives include selected pharmacologic agents
perioperative exacerbations of pulmonary dysfunction. which modulate pain, such as amitriptyline or carba-
Respiratory disease as a cause of late mortality (after the mazepine, the intrathecal administration of analgesics,
first anniversary of injury) has recently been shown to be biofeedback, dorsal column stimulators, and surgical ab-
increasing. lation of the dorsal roots of the spinal cord. Disabling
spasticity may be treated either with oral baclofen or
an implanted intrathecal baclofen pump. Serious conse-
AUTONOMIC DYSREFLEXIA quences including rhabdomyolysis and renal dysfunction
have been reported after the acute withdrawal of intrathe-
Autonomic dysreflexia is an uninhibited sympathetic cal baclofen.
nervous system response to a variety of noxious stimuli. It Occasionally, exacerbations of pain are related to
most commonly occurs in persons with SCI at the level of the development of syringomyelia. This is a chronic,
T6 or above. Episodes are often unpredictable during progressive condition in which a fluid-filled cystic cavity
the first year after injury and may occur throughout or syrinx develops in the central part of the spinal
an individuals life.49 Untreated, massive hypertension cord and is most prominent several years after injury.
may result in intracerebral hemorrhage, myocardial Additional neurologic deficits may become manifest, as
infarction, and even death. The reflex becomes evident well as loss of reflexes. Following MRI, the syringomyelia
following resolution of the spinal shock phase of injury. may be treated with a diverting shunt between the syrinx
Often, the syndrome is initiated by relatively innocuous and subarachnoid space. Early recognition and prompt
stimulations such as bladder distention, defecation, or treatment is important for a successful outcome.52
extracorporeal shockwave lithotripsy below the level of Additional concerns related to the care of a patient
the cord lesion. Autonomic dysreflexia is manifested with a chronic SCI often include the presence of orthosta-
by headache, flushing, and diaphoresis above the level sis and hypovolemia, the possibility of substance abuse,
of injury, and elevated blood pressure, bradycardia, and an increased incidence of delayed gastric emptying
nasal congestion, chills without fever, and possibly when compared to controls. Breathlessness is a com-
bronchospasm. Therapy involves discontinuation of the mon complaint in subjects with chronic SCI, and airway
noxious stimuli, as well as blood pressure control with an hyperreactivity following the inhalation of methacholine
arteriolar vasodilator such as nitroprusside. or histamine has recently been demonstrated in persons
with chronic cervical cord injuries. 2 -Agonist therapy
is helpful in ameliorating respiratory symptoms. Some
PREGNANCY patients may also be dependent on partial ventilatory
support and diaphragmatic pacing in the perioperative
Optimal anesthetic management of the pregnant patient period.10
with an SCI involves careful monitoring of respiratory
function and preterm labor during the pregnancy and
initiation of adequate labor analgesia such as an epidural
or spinal anesthetic to avoid autonomic dysreflexia. Owing KEY POINTS
to the physiologic changes of pregnancy, such as increased
blood volume and decreased functional residual capacity, 1. Therapy in the acute phase of an SCI is targeted at
a patient with a high level of quadriplegia may need to be reducing secondary injury through the prevention of
electively ventilated in the latter stages of pregnancy. Good ischemia and limiting the effects of cell breakdown.
experience exists with the use of epidural analgesia with 2. Methylprednisolone therapy, if it is chosen, should
infusions of local anesthetics in this patient population in be initiated within 3 hours of injury.
CHAPTER 28/SPINAL CORD INJURY 397
stylet, GlideScope, and Macintosh laryngoscope. Anesth 40. Kouchoukos NT, Masetti P, Rokkas CK, et al. Safety
Analg. 2005;101:910. and efficacy of hypothermic cardiopulmonary bypass and
29. Sahin A, Salman MA, Erden IA, et al. Upper cervical vertebrae circulatory arrest for operations on the descending thoracic
movement during intubating laryngeal mask, fiberoptic and thoracoabdominal aorta. Ann Thorac Surg. 2001;72:
and direct laryngoscopy: A video-fluoroscopic study. Eur 699.
J Anaesthesiol. 2004;21:819. 41. Raja SG. Pump or no pump for coronary artery bypass:
30. Banoub M, Tetzlaff J, Schubert A. Pharmacologic and Current best available evidence. Tex Heart Inst J. 2005;32:489.
physiologic influences affecting sensory evoked potentials. 42. Coselli JS, LeMaire SA, Conklin LD, et al. Left heart bypass
Anesthesiology. 2003;99:716. during descending thoracic aortic aneurysm repair does
31. Svensson LG, Crawford ES, Hess KR, et al. Experience not reduce the incidence of paraplegia. Ann Thorac Surg.
with 1509 patients undergoing thoracoabdominal aortic 2004;77:1298.
operations. J Vasc Surg. 1993;17:357. 43. Coselli JS, Lemaire SA, Koksoy C, et al. Cerebrospinal fluid
32. Tabayashi K. Spinal cord protection during thoracoabdomi- drainage reduces paraplegia after thoracoabdominal aortic
nal aneurysm repair. Surg Today. 2005;35:1. aneurysm repair: Results of a randomized clinical trial. J Vasc
33. Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal Surg. 2002;35:631.
aortic aneurysms: Preoperative and intraoperative factors 44. Cambria RP, Davison JK, Carter C, et al. Epidural cool-
determining immediate and long-term results of operations ing for spinal cord protection during thoracoabdominal
in 605 patients. J Vasc Surg. 1986;3:389. aneurysm repair: A five-year experience. J Vasc Surg. 2000;31:
34. Biglioli P, Roberto M, Cannata A, et al. Upper and lower 1093.
spinal cord blood supply: The continuity of the anterior 45. Svensson LG, Hess KR, DAgostino RS, et al. Reduction of
spinal artery and the relevance of the lumbar arteries. neurologic injury after high-risk thoracoabdominal aortic
J Thorac Cardiovasc Surg. 2004;127:1188. operation. Ann Thorac Surg. 1998;66:132.
35. Jacobs MJ, de Mol BA, Elenbaas T, et al. Spinal cord blood 46. Tefera G, Acher CW, Wynn MM. Clamp and sew techniques
supply in patients with thoracoabdominal aortic aneurysms. in thoracoabdominal aortic surgery using naloxone and CSF
J Vasc Surg. 2002;35:30. drainage. Semin Vasc Surg. 2000;13:325.
36. Ohtsubo S, Itoh T, Okazaki Y, et al. Selective perfusion 47. Acher CW, Wynn MM. Thoracoabdominal aortic aneurysm.
of preoperatively identified artery of Adamkiewicz during How we do it. Cardiovasc Surg. 1999;7:593.
repair of thoracoabdominal aortic aneurysm. J Thorac 48. Kunihara T, Matsuzaki K, Shiiya N, et al. Naloxone lowers
Cardiovasc Surg. 2004;127:272. cerebrospinal fluid levels of excitatory amino acids after
37. Cottrell JE, Smith DS. Anesthesia and neurosurgery, 4th ed. thoracoabdominal aortic surgery. J Vasc Surg. 2004;40:681.
St. Louis: Mosby; 2001. 49. Bycroft J, Shergill IS, Coong EAL, et al. Autonomic dysre-
38. Weigang E, Hartert M, Siegenthaler MP, et al. Neuro- flexia: A medical emergency. Postgrad Med J. 2005;81:232.
physiological monitoring during thoracoabdominal aortic 50. Atterbury JL, Groome LJ. Pregnancy in women with spinal
endovascular stent graft implantation. Eur J Cardiothorac cord injuries. Nurs Clin North Am. 1998;33:603.
Surg. 2006;29:392. 51. Siddall PJ, Loeser JD. Pain following spinal cord injury.
39. Weigang E, Hartert M, Sircar R, et al. Setup of neurophysio- Spinal Cord. 2001;39:63.
logical monitoring with tcMEP/SSEP during thoracoabdom- 52. Bodley R. Imaging in chronic spinal cord injuryindications
inal aneurysm repair. Thorac Cardiovasc Surg. 2005;53:28. and benefits. Eur J Radiol. 2002;42:135.
CHAPTER OPHTHALMOLOGIC
29
COMPLICATIONS
M. Tariq Bhatti
A
went left total hip arthroplasty for post- of local or general anesthesia during ocular or nonoc-
traumatic degenerative joint disease due to ular surgery. Clinicians should be familiar with ocular
a motor vehicle accident. Preoperatively, and systemic risk factors that may place a patient at risk
his hemoglobin was 15.7 g per dL, and of a vision-threatening or life-threatening adverse event
blood pressure (BP) was 123/65 mm Hg. during ocular or nonocular surgery.
The operative procedure under general anesthesia lasted The anatomy and physiology of the eye and its
2.5 hours. Thirty minutes into the procedure, the BP fell associated structures is a very complex organ system with
to 90/50 mm Hg for 30 minutes. At the conclusion of the intricate anatomic structures, an elaborate physiologic
case, the estimated blood loss was 1000 mL, and a total of system, and a wide array of pathologic processes. The
2,900 mL of crystalloid replacement fluid was given. On visual system can be divided into three components:
postoperative day 1, hemoglobin was 10.6 g per dL (nor- Optical, retinocortical and integrative.1 The following
mal 14.0 to 18 g per dL). The patient did well during his section is intended as a brief review of the pertinent
hospital stay. However, on postoperative day 3, the day of anatomy and physiology of the eye for the practicing
discharge, he noted blurred vision in his right eye, which
anesthesiologist. The interested reader is encouraged to
was attributed to the residual effects of general anesthe-
refer to a more detailed text on the subject.2
sia and postoperative pain medications. Hemoglobin had
decreased to 8.5 g per dL, but the BP remained stable in
the postoperative period.
Over the next week, the patients vision remained EYE OR GLOBE
blurred in the right eye, for which he underwent an oph-
thalmologic examination. Visual acuity was 20/20 in each The eye is composed of two modified spheres joined
eye, and his pupillary reflexes showed a right relative af- together, with the anterior sphere comprising the cornea,
ferent pupillary defect (Marcus Gunn pupil). Automated and the posterior sphere comprising the sclera. The
perimetry showed an inferior altitudinal visual field defect junction of the cornea with the sclera is termed the limbus,
on the right (see Fig. 29.1) but was normal on the left. which is clearly visible on external inspection of the eye.
Funduscopy revealed pallid swelling of the superior por- There is marked individual anatomic variation of the axial
tion of the optic disc, with hemorrhage within the retinal length of the globe, but on average it measures 24.0 mm.
nerve fiber layer (NFL) on the right, and a congenitally Each eye is located within the anterior eye socket and
anomalous optic disc on the left (not shown). A diagnosis occupies only one fifth of the orbital volume (Fig. 29.2).
of anterior ischemic optic neuropathy (AION) in the right There are several internal chambers of the eye. The
eye was made. Repeat hemoglobin was 11.9 g per dL. anterior chamber, filled with aqueous humor, is the
Eight weeks later, visual function remained stable, and space between the posterior cornea and iris-lens plane.
funduscopy revealed resolution of the optic disc edema The posterior chamber is bounded anteriorly by the iris,
on the right, with superior sectoral optic disc pallor. laterally by the ciliary body, and posteriorly by the anterior
vitreous face. The volume of the posterior chamber is
small compared with that of the anterior chamber, 60 L
and 250 L, respectively. More than two thirds of the
What Is the Relevant eye volume is formed by the vitreous cavity. In the
Ophthalmic Anatomy and nonvitrectomized (nonsurgical) eye the vitreous cavity is
filled with vitreous humor, a gel composed of 98% water,
Physiology? hyaluronic acid, and collagenous fibrils.
The three layers of the eye, from outer to inner, are:
A basic knowledge and understanding of the anatomy and (i) The corneosclera, (ii) the uveal tract, and (iii) the
physiology of the human eye will help anesthesiologists in retina. The cornea is an elliptic-shaped structure that is
399
400 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 29.1 Automated visual field perimetry demonstrating an inferior visual field defect in the
right eye.
Limbus
C
Ophthalmic artery
Choroid Central
Sclera retinal
artery
Cornea
Retina
Lens
Canal of Anterior
Schlemm chamber
Optic
V itr e o u s nerve
Ciliary body
A B
FIGURE 29.2 Illustration of the eye. A: Magnified view of the anterior segment of the eye. Arrows
represent the flow of aqueous humor. B: Sagittal view of the eye and its major structures. C: External
drawing of the eye and ocular adnexa.
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 401
FIGURE 29.5 Anatomy of the bony orbit and structures within the orbital apex.
axons of each optic tract synapse at the lateral geniculate the maxillary bone of the nasal orbit, all the extraoc-
nucleus. The retrogeniculate pathway begins with the ular muscles arise from the orbital apex. The extraoc-
optic radiations located within the temporal and parietal ular muscles are connected to each other and various
lobes, traveling to and synapsing in the primary visual other orbital structures by a complex orbital connec-
area of the occipital lobe or striate cortex. The integrative tive tissue system. Conceptually, the four recti muscles
process of vision is accomplished by the transmission form a cone-shaped compartment within the orbit be-
of visual impulses from the striate cortex to various tween the orbital apex and the eye. On the basis of
vision-associated regions within the occipital, parietal, anatomic studies, the division of the orbit into an in-
and temporal lobes. traconal and extraconal space has not been validated,
but remains a good practical concept to keep in mind.
The extraocular muscles are innervated by three ocu-
ORBIT lar motor cranial nerves: The third (oculomotor), fourth
(trochlear) and sixth (abducens) nerves. Disturbances of
The paired bony orbits are pyramidal in shape, covered the orbital facial system, dysfunction of the extraocu-
by the orbital periosteum (periorbita), and contain the lar muscles, or neural interruption to the extraocular
eye, extraocular muscles, orbital fat, connective tissue, muscles can result in a misalignment of the two eyes
lacrimal gland, nerves, veins, and arteries. The orbital (strabismus) and the subjective complaint of double
apex comprises the optic foramen (opening to the optic vision.
canal), superior orbital fissure and inferior orbital fissure.
It is the entry and exit point of the major neurovascular
structures of the eye and orbit, and the site of the annulus
of Zinn, the origin of the rectus muscles. Each orbit
VASCULAR SUPPLY
is composed of seven bones forming four walls (see OF THE EYE, ORBIT,
Fig. 29.5). AND OPTIC NERVE
The main arterial supply of the eye and orbit is derived
EXTRAOCULAR MUSCLES from the ophthalmic artery, the first intracranial branch of
the internal carotid artery (see Fig. 29.6). The ophthalmic
There are six extraocular muscles that move the eye artery enters the orbit in partnership with the optic nerve
through the nine cardinal directions of gaze. The four through the optic canal in the orbital apex.
recti muscles are the superior, inferior, medial, and lat- The blood supply to the proximal and distal segments
eral rectus muscles; and the two oblique muscles are of the optic nerve is quite variable from person to
the superior oblique and inferior oblique muscles. Aside person, but is derived from the branches of the retinal
from the inferior oblique muscle, which originates from vasculature, choroidal vasculature, and ophthalmic artery
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 403
Supratrochlear V.
Supratrochlear A.
Ophthalmic A
Cavernous sinus
Interal carotid A
FIGURE 29.6 Vascular supply (arterial and venous) of the orbit. A, artery. V, vein.
(see Fig. 29.7).3 From anterior to posterior, the optic the retrolaminar portion of the optic nerve head. The
nerve head can be divided into four regions: The optic arterial supply of the intraorbital portion of the optic
disc (surface of optic nerve head), prelaminar, LC, nerve is derived from pial branches of the ophthalmic
and postlaminar. The optic disc is supplied by the artery.4 These vascular networks have watershed zones
and are vulnerable to occlusion due to small vessel
retinal arterioles. Segmental centripetal branches from
disease, hypoperfusion, and, rarely, embolic disease, all
the peripapillary choroid supply the prelaminar portion of
of which can result in ischemic disorders of the optic
the optic nerve head. The arterial supply of the LC region nerve, known as ischemic optic neuropathy. Depending
of the optic nerve head is derived from the centripetal on the vascular network affected, the clinical presentation
branches of the short posterior ciliary arteries, which can be either an AION or a posterior ischemic optic
also contribute to the circle of Haller and Zinn (CHZ). neuropathy (PION).
Arterial branches from CHZ, the peripapillary choroid Blood flow to the optic nerve head is determined
and, infrequently, the central retinal artery (CRA) supply by resistance to blood flow, BP and intraocular pressure
C
Retinal vein R S Col.Br.
Retinal artery NFL
PCA R
D A C RA
Pia
S PLR
OD ON
PR LC
LC
SAS
CRV Branch
Cilio retinal artery PCA CRA CRA
PCA P ON
A B C
FIGURE 29.7 Schematic drawing of the blood supply to the optic nerve and optic nerve head.
A, arachnoid; C, choroid; Col Br, collateral branches; CRV, central retinal vein; D, dura; LC, lamina
cribrosa; NFL, surface nerve fiber layer of disc; OA, ophthalmic artery; OD, optic disc; ON, optic
nerve; P, pia; PCA, posterior ciliary artery; PR and PLR, prelaminar region; R, retina; RA, retinal
arteriole; S, sclera; SAS, subarachnoid space. (Reprinted with permission, Hayreh SS. The blood
supply of the optic nerve head and the evaluation of itmyth and reality. Prog Retin Eye
Res. 2001;20:563.)
404 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
(IOP). Blood flow to the optic nerve head can be calculated INTRAOCULAR PRESSURE:
by the formula:
AQUEOUS HUMOR
Blood Flow = Perfusion Pressure/Resistance to Flow PRODUCTION
The perfusion pressure (PP) of the optic nerve head is AND CIRCULATION
determined by the mean arterial pressure (MAP) within The aqueous humor is a clear fluid that provides nutrients
the optic nerve head and the IOP as stated by the following and is a means of exchanging gases and metabolites to
equation: the anterior structures of the eye. Once secreted from the
PP = MAP IOP ciliary processes of the ciliary body, the aqueous humor
circulates into the posterior chamber, traveling through
In addition, the resistance of blood flow to the optic nerve the pupil, and then into the anterior chamber (Fig. 29.2).
head is influenced by the condition of the vessels (lumen It is drained out of the eye through the structures within
size and integrity), autoregulation, endothelial-derived the anterior chamber angle, known as the trabecular
vasoactive substances, and flowability of blood.5,6 meshwork and Schlemm canal, to eventually flow into
To a greater extent than the arterial system, the the episcleral and scleral venous vessels. Approximately
venous drainage system of the eye is anatomically highly 15% of the aqueous humor drains through an alternative
variable (Fig. 29.6). The two sources that drain the eye or unconventional pressure-independent system known
are the central retinal vein (CRV) for the inner retina as the uveoscleral drainage pathway.
and the vortex veins for the uveal tract. The venous
Normal IOP ranges between 14 and 22 mm Hg (mean
drainage of the optic nerve head is primarily into the
of 16 mm Hg) but are highly individually variable. The
CRV, with the prelaminar portion draining into the
control and regulation of IOP is not fully understood,
peripapillary choroidal veins.3 The multitude of venous
but there appear to be many factors involved, includ-
orbital tributaries ultimately drain into the cavernous
ing diurnal variation, cardiorespiratory cycle, age, race,
sinus through the superior and inferior ophthalmic veins.
genetics, episcleral venous pressure, BP, and transmural
The central retinal venous pressure may have an influence
pressure across ocular blood vessels. Any external pres-
on the perfusion pressure and ultimately the blood flow
sure on the eye can raise the IOP. Furthermore, acute
of the optic nerve head.5
increases in central venous pressure (CVP), as can occur
with Valsalva-type maneuvers (e.g., coughing, straining,
vomiting, or endotracheal intubation) can result in a rapid
SENSORY INNERVATION rise in IOP.
OF THE EYE AND ORBIT The generation of IOP is based on the production
and drainage rate of aqueous humor as described by the
The primary sensory supply of the eye, the ocular adnexa, following equation:7
and orbit is from the first division (ophthalmic) of
the trigeminal nerve (see Fig. 29.8). After entering the IOP = R(F Fu) + Pv
orbit through the superior orbital fissure, the ophthalmic
nerve divides into the frontal, lacrimal, and nasociliary where F = aqueous inflow, Fu = unconventional aqueous
nerves. outflow, Pv = episcleral venous pressure and R = outflow
FIGURE 29.8 Sensory supply of the orbit and eye from the trigeminal nerve.
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 405
resistance (R = 1/C). C represents the outflow facility, the preoperative evaluation of patients undergoing com-
which is influenced by F and IOP (C = F/IOP). An mon elective eye surgeries and identifying patients at low
increase in IOP results in a collapse of Schlemm canal risk of perioperative complications.12
and thereby a decrease in outflow facility. Increases in Patients undergoing ocular surgery who are antico-
IOP also influence the production of aqueous humor by agulated can pose a challenge to the ophthalmic surgeon
decreasing the inflow rate. and anesthesiologist. Because of the fear of a retrobulbar
IOP is also influenced by the hydrostatic pressure of or intraocular hemorrhage during surgery, the discon-
the ocular blood vessels and can be summarized by the tinuation of anticoagulation or antiplatelet therapy is
equation: sometimes recommended.13 On the other hand, the risk
of discontinuing treatment places the patient at risk of
IOP = K([OpaqOPpl] + CP) developing a stroke, myocardial infarction, or pulmonary
embolism, depending on the reason for the therapy. In
Where K = coefficient of outflow, Opaq = aqueous humor a prospective cohort study of more than 19,000 patients
osmotic pressure, Oppl = plasma osmotic pressure and undergoing cataract surgery, the continued use of an-
CP = capillary pressure. Clinically, IOP can be reduced tiplatelet or anticoagulation therapy during surgery did
effectively and rapidly by increasing the osmotic pressure not show an increase in the frequency of ocular hem-
of the plasma with the systemic administration of orrhagic events nor an increased frequency of adverse
hypertonic solutions such as mannitol. medical events compared to those patients who dis-
As mentioned earlier, the IOP contributes to the continued medication before surgery.14 Although there
perfusion pressure of the optic nerve head. Elevations does not appear to be a significant increase in the risk
in IOP can lead to decreases in blood flow and ischemic of hemorrhagic events with injectable orbital regional
disorders of the retina or optic nerve. anesthesia in those patients on chronic antiplatelet or an-
ticoagulation therapy,15 a careful evaluation of the risks
and benefits of discontinuing antiplatelet or anticoagu-
lation therapy should be made on a case-by-case basis
What Are the Anesthestic and, if necessary, after consultation with the treating
physician.
Considerations in Ophthalmic Because many diseases of the eye are manifestations
Surgery? of systemic genetic and metabolic disorders, especially
in the pediatric age group, the anesthesiologist should
be aware of these associations and proper steps taken
to assure the successful delivery of anesthesia and
PREOPERATIVE EVALUATION performance of surgery (see Table 29.1).16 18
OF THE OPHTHALMIC A wide variety of ophthalmic drugs (topical, intracam-
eral, and systemic) are used in the medical and surgical
PATIENT treatment of ocular diseases that are relevant to the anes-
Before eye surgery, or for that matter any surgical pro- thetic management of the ophthalmic surgical patient. In
cedure, it is very important for the anesthesiologist to be particular, some of the glaucoma medications can have
familiar with the patients medical history and the inges- serious systemic adverse effects, as well as potentiate
tion of topical ocular or systemic medications. Routine the effects of systemic medications, and therefore special
laboratory testing has not been shown to increase the attention should be made to note their use during the pre-
safety of cataract surgery;8 however, laboratory and an- operative evaluation (see Table 29.2).19 Conversely, some
cillary testing (e.g., electrocardiogram, chest radiograph, of the systemic medications administered during surgery,
etc.) should be performed on the basis of the ophthalmic in particular the general inhalants, can affect the IOP (see
procedure scheduled, the type of anesthesia to be used, Table 29.3).20
and the patients current health status and past medical
history. Most eye surgeries are elective cases; therefore, if
the patient is found to be medically unstable, the surgery OPHTHALMIC SURGERY
should be postponed so that the appropriate and neces-
sary medical evaluation and treatment can be performed AND CHOICE OF
to optimize the patients health before proceeding with ANESTHETIC TECHNIQUE
surgery. On the basis of several older studies, the periop-
erative mortality rate associated with eye surgery is low A general familiarity of the broad categories of ophthalmic
and has been quoted in the range of 0.06% to 0.18%, with surgeries is helpful in understanding and providing oph-
the greatest risk factor of death occurring in patients with thalmic anesthesia. Ophthalmic surgery can be broadly
serious preexisting medical conditions.9 In patients under- divided into intraocular and extraocular surgery. Owing
going cataract surgery, the incidence and prevalence of to the unique nature of many ophthalmic surgical pro-
systemic illnesses is increased, in particular hypertension, cedures, anesthetic care should be custom-tailored for
diabetes mellitus and genitourinary diseases.10,11 Self- each patient. Some of the more common ophthalmic pro-
administered health questionnaires are being developed cedures performed include cataract extraction (often by
to assist physicians in improving the cost-effectiveness of phacoemulsification), penetrating keratoplasty (corneal
406 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Reprinted with permission. McGoldrick KE, ed. Anesthesia for ophthalmic and otolaryngologic surgery. Philadelphia: WB Saunders;
1992:217217.
transplantation), strabismus surgery, glaucoma surgery The choice of anesthetic technique for ophthalmic
(trabeculectomy or filtering surgery), oculoplastic surgery, surgery is dependent on physician preference and skill,
orbital surgery, vitreoretinal surgery, and traumatic open as well as patient preference and cooperation. In some
globe repair.21 conditions, general anesthesia may be the anesthetic
Aside from the routine concern of the overall health technique of choice, although, in most cases, ophthalmic
and safety of the patient, specific operative and anesthetic surgery can be done under topical or orbital regional
conditions critical to the safety and successful outcome anesthesia (see Tables 29.4 and 29.5).22 In a select group of
of patients who have undergone ophthalmic surgery are patients undergoing surgery by skilled cataract surgeons,
immobility of the eye and eyelids (akinesia), profound oc- the complete avoidance of topical or regional anesthesia
ular analgesia, hemostasis, avoidance of the oculocardiac has been demonstrated to be a feasible option.23
reflex, as well as uneventful induction, maintenance and There is a myriad of local anesthetic solutions that
emergence from general anesthesia. To maintain sterility are used for local ophthalmic anesthesia. The amide-
of the eye during surgery, there may often not be direct linked agents (lidocaine, bupivacaine, and mepivacaine)
access to the patients airway; therefore, it is vital that are routinely used for ophthalmic surgery. The advantages
the anesthesiologist remain vigilant to any problems or of amide-linked agents over the ester-linked agents are
concerns the surgeon or patient may raise during surgery. increased stability, hypoallergenicity, and longer half-life.
408 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 29.2 Summary of Nonocular Interactions Between Glaucoma Medications and Systemic Drugs
Documentation
Interaction
Glaucoma Quality of
Medication Systemic Drug Additive Antagonistic Potential Result References
-Adrenergic Anesthetic agents X Systemic hypotension Poora
antagonist (inhalational)
Hypoglycemic agents X A. Retard hypoglycemic rebound None
B. Mask hypoglycemic symptoms Poorb
C. Produce hypoglycemia Poorb,c
-Adrenergic X Increased toxic effects of Poord,e,f
-antagonists
Calcium channel blockers X Cardiac depression Poorg,h
Cholesterol-lowering X Decrease high-density lipoprotein Goodi,j,k
medication cholesterol
Cholinesterase inhibitors X Weakness of striated muscle Fairl,m
Clonidine X Systemic hypertensive rebound Poorn
after clonidine withdrawal
Digitalis glycosides X Cardiac depression Fairn,o,p,q,r,s
Fentanyl derivatives X Increased toxic effects of fentanyl None
Phenothiazines X Increased serum levels of None
-blocker and phenothiazine
with potential toxide side
effects
Prednisone X Increased serum potassium Poort
Quinidine X Cardiac depression Good u,v
a Mishra P, Calvey TN, Williams NE, et al. Intraoperative bradycardia and hypotension associated with timolol and pilocarpine eye drops. Br J Anaesth.
1983;55:8979.
b Velde TM, Kaiser FE. Ophthalmic timolol treatment causing altered hypoglycemic response in a diabetic patient. Arch Intern Med. 1983;143:1627.
c Angelo-Nielsen K. Timolol topically and diabetes mellitus. JAMA. 1980;244:2263.
d Batchelor ED, ODay DM, Shand DG, et al. Interaction of topical and oral timolol in glaucoma. Ophthalmology. 1979;86:605.
e Blondeau P, Cote M, Tetrault L. Effect of timolol eyedrops in subjects receiving systemic propanolol therapy. Can J Ophthalmol 1983;18:1821.
f Chamberlain TJ. Myocardial infarction after ophthalmic betaxolol. N Engl J Med. 1989;321:134245.
g Pringle SD, McEwen CJ. Severe bradycardia due to interaction of timolol eye drops and verapamil. BMJ. 1987;294:1556.
h Sinclair NI, Benzie JL. Timolol eye drops and verapamilA dangerous combination. Med J. 1983;1:548.
i Sacks FM, Dzau VJ. Adrenergic effects on plasma lipoprotein metabolism. Speculation on mechanisms of action. Am J Med. 1986;80(Suppl 2A):7181.
j Freedman SJ, Freedman NJ, Shields MB, et al. Effects of ocular carteolol and timolol on plasma high-density lipoprotein cholesterol level. Am J Ophthalmol.
1993;116:60011.
k Coleman AL, Diehl DLC, Jampel HD, et al. Topical timolol decreases plasma high-density lipoprotein cholesterol level. Arch Ophthalmol. 1990;108:126063.
l Shavitz SA. Timolol and myasthenia gravis. JAMA. 1979;242:16112.
m Verkijk A. Worsening of myasthenia gravis with timolol maleate eye drops. Ann Neurol. 1985;17:21112.
n Bailey RR, Neale TJ. Rapid clonidine withdrawal with blood pressure overshoot exaggerated by beta-blockade. BMJ. 1976;17:9423.
o LeWinter MW, Crawford MH, ORourke RA, et al. The effects of oral propanolol, digoxin and combination therapy on the resting and exercise electrocardiogram.
Am Heart J. 1977;93:2029.
p Lowenthal DT, Porter S, Saris SD, et al. Clinical pharmacology, pharmacodynamics and interactions with esmolol. Am J Cardiol. 1985;56:14F18F.
q Yamaoki K, Kakui M, Seko Y, et al. A case of digitalis poisoning after syncope due to beta-blocker eyedrops. Jpn Circ J. 1988;52:43.
r Rynne MV. Ophthalmic medication complicating systemic disease. J Maine Med Assoc. 1980;71:82.
s Ball S. Congestive heart failure from betaxolol. Arch Ophthalmol. 1987;105:320.
t Swanson ER. Severe hyperkalemia as a complication of timolol, a topically applied beta-adrenergic antagonist. Arch Intern Med. 1986;146:122021.
u Edeki TI, He II, Wood AJJ. Pharmacogenetic explanation for excessive -blockade following timolol eye drops: Potential for oral-ophthalmic drug interaction.
JAMA. 1995;274:161113.
v Dinai Y, Sharir M, Naveh N, et al. Bradycardia induced by interaction between quinidine and ophthalmic timolol. Ann Intern Med. 1985;103:8901.
w Leier CV, Baker ND, Weber PA. Cardiovascular effects of ophthalmic timolol. Ann Intern Med. 1986;104:1979.
x Munroe WP, Rindone JP, Kershner RM. Systemic side effects associated with the administration of timolol. Drug Intell Clin Pharm. 1985;19:859.
y Giudicelli JF, Chauvin M, Thuillez C, et al. -Adrenoceptor blocking effects and pharmacokinetics of betaxol (SL 75212) in man. Br J Clin Pharmacol.
1980;10:419.
z Fraunfelder FT, Meyer SM. Systemic adverse reactions to glaucoma medications. Int Ophthalmol Clin 1989;29:1436.
aa Guzman C. Exacerbation of bronchorrhea by topical timolol. Am Rev Respir Dis. 1980;121:899900.
bb Schoene RB, Martin TR, Charan NB, et al. Timolol-induced broncho-spasm in asthmatic bronchitis. JAMA. 1981;245:14601.
cc Charan NB, Lakshminarayan S. Pulmonary effects of topical timolol. Arch Intern Med. 1980;140:8434.
dd Francois J, Verbraeken H. Danger of collyrium with 2% levorenone (Abstract). Bull Soc Belge Ophtalmol. 1979;185:99102.
ee Smith RB, Douglas H, Petruscak J, et al. Safety of intraocular adrenaline with halothane anesthesia. Br J Anaesth. 1972;44:131417.
ff Boakes AJ, Laurence DR, Teoh PC, et al. Interactions between sympathomimetic amines and antidepressant agents in man. BMJ. 1973;1:3115.
gg Elis J, Lawrence DR, Mattie H. Modification by monoamine oxidase inhibitors of the effect of some sympathomimetics on blood pressure. BMJ. 1967;2:758.
hh Horwitz D, Goldberg LI, Sjoerdsma A. Increased blood pressure responses to dopamine and norepinephrine produced by monoamine oxidase inhibitors in
man. J Lab Clin Med. 1960;56:74753.
ii King MH, Richards DW. Near syncope and chest tightness after administration of apraclonidine before argon laser iridotomy. Am J Ophthalmol. 1990;110:3089.
jj Zsigmoid EK, Eilderton TE. Abnormal reaction to procaine and succinylcholine in a patient with inherited atypical plasma cholinesterase. Can Anaesth Soc J.
1968;15:498500.
kk Eilderton TE, Farmati O, Zsigmoid EK. Reduction in plasma cholinesterase levels after prolonged administration of echothiophate iodide eyedrops. Can Anaesth
Soc J. 1968;15:2916.
ll McGavi D. Depressed levels of pseudo-cholinesterase with echothiophate iodide eye drops. Lancet. 1975;2:272.
mm Cavallaro RJ, Krunperman LW, Kugle F. Effect of echothiophate therapy on metabolism of succinylcholine in man. Anesth Analg. 1968;47:5704.
nn Boada JE. Severe mixed acidosis by combining therapy with acetazolamide and timolol eyedrops. Eur J Respir Dis. 1986;68:2268.
oo Keogh A, Esmore D, Spratt P, et al. Acetazolamide and cyclosporine. Transplantation. 1988;46:4789.
pp Thomsen K, Schor M. Renal lithium excretion in man. Am J Physiol. 1968;215:8237.
qq Syversen GB, Morgan JP, Weintraub M, et al. Acetazolamide-induced interference with primidone absorption. Arch Neurol. 1977;34:804.
rr Anderson CJ, Kaufman PL, Sturm RJ, et al. Toxicity of combined therapy with carbonic anhydrase inhibitor and aspirin. Am J Ophthalmol. 1978;86:5169.
ss Cowan RA, Hartnell GG, Lowdel CP, et al. Metabolic acidosis induced by carbonic anydrase inhibitors and salicylates in patients with normal renal function.
BMJ. 1984;289:347.
tt Sweeney KR, Chapron DJ, Brandt JL, et al. Toxic interaction between acetazolamide and salicylate: Case reports and a pharmacokinetic explanation. Clin
Pharmacol Ther. 1987;41:67.
Reproduced with permission. Albert DM, Jakobiec FA, Azar DT, eds. Principles and practice of ophthalmology, 2nd ed. Philadelphia: WB Saunders; 2000.
410 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 29.3 Effects of Anesthetic Agents, Techniques inadvertent intravascular injection or direct spread of the
and Adjuvant Drugs on Intraocular Pressure agent into the central nervous system, resulting in mental
status changes, seizure activity, coma, or cardiovascular
Decreased IOP Increased IOP collapse. Theoretically, the use of epinephrine in local
anesthetic solution for orbital regional anesthesia may
Inhalational agents Ketamine (possibly)
promote vasconstriction of the ocular blood vessels,
Barbiturates, propofol, Hypoxemia
resulting in reduced blood flow and ocular ischemia.
etomidate, narcotics
Hyaluronidase is very frequently added to the anes-
Neuroleptic agents Hypercapnia
thetic solution to reduce the induction time and increase
Nondepolarizing muscle Succinylcholine
the effectiveness of anesthesia and akinesia. In addition,
relaxants (unless alveolar
the depolymerization of hyaluronic acid (a major compo-
hypoventilation occurs)
nent of orbital connective tissue) allows a lower volume
Hyperventilation
of anesthetic solution to be used and a more rapid dis-
Hypothermia
tribution of the solution within the orbital tissue, thereby
IOP, intraocular pressure. reducing the duration and degree of proptosis and ele-
vated intraorbital pressure.25,26 Few complications have
been reported with the use of hyaluronidase. An allergic
In many cases, a combination of an agent with quicker reaction resulting in an orbital inflammatory syndrome
onset, such as lidocaine, and a longer-acting agent, such as
has been described in five patients.27 During times of
bupivacaine, is given in a 1:1 mixture. Also in many cases,
limited supply of hyaluronidase, clusters of strabismus
a premixed local anesthetic solution with epinephrine
cases have been reported and postulated to have occurred
is used to reduce bleeding, prolong the duration of the
because of either the increased contact time of the anes-
anesthesia, and improve the effectiveness of the anesthesia
thetic solution with the extraocular muscle, resulting in
desired.24 Only a small amount of epinephrine is required,
myotoxicity or focal increased intraorbital pressure, in
and the final concentration should not be greater than
turn leading to an extraocular muscle compartmental
5 g/mL (1:200,000). The total dose should not exceed
syndrome.28 Increasing the pH of the anesthetic solution
0.1 mg.
by the addition of sodium bicarbonate improves the ef-
The three types of adverse effects that may occur from
fectiveness of anesthesia and reduces injection pain. Its
the tissue infiltration of local anesthetic solution are an
use in orbital regional anesthesia has been studied with
allergic reaction, tissue toxicity, and systemic toxicity.24
mixed results and is not commonly used.29
A true allergic reaction to local anesthetic solutions is
rare, and the use of preservative-free preparations may
avoid many problems. Neurotoxicity and myotoxicity of Topical Anesthesia
local anesthetic solutions may explain some of the cases of
optic nerve dysfunction and strabismus following orbital Topical anesthesia refers to the administration of oph-
regional anesthesia. Systemic toxicity may occur from the thalmic anesthetic drops directly onto the eye. With the
recent advances in intraocular lens design and micro- Although several studies have suggested that topical
surgical techniques, topical anesthesia (with or without anesthesia is associated with an equivalent level of pain
intracameral application) is becoming a popular choice control during surgery compared with injectable orbital
in the surgical management of patients requiring cataract regional anesthesia,32 other studies have refuted this ob-
extraction.30 In a 2003 survey of cataract and refractive servation.33 A literature review of the randomized trials of
surgeons, topical anesthesia was used in 61% of cases, orbital regional anesthesia found that injectable anesthe-
an increase from 8% in 1995.31 Topical anesthetic agents sia provided better pain control than topical anesthesia
used in modern day cataract surgery include 0.75% bupi- during the surgical procedure but was more commonly
vacaine, 0.5% tetracaine, and 2% or 4% lidocaine. The reported to be painful on administration.34 The pain or
main advantage of topical anesthesia is avoiding the risks discomfort experienced by a patient undergoing surgery
associated with injectable orbital regional anesthesia (ex- with topical anesthesia is often due to manipulation of
plained further in the text). Other benefits of performing the iris and ciliary body. The intracameral application
surgery with topical anesthesia are: of unpreserved 1% lidocaine has been shown to improve
patient comfort.35
Its safe use in anticoagulated patients Complications of topical anesthesia are confined to
Decreased need for intravenous anxiolytic medications the eye and include corneal epithelial toxicity, difficult
Rapid onset of action operative conditions due to lack of eyelid and globe akine-
Intraoperative patient cooperation sia, and intraoperative pain or discomfort. Intracameral
Immediate return of vision lidocaine may result in temporary blindness due to retinal
Lack of periorbital bruising or optic nerve toxicity.36 The application of intracam-
Absence of postoperative diplopia and eral lidocaine does not result in any detectable systemic
Avoidance of a postoperative eye patch levels.37
412 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Reprinted with permission. Smith GB, Hamilton RC, Carr CC, eds. Ophthalmic anaesthesia. A practical handbook, 2nd ed. London: Oxford
University Press; 1996.
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 413
a decrease in visual acuity, visual field loss, impaired to subconjunctival antibiotic injection. The proposed
color vision, and a relative afferent pupillary defect as mechanisms of diplopia from orbital regional anesthe-
detected by the swinging flashlight test. The orbital depth sia are direct muscle trauma, Volkmann-like contracture,
and length of the needle have been shown to be critical and anesthetic myotoxicity.9294 The most common ex-
risk factors for optic nerve injury during retrobulbar or traocular muscle injured from orbital regional anesthesia
peribulbar anesthesia.8587 Cadaveric studies have shown is the inferior rectus muscle, although the superior rectus
anatomic variation in the distance, ranging from 42 mm and inferior oblique can also be injured.65,95 Capo et al.
to 57 mm (average 48 mm), between the junction of the were able to demonstrate contact of a 1.5 in. needle to the
lateral third and middle two thirds of the inferior orbital superior or inferior rectus muscles from an inferotempo-
rim (transcutaneous site of injection) to the lateral margin ral retrobulbar injection site.65 The study also concluded
of the optic foramen.8587 Katsev et al. demonstrated that that compared with retrobulbar anesthesia, peribulbar
a 38-mm length needle could penetrate the optic nerve anesthesia carried a 4.8-fold greater chance of damaging
7 mm in front of the optic foramen in 20% of the ca- the inferior rectus muscle than the superior rectus mus-
daveric orbits they studied.85 It is recommended that a cle. Damage to the inferior rectus muscle may occur if
needle no longer than 35 mm, preferably a 31.5 mm, be the needle enters the orbit medial to the junction of the
used for retrobulbar anesthesia.85 The position of the eye lateral third and middle two thirds of the inferior orbital
within the orbit is another important factor that should rim and the tip of the needle has not been adequately
be considered during the administration of retrobulbar elevated from the orbital floor.96 The downgaze position
or peribulbar anesthesia (see Fig. 29.9). Atkinson, in his shifts the inferior oblique backward, closer to the orbital
original descriptions of performing retrobulbar anesthe- floor behind the orbital rim, placing it in harms way of
sia, recommended that the eye be positioned up and in.88 the needle.97
Computed tomography and magnetic resonance imaging Postoperative ptosis has been attributed mainly to
can show the position of the optic nerve in the various trauma to the superior rectus muscle complex. The factors
positions of gaze. Using these imaging techniques, the responsible for postoperative ptosis are injection of the
position advocated by Atkinson was shown to put the anesthetic solution into the eyelid or at the superior orbital
optic nerve on stretch, and move it into a down-and-out rim, bulbar massage, forceps grasping or bridle suture
position within the path of the needle during retrobulbar traction of the superior rectus muscle, large superior
anesthesia. Therefore, the preferred eyeball position dur- conjunctival flap, and prolonged postoperative patching
ing retrobulbar anesthesia is either looking straight ahead and edema.98
or in a down-and-in position.89,90 Inadvertent entry into the maxillary sinus during
Strabismus is defined as a misalignment of the eyes retrobulbar anesthesia can occur if there is either a
usually caused by extraocular imbalance, often result- natural, iatrogenic, or traumatic defect of the orbital
ing in the subjective complaint of diplopia or double floor. The systemic complications associated with retro-
vision. Strabismus following intraocular surgery has been bulbar anesthesia can be life-threatening; therefore, such
well described in the literature and, in some cases, may problems should be recognized promptly, and supportive
not be related to local anesthesia or surgery, but rather therapy initiated immediately if a poor outcome is to be
from factors such as a preexisting strabismus condi- avoided. Central nervous system complications following
tion present before surgery, optical aberration, or the retrobulbar anesthesia have been well described in the
occlusive ocular condition for which the patient is hav- literature and can result in mental status changes,
ing surgery.91 Suggested surgical causes of strabismus shivering, apnea, seizures, coma, nausea, vomiting, and
following ocular surgery are placement of a bridle su- cardiopulmonary arrest. The incidence of respiratory
ture, conjunctival scarring, and inflammatory reaction arrest from retrobulbar anesthesia has been reported
A B C
FIGURE 29.9 Positions of eye during retrobulbar injection and the risk of injury to optic nerve and
inferior oblique muscle. A: In upgaze, the optic nerve drapes downward in the potential line of
injury of the needle. B: In downgaze, the optic nerve swings upward but the inferior oblique moves
anteriorally close to the path of the needle. C: In primary gaze, the optic nerve and inferior oblique
are out of the path of the needle.
416 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
to be 0.09% to 0.79%, with a higher rate occurring with peribulbar anesthesia because the anesthetic solution
in cases associated with larger volumes of anesthetic will diffuse anteriorly into the eyelids.
solution injected.99 The incidence of brain stem anesthesia
following retrobulbar anesthesia ranges between 1:350 Special Techniques
and 1:500.100,101 In most of these cases, the mechanism
is thought to be due to direct spread of the anesthetic There are many techniques to establish eyelid anesthesia
agent to the central nervous system, but in some cases and akinesia (see Fig. 29.10).
it may occur because of intravascular injection of the
anesthestic solution, especially if the signs and symptoms
occur within seconds of the injection.69,102 104 Radiologic
Van Lint
and cadaver studies have demonstrated the spread of The Van Lint method involves an injection at the lateral
anesthetic solution from the intraorbital subdural space orbital rim with redirection of the needle in a cephalad and
of the optic nerve to the central nervous system during caudad direction to form a V. The modified Van Lint
retrobulbar anesthesia.105,106 In one case, the anesthetic method is performed with the injection one centimeter
agent was detected in the cerebral spinal fluid of a patient away from the lateral orbital rim, followed by the same
who developed respiratory arrest following retrobulbar redirection of the needle.
anesthesia.106 The onset of symptoms from the time of
the retrobulbar anesthesia can range from 2 minutes OBrien
to 40 minutes. The severity of the complication cannot
The OBrien method involves injection of the anesthetic
be predicted; therefore, life support measures should be
solution 1 cm anterior to the tragus of the ear, over the
available and instituted where deemed necessary.
condyle of the mandible. A modification of the OBrien
Contralateral amaurosis and akinesia is indicative
method can be performed by injecting over the condyle
that the local anesthetic solution has spread beyond the
of the temporomanibular joint, with redirection of the
orbit and into the central nervous system. The postulated
needle towards the lateral canthus.
mechanism is accidental penetration of the optic nerve
sheath and injection of the solution into the subdural or
subarachnoid space (SAS) traveling through the ipsilateral Atkinson
optic nerve to the optic chiasm, contralateral optic nerve, The Atkinson method is performed by injecting over the
and upper brainstem.107,108 zygomatic arch.
of the mandibular ramus, the area in which the main his or her presence in the operating room is important
trunk of the facial nerve exits the stylomastoid foramen, to provide essential life support if a serious systemic
resulting in complete hemiakinesia of the face. complication, were to occur during orbital regional anes-
Complications Minor complications associated thesia.22
with many of these techniques are injection pain and Premedication with a sedative agent before perform-
eyelid ecchymosis. Visual loss is rare after local injection ing orbital regional anesthesia is often helpful to decrease
into the eyelid but can occur from intravascular injection the anxiety level of the patient and avoid the discomfort
of the anesthetic solution into a peripheral arterial branch associated with the procedure. Rapid onset and short
with retrograde flow into the retinal circulation.111 Diffu- duration agents such as midazolam, methohexital, or
sion of the local anesthetic solution from the eyelid into propofol are some of the commonly used sedatives. It
the orbit can result in pupil and extraocular muscle dys- is preferable not to have the patient heavily sedated dur-
function.112 Globe perforation can occur during injection ing surgery so that the patient can remain mentally alert
of local anesthesia directly into the eyelid.113 The Van and cooperative, thereby preventing unpredictable head
Lint technique may result in ptosis. The modified OBrien movements from occurring during surgery. As outlined
technique can result in prolonged facial nerve paralysis. by the American Society of Anesthesiologists, the patients
The Atkinson technique carries a higher failure rate of oxygenation, ventilation, and circulation should be con-
establishing facial anesthesia because of the anatomic tinuously monitored from the time of the orbital regional
variability of the branches of the seventh nerve. anesthesia to the conclusion of the surgery.118
It is no surprisegiven the close anatomic relation of General anesthesia is the technique of choice in cases
the vagus, glossophyarngeal nerves, and spinal accessory of an open globe due to ocular trauma. Injectable orbital
nerves to the sylomastoid foramenthat the highest risk regional anesthesia is avoided because of its associated
of major complications is associated with the Nadbath- increase in IOP and risk of intraocular content extrusion.
Rehman technique. Cases of prolonged facial nerve palsy, The use of the depolarizing neuromuscular blocking agent,
dysphonia, laryngospasm, vocal cord paralysis, dysphagia, succinylcholine, has been traditionally considered a con-
aspiration, and respiratory distress or arrest have been traindication in cases of an open globe due to its potential
reported.110,114 To avoid such complications, a short to promote intraocular content extrusion from contrac-
needle (<12 mm), a small volume of anesthetic solution tion of the extraocular muscles and subsequent elevation
(<3 mL), and avoiding the use of hyaluronidase have in IOP. The clinical evidence of such an effect of succinyl-
been recommended, as well as directing the needle in an choline on the open globe is, however, not very strong and
anterocephalad direction rather than perpendicular to the is based on anecdotal reports in the literature.119,120
skin.110 Thin patients may be particularly vulnerable to To maintain anatomic attachment of the retina
the potential complications associated with the Nadbath- following vitreoretinal surgery, a long-acting gas such as
Rehman technique.115 perfluoropropane may be injected into the vitreous cavity
to serve as an internal tamponade. Following surgery,
patients are often warned not to ascend to high altitude
areas or travel by air due to the risk of an increase in
What Is the Role of the the size of the gas bubble. Because of its high solubility
Anesthesiologist During characteristics, nitrous oxide can increase the size of an
intraocular gas bubble, resulting in extreme elevations
Ophthalmic Surgery? in IOP and permanent visual loss due to CRA occlusion.
Depending on the type, concentration, and volume of
In most operating rooms, the ophthalmologist performs gas, the intraocular gas bubble may remain in the eye
the local anesthesia in preparation for ocular surgery; for as long as 2 to 3 months. If nitrous oxide-based
but in some high volume practices the duty of perform- anesthesia is contemplated, it is very important to confirm
ing the local anesthesia falls upon the anesthesiologist.22 that the intraocular gas bubble has been totally absorbed
Several articles have suggested a higher frequency of or will be not be injected into the eye during vitreoretinal
ocular complications during the delivery of local anes- surgery.121,122 If nitrous oxide is inadvertently used in
thesia by nonophthalmologists,67,73,76 although the true the presence of an intraocular gas bubble, it should
incidence of complications associated with injectable or- be stopped immediately, 100% oxygen given, and an
bital regional anesthesia delivered by ophthalmologists emergency ophthalmologist consult obtained.123
compared with nonophthalmologists is unknown.38 Some Nasopharyngeal secretion contamination of the sur-
have argued that inadequate understanding of the or- gical field may be a source of postoperative infection
bital anatomy, deficiency in training and skill, and the (endophthalmitis). To avoid such a complication, sev-
lack of clinical ability to assess the eye should preclude eral strategies have been recommended. They include the
a nonophthalmologist from performing orbital regional preoperative use of anticholinergics, perinasal skin ster-
anesthesia.116 The advantages cited for an anesthesiolo- ilization, placement of nasal packing, meticulous drape
gist performing injectable orbital regional anesthesia are adhesion around the nose, adjustment of neck flexion, and
the complete and continuous care of the patient, ensuring monitoring for secretions arising around the endotracheal
sufficient time to achieve adequate anesthesia and akine- tube.124,125
sia, and improving operating room efficiency.117 Even if The delivery of supplemental oxygen by either nasal
the anesthesiologist does not perform the local anesthesia, cannula or indirect oxygenation of ambient air at flow
418 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
rates of 2 L per minute prevents hypoxemia in patients of an ocular complication during or following surgery.
covered by the surgical drape.126 The present plastic ma- A complete neuro-ophthalmic examination with formal
terial and design of ophthalmic surgical drapes results in visual field testing will aid in determining the location of
the accumulation of carbon dioxide during ophthalmic injury. Ancillary testing such as intravenous fluorescein
surgery with the delivery of supplemental oxygen.127 To angiography, computed tomography, and magnetic reso-
prevent the rebreathing of accumulated carbon dioxide nance imaging may be required to evaluate and confirm
under the surgical drape and its subsequent unwanted the suspected site of injury. Once visual loss has occurred,
effects of tachypnea and elevated IOP, a suction de- curative treatment is often not possible; therefore, the ma-
vice has been recommended for all patients undergoing jor goal of treatment is the prevention of complications
ophthalmic surgery.128 Supplemental, open oxygen ad- by promptly recognizing, avoiding, and correcting pre-
ministration can also lead to oxygen accumulation under operative and intraoperative risk factors. In some cases,
the drapes and, absent special precautions,129 risks fire if predisposing risk factors may not be amenable to mod-
cautery is used. ification, or there may be multiple risk factors involved.
Furthermore, controlling one risk factor may potentiate
Malignant Hyperthermia the effect of another, for example, avoiding intraoperative
hypotension to better facilitate intraoperative visualiza-
Malignant hyperthermia is another anesthetic complica- tion may result in greater surgical blood loss.
tion. It is most commonly seen in whites of northern Postoperative visual loss has been reported following
European heritage; patients with Evans myopathy, King- a wide variety of nonocular surgical procedures (without
Denborough syndrome, and central-core disease130 and direct visual pathway trauma):
patients with a biopsy proven family history appear to be
particularly susceptible. Patients with congenital ptosis Spinal surgery134142
and strabismus may also have a greater risk of malignant Cardiac surgery143175
hypertension.131 General surgery including abdominal procedures146,147
The modern day mortality rate of malignant hyper- Head and neck surgery148
thermia is <5% due to advancements in monitoring, Liposuction149
detection, and treatment of the syndrome. Once malig- Hip replacement surgery150
nant hyperthermia is recognized, the offending anesthetic Neurosurgery151
agent should be discontinued immediately, hyperventila- Shoulder surgery152
tion with 100% oxygen instituted, and 2.5 mg per kg of Dental procedures153
dantrolene given intravenously. Any cardiac arrhythmias Epidural puncture146
and metabolic disturbances should also be corrected with- Obstetric surgery146
out delay. The hyperthermia should be corrected by active Transurethral prostatectomy146,154
cooling. The exact incidence of postoperative visual loss is not
known but, on the basis of several large retrospective stud-
Ocular Injury ies, ranges between 0.000008% to 1%.132,135,141,143,155 158
The American Society of Anesthesiologists established a
The most common ocular injury after nonocular surgery national registry in an attempt to determine and assess the
is a corneal abrasion. This may occur from either the risks of postoperative visual loss from nonocular surgery.
inadvertent contact of the cornea by various objects while At the time of writing of this chapter, 79 cases have been
preparing the patient for surgery or exposure (drying) collected, with prone position spine surgery (67%) and
from incomplete eyelid closure (lagophthalmos) during cardiac surgery (22%) being the most common surgeries
surgery. The development of a corneal abrasion may be associated with postoperative visual loss. Ischemic optic
aggravated by the decrease in tear production rate and neuropathy was the most common clinical presentation
poor tear film stability associated with general anesthesia. of postoperative visual loss.159
An aqueous or paraffin-based ointment applied to the eye
or closing the eyelid with tape will lessen the risk of a
corneal abrasion.132,133
CLINICAL MANIFESTATIONS
In general, postoperative visual loss may occur as the
result of retinal vascular occlusion (arterial or venous),
How Is Perioperative Visual optic nerve dysfunction, or cortical blindness.146 The
Loss Associated with clinical manifestations of visual loss are characterized
by the site of injury and depend on a variety of factors,
Nonocular Surgery? including mechanism of injury, type of nonocular surgery,
and preexisting anatomic ocular vulnerability.
Perioperative visual loss associated with nonocular
surgery can result from insult to the cornea, retina, an- Retinal Vascular Occlusion
terior visual pathway (optic nerve), or posterior visual
pathway (occipital cortex). Consultation with an ophthal- Retinal vascular occlusion is an uncommon cause of
mologist should be obtained in all patients suspected postoperative visual loss. In approximately 75% of the
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 419
cases reported in the literature, symptomatic branch pathomechanism by which this occurs is unknown. Is-
retinal or CRA occlusion occurred during sinonasal or chemia to the visual pathways can occur from poor
spinal surgery.146,160,161 Retinal artery occlusion com- perfusion pressure, increased vascular resistance, or de-
monly presents with sudden, painless loss of vision, or creased blood oxygenation.156 Currently, it is not possible
a unilateral visual field defect corresponding to the vas- to predict which patient will develop postoperative visual
cular territory of the occluded vessel.162 In most cases, a loss and, although there is no proven cause, several asso-
dilated fundus examination will reveal retinal edema and ciated risk factors have been identified. The mechanism
a cherry-red spot. Occasionally, an embolic plaque may of postoperative ION appears to be multifactorial, and
be visible in the arterial system. Rarely, a CRV occlusion the accumulation of risk factors may be important in
can develop in the perioperative period.141,163 Unlike a the development of visual loss (see Fig. 29.11). A retro-
retinal artery occlusion, a retinal vein occlusion is asso- spective review of ION following spinal surgery identified
ciated with significant retinal hemorrhages and vascular the potential risk factors for visual loss to be prolonged
dilatation and tortuosity. prone position surgery, decreased ocular perfusion pres-
sure, blood loss, anemia, and the patients fluid status
Ischemic Optic Neuropathy (see Table 29.7).169 A retrospective review of seven pa-
tients with postoperative PION noted an association with
Ischemic optic neuropathy (ION) is the most common intraoperative blood loss with anemia, intraoperative hy-
cause of visual loss following nonocular surgery. Optic potension, and facial edema.170 Postoperative visual loss
nerve dysfunction often presents with loss of central may occur in what otherwise would have been considered
visual acuity, visual field defect, dyschromatopsia, and a a routine case. For example, PION was reported to occur
relative afferent pupillary defect. Excluding glaucomatous in a young normal patient during spinal surgery with no
optic neuropathy, ION is the most common cause of known preoperative vascular risk factors or significant
optic nerve dysfunction in the elderly.164 It results from intraoperative blood loss, hypotension, or anemia.171
vascular insufficiency to the optic nerve and, depending Johnson et al. described the postmortem histopatho-
on the pattern of vascular involvement, manifests in two logic findings of the optic nerves in a patient with bilateral
forms: The more common, AION and PION. AION is blindness due to PION following repeated gastrointestinal
characterized by optic nerve edema, a consequence of hemorrhages requiring exploratory laporatomy. The in-
vascular compromise from the small branches of the traorbital portions of the optic nerve demonstrated infarc-
posterior ciliary arteries supplying the optic nerve head.3 tion with the distal and proximal ends relatively spared.
PION is manifested by optic nerve dysfunction in the On the basis of their clinicopathologic observations and
setting of a normal-appearing optic nerve, progressing to review of the literature, the authors concluded that, in
optic nerve pallor in 4 to 6 weeks. It results from vascular general, cerebral infarction can occur from significant
insufficiency of the intraorbital portion of the optic nerve hypotension in the absence of anemia or atherosclerotic
supplied by the pial branches of the ophthalmic artery.3,4 risk factors, AION from brief hypotension and preexisting
Postoperative AION is most frequently associated with atherosclerotic risk factors, and PION from hypotension
spinal and cardiac surgeries, whereas PION is associated and anemia.172
with spinal and head and neck surgeries.146,165 In a
retrospective study of 28 patients with postoperative
PION, 54% of the cases were bilateral and simultaneous; Patient Risk Factors
compared with eyes with nonarteritic PION, those with
Patient systemic factors may play a role in the de-
postoperative PION was associated with a greater degree
velopment of postoperative visual loss. Peripheral vas-
of initial visual loss, less chance of visual recovery, and a
cular disease, diabetes mellitus, hypertension, and
larger risk of profound final visual acuity loss.166
tobacco smoking are characteristics that are seen
Bilateral insult to the occipital lobes results in
more frequently in patients with postoperative visual
cortical blindness; most cases have been described after
loss.134,139,143,156,165,172 174 Patients with these systemic
cardiac surgery.146 Occasionally, neck167,168 and spinal
conditions may be predisposed to ischemia because of in-
surgery136 can be complicated by cortical blindness.
creased arterial resistance secondary to atherosclerotic
The neurologic examination is often normal, except
disease. In addition, some reports have theorized an
for profound visual loss, confusion, disorientation, and
anatomic variation because of a small cup-to-disc ratio
sometimes denial of visual problems (Anton syndrome).
of the optic nerve, watershed vascular zones within the
Optokinetic nystagmus is absent with intact pupillary
optic nerve, and vessel caliber as contributing factors in
function and normal-appearing retina and optic nerve.
the development of postoperative visual loss.134,143,147
MECHANISMS Hypotension
AND POTENTIAL RISK The perfusion pressure of the eye is dependent on the
FACTORS OF POSTOPERATIVE systemic mean arterial BP and IOP. IOP is influenced
VISUAL LOSS by the episcleral venous pressure, which in turn is influ-
enced by the CVP. On the basis of this relation, nocturnal
Postoperative visual loss occurs in most cases because hypotension has been suggested to influence the develop-
of ischemia to the visual pathways, but the exact ment of AION.175 In some cases, the aggressive treatment
420 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Perioperative Deliberate
Hypovolemia
blood loss hypotension
Increased
venous
Perioperative
pressure
anemia Other causes
Autoregulation
? dysfunction ?
PION
FIGURE 29.11 Flow chart of proposed pathogenesis of perioperative posterior ischemic optic
neuropathy. PION, posterior ischemic optic neuropathy; IJV, internal jugular vein. (Reprinted with
permission. Buono LM, Foroozan R. Perioperative posterior ischemic optic neuropathy. Surv
Ophthalmol. 2005;50:23.)
of malignant hypertension can result in blindness,176,177 Intraoperative Blood Loss and Anemia
and therefore it is not surprising to find an association
between intraoperative hypotension and postoperative vi- The prevention and correction of acute intraoperative
sual loss.146 In a study of six patients with postoperative blood loss can be challenging. Concern over the adverse
ION following a variety of major surgical procedures, in- effects and risks of blood transfusion has caused some
traoperative hypotension was noted to be a significant clinicians to accept relatively low blood counts in the pe-
factor in the development of visual loss, with a 25% to rioperative period.178 The association of visual loss and
46% decrease in the mean BP from preoperative levels acute hemorrhage has been well known for centuries.179
and the duration of hypotension ranging from 15 to 120 Aside from decreasing the oxygen-carrying capacity of
minutes.173 All these patients had a period of signifi- blood to the eye, acute hemorrhage has been suggested to
cant anemia, hemoglobin <8 gm per dL, associated with result in vasoconstriction of the blood vessels supplying
the hypotension. In a retrospective, time-matched, case the choroid and optic nerve head through an increase in
control study of 17 patients with ION following cardiopul- renin production and other endogenous vasoconstrictor
monary bypass surgery, presurgical and postsurgical BPs substances.179 In a case-matched study of 28 patients with
were not found to be a statistically significant factor for visual loss following spinal surgery, the visual loss group
postoperative visual loss.143 The concern over deliber- suffered on average a 2,720 mL greater blood loss than
ate hypotension during spinal surgery precipitating visual the control group.140 Furthermore, a study of ION follow-
loss has been addressed by several reports,134,137,140,141 yet ing cardiopulmonary bypass surgery found that 76.5% of
the optimal BP range has not been determined.156 Other patients with visual loss had a postoperative hemoglobin
risk factors, such as anemia or cardiopulmonary bypass, value of 8.5 g per dL or less compared with only 41.2%
appear to be necessary in combination with hypotension of control patients.143 Intraoperative hemodilution can
for the development of ION.141,156,157,173 also decrease the oxygen-carrying capacity of the blood
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 421
TABLE 29.7 Patient Characteristics of Ischemic Optic Neuropathy Associated With Spinal Surgery as Reported in the
Literature
(continued)
422 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Time of Symptoms
Upon awakening 0% 59%
Within 24 h 40% 29%
Postoperative day 25 20% 12%
Postoperative day 59 20% 0%
>Postoperative day 9 20% 0%
Ocular Findings
Pupil
Afferent pupil defect 100% 33%
Nonreactive pupil 0% 12%
Funduscopy
Disc edema 100% 0%
Normal 0% 100%
Hemorrhage 20% 0%
Cherry-red macula 0% 0%
Retinal ischemia 0% 0%
Visual defects
Altitudinal defects 0% 27%
Central scotoma 0% 6%
Vision loss
Monocular involvement 60% 53%
Binocular involvement 40% 47%
% Both eyes involvement equal 20% 24%
Postoperative care
% Receiving intervention 20% (1) 29% (5)
Outcomes
Some improvement at followup 60% 65%
No improvement at followup 20% 24%
Worsening at followup 0% 0%
Not indicated 20% 80% (5)
% Intervention with improvement 100% (1) 80% (5)
% No intervention with improvement 33% (2) 58% (7)
Note: The numbers in parenthesis adjacent to mean or median values are the numbers of patients for which data were provided.
a Dilger JA, Tetzlaff JE, Bell GR, et al. Ischaemic optic neuropathy after spinal fusion. Can J Anaesth. 1998;45:6366.
b Katz DM, Trobe JD, Cornblath WT, et al. Ischemic optic neuropathy after lumbar spine surgery. Arch Ophthalmol. 1994;112:925931.
c Stevens WR, Glazer PA, Kelley SD, et al. Ophthalmic complications after spinal surgery. Spine. 1997;22:13191324.
d Abraham M, Sakhuja N, Sinha S, et al. Unilateral visual loss after cervical spine surgery. J Neurosurg Anesthesiol. 2003;15:319322.
e
Alexandrakis G, Lam BL. Bilateral posterior ischemic optic neuropathy after spinal surgery. Am J Ophthalmol. 1999;127:354355.
f
Brown RH, Schauble JF, Miller NR. Anemia and hypotension as contributors to perioperative loss of vision. Anesthesiology. 1994;80:222226
g
Dunker S, Hsu HY, Sebag J, et al. Perioperative risk factors for ischemic optic neuropathy. J Am Coll Surg. 2002;194:705710.
h
Lee LA, Lam AM. Unilateral blindness after prone lumbar surgery. Anesthesiology. 2001;95:793795.
i
Lee AG. Ischemic optic neuropathy following lumbar spine surgery. J Neurosurg. 1995;83:348349.
j Murphy MA. Bilateral posterior ischemic optic neuropathy after lumbar spine surgery. Ophthalmology. 2003;110:14541457.
k Roth S, Thisted RA, Erickson JP, et al. Eye injuries after non-ocular surgery: A study of 60, 965 anesthetics from 19881992. Anesthesiology.
1996;85:10201027.
AION, anterior ischemic optic neuropathy; PION, posterior ischemic optic neuropathy.
Reproduced with permission. Ho VT, Newman NJ, Song S, et al. Ischemic optic neuropathy following spine surgery. J Neurosurg
Anesthesiol. 2005;17:38.
to intolerable levels in some surgical patients.180 The in- does not affect the oxygen delivery capacity of the blood
cidence of non-RBC transfusion was found to be more to the optic nerve head or choroid.146
frequent in patients with visual loss following cardiopul-
monary bypass surgery.143 However, the effect, if any, of
Intraocular Pressure
hemodilution on ocular perfusion is not established.156,181 The causative effect of IOP in the development of ION
Experimental studies suggest isovolumic hemodilution is debatable.146,182 A rise in IOP can result in a decrease
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 423
in the perfusion pressure to the optic nerve head.5 An vision, including complement activation, release of
intraoperative increase in IOP may occur from car- inflammatory mediators, exogenous and endogenous cat-
diopulmonary bypass, prone positioning, intravascular echolamines, and elevated IOP secondary to colloid-
and extravascular fluid shifts, and direct orbital pres- crystalloid transfusion.143145,158
sure.144,156,183,184 A significant elevation in IOP due to
direct ocular compression from a poorly positioned head
rest has been associated with CRA occlusion.140,161,185
Further supporting a major role of direct compression in KEY POINTS
the pathomechanism of postoperative CRA occlusion, as
1. The anatomy and physiology of the afferent and
compared to postoperative ION, is the lower frequency of
efferent visual system is intricate and complex.
blood loss, lack of anemia, and shorter duration of prone
The branches of the ophthalmic artery provide the
position surgery in cases of CRA occlusion.186
vascular supply to the retina and optic nerve.
2. IOP is generated by the production, circulation,
Emboli and drainage of aqueous humor. The IOP can be
Stroke is a well recognized adverse outcome after cardiac influenced by systemic medications, CVP, orbital
surgical procedures.187 In most cases, stroke occurs from congestion, orbital regional anesthesia, and external
cardiogenic emboli.188 Unlike the cerebral circulation, direct pressure.
the retina allows for a unique opportunity to study the 3. The mean arterial BP within the optic nerve head
vascular system without the use of expensive or invasive and the IOP determines the perfusion pressure of
diagnostic testing.189 Clinical observation and intravenous the optic nerve head. The perfusion pressure of the
fluorescein angiography have identified asymptomatic optic nerve head is influenced by the IOP, which is
retinal microemboli in a substantial proportion of patients influenced by the episcleral venous pressure which
undergoing cardiac surgery.190192 In a prospective study in turn, is influenced by the CVP.
of 314 patients who underwent coronary bypass surgery, 4. Topically applied glaucoma medications can interact
17.3% of the patients were noted postoperatively to have with inhalational anesthetic agents and systemic
cotton-wool spots, a sign of microemboli, and 2.6% had medications.
visible emboli in the retinal circulation.193 The origin of 5. Retrobulbar and peribulbar anesthesia can result in
emboli may be atherosclerotic disease, thromboembolic serious ocular and life-threatening complications,
disease, or air, fat or platelet aggregation.187,189,194 such as orbital hemorrhage, globe penetration, optic
nerve trauma, and brainstem anesthesia.
6. Facial nerve block, in particular the Nadbath-
Other Risk Factors Rehman technique, can result in prolonged facial
nerve palsy, laryngospasm, vocal cord paralysis,
As mentioned earlier, spinal surgery in the prone position,
aspiration, and respiratory distress.
with a poorly positioned headrest, can result in direct
7. The etiopathogenesis of postoperative visual loss
compression on the eyeball, leading to extreme elevations
following nonocular surgery is multifactorial. Hy-
of IOP and CRA occlusion.160,161,184,185,195,196 In most
potension, anemia, intraoperative blood loss, prone
cases of ION, significant periorbital edema is often
positioning, orbital congestion, and elevated IOP
evident. When extreme, the orbit essentially becomes like
appear to be major contributing factors in the devel-
a closed compartment, which decreases arterial inflow
opment of postoperative visual loss.
and impedes venous drainage of the orbital contents.
8. A variety of nonocular surgeries have been associ-
Placing the patient in a prone position has been shown
ated with postoperative visual loss, with the most
to increase the IOP,183 but a slight change in the
common being spinal surgery, heart surgery, and
inclination of the table (10 degree reverse Trendelenburg
head and neck surgery.
position) can reduce the increased IOP effect of prone
positioning.197 9. Direct orbital pressure from a poorly positioned
Radical neck surgery with internal jugular vein (IJV) headrest during prone position surgery can result
ligation can impede venous drainage, thereby contributing in an orbital compartmental syndrome and a CRA
to ischemia in the intraorbital portion of the optic occlusion.
nerve, resulting in PION.146,148,198 Surgical procedures 10. Ischemic optic neuropathy is the most common
requiring systemic hypothermia may decrease blood flow presentation of postoperative visual loss following
and increase blood viscosity, resulting in ischemia to the nonocular surgery.
eye.145,156,158 The exogenous use of vasopressors during
surgery may compromise the blood flow to the optic nerve REFERENCES
and contribute to the development of postoperative visual
1. Trobe JD. The neurology of vision. Contemporary Neurology
loss.144,146,199
Series. New York: Oxford University Press; 2001.
Several reports have implicated extended car- 2. Oyster CW. The human eye: Structure and function. Sunder-
diopulmonary bypass time as a possible risk factor land, MA: Sinauer Associates; 1999.
in the development of postoperative ION.143,145,158,174 3. Hayreh SS. The blood supply of the optic nerve head and
Various mechanisms have been theorized to explain the evaluation of it - myth and reality. Prog Retin Eye Res.
the deleterious effect of cardiopulmonary bypass on 2001;20:5633.
424 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
4. Isayama Y, Hiramatsu K, Asakura S, et al. Posterior 24. Hamilton RC. The local anaesthetics and adjuvant drugs.
ischemic optic neuropathy. I. Blood supply of the optic In: Smith GB, Hamilton RC, Carr CA, eds. Ophthalmic
nerve. Ophthalmologica. 1983;186:197. anaesthesia. A practical handbook, 2nd ed. London: Arnold;
5. Hayreh SS. Blood flow in the optic nerve head and factors 1996:84.
that may influence it. Prog Retin Eye Res. 2001;20:595. 25. Nicoll JM, Treuren B, Acharya PA, et al. Retrobulbar
6. Hayreh SS. Retinal and optic nerve head ischemic disorders anesthesia: The role of hyaluronidase. Anesth Analg. 1986;
and atherosclerosis: Role of serotonin. Prog Retin Eye Res. 65:1324.
1999;18:191. 26. Hulbert MF, Yang YC, Pennefather PM, et al. Pulsatile ocu-
7. Johnson M, Erickson K. Aqueous humor and the dynamics lar blood flow and intraocular pressure during retrobulbar
of its flow. Mechanisms and routes of aqueous humor injection of lignocaine: Influence of additives. J Glaucoma.
drainage. In: Albert DM, Jakobiec FA, Azar DT, et al. 1998;7:413.
eds. Principles and practice of ophthalmology,2nd ed. Vol 4. 27. Kempeneers A, Dralands L, Ceuppens J. Hyaluronidase in-
Philadelphia: WB Saunders; 2000:2577. duced orbital pseudotumor as complication of retrobulbar
8. Schein OD, Katz J, Bass EB, et al. Study of Medical Testing anesthesia. Bull Soc Belge Ophtalmol. 1992;243:159.
for Cataract Surgery. The value of routine preoperative 28. Brown SM, Coats DK, Collins ML, et al. Second cluster
medical testing before cataract surgery. N Engl J Med. of strabismus cases after periocular anesthesia without
2000;342:168. hyaluronidase. J Cataract Refract Surg. 2001;27:1872.
9. Coley CM, Hughes RA. The primary care physicians and 29. McLure HA. Pharmacology for regional ophthalmic anaes-
medical consultants approach to the surgical patient. In: thesia. In: Kumar CM, Dodds C, Fanning GL, eds. Oph-
Albert DM, Jakobiec FA, Azar DT, et al. eds. Principles and thalmic anaesthesia. Lisse, The Netherlands: Swets and
practice of ophthalmology, 2nd ed. Vol 5. Philadelphia: WB Zeitlinger; 2002:37.
Saunders; 2000:4493. 30. Naor J, Slomovic AR. Anesthesia modalities for cataract
10. Neima D, Ramsey MS. Systemic illnesses in cataract surgery. Curr Opin Ophthalmol. 2000;11:7.
patients: 1. Incidence. Can J Ophthalmol. 1987;22:165. 31. Leaming DV. Practice styles and preferences of ASCRS
11. Neima D, Ramsey MS. Systemic illnesses in cataract members2003 survey. J Cataract Refract Surg. 2004;30:
patients: 2. Prevalence. Can J Ophthalmol. 1987;22:168. 892.
12. Reeves SW, Tielsch JM, Katz J, et al. A self-administered 32. Sauder G, Jonas JB. Topical versus peribulbar anaesthesia
health questionnaire for the preoperative risk stratification for cataract surgery. Acta Ophthalmol Scand. 2003;81:596.
of patients undergoing cataract surgery. Am J Ophthalmol. 33. Katz J, Feldman MA, Bass EB, et al. The Study of Medical
2003;135:599. Testing for Cataract Surgery study team. Injectable versus
13. Konstantatos A. Anticoagulation and cataract surgery: topical anesthesia for cataract surgery: Patient perceptions
A review of the current literature. Anaesth Intensive Care. of pain and side effects. Ophthalmology. 2000;107:2054.
2001;29:11. 34. Friedman DS, Bass EB, Lubomski LH, et al. Synthesis of
14. Katz J, Feldman MA, Bass EB, et al. Risks and benefits the literature on the effectiveness of regional anesthesia for
of anticoagulant and antiplatelet medication use before cataract surgery. Ophthalmology. 2001;108:519.
cataract surgery. Ophthalmology. 2003;110:1784. 35. Crandall AS, Zabriskie NA, Patel BC, et al. A comparison
15. Kallio H, Paloheimo M, Maunuksela EL. Haemorrhage of patient comfort during cataract surgery with topical
and risk factors associated with retrobulbar/peribulbar anesthesia versus topical anesthesia and intracameral
block: A prospective study in 1383 patients. Br J Anaesth. lidocaine. Ophthalmology. 1999;106:60.
2000;85:708. 36. Hoffman RS, Fine IH. Transient no light perception visual
16. McGoldrick KE. Anesthesia for ophthalmic and otolaryngo- acuity after intracameral lidocaine injection. J Cataract
logic surgery. Philadelphia: WB Saunders; 1992. Refract Surg. 1997;23:957.
17. Kent CJ, Jakobiec FA. Ophthalmic manifestations of 37. Wirbelauer C, Iven H, Bastian C, et al. Systemic levels
some metabolic and endocrine disorders. In: Albert DM, of lidocaine after intracameral injection during cataract
Jakobiec FA, Azar DT, et al. eds. Principles and practice of surgery. J Cataract Refract Surg. 1999;25:648.
ophthalmology, 2nd ed. Vol 5. Philadelphia: WB Saunders; 38. Wong DH. Regional anaesthesia for intraocular surgery.
2000:4730. Can J Anaesth. 1993;40:635.
18. Cooper LL, Smith LEH. Inherited metabolic disease with 39. Hamilton RC. Techniques of orbital regional anesthesia.
pediatric ocular manifestations. In: Albert DM, Jakobiec FA, In: Smith GB, Hamilton RC, Carr CA, eds. Ophthalmic
Azar DT, et al. eds. Principles and practice of ophthalmology, anaesthesia. A practical handbook, 2nd ed. London: Arnold;
2nd ed, Vol 5. Philadelphia: WB Saunders; 2000:4416. 1996:104.
19. Gerber SL, Cantor LB. Systemic side effects and interac- 40. Edge R, Navon S. Scleral perforation during retrobulbar
tions of glaucoma medications. In: Albert DM, Jakobiec FA, and peribulbar anesthesia: Risk factors and outcome in
Azar DT, et al. eds. Principles and practice of ophthalmology, 50,000 consecutive injections. J Cataract Refract Surg.
2nd ed. Vol 4. Philadelphia: WB Saunders; 2000:2916. 1999;25:1237.
20. McGoldrick KE. Principles of ophthalmic anesthesia. J Clin 41. Hamilton RC, Gimbel HV, Strunin L. Regional anaesthesia
Anesth. 1989;1:297. for 12,000 cataract extraction and intraocular lens implan-
21. Spaeth GL. Ophthalmic surgery: Principles and practice, 3rd tation procedures. Can J Anaesth. 1988;35:615.
ed. Philadelphia: WB Saunders; 2003. 42. Davis DB, Mandel MR, II. Efficacy and complication rate
22. Hamilton RC. Regional versus general anesthesia. In: of 16,224 consecutive peribulbar blocks. A prospective
Smith GB, Hamilton RC, Carr CA, eds. Ophthalmic anaes- multicenter study. J Cataract Refract Surg. 1994;20:327.
thesia. A practical handbook, 2nd ed. London: Arnold; 43. Murdoch IE. Peribulbar versus retrobulbar anaesthesia.
1996:75. Eye. 1990;4(Pt 3):445.
23. Pandey SK, Werner L, Apple DJ, et al. No-anesthesia 44. Shriver PA, Sinha S, Galusha JH. Prospective study of
clear corneal phacoemulsification versus topical and topical the effectiveness of retrobulbar and peribulbar anesthesia
plus intracameral anesthesia. Randomized clinical trial. for anterior segment surgery. J Cataract Refract Surg.
J Cataract Refract Surg. 2001;27:1643. 1992;18:162.
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 425
45. Petersen WC, Yanoff M. Subconjunctival anesthesia: An 68. Sharma A, Gupta A, Bandyopadhyay S, et al. Necrosis of the
alternative to retrobulbar and peribulbar techniques. Oph- eyelids and sclera after retrobulbar anesthesia. J Cataract
thalmic Surg. 1991;22:199. Refract Surg. 2003;29:842.
46. Yanoff M, Redovan EG. Anterior eyewall perforation 69. Feitel ME, Krupin T. Retrobulbar anesthesia. Ophthalmol
during subconjunctival cataract block. Ophthalmic Surg. Clin North Am. 1990;3:83.
1990;21:362. 70. Hamilton RC: The complications of orbital regional anaes-
47. Swan KC. New drugs and techniques for ocular anesthesia. thesia. In: Smith GB, Hamilton RC, Carr CA, eds. Oph-
Trans Am Acad Ophthalmol Otolaryngol. 1956;60:368. thalmic anaesthesia. A practical handbook, 2nd ed. London:
48. Turnbull CS. The hydochlorate of cocaine, a judicious. opin- Arnold; 1996:148.
ion of its merits. (Editorial). Med Surg Rep. 1884;29:628. 71. Cionni RJ, Osher RH. Retrobulbar hemorrhage. Ophthal-
49. Stevens JD. A new local anesthesia technique for cataract mology. 1991;98:1153.
extraction by one quadrant sub-Tenons infiltration. Br J 72. Girard LJ, Donaldson K. Combined peribulbar and retrob-
Ophthalmol. 1992;76:670. ulbar block anesthesia and antibiotic-steroid injection: A
50. Patton N, Malik TY, Aslam TM, et al. Effect of volume simple method to prevent retrobulbar hemorrhage, globe
used in sub-Tenons anaesthesia on efficacy and intraocular perforation, or infection. Ophthalmic Surg Lasers. 1997;28:
pressure: A randomized clinical trial of 3 mL versus 5 mL. 251.
Clin Experiment Ophthalmol. 2004;32:488. 73. Hay A, Flynn HW Jr, Hoffman JI, et al. Needle penetration
51. Patton N, Malik TY, Aslam TM. Sub-Tenons anesthetic of the globe during retrobulbar and peribulbar injections.
administration for cataract surgery: How much stays in? Ophthalmology 1991;98:1017.
Anesth Analg. 2005;101:1012. 74. Duker JS, Belmont JB, Benson WE, et al. Inadvertent globe
52. Lai MM, Lai JC, Lee WH, et al. Comparison of retrobulbar perforation during retrobulbar and peribulbar anesthesia.
and sub-Tenons capsule injection of local anesthetic in Patient characteristics, surgical management, and visual
vitreoretinal surgery. Ophthalmology. 2005;112:574. outcome. Ophthalmology. 1991;98:519.
53. Ripart J, Lefrant JY, Vivien B, et al. Ophthalmic regional 75. Lincoff H, Zweifach P, Brodie S, et al. Intraocular injection
anesthesia: Medial canthus episcleral (sub-tenon) anesthe- of lidocaine. Ophthalmology. 1985;92:1587.
sia is more efficient than peribulbar anesthesia: A double- 76. Grizzard WS, Kirk NM, Pavan PR, et al. Perforating ocular
blind randomized study. Anesthesiology. 2000;92:1278. injuries caused by anesthesia personnel. Ophthalmology.
54. Feibel RM, Guyton DL. Transient central retinal artery 1991;98:1011.
occlusion after posterior sub-Tenons anesthesia. J Cataract 77. Klein ML, Jampol LM, Condon PI, et al. Central retinal
Refract Surg. 2003;29:1821. artery occlusion without retrobulbar hemorrhage after
55. Olitsky SE, Juneja RG. Orbital hemorrhage after the admin- retrobulbar anesthesia. Am J Ophthalmol. 1982;93:573.
istration of sub-Tenons infusion anesthesia. Ophthalmic 78. Lemagne JM, Michiels X, Van Causenbroeck S, et al.
Surg Lasers. 1997;28:145. Purtscher-like retinopathy after retrobulbar anesthesia.
56. Frieman BJ, Friedberg MA. Globe perforation associated Ophthalmology. 1990;97:859.
with subtenons anesthesia. Am J Ophthalmol. 2001;131: 79. Sullivan KL, Brown GC, Forman AR, et al. Retrobulbar
520. anesthesia and retinal vascular obstruction. Ophthalmology.
57. Jaycock PD, Mather CM, Ferris JD, et al. Rectus muscle 1983;90:373.
trauma complicating sub-Tenons local anaesthesia. Eye. 80. Jay WM, Carter H, Williams B, et al. Effect of applying
2001;15:583. the Honan intraocular pressure reducer before cataract
58. Kim SK, Andreoli CM, Rizzo JF III, et al. Optic neuropathy surgery. Am J Ophthalmol. 1985;100:523.
secondary to sub-tenon anesthetic injection in cataract 81. McDonnell PJ, Quigley HA, Maumenee AE, et al. The Honan
surgery. Arch Ophthalmol. 2003;121:907,. intraocular pressure reducer. An experimental study. Arch
59. Davis DB, Mandel MR II. Posterior peribulbar anesthesia: Ophthalmol. 1985;103:422.
An alternative to retrobulbar anesthesia. Indian J Ophthal- 82. Coupland SG, Deschenes MC, Hamilton RC. Impairment of
mol. 1989;37:59. ocular blood flow during regional orbital anesthesia. Can J
60. Davis DB, Mandel MR II. Peribulbar anesthesia. A review of Ophthalmol. 2001;36:140.
technique and complications. Ophthalmol Clin North Am. 83. Watkins R, Beigi B, Yates M, et al. Intraocular pressure and
1990;3:101. pulsatile ocular blood flow after retrobulbar and peribulbar
61. Ripart J, Lefrant JY, de La Coussaye JE, et al. Peribulbar anaesthesia. Br J Ophthalmol. 2001;85:796.
versus retrobulbar anesthesia for ophthalmic surgery: 84. Huber KK, Remky A. Colour Doppler imaging before and af-
An anatomical comparison of extraconal and intraconal ter retrobulbar anaesthesia in patients undergoing cataract
injections. Anesthesiology. 2001;94:56. surgery. Graefes Arch Clin Exp Ophthalmol. 2005;243:
62. Ropo A, Nikki P, Ruusuvaara P, et al. Comparison of 1141.
retrobulbar and periocular injections of lignocaine by 85. Katsev DA, Drews RC, Rose BT. An anatomic study
computerised tomography. Br J Ophthalmol. 1991;75:417. of retrobulbar needle path length. Ophthalmology. 1989;
63. Puustjarvi T, Purhonen S. Permanent blindness following 96:1221.
retrobulbar hemorrhage after peribulbar anesthesia for 86. Karampatakis V, Natsis K, Gigis P, et al. The risk of optic
cataract surgery. Ophthalmic Surg. 1992;23:450. nerve injury in retrobulbar anesthesia: A comparative study
64. Vindhya PK, Sheets JH. Bilateral block with unilateral of 35 and 40 mm retrobulbar needles in 12 cadavers. Eur J
peribulbar block. Ophthalmic Surg. 1992;23:69. Ophthalmol. 1998;8:184.
65. Capo H, Roth E, Johnson T, et al. Vertical strabismus after 87. Karampatakis V, Natsis K, Gigis P, et al. Orbital depth
cataract surgery. Ophthalmology. 1996;103:918. measurements of human skulls in relation to retrobulbar
66. Zahl K, Nassif JM, Meltzer MA, et al. Simulated peribulbar anesthesia. Eur J Ophthalmol. 1998;8:118.
injection of anesthetic. Ann Ophthalmol. 1991;23:114. 88. Atkinson WS. The development of ophthalmic anesthesia.
67. Rinkoff JS, Doft BH, Lobes LA. Management of ocular Am J Ophthalmol. 1961;51:1.
penetration from injection of local anesthesia preceding 89. Unsold R, Stanley JA, DeGroot J. The CT-topography of
cataract surgery. Arch Ophthalmol. 1991;109:1421. retrobulbar anesthesia. Anatomic-clinical correlation of
426 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
complications and suggestion of a modified technique. 112. Perlman JP, Conn H. Transient internal ophthalmoplegia
Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1981;217: during blepharoplasty. A report of three cases. Ophthal Plast
125. Reconstr Surg. 1991;7:141.
90. Liu C, Youl B, Moseley I. Magnetic resonance imaging of 113. Zaturansky B, Hyams S. Perforation of the globe during the
the optic nerve in extremes of gaze. Implications for the injection of local anesthesia. Ophthalmic Surg. 1987;18:585.
positioning of the globe for retrobulbar anaesthesia. Br J 114. Lindquist TD, Kopietz LA, Spigelman AV, et al. Compli-
Ophthalmol. 1992;76:728. cations of Nadbath facial nerve block and a review of the
91. Hamed LM. Strabismus presenting after cataract surgery. literature. Ophthalmic Surg. 1988;19:271.
Ophthalmology. 1991;98:247. 115. Cofer HF. Cord paralysis after Nadbath facial nerve block.
92. Rainin EA, Carlson BM. Postoperative diplopia and ptosis. Arch Ophthalmol. 1986;104:337.
A clinical hypothesis based on the myotoxicity of local 116. Lichter PR. The relative value of quality care. Ophthalmol-
anesthetics. Arch Ophthalmol. 1985;103:1337. ogy. 1991;98:1151.
93. Hamed LM, Mancuso A. Inferior rectus muscle contracture 117. Rubin AP. Anaesthesia for cataract surgerytime for
syndrome after retrobulbar anesthesia. Ophthalmology. change? Anaesthesia. 1990;45:717718.
1991;98:1506. 118. Standards for basic anesthesia monitoring. Directory of mem-
94. Carlson BM, Emerick S, Komorowski TE, et al. Extraocular bers. Park Ridge: American Society of Anesthesiologists;
muscle regeneration in primates. Local anesthetic-induced 2005.
lesions. Ophthalmology. 1992;99:582. 119. Vachon CA, Warner DO, Bacon DR. Succinylcholine and
95. Hunter DG, Lam GC, Guyton DL. Inferior oblique muscle the open globe. Tracing the teaching. Anesthesiology. 2003;
injury from local anesthesia for cataract surgery. Ophthal- 99:220.
mology. 1995;102:501. 120. McGoldrick KE. The open globe: Is an alternative to
96. Hamilton RC. Complications of ophthalmic regional anaes- succinylcholine necessary? J Clin Anesth. 1993;5:1.
thesia. In: Kumar A, Dodds C, Fanning GL, eds. Ophthalmic 121. Seaberg RR, Freeman WR, Goldbaum MH, et al. Permanent
anaesthesia. Lisse, The Netherlands: Sweets and Zeitlinger; postoperative vision loss associated with expansion of
2002:181. intraocular gas in the presence of a nitrous oxide-containing
97. Dutton JJ, Hasan SA, Edelhauser HF, et al. Anesthesia for anesthetic. Anesthesiology. 2002;97:1309.
intraocular surgery. Surv Ophthalmol. 2001;46:172. 122. Lee EJ. Use of nitrous oxide causing severe visual loss
98. Kaplan LJ, Jaffe NS, Clayman HM. Ptosis and cataract 37 days after retinal surgery. Br J Anaesth. 2004;93:464.
surgery. A multivariant computer analysis of a prospective 123. Vote BJ, Hart RH, Worsley DR, et al. Visual loss after use
study. Ophthalmology. 1985;92:237. of nitrous oxide gas with general anesthetic in patients
99. Wittpenn JR, Rapoza P, Sternberg P Jr, et al. Respiratory with intraocular gas still persistent up to 30 days after
arrest following retrobulbar anesthesia. Ophthalmology. vitrectomy. Anesthesiology. 2002;97:1305.
1986; 93:867. 124. Nagle F, Cooper RL. Nasal secretions in ocular surgery
100. Nicoll JM, Acharya PA, Ahlen K, et al. Central nervous under general and local anesthesia. Ophthalmic Surg. 1993;
system complications after 6000 retrobulbar blocks. Anesth 24:13.
Analg. 1987;66:1298. 125. Kuhn F, Morris R, Witherspoon CD, et al. Nasal secretion
101. Hamilton RC. Brain-stem anesthesia as a complication and intraocular surgery. Ophthalmic Surg. 1993;24:780.
of regional anesthesia for ophthalmic surgery. Can J 126. Schlager A, Luger TJ. Oxygen application by a nasal probe
Ophthalmol. 1992;27:323. prevents hypoxia but not rebreathing of carbon dioxide in
102. Javitt JC, Addiego R, Friedberg HL, et al. Brain stem patients undergoing eye surgery under local anaesthesia. Br
anesthesia after retrobulbar block. Ophthalmology. 1987;94: J Ophthalmol. 2000;84:399.
718. 127. Schlager A. Accumulation of carbon dioxide under oph-
103. Meyers EF, Ramirez RC, Boniuk I. Grand mal seizures after thalmic drapes during eye surgery: A comparison of three
retrobulbar block. Arch Ophthalmol. 1978;96:847. different drapes. Anaesthesia. 1999;54:690.
104. Beltranean HP, Vega MJ, Kirk N, et al. Inadvertent 128. Inan UU, Sivaci RG, Ozturk F. Effectiveness of oxygenation
intravascular bupivacaine injection following retrobulbar and suction in cataract surgery: Is suction of CO2-enriched
block. Report of three cases. Reg Anesth. 1981;6:149. air under the drape during cataract surgery necessary? Eye.
105. Kaufer G, Augustin G. Orbitography. Report of a compli- 2003;17:74.
cation with use of water-soluble contrast material. Am J 129. Bhananker SM, Posner KL, Cheney FW, et al. Injury and
Ophthalmol. 1966;61:795. liability associated with monitored anesthesia care. A closed
106. Kobet KA. Cerebral spinal fluid recovery of lidocaine and claims analysis. Anesthesiology. 2006;104:228.
bupivacaine following respiratory arrest subsequent to 130. Denborough M. Malignant hyperthermia. Lancet. 1998;
retrobulbar block. Ophthalmic Surg. 1987;18:11. 352:1131.
107. Antoszyk AN, Buckley EG. Contralateral decreased visual 131. Carr CA. Strabismus. In: Smith GB, Hamilton RC, Carr CA,
acuity and extraocular muscle palsies following retrobulbar eds. Ophthalmic anaesthesia. A practical handbook, 2nd ed.
anesthesia. Ophthalmology. 1986;93:462. London: Arnold; 1996:225.
108. Friedberg HL, Kline OR Jr. Contralateral amaurosis af- 132. Roth S, Thisted RA, Erickson JP, et al. Eye injuries after
ter retrobulbar injection. Am J Ophthalmol. 1986;101: nonocular surgery. A study of 60,965 anesthetics from 1988
688. to 1992. Anesthesiology. 1996;85:1020.
109. Batterbury M, Wong D, Williams R, et al. Peribulbar 133. White E, Crosse MM. The aetiology and prevention of
anaesthesia: Failure to abolish the oculocardiac reflex. Eye. peri-operative corneal abrasions. Anaesthesia. 1998;53:
1992;6(Pt 3):293. 157.
110. Zahl K. Blockade of the orbicularis oculi. Ophthalmol Clin 134. Cheng MA, Sigurdson W, Tempelhoff R, et al. Visual loss
North Am. 1990;3:93. after spine surgery: A survey. Neurosurgery. 2000;46:6250;
111. Brancato R, Pece A, Carassa R. Central retinal artery discussion 6301.
occlusion after local anesthesia for blepharoplasty. Graefes 135. Roth S, Barach P. Postoperative visual loss: Still no
Arch Clin Exp Ophthalmol. 1991;229:593. answersyet. Anesthesiology. 2001;95:575.
C H A P T E R 2 9 / O P H T H A L M O L O G I C C O M P L I C AT I O N S 427
136. Huber JF, Grob D. Bilateral cortical blindness after lumbar 161. Wolfe SW, Lospinuso MF, Burke SW. Unilateral blindness
spine surgery. A case report. Spine. 1998;23:1807. as a complication of patient positioning for spinal surgery.
137. Katz DM, Trobe JD, Cornblath WT, et al. Ischemic optic A case report. Spine. 1992;17:600.
neuropathy after lumbar spine surgery. Arch Ophthalmol. 162. Recchia FM, Brown GC. Systemic disorders associated with
1994;112:925. retinal vascular occlusion. Curr Opin Ophthalmol. 2000;
138. Lee AG. Ischemic optic neuropathy following lumbar spine 11:462.
surgery. Case report. J Neurosurg. 1995;83:348. 163. Quinlan MF, Salmon JF. Ophthalmic complications after
139. Lee LA, Lam AM. Unilateral blindness after prone lumbar heart transplantation. J Heart Lung Transplant. 1993;12:252.
spine surgery. Anesthesiology. 2001;95:793. 164. Beri M, Klugman MR, Kohler JA, et al. Anterior ischemic
140. Myers MA, Hamilton SR, Bogosian AJ, et al. Visual loss as a optic neuropathy. VII. Incidence of bilaterality and various
complication of spine surgery. A review of 37 cases. Spine. influencing factors. Ophthalmology. 1987;94:1020.
1997;22:1325. 165. Buono LM, Foroozan R. Perioperative posterior ischemic
141. Stevens WR, Glazer PA, Kelley SD, et al. Ophthalmic optic neuropathy: Review of the literature. Surv Ophthal-
complications after spinal surgery. Spine. 1997;22:1319. mol. 2005;50:15.
142. Alexandrakis G, Lam BL. Bilateral posterior ischemic 166. Sadda SR, Nee M, Miller NR, et al. Clinical spectrum of
optic neuropathy after spinal surgery. Am J Ophthalmol. posterior ischemic optic neuropathy. Am J Ophthalmol.
1999;127:354. 2001;132:743.
143. Nuttall GA, Garrity JA, Dearani JA, et al. Risk factors for 167. Leuthardt R, Petralli C, Lutschg J, et al. Cortical blindness:
ischemic optic neuropathy after cardiopulmonary bypass: An unusual complication after removal of a ganglioneu-
A matched case/control study. Anesth Analg. 2001;93: roma of the neck. Childs Nerv Syst. 2001;17:356.
1410. 168. Raj P, Moore PL, Henderson J, et al. Bilateral cortical
144. Shapira OM, Kimmel WA, Lindsey PS, et al. Anterior blindness: An unusual complication following unilateral
ischemic optic neuropathy after open heart operations. neck dissection. J Laryngol Otol. 2002;116:227.
Ann Thorac Surg. 1996;61:660. 169. Ho VT, Newman NJ, Song S, et al. Ischemic optic neu-
145. Busch T, Sirbu H, Aleksic I, et al. Anterior ischemic optic ropathy following spine surgery. J Neurosurg Anesthesiol.
neuropathy: A complication after extracorporal circulation. 2005;17:38.
Ann Thorac Cardiovasc Surg. 1998;4:354. 170. Dunker S, Hsu HY, Sebag J, et al. Perioperative risk factors
146. Roth S, Gillesberg, I. Injuries to the visual system and other for posterior ischemic optic neuropathy. J Am Coll Surg.
sense organs. In: Benumof JL, Saidman LJ, eds. Anesthesia 2002;194:705.
and perioperative complications. 2nd ed. St Louis: Mosby; 171. Murphy MA. Bilateral posterior ischemic optic neuropathy
1999:377. after lumbar spine surgery. Ophthalmology. 2003;110:
147. Rizzo JF III, Lessell S. Posterior ischemic optic neuropathy 1454.
during general surgery. Am J Ophthalmol. 1987;103:808. 172. Johnson MW, Kincaid MC, Trobe JD. Bilateral retrob-
148. Fenton S, Fenton JE, Browne M, et al. Ischaemic optic ulbar optic nerve infarctions after blood loss and hy-
neuropathy following bilateral neck dissection. J Laryngol potension. A clinicopathologic case study. Ophthalmology.
Otol. 2001;115:158. 1987;94:1577.
149. Minagar A, Schatz NJ, Glaser JS. Liposuction and ischemic 173. Brown RH, Schauble JF, Miller NR. Anemia and hy-
optic neuropathy. Case report and review of literature. potension as contributors to perioperative loss of vision.
J Neurol Sci. 2000;181:132136. Anesthesiology. 1994;80:222.
150. Bonardo P, Ebner R, Martinez O, et al. Bilateral and 174. Kalyani SD, Miller NR, Dong LM, et al. Incidence of and
simultaneous ischemic optic neuropathy following hip risk factors for perioperative optic neuropathy after cardiac
replacement surgery. Neurologia. 2000;15:258. surgery. Ann Thorac Surg. 2004;78:34.
151. Hollenhorst RW, Svien JH, Benoit CF. Unilateral blindness 175. Hayreh SS, Zimmerman MB, Podhajsky P, et al. Nocturnal
occurring during anesthesia for neurosurgical operations. arterial hypotension and its role in optic nerve head and
Arch Ophthalmol. 1954;52:819. ocular ischemic disorders. Am J Ophthalmol. 1994;117:
152. Bhatti MT, Enneking FK. Visual loss and ophthalmoplegia 603.
after shoulder surgery. Anesth Analg. 2003;96:899. 176. Cove DH, Seddon M, Fletcher RF, et al. Blindness after
153. van der Bijl P, Meyer D. Ocular complications of dental treatment for malignant hypertension. Br Med J. 1979;2:245.
local anaesthesia. SADJ. 1998;53:235. 177. Connolly SE, Gordon KB, Horton JC. Salvage of vision
154. Barletta JP, Fanous MM, Hamed LM. Temporary blindness after hypotension-induced ischemic optic neuropathy. Am
in the TUR syndrome. J Neuroophthalmol. 1994;14:6. J Ophthalmol. 1994;117:235.
155. Warner ME, Warner MA, Garrity JA, et al. The frequency 178. Goodnough LT, Brecher ME, Kanter MH, et al. Transfusion
of perioperative vision loss. Anesth Analg. 2001;93:1417. medicine. First of two partsblood transfusion. N Engl J
156. Williams EL, Hart WM Jr, Tempelhoff R. Postoperative Med. 1999;340:438.
ischemic optic neuropathy. Anesth Analg. 1995;80:1018. 179. Hayreh SS. Anterior ischemic optic neuropathy. VIII.
157. Chang SH, Miller NR. The incidence of vision loss due to Clinical features and pathogenesis of post-hemorrhagic
perioperative ischemic optic neuropathy associated with amaurosis. Ophthalmology. 1987;94:1488.
spine surgery: The Johns Hopkins Hospital Experience. 180. Jaben SL, Glaser JS, Daily M. Ischemic optic neuropathy
Spine. 2005;30:1299. following general surgical procedures. J Clin Neuroophthal-
158. Sweeney PJ, Breuer AC, Selhorst JB, et al. Ischemic optic mol. 1983;3:239.
neuropathy: A complication of cardiopulmonary bypass 181. Roth S, Roizen M. Optic nerve injury: Role of the
surgery. Neurology. 1982;32:560. anesthesiologist? Anesth Analg. 1996;82:435.
159. Lee LA. POVL registry reports preliminary data. APSF 182. Potarazu SV. Ischemic optic neuropathy: Models for
Newsl. 2003;18:2122. mechanism of disease. Clin Neurosci. 1997;4:264.
160. Locastro A, Novak KD, Biglan AW. Central retinal artery oc- 183. Cheng MA, Todorov A, Tempelhoff R, et al. The effect of
clusion in a child after general anesthesia. Am J Ophthalmol. prone positioning on intraocular pressure in anesthetized
1991;112:91. patients. Anesthesiology. 2001;95:1351.
428 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
184. Leibovitch I, Casson R, Laforest C, et al. Ischemic orbital 193. Shaw PJ, Bates D, Cartlidge NE, et al. Neuro-
compartment syndrome as a complication of spinal surgery ophthalmological complications of coronary artery bypass
in the prone position. Ophthalmology. 2006;113:105. graft surgery. Acta Neurol Scand. 1987;76:1.
185. Kumar N, Jivan S, Topping N, et al. Blindness and rectus 194. Moody DM, Bell MA, Challa VR, et al. Brain microem-
muscle damage following spinal surgery. Am J Ophthalmol. boli during cardiac surgery or aortography. Ann Neurol.
2004;138:889. 1990;28:477.
186. Lee LA, Roth S, Posner KL, et al. An analysis of 71 spine caes 195. Grossman W, Ward WT. Central retinal artery occlusion
with ischemic optic neuropathy from the ASA postoperative after scoliosis surgery with a horseshoe headrest. Case
visual loss registry. Anesthesiology. 2005;103:A1. report and literature review. Spine. 1993;18:1226.
187. Martin WR, Hashimoto SA. Stroke in coronary bypass 196. Halfon MJ, Bonardo P, Valiensi S, et al. Central retinal
surgery. Can J Neurol Sci. 1982;9:21. artery occlusion and ophthalmoplegia following spinal
188. Hart R, Hindman B. Mechanisms of perioperative cerebral surgery. Br J Ophthalmol. 2004;88:1350.
infarction. Stroke. 1982;13:766. 197. Ozcan MS, Praetel C, Bhatti MT, et al. The effect of body in-
189. Gilman S. Neurological complications of open heart clination during prone positioning on intraocular pressure
surgery. Ann Neurol. 1990;28:475. in awake volunteers: A comparison of two operating tables.
190. Blauth C, Arnold J, Kohner EM, et al. Retinal microem- Anesth Analg. 2004;99:1152.
bolism during cardiopulmonary bypass demonstrated by 198. Pazos GA, Leonard DW, Blice J, et al. Blindness after
fluorescein angiography. Lancet. 1986;2:837. bilateral neck dissection: Case report and review. Am J
191. Blauth CI, Arnold JV, Schulenberg WE, et al. Cerebral Otolaryngol. 1999;20:340.
microembolism during cardiopulmonary bypass. Retinal 199. Lee LA, Nathens AB, Sires BS, et al. Blindness in the
microvascular studies in vivowith fluorescein angiography. intensive care unit: Possible role for vasopressors? Anesth
J Thorac Cardiovasc Surg. 1988;95:668. Analg. 2005;100:192.
192. Williams IM. Intravascular changes in the retina during
open-heart surgery. Lancet. 1971;2:1260.
D . R E N A L DY S F U N C T I O N
A N D D I SO R D E R S : F L U I D S
A N D E L EC T R O LY T E S .
CHAPTER OLIGURIA
CASE SUMMARY
What Should We Know About
48-year-old, 90-kg man with a 24-hour pe-
Oliguria?
A
riod of severe vomiting and abdominal pain
undergoes emergency surgery for a small
bowel obstruction. His baseline serum elec- All we really know for certain about the kidney is
trolytes include Na+ 145 mEq per L, Cl that it produces urine.1 This statement was made by
90 mEq per L, K+ 3.1 mEq per L. Before the eminent twentieth century renal physiologist and
surgery, he receives 3.5 L of 5% dextrose in 0.45% saline. philosopher, Homer Smith, who thus characterized the
During the 4-hour operative procedure for exploratory la- complexity of urine formation. Normal urine production
parotomy, lysis of adhesions, and a 1-ft segmental small reflects the delicate balance between renal and extrarenal
bowel resection for vascular compromise, the anesthe- factors. This balance is often disrupted in multitrauma
siologist administers 4.2 L of lactated Ringers solution. patients or those undergoing major surgery. The result is
Total urine output during surgery is 200 mL. Postoper- decreased urine output, or oliguria.
atively, over the first 18 hours, he receives 4.5 L of 5% Historically, and often in present times, oliguria is
dextrose in 0.45% saline. His urine output is 425 mL, but viewed as a reflection of dysfunctional kidneys, or even
for the last 5 hours averages <30 mL per hour. Serum renal failure. Therapy commonly is directed to restoration
electrolytes are measured and reveal Na+ 119 mEq per L, of a normal urinary output. Frequently overlooked is the
Cl 87 mEq per L, and K+ 2.9 mEq per L. He is irritable fact that oliguria may represent an adaptive mechanism
and lethargic and follows commands only sporadically. by the kidneys to restore homeostasis.2,3 However, renal
His blood pressure is 90/65 mm Hg, and his pulse rate compensatory mechanisms are limited, and overt renal
is 120 bpm. The surgeon orders furosemide, 40 mg in- insufficiency may occur over a short period of time
travenously, but over the next 2 hours he only urinates (minutes to hours). As a result, the window for therapeutic
45 mL. A nephrologist is consulted because of the oliguria intervention is limited.4,5
and apparent alteration in renal function. After examining
the patient and reviewing the chart, he recommends fluid
restriction and the administration of 500 mL of 3% saline GLOMERULAR FILTRATION
over the subsequent 5-hour period. The patients mental
status gradually improves, and his urine output increases Evaluation and treatment of oliguria requires an under-
to 50 to 60 mL per hour, with a rise in serum Na+ to standing of the mechanisms involved in urine production,
130 mEq per L. Five days later, he was discharged after the risk factors that lead to oliguria, the pathophysiologic
an otherwise uneventful recovery. entities responsible for decreased urine output, and the
429
430 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
goals of therapy. Prevention of acute renal failure, as op- the release of atrial and brain natriuretic peptides (BNPs)
posed to renal dysfunction, is essential to prevent the poor from the left atrial and ventricular endocardium in re-
associated prognosis.6,7 sponse to increased filling pressures.13 These endogenous
One hundred and sixty to 180 L of water, each peptides are important renal vasodilators, which also in-
containing approximately 300 mOsm of solute, are filtered crease the glomerular filtration rate (GFR) and sodium
daily through the glomeruli of a normal adult in a excretion.
24-hour period. Oliguria in an adult with normally The response to renal hypoperfusion includes affer-
functioning kidneys is usually defined by a urine output ent arteriolar dilatation combined with efferent arteriolar
<400 mL per day (the amount of maximally concentrated constriction that increases filtration fraction and activa-
urine required to excrete the normal daily nitrogenous tion of the renin-angiotensin aldosterone axis, followed
waste).8 However, patients who receive osmotic diuretics by increased sodium reabsorption and efferent arteriolar
or have severe hyperglycemia may be oliguric despite a resistance.1417 Prostaglandins D2 , I2 , and E2 produced
high urinary output. Individuals with preexisting renal in the kidneys autoregulate local blood flow by acting
dysfunction may be unable to concentrate the urine as renal vasodilators. PGE2 y counteracts the renal vaso-
effectively and therefore require a higher urinary output constrictive effects of angiotensin and norepinephrine to
to maintain homeostasis. preserve renal blood flow.18 Certain nonsteroidal anti-
Abnormalities in extracellular fluid volume, or os- inflammatory agents (NSAIDs) decrease the synthesis
molality, elicit increases or decreases in urine output. of PGE2 by inhibiting the activity of cyclooxygenase,
Marked reduction or cessation of urinary output in a thereby increasing the risk of renal ischemia in susceptible
trauma patient during or after surgery can occur sud- patients.19
denly or over a period of several hours to days. It may be
the harbinger of acute renal failure. Correct diagnosis and
treatment are paramount to prevent significant morbidity
and mortality. How Is Oliguria Classified?
If oliguria is associated with renal insufficiency,
a measurable decrease in renal function is present,
Oliguria may be prerenal (hypovolemia, inadequate renal
but serum biochemical values are normal. As renal
perfusion), renal (intrinsic), and postrenal (obstruction,
failure supervenes, the function deteriorates to such an
extravasation) (see Table 30.1).20
extent that abnormal serum biochemical values result.
Significant oliguria implies a state of renal failure, because
a urine output of 400 mL per day cannot excrete the
average daily solute load of 650 to 750 mOsm (normal PRERENAL
diet) in a maximally concentrated urine (1.2 mOsm per
mL). Because renal failure may also be associated with a Renal Hypoperfusion
high urine output, oliguric and polyuric states represent
the extremes of a continuum. A diagnosis of renal failure This condition may be caused by inadequate circulating
must take into account the quantity and quality of urine. blood volume, the administration of some anesthetic
agents, and the extremity compartment syndrome21
following vascular or trauma surgery. Cardiac failure and,
rarely, renal artery thrombosis may be responsible. With
RENAL BLOOD FLOW minimal or moderate depression of renal blood flow, a
The kidneys receive approximately one fifth the blood vol- compensatory increase in filtration fraction results, and
ume of the total cardiac output; however, their oxygen GFR and urine formulation remain relatively unaffected.
consumption is rather low (approximately 10% of total When marked depression of renal blood flow occurs, GFR,
body oxygen consumption).9 The primary determinants of urine formation, and electrolyte excretion are significantly
renal blood flow are cardiac output, renal perfusion pres- reduced.
sure, and local hemodynamic factors such as glomerular
afferent and efferent arteriolar tone. Anesthetic Agents
Ninety to 95% of renal blood flow is to the cortex,
where most glomeruli are located. Conversely, the medulla Potent fluorinated volatile anesthetics cause dose-related
receives only a 5% to 10% of flow.10 Medullary cells have myocardial depression and peripheral vasodilation, prob-
a higher oxygen extraction than cortical cells (80% vs. ably by altering calcium flux within myocardial cells and
20%),11 which, in combination with a lower blood supply, the arteriolar smooth muscle. Associated renal changes in-
places them at a higher risk for hypoxic damage, despite clude decreased renal blood flow, GFR, sodium, potassium
what would be perceived as adequate total renal blood and chloride excretion, and decreased or increased urine
flow. osmolality. Historically, proposed mechanisms for the de-
Renal blood flow is constant over a wide range of crease in urinary output included increase in aldosterone
mean arterial pressures. The same degree of hypotension and 17-hydroxy corticosteroids; increased catecholamine
associated with hypovolemia causes a greater reduction in secretion (which was known to occur during ether and
renal blood flow than hypotension caused by impaired left cyclopropane anesthesia); and increase in antidiuretic
ventricular (LV) function.12 This difference results from hormone (ADH) secretion. However, the urine solute was
CHAPTER 30/OLIGURIA 431
Modified from Dooley JR, Mazze RI. Oliguria. In: Gravenstein N, Kirby RR, eds. Complications in anesthesiology.
2nd ed. Philadelphia: Lippincott-Raven, 1996;484.
reduced by 80% to 85%, and the antidiuresis was not in- with vinblastine, bleomycin, or cisplatin.29 Oliguria and
variably associated with hyperosmotic urine. Therefore, azotemia have occurred after captopril therapy30,31 or
increased ADH secretion was not likely to have been following treatment with amphotericin B. In the latter
responsible. situation, renal vasoconstriction is thought to occur
General anesthesia may abolish the autoregulation concomitantly with nephrotoxicity.
that normally maintains renal perfusion in the face of
hypotension, thereby leading to decreased renal blood
flow and GFR. Although degradation of sevoflurane to Other Factors
compound A that is nephrotoxic in rats has been demon-
Tubular obstruction and backflow of filtrate into damaged
strated,22 low-flow isoflurane and sevoflurane do not ap-
tubules may be causative factors.32 Most common in the
pear to alter renal function in patients with preexisting
pathogenesis of acute renal failure is the suppression of
stable renal disease.23 With the exception of methoxyflu-
glomerular filtration.33 Light and electron microscopic
rane, which was associated with high output renal failure
studies of glomeruli generally failed to reveal structural
and no longer is used, anesthetic-induced changes of renal
abnormalities; hence, reduced glomerular filtration likely
function are readily reversed when the agent is discontin-
is caused by vasomotor phenomena.34
ued.2427 However, the antidiuresis may persist well into
the postoperative period and argues against any persistent
anesthetic agent role.
RENAL
Reduction of Circulating Blood Volume
Causes
Oliguria caused by blood loss, fluid sequestration into
a surgical or traumatic third space, dehydration, or gas- Nephrotoxic Agents
trointestinal loss is a complex problem. If hypovolemia is Ionic and nonionic contrast materials used in digital vas-
sufficient to cause renal ischemia, functional lesions ac- cular imaging and selective renal angiography can impair
quire a renal morphologic component. In extreme cases, renal function. NSAIDS such as phenylbutazone, ibupro-
simple prerenal oliguria is transformed into oliguric renal fen, and indomethacin; and aminoglycoside antibiotics
failure.28 The anesthesia provider in such circumstances may lead to oliguria and renal insufficiency. The latter
must administer sufficient blood or fluid to replete the in- agents account for 5% to 10% of all hospital-acquired
travascular volume and restore renal blood flow and GFR. acute renal failure.3540
renal failure when hemoglobin enters the glomeruli.41,42 ultrasound has been used to evaluate intrarenal blood
In addition, when incompatible blood is transfused, flow.56 Surrogates of renal blood flow include measure-
disseminated intravascular coagulation results, with fibrin ment of arterial blood pressure, LV preload, and cardiac
deposition in renal tubules. Red cell membranes are output.
thought to initiate the coagulation process, ultimately
leading to a decrease in platelets, fibrinogen, and factors II, Arterial Blood Pressure
V, and VII.41 Myoglobinuria following extensive, crushing-
type muscle injuries also may contribute to oliguric renal Mean arterial blood pressure provides an estimate of renal
failure.41,42 The mechanism is probably similar as that for perfusion pressure, whereas the variation in systolic blood
hemoglobin. pressure (SBP) during positive-pressure ventilation may
provide information on effective circulating volume.57
TABLE 30.2 Relation Between Serum Creatinine TABLE 30.4 Urinary Indices and Oliguria
and Glomerular Filtration Rate
Parameter Prerenal Renal
Glomerular Filtration Rate U/P Osm >1.5 1
Creatinine (mg/dL) (mL/min) C Osm >1 0
1 100 U/P Cr >30 <10
2 50 C H2 O <0 0
4 25 FENa (%) <1 >2
8 12.5 RFI <1 >2
16 6.25
U/P Osm, ratio of urinary and plasma osmolarity; C Osm, Osmolar
clearance (U/P Osm urinary volume [mL/min]); U/P Cr, ratio
of urinary/plasma creatinine; C H2 O, free water clearance (Urinary
Serum creatinine (SCr) also is inversely proportional volume [mL/min]C Osm); FENa, fraction of excretion of sodium
to GFR. Creatinine production and release are related to (U/P Na+ /U/P Cr); RFI, renal failure index (U/P Na+ /Scr).
muscle mass. It is freely filtered by the glomeruli with
very little secretion or reabsorbtion by the tubules. Serum
creatinine GFR is a constant in steady state conditions. and high osmolar clearance, also suggest a prerenal
If GFR is halved, SCr doubles (see Table 30.2). If GFR etiology (see Table 30.4), with the exception of hepatorenal
ceases, the SCr rises approximately 0.5 to 1.0 mg/dL/day. syndrome. This form of renal failure is due to local
The increase is much greater following severe trauma redistribution of blood flow and is invariably associated
with resultant muscle damage, or in rhabdomyolysis. with a very low urinary sodium concentration. The
Loss of muscle tissue with renal failure may result in fractional excretion of sodium, the renal failure index, and
a deceptively low SCr. In this setting, BUN and SCr values other urinary indices for differentiating prerenal failure
do not accurately reflect acute changes in GFR. Conditions from acute tubular necrosis are of limited clinical value in
associated with acute elevations in BUN and creatinine the operating room. In the absence of lower urinary tract
are summarized in Table 30.3. trauma, a positive heme result on a dipstick suggests the
presence of free hemoglobin or myoglobin in the urine.
The latter is characterized by the absence of red cells on
urinalysis.
URINALYSIS
Urine Flow
RADIOGRAPHIC STUDIES
Causes of complete anuria include cortical necrosis,
vascular accident, glomerulitides, vasculitides, and severe Sonography, Computed Tomography,
intra-abdominal hypertension.72 Urinary tract obstruction Radionuclide Studies
is usually incomplete when the obstructing body changes
position. Sonography and computed tomography are integral
to the evaluation of oliguria.73,74 Radionuclide studies
are useful in the diagnosis of renal artery stenosis,
Urinary Composition chronic pyelonephritis, and lymphomatous infiltration of
Amber color urine with a high specific gravity (1.030) the kidney. These studies may also provide functional
suggests preservation of the concentrating ability of the information for each kidney without invasive measures.74
kidney and points towards a prerenal cause. Urinary
indices, such as a low fraction excretion of sodium
TABLE 30.5 Preoperative Risk Factors for Postoperative As a general rule, replacement fluid should match the
Oliguria and Renal Failure type of fluid that has been lost. A variety of solutions are
available (see Table 30.7). In surgery, shock, and trauma
Patient-Related Elderly exsanguination necessitates blood volume replacement.
Diabetes mellitus When blood loss is less severe, or when hypovolemia
Cardiac dysfunction has been present for a prolonged period, options other
Peripheral vascular disease than blood replacement alone should be considered. Of
Sepsis liver failure major importance is the avoidance of large amounts of
Preexisting renal dysfunction hypotonic solutions, which by themselves can lead to
Burns significant oliguria.
Procedure-Related Aortic surgery Surgical trauma is associated with a decrease in
Cardiopulmonary bypass functional extracellular fluid volume (FECV), which
Liver or renal transplant roughly equates to the interstitial fluid. The magnitude
Trauma surgery of fluid shifts (into the third space) and the management
Biliary surgery of fluid and electrolyte therapy in the surgical patient have
Renal Insults Severe hypovolemia been subjects of inquiry and debate for almost 100 years.
Hypotension In 1913, Trout75 commented thus:
Embolism Even we surgeons know of the wonderful improve-
Nephrotoxins ment in some patients with nephritis when placed on a
Endogenous: Bilirubin, hemoglobin, salt-free diet, and all of us realize there is a transient renal
uric acid, myoglobin irritation or possibly a nephritis following the majority of
Enxogenous: NSAIDs, anesthetics and infections. . . . It is true, sodium chloride
aminoglycosides, cyclosporin, is the least toxic of the group of similar metal chlorides,
amphotericin B, contrast dyes but even at that it is a poison to all people when given in
large doses, and occasionally very toxic in small doses to
NSAIDs, nonsteroidal anti-inflammatory agents. a certain class of cases.
Approximately 30 years later, Coller and his col-
leagues76 stated:
renal effects of the contrast material. In addition, several The experience of years regarding the toxicity of
other measures have been tested with varying success (see isotonic sodium chloride solution has been forgotten. . . .
Table 30.6). Because of the relatively high incidence of salt intolerance
Repletion of the intravascular volume, as well as that following a general anesthesia, it is felt that no isotonic
of other body compartments, is essential if a significant saline or Ringers solution should be given during the
deficit is present. Of great importance is the choice of day of operation and during the first two post-operative
fluid. days. The fluid requirement is met with glucose solution.
If a significant loss of extracellular fluid occurs during
Balanced Electrolyte Solutions the above period, it is replaced with 0.5 per cent sodium
chloride solution to which 50 Gm per liter of dextrose
Restoration of circulating blood volume to values as near has been added. Isotonic saline solution or Ringers
normal as possible is a major goal in patients with solution is used to replace extracellular fluid loss after
hypovolemia, regardless of the etiology of the deficit. the postoperative urinary suppression has disappeared,
During surgery, additional fluid losses may occur owing usually after the second postoperative day.
to hemorrhage, formation of a surgical third space, Collers influence was powerful and pervasive. For the
or evaporative loss from exposed intestinal surfaces. next 20 years, surgical fluid therapy was restricted largely
Postoperative hypovolemia should be avoided as carefully to the administration of glucose solutions, and sodium
as during the preoperative or intraoperative periods. chloride solutions were relegated to the management of
conditions such as acute Addisonian crisis.
By 1952, however, equally powerful figures in the
TABLE 30.6 Preventive Strategies During Administration surgical world were challenging these concepts. Zimmer-
of Contrast Dye man and Wangensteen77 commented on the problem of
hypoosmolality in patients treated with large amounts of
Vigorous hydration before, during, and after the salt-free intravenous solutions:
procedure Most complications of this nature (convulsions fol-
Use of low contrast volume lowing dilution of the extracellular electrolytes) could
Prevention of hypotension be prevented by the avoidance of excessive amounts of
Avoidance of mannitol and loop diuretics in the glucose solution at the time of surgery and through the
presence of renal dysfunction days immediately following surgery. Daily estimations
Administration of DA1 receptor agonists (e.g., of the body weight and determinations of serum elec-
fenoldopam) trolyte levels before actual disturbances develop should
make it possible to take measures to prevent this com-
DA1, dopaminergic type 1. plication before it occurs. Hypertonic saline solution has
CHAPTER 30/OLIGURIA 435
been of great value in rapidly restoring the normal salt- by reduced lactate values. An early increase in urine
water relation of the extracellular fluid once seizures have output, decreased fluid retention, and improved late
occurred. pulmonary function also are demonstrable. Currently,
only 3% solutions are available off-the-shelf in the
Hypertonic Solutions United States.
Because electrolytes freely cross capillary mem-
Hypertonic saline solutions are beneficial in resuscita- branes, the fluid is divided between the intravascular and
tion from shock/trauma and in many surgical procedures extravascular compartments according to their relative
(see Table 30.8). Because the osmolality of the admin- volumes. Although hypertonic saline solutions increase
istered solution exceeds that of intracellular water, and the intravascular volume more than would the same vol-
because sodium and chloride ions cannot freely cross cell ume of a balanced salt solution, they do so at the expense
membranes, the extracellular fluid becomes slightly hy- of a decreased intracellular volume. If large volumes of
perosmolar. A gradient for water to pass from the cells previously administered balanced electrolyte solutions
into the extravascular compartment is established, and have already increased intracellular volume (remember
the extracellular and intravascular volumes are expanded that most are, in effect, slightly hypotonic), hypertonic
significantly. saline is therapeutic. If not, cellular dehydration can
With respect to survival, 7.5% saline has been shown result.
to be more beneficial than 0.9%, 5%, or 10% saline Compared to isotonic solutions, the lesser volumes are
solutions. Improved tissue perfusion occurs, as indicated associated with equivalent or improved systemic blood
pressure, cardiac output, and survival in experimental
animals.78 A positive inotropic effect has been consistently
TABLE 30.8 Potential Indications For Hypertonic Saline noted. Restoration of normal cellular transmembrane
potential is enhanced, indicating a reversal of the cellular
Major surgical procedures abnormalities induced by hemorrhagic shock. As long
Aortic reconstructive surgery as 24 hours after the shock episode, blood pressure is
Radical cancer surgery (prostatectomy, hysterectomy, maintained more effectively than with lactated Ringers
vulvectomy, and node dissections, or any proce- solution alone.
dure in which extensive blood loss and fluid Increased myocardial contractility and a reduced
shifts may be anticipated) systemic vascular resistance probably are contributory.
Shock (primarily hemorrhagic, but also septic) An early increase in urine output, decreased fluid re-
Slow correction of symptomatic, chronic hyponatremia tention, and improved late pulmonary function are also
TURP syndrome demonstrable. Hypertonic saline solutions appear equally
Reduction in perioperative weight gain and peripheral efficacious, whether administered peripherally through
edema the cephalic vein and femoral artery or centrally in
Reduction in ICP (trauma, subarachnoid hemorrhage, the superior vena cava. Gross and histologic inspection
surgery) demonstrates no vascular damage.
RescueFlow (BioPhausia, Knivsta, Sweden) is a
TURP, transurethral resection of the prostate; ICP, intracranial 250-mL solution containing 7.5% sodium chloride and 6%
pressure. dextran 70. According to the manufacturer, it represents
436 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
a new concept drug for Small Volume Resuscitation, TABLE 30.10 Decreasing Ability to Excrete an
is currently registered in at least 14 European countries, Administered Water Load Following Acute Water
and is recommended as a volume substitution solution in Intoxication and Progressive Hyponatremia
trauma. However, this is not a new concept (as claimed
by the manufacturer), having been promulgated in the Administered Water
early 1970s for burn therapy. Nevertheless, a prepackaged Serum Sodium (mMol/L) Load Excreted (%)
resuscitation solution combining the advantages of crys- 140 71
talloid and colloid is certainly convenient and probably 135139 60
useful. 130134 48
The increased intravascular volume provided by 125129 43
250 mL of RescueFlow is alleged to be two to three 120124 43
times the infused volume, equivalent to the increase 110119 30
in volume resulting from intravenous administration
of 3 L of crystalloid solution. Treatment benefits have
been observed in patients with severe injuries such as
penetrating injury requiring surgery and for patients certain level of sodium deficiency when the subjects drank
requiring intensive care. Application for Food and Drug copious amounts of water.
Association (FDA) approval in the United States has been In another study of salt depletion without dehydra-
submitted. tion,82 hyponatremia caused a 30% to 80% reduction in
GFR. A large infusion of 5% dextrose in water (D5 W)
Hypotonic Solutions when the subjects were already hyponatremic resulted
in further reduction of GFR and did not increase urine
Although the administration of hypertonic crystalloid so- flow. In spite of hyponatremia (10 to 30 mEq) below nor-
lutions can produce hypernatremia, this condition will mal and overhydration, the urine remained hyperosmotic
not have an adverse effect on renal function and oliguria. compared to plasma. Treatment with hypertonic saline
Conversely, if hypotonic solutions are administered in resulted in a return to normal renal blood flow and GFR
large amounts (as frequently happens following resusci- and led to the excretion of large volumes of hypotonic
tation when 0.45% saline or dextrose solution is ordered), urine.
the resultant hyponatremia and water intoxication may Physiologic explanations for these experimental ob-
shut down the kidneys and lead to hyponatremia and servations soon followed. The maximum range of urinary
the water intoxication syndrome (see Table 30.9). This concentration and dilution by healthy, nonstressed kid-
situation was described by Bristol,79 who in 1951 showed neys is 1 mOsm of solute in as little as 0.7 mL or as
that the ability to excrete an administered waterload de-
much as 10 mL of urine. Body osmolality for an individ-
creased with progressive hyponatremia (see Table 30.10).
ual eating a regular diet with normal solute production
Numerous additional studies before and after those of
can be maintained in a normal range with an intake of
Bristol confirmed the detrimental effects of hypona-
1,120 mL/m2 /24 hours to 8,800 mL/m2 /24 hours. How-
tremia on renal function in experimental animals and
ever, if the individual receives only intravenous D5 W,
patients.8082
normal osmolality can be maintained with a maxi-
In one study,81 subjects were made sodium-deficient
mum volume of 2,800 mL/m2 /24 hours because of a
by a salt-free diet and sweat-box treatments, and were
lack of solute for urine formation. Any amount above
given unlimited tap water to drink. Initially, salt and
that limit will be associated with progressive hypona-
water were lost together, and normal body osmolality was
tremia.
maintained at the expense of a decrease in total body
A postoperative stressed surgical patient can maxi-
water (approximately 2 L). With continued sodium loss,
mally dilute urine to 1 Osm in a maximum of 1.2 to
osmolality was sacrificed to maintain body water content
at its now decreased volume. The GFR fell by 30%, and 1.6 mL of urine. If this individual receives only D5 W
it became extremely difficult to evoke a diuresis after a with an average total solute of 200 mOsm per 24 hours,
only approximately 320 mL/m2 /24 hours can be excreted
(average of 500 mL per 24 hours in a typical 70 kg pa-
TABLE 30.9 Signs and Symptoms of Acute Water tient). Larger volumes will lead inexorably to dilutional
Intoxication hyponatremia, the magnitude of which is dependent on
the amount infused.
Weakness The salt and water intolerance in postopera-
Lethargy tive surgical patients described by Coller in 1944,76
Disorientation which was associated with weakness, lethargy, disori-
Nausea, vomiting entation, nausea, vomiting, abdominal distention, olig-
Abdominal distention uria, and coma, are the classic signs of free water
Oliguria excess and hyponatremia. They occur because of in-
Coma appropriate fluid and electrolyte therapy and can be
Death prevented with the administration of balanced electrolyte
solutions.
CHAPTER 30/OLIGURIA 437
+, increased urinary loss; , decreased or no urinary loss; Cl, chloride; FENa, fractional excretion of sodium (sodium clearance/creatinine
clearance 100); H+ , hydrogen ion; HCO3 , bicarbonate; K, potassium; MTAL, medullary thick ascending loop of Henle; Na, sodium.
From Sladen RN. Diuretics. In: Bovill JG, Howie MB, eds. Clinical pharmacology for anaesthetists. London: WB Saunders, 1999;281.
CHAPTER 30/OLIGURIA 439
patient to void within 30 minutes of arrival in the 8. Avoidance of hypovolemia and maintenance of ade-
PACU, reported an incidence of 16%. Age 50 years, quate renal perfusion are the cornerstones of therapy
the amount of intraoperative fluid administered (750 to preserve renal function in high-risk patients.
mL), and bladder volume upon PACU arrival (270
mL) were found independently to increase the risk of
REFERENCES
urinary retention. Bladder volume was measured by a
portable ultrasound device and was found to correlate 1. Smith HW. Renal physiology between two wars. In:
closely to subsequently measured urine drainage. Because Lawrence, KS, ed. Lectures on the kidney, University of
bladder overdistention can be associated with permanent Kansas, 1943;75.
detrusor damage and prolonged micturition difficulties, 2. Thurau K, Boylan JW. Acute renal success: The unex-
pected logic of oliguria in acute renal failure. Am J Med.
the investigators suggest that ultrasonic measurement of
1976;61:308.
urine volume should be performed upon PACU admission 3. Wilson WC, Aronson S. Oliguria: A sign of renal success
and before discharge. or impending renal failure? Anesthesiol Clin North America.
2001;19:841.
4. Anderson RJ, Schrier RW. Acute renal failure. In:
RELIEF OF RENAL Schrier RW, Gottschalk CW, eds. Diseases of the kidney.
Boston: Little, Brown and Company, 1997:1069.
OBSTRUCTION 5. Star RA. Treatment of acute renal failure. Kidney Int.
1998;54:1817.
In postobstructive uropathy, prompt intervention is essen- 6. Thadhani R, Pascual M, Bonventre JV. Acute renal failure.
tial, because prolonged ureteral obstruction may result N Engl J Med. 1996;334:1448.
in irreversible loss of renal function. The degree of re- 7. Wardle EN. Acute renal failure and multiorgan failure.
nal damage is related to the degree of obstruction, its Nephron. 1994;66:380.
duration, and the presence or absence of infection.35 Ther- 8. Dooley JR, Mazze RI. Oliguria. In: Gravenstein N, Kirby RR,
apy may be relatively simple, including single in-and-out eds. Complications in anesthesiology, 2nd ed. Philadelphia:
Lippincott-Raven, 1996:479.
catheterization of the bladder, or, for more severe cases,
9. Sladen RN. Effect of anesthesia and surgery on renal
insertion of an indwelling catheter for 5 to 7 days. function. Crit Care Clin. 1987;3:373.
More complicated cases often require surgical inter- 10. Aukland K. Methods for measuring renal blood flow: Total
vention including percutaneous nephrostomy or supra- flow and regional distribution. Annu Rev Physiol. 1980;
pubic cystostomy. Massive diuresis occasionally follows 42:543.
relief of obstruction. Signs of hypovolemia and electrolyte 11. Brezis M, Rosen S. Hypoxia of the renal medulla-its
imbalance must be monitored closely. Hypotension some- implications for disease. N Engl J Med. 1995;332:647.
times follows rapid decompression of an overdistended 12. Gorfinkel HS, Szidon JP, Hirsch LJ, et al. Renal perfor-
bladder. mance in experimental cardiogenic shock. Am J Physiol.
1972;222:1260.
13. Christensen G. Cardiovascular and renal effects of atrial
natriuretic factor. Scand J Clin Lab Invest. 1993;53:
203.
KEY POINTS 14. Schrier RW. Effects of the adrenergic nervous system
1. Acute oliguria is a common finding in patients and catecholamines on systemic and renal hemodynamics,
sodium and water excretion and renin secretion. Kidney Int.
undergoing major surgery or after significant trauma.
1974;6:291.
2. In most surgical patients, renal hypoperfusion and 15. Boventure JV. Mechanism of ischemic acute renal failure.
nephrotoxins play a significant role in the develop- Kidney Int. 1993;43:1160.
ment of acute oliguria. 16. Bersten AD, Holt AW. Vasoactive drugs and the importance
3. Renal failure may be associated with a high urine out- of renal perfusion pressure. New Horiz. 1995;3:650.
put; therefore, oliguric and polyuric states represent 17. Robertson GL. Thirst and vasopressin function in normal
the extremes of a continuum. and disordered states of water balance. J Lab Clin Med.
4. Prompt restoration of adequate renal perfusion and 1983;101:351.
limiting the effects of toxins on the nephron are 18. Anggard E, Oliw E. Formation of prostaglandins in the
kidney. Kidney Int. 1981;19:771.
essential components of therapy that should be
19. Garella S, Matarese RA. Renal effects of prostaglandins and
carried out by anesthesia providers before irreversible clinical adverse effects of nonsteroidal anti-inflammatory
damage occurs. agents. Medicine. 1984;63:165.
5. Clinicians must rely on frequent evaluation of hemo- 20. Harrington JT, Cohen JC. Acute oliguria. N Engl J Med.
dynamic status and indirect measures of renal func- 1975;292:89.
tion for the prevention and timely management of 21. Mendieta JM, Rubio J, Elias M, et al. Bilateral compartment
oliguria. syndrome after surgical perfusion for chronic ischemia of
6. Strategies, such as the indiscriminate use of diuretics the lower extremities. Rev Esp Anestesiol Reanim. 2004;51:
to convert patients to a nonoliguric state or the use 155.
22. Gonsowski CT, Laster MJ, Eger EI, et al. Toxicity of
of low-dose dopamine, have failed to demonstrate a
compound A in rats. Effect of a 3-hour administration.
favorable outcome. Anesthesiology. 1994;80:556.
7. Osmotic diuretics and DA1 receptor agonists may 23. Conzen PF, Kharasch ED, Czemer SF, et al. Low-flow
help preserve renal function in some patients. sevoflurane compared with low-flow isoflurane anesthesia
440 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
in patients with stable renal insufficiency. Anesthesiology. 45. Oray-Schrom P, St Martin D, Bartelloni P, et al. Giant
2002;97:578. nonpancreatic pseudocyst causing acute anuria. J Clin
24. Habif DV, Papper EM, Fitzpatrick HF, et al. The renal Gastroenterol. 2002;34:160.
and hepatic blood flow, glomerular filtration rate, and 46. Charles AH. Some hazards of pelvic surgery. Proc R Soc
urinary output of electrolytes during cyclopropane, ether, Med. 1967;60:656.
and thiopental anesthesia, operation, and the immediate 47. Mate-Kole MO, Yeboah ED, Affram RK, et al. Anuric acute
postoperative period. Surgery. 1951;30:241. renal failure due to bilateral accidental ureteric ligation
25. Deutsch S, Goldberg M, Stephen GW, et al. Effects of during abdominal hysterectomy. Int J Gynaecol Obstet.
halothane anesthesia on renal function in normal man. 1993;41:67.
Anesthesiology. 1966;27:793. 48. Chisholm GD, Shackman R. Malignant obstructive
26. Mazze RI, Cousins MJ, Barr GA. Renal effects and uraemia. Br J Urol. 1968;40:720.
metabolism of isoflurane in man. Anesthesiology. 1974;40: 49. Wanuck S. Carcinoma of pancreas causing ureteral ob-
536. struction. J Urol. 1973;110:395.
27. Cousins MJ, Greenstein LR, Hitt BA, et al. Metabolism 50. Fall M, Petterson S. Ureteral complications after intra-
and renal effects of enflurane in man. Anesthesiology. 1976; venous formalin instillation. J Urol. 1979;122:160.
44:44. 51. Elem B, Sinha SN. Ureterocele in bilharziasis of the urinary
28. Barry KG, Malloy JP. Oliguric renal failure. JAMA. tract. Br J Urol. 1981;53:428.
1962;179:510. 52. Biggers R, Edwards J. Anuria secondary to bilateral
29. Harrel RM, Sibley R, Vogelzang NJ. Renal vascular ureteropelvic fungus balls. Urology. 1980;15:161.
lesions after chemotherapy with vinblastine, bleomycin, 53. Mundy AR, Podesta ML, Bewick M, et al. The urolog-
and cisplatin. Am J Med. 1982;73:429. ical complications of 1000 renal transplants. Br J Urol.
30. Hricik DE, Browning PJ, Kopelman R, et al. Captopril- 1981;53:397.
induced functional renal insufficiency in patients with 54. Mariani AJ, Barrett DM, Kurtz SB, et al. Bilateral localized
bilateral renal artery stenosis or renal artery stenosis in amyloidosis of the ureter presenting with anuria. J Urol.
a solitary kidney. N Engl J Med. 1983;308:373. 1978;120:757.
31. Farrow PR, Wilkinsin R. Reversible renal failure during 55. Derrick F Jr, Kretkowski RC. Trauma to kidney, ureter,
treatment with captopril. Br Med J. 1979;1(6179):1680. bladder and urethra. Diagnosis and management. Postgrad
32. Ruiz-Guinazu A, Coelho JB, Paz RA. Methemoglobin- Med. 1974;55:183.
induced acute renal failure in the rat: in vivo observation, 56. Garwood S, Davis E, Harris S. Intraoperative trans-
histology and micropuncture measurements of intratubular esophageal ultrasonograpy can measure renal blood flow.
and postglomerular vascular pressures. Nephron. 1967; J Cardiothorac Vasc Anesth. 2001;15:65.
4:257. 57. Perel A, Pizov R, Cotev S. The systolic pressure variation
33. Flanigan WJ, Oken DE. Renal micropuncture study of the is a sensitive indicator of hypovolemia in ventilated dogs
development of anuria in the rat with mercury-induced subjected to gradual hemorrhage. Anesthesiology. 1987;67:
renal failure. J Clin Invest. 1965;44:449. 498.
34. Olsen ST, Skjoldborg H. The fine structure of the renal 58. Fontes ML, Bellows W, Ngo L, et al. Assessment of
glomerulus in acute anuria. Acta Pathol Microbiol Scand. ventricular function in critically ill patients: Limitation of
1967;70:205. pulmonary artery catheterization. Institutions of the McSPI
35. Parfrey PS, Griffiths SM, Barrett BJ, et al. Contrast material- group. J Cardiothorac Vasc Anesth. 1999;13:521.
induced renal failure in patients with diabetes mellitus, 59. Benjamin E, Griffin K, Lebowitz AB, et al. Goal-directed
renal insufficiency, or both. A prospective controlled study. transesophageal echocardiography performed by inten-
N Engl J Med. 1989;320:143. sivists to assess left ventricular function: Comparison
36. Manske CL, Sprafka JM, Strony JT, et al. Contrast with pulmonary artery catheterization. J Cardiothorac Vasc
nephropathy in azotemic diabetic patients undergoing Anesth. 1998;12:10.
coronary angiography. Am J Med. 1990;89:615. 60. Greim RC, Roewer N, Apfel G, et al. Relationship of
37. Walshe JJ, Venuto RC. Acute oliguric renal failure induced echocardiographic preload indices to stroke volume in
by indomethacin: Possible mechanisms. Ann Intern Med. critically ill patients with normal and low cardiac index.
1979;91:47. Intensive Care Med. 1997;23:411.
38. Brandstetter RD, Mar DD. Reversible oliguric renal failure 61. San Fillipo AJ, Weyman AE. The role of echocardiography
associated with ibuprofen treatment. Br Med J. 1978; in managing critically ill patients. J Crit Illn. 1988;
2(6146):1194. 3:27.
39. Hou SH, Burshinsky DA, Wish JB, et al. Hospital- 62. Parker MM, Cunnion RE, Parillo JE. Echocardiography
acquired renal insufficiency: A prospective study. Am J and nuclear imaging in the critical care unit. JAMA. 1985;
Med. 1983;74:243. 254:2935.
40. Birndor N. DIC and renal failure. J Lab Invest. 1971;24:314. 63. Polaert JI, Trouerbach J, De Buyzere M, et al. Evaluation
41. Huang KC, Lee TS, Lin YM, et al. Clinical features of transesophageal echocardiography as a diagnostic and
and outcome of crush syndrome caused by the Chi-Chi therapeutic aid in a critical care setting. Chest. 1995;107:
earthquake. J Formos Med Assoc. 2002;101:249. 774.
42. Better OS. The crush syndrome revisited (19401990). 64. Tabata T, Thomas JD, Klein AL. Pulmonary venous flow by
Nephron. 1990;55:97. Doppler echocardiography: Revisited 12 years later. J Am
43. Grubben AC, van Baardwijk AA, Broering DC, et al. Coll Cardiol. 2003;41:1243.
Pathophysiology and clinical significance of the abdominal 65. Jaffe MB. Partial CO2 rebreathing cardiac outputoperating
compartment syndrome. Zentralbl Chir. 2001;126:605. principles of the NICO system. J Clin Monit Comput. 1999;
44. Chang MC, Miller PR, DAgostino R Jr, et al. Effects of 15:387.
abdominal decompression on cardiopulmonary function 66. Haryadi D, Orr JA, Kuck K, et al. Partial CO2 rebreathing
and visceral perfusion in patients with intra-abdominal indirect Fick technique for non-invasive measurement of
hypertension. J Trauma. 1998;44:440. cardiac output. J Clin Monit Comput. 2000;16:361.
CHAPTER 30/OLIGURIA 441
67. Laupland KB, Bands CJ. Utility of esophageal Doppler as a unanesthetized human volunteers. Anesthesiology. 1993;79:
minimally invasive hemodynamic monitor: A review. Can J 685.
Anaesth. 2002;49:393. 92. MacGregor DA, Butterworth JF IV, Zaloga CP, et al. Hemo-
68. Zollner C, Haller M, Weis M, et al. Beat-to-beat measure- dynamic and renal effects of dopexamine and dobutamine
ment of cardiac output by intravascular pulse contour in patients with reduced cardiac output following coronary
analysis: A prospective criterion standard study in patients artery bypass grafting. Chest 1994;106:835.
after cardiac surgery. J Cardiothorac Vasc Anesth. 2000;14: 93. Sherry E, Tooley MA, Bolsin SN, et al. Effect of dopexamine
125. hydrochloride on renal vascular resistance index and
69. Odenstedt H, Stenqvist O, Lundin S. Clinical evaluation haemodynamic responses following coronary artery bypass
of a partial CO2 rebreathing technique for cardiac output graft surgery. Eur J Anaesthesiol. 1997;14:184.
monitoring in critically ill patients. Acta Anaesthesiol Scand. 94. Allison NL, Cubb JW, Ziemniak JA, et al. The effects of
2002;46:152. fenoldopam, a dopaminergic agonist on renal hemodynam-
70. Marik PE. Pulmonary artery catheterization and esophageal ics. Clin Pharamacol Ther. 1987;41:282.
Doppler monitoring in the ICU. Chest. 1999;116:1085. 95. Aronson S, Goldberg LI, Glock D. Effects of fenoldopam on
71. Rauch H, Muller M, Fleischer F, et al. Pulse contour analysis renal blood flow and systemic hemodynamics during isoflu-
versus thermodilution in cardiac surgery patients. Acta rane anesthesia. J Cardiothorac Vasc Anesth. 1991;5:29.
Anaesthesiol Scand. 2002;46:424. 96. Murphy MB, McCoy CE, Weber RR, et al. Augmentation
72. Swann RC, Merrill JP. The clinical course of acute renal of renal blood flow and sodium excretion in hypertensive
failure. Medicine. 1953;32:215. patients during blood pressure reduction by intravenous
73. Balfe DM, McClennan BL. CT of the retroperitoneum in uro- administration of the dopamine agonist fenoldopam. Cir-
surgical disorders. Surg Clin North America. 1982;62:919. culation. 1987;76:1312.
74. Sherman RA, Byun KJ. Nuclear medicine in acute and 97. Garwood S, Hines R. Perioperative renal preservation.
chronic renal failure. Semin Nucl Med. 1982;12:265. Dopexamine and fenoldopam-new agents to augment renal
75. Trout HH. Protoclysis - an experimental study. Surg Gynecol performance. Semin Anesth Periop Med Pain. 1998;17:
Obstet. 1913;16:560. 308.
76. Coller FA, Campbell KN, Vaughan HH, et al. Postoperative 98. Singer I, Epstein M. Potential of dopamine A-1 agonists in
salt intolerance. Ann Surg. 1944;119:533. the management of acute renal failure. Am J Kidney Dis.
77. Chiara O, Pelosi P, Brazzi L, et al. Resuscitation from 1998;31:743.
hemorrhagic shock: Experimental model comparing nor- 99. Rankin AJ. Mechanisms for the release of atrial natriuretic
mal saline, dextran, and hypertonic saline solutions. Crit peptide. Can J Physiol Pharmacol. 1987;65:1673.
Care Med. 2003;31:1915. 100. Levin ER, Gardner DG, Samson WK. Natriuretic peptides.
78. Zimmerman B, Wangensteen OH. Water intoxication in N Engl J Med. 1998;339:321.
surgical patients. Surgery. 1952;11:654. 101. Conger JD, Falk SA, Hammond WS. Atrial natriuretic
79. Bristol WR. Relation of sodium chloride depletion to urine peptide and dopamine in established acute renal failure
excretion and H2 O intoxication. Am J Sci. 1951;221:412. in the rat. Kidney Int. 1991;40:21.
80. Ariel IM, Miller F. The effects of hypochloremia upon renal 102. Seki G, Suzuki K, Nonaka T, et al. Effects of atrial
function in surgical patients. Surgery. 1950;28:552. natriuretic peptide on glycerol-induced acute renal failure
81. McCance RA. Experimental sodium chloride depletion in in the rat. Jpn Heart J. 1992;33:383.
man. Proc R Soc Lond B Biol Sci. 1935;119:245. 103. Allgren RL, Marbury TC, Rahman SN, et al. Auriculin
82. Stokes JM, Bernard HR, Balfour J. Effects of experimental Anaritide Acute Renal Failure Study Group. Anaritide in
electrolyte depletion upon renal water and solute excretion. acute tubular necrosis. Auriculin Anaritide Acute Renal
Arch Surg. 1962;85:540. Failure Study Group. N Engl J Med. 1997;336:828.
83. Philip W, Perdue MD, Blaser JR, et al. Renal dose 104. Dhingra H, Roongsritong C, Kurtzman NA. Brain natri-
dopamine in surgical patients. Dogma or science? Ann uretic peptide: Role in cardiovascular and volume home-
Surg. 1998;227:470. ostasis. Semin Nephrol. 2002;22:423.
84. Holmes CL, Walley KR. Bad medicine. Low-dose dopamine 105. Abraham WT, Lowes BD, Ferguson DA, et al. Systemic
in the ICU. Chest. 2003;123:1266. hemodynamic, neurohormonal, and renal effects of a steady
85. Juste RN, Moran L, Hooper J, et al. Dopamine clearance in state infusion of human brain natriuretic peptide in patients
critically ill patients. Intensive Care Med. 1998;24:1217. with hemodynamically decompensated heat failure. J Card
86. Marik PE. Low-dose dopamine in critically ill oliguric Fail. 1998;4:37.
patients: The influence of the renin-angiotensin system. 106. Schrier RW. Hormones and hemodynamics in heart failure.
Heart Lung. 1993;22:171. N Engl J Med. 1999;341:577.
87. MacCannell KL, Giraud GD, Hamilton PL. Haemodynamic 107. Rayburn BK, Bourge R. Nesiritide: A unique therapeutic
responses to dopamine and dobutamine infusions as a cardiac peptide. Rev Cardiovasc Med. 2001;2(Suppl 2):S25.
function of duration of infusion. Pharmacology. 1983;26:29. 108. Lindner A. Synergism of dopamine and furosemide in
88. Stephan H, Sonntag H, Henning H, et al. Cardiovascular diuretic-resistant, oliguric acute renal failure. Nephron.
and renal haemodynamic effects of dopexamine: Compari- 1983;33:121.
son with dopamine. Br J Anaesth. 1990;65:380. 109. Kramer HJ, Schumann J, Wassermann C, et al.
89. Jamison M, Widerhorn J, Weber L, et al. Central and Prostaglandin-independent protection by furosemide from
renal hemodynamic effects of a new agonist at peripheral oliguric ischemic renal failure in conscious rats. Kidney Int.
dopamine- and beta-2 adrenoreceptors (dopexamine) in 1980;17:455.
patients with heart failure. Am Heart J. 1989;117:607. 110. Mehta RL, Pascual MT, Soroko S, et al Picard Study Group.
90. Foulds RA. Clinical development of dopexamine hydro- Diuretics, mortality, and nonrecovery of renal function in
choloride (Dopacard) and an overview of its hemodynamic acute renal failure. JAMA 2002;288:2547.
effects. Am J Cardiol. 1988;62:41C. 111. Valdes ME, Landau SE, Shah DM, et al. Increased glomeru-
91. Olsen NV, Lund J, Jensen PF, et al. Dopamine, dobu- lar filtration rate following mannitol administration in man.
tamine and dopexamine: A comparison of renal effects in J Surg Res. 1979;26:473.
442 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
112. Better OS, Rubinstein I, Winaver JM, et al. Mannitol therapy 116. Ruggenenti P, Perico N, Mosconi L, et al. Cal-
revisited (19401997). Kidney Int. 1997;51:886. cium channel blockers protect transplant patients from
113. Elwell RJ, Spencer AP, Eisele G. Combined furosemide cyclosporine-induced daily renal hypoperfusion. Kidney Int.
and human albumin treatment for diuretic-resistant edema. 19993;43:706.
Ann Pharmacother. 2003;37:695. 117. Roux S, Qiu C, Sprecher U, et al. Protective effects of
114. Tanahashi M, Hara S, Yoshida M, et al. Effects of rolipram endothelin receptor antagonists in dogs with aortic cross-
and cilostamide on renal functions and cyclic AMP release clamping. J Cardiovasc Pharmacol. 1999;34:199.
in anesthetized dogs. J Pharmacol Exp Ther. 1999;289: 118. Brophy DF. Role of N-acetylcysteine in the prevention
1533. of radiocontrast-induced nephropathy. Ann Pharmacother.
115. Parker MR, Willatts SM. A pilot study to investigate the 2002;36:1466.
effects of an infusion of aminophylline on renal function 119. Keita H, Diouf E, Tubach F, et al. Predictive factors of early
following major abdominal surgery. Anaesthesia. 2001; postoperative urinary retention in the postanesthesia care
56:670. unit. Anesth Analg. 2005;101:592.
CHAPTER ACUTE PERIOPERATIVE KIDNEY
31
INJURY
A
surgery for an abdominal aortic aneurysm. 4 days later, and he was discharged from the surgical ICU
Baseline electrolytes were Na+ 138 mEq to the surgical floor.
per L, Cl 96 mEq per L, and K+ 3.9 mEq
per L. His baseline creatinine was 0.8 mg
per dL and blood urea nitrogen (BUN) was
22 mg per dL. He was NPO for 10 hours preoperatively.
During the 4-hour surgery, he received 5,400 mL of normal What Risk Factors for Acute
saline, four units of packed red blood cells, and two units Kidney Injury Were Present?
of fresh frozen plasma. He had a 10-minute period of
hypotension with a systolic blood pressure of 85 mm Hg,
The most important risk factor for perioperative acute kid-
which responded to intravenous fluids. Infrarenal aortic
ney injury (AKI) is preexistent renal insufficiency. This pa-
cross-clamping was used for approximately 25 minutes.
tient had multiple risk factors for perioperative AKI, but he
His total urine output (UO) was 500 mL during the
did not have a history of renal insufficiency, and his base-
surgery.
line BUN and creatinine levels were within normal limits.
Postoperatively, he was transferred to the surgical
Other well known risk factors are type 1 diabetes mellitus,
intensive care unit (ICU) intubated with mechanical ven-
age more than 65 years, major vascular surgeries such
tilation. He was stable overnight with minimal volume
as coronary artery bypass graft (CABG), and major aor-
requirements and a UO of more than 0.5 mL/kg/hour.
He began to have hypotensive episodes (systolic blood tic procedures, as well as recent exposure to nephrotoxic
pressure <90 mm Hg), tachycardia (heart rate [HR] agents, such as radiocontrast dye, bile pigments, amino-
>110 bpm), with a central venous pressure (CVP) of glycoside antibiotics, and nonsteroidal anti-inflammatory
2 mm Hg. Hypotension persisted, with a CVP of 3 to drugs (NSAIDs). Hemodynamic instability during the pe-
4 mm Hg, even after 3 L of intravenous 0.9% saline rioperative period further increases the risk for AKI.
solution boluses. Significant abdominal distension oc-
curred, and rectal examination revealed bloody stool.
Colonoscopy showed ischemic changes in the sigmoid
colon, and he was taken back to the operating room What Is Perioperative Acute
urgently for exploratory laparotomy.
Sigmoid colectomy and colostomy were performed.
Kidney Injury?
The patient had hypotensive episodes during the 2 hours
of surgery, requiring 5 L of intravenous crystalloids, and AKI is a syndrome characterized by an abrupt and
one liter of colloid. Postoperatively, his UO decreased to sustained decrease in renal function. This change results
<0.25 mL/kg/hour. He was given 40 mg of intravenous in the retention of nitrogenous (i.e., urea and creatinine)
furosemide, but his UO did not increase. Aggressive and nonnitrogenous waste products and alteration of
intravascular volume resuscitation increased the CVP the kidneys ability to regulate water, electrolytes, and
to 12 mm Hg and systolic blood pressure to more than proton balance. The kidneys role in the initiation and
100 mm Hg. UO increased to more than 0.5 mL/kg/hour. maintenance of dysoxic injury is also intriguing. However,
The patients electrolytes at that time were Na+ 142 mEq we are unable to measure and quantify the accumulation
per L, Cl 114 mEq per L, and K+ 4.0 mEq per L; BUN of various, poorly defined, middle and large molecules
was 40 mg per dL and creatinine was 1.4 by the next day. with postulated toxic effects (uremic toxins), as well
443
444 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 31.1 Common Causes of Acute Kidney Injury (AKI) among Surgical Patients
1. Multifactorial b. Toxic ATN
2. Prerenal Endogenous toxins
a. Hypovolemia Exogenous toxins
(1) Preoperative Radiocontrast nephropathy
Fluid depletion secondary to poor intake Drugs
(routine nil by mouth) Nonsteroidal anti-inflammatory drugs
(2) Effects of surgery Antibiotics
Blood losses Aminogycosides
Extracellular fluid losses Amphothericin
Insensible losses (major abdominal Antivirals (acyclovir)
surgery up to 10 mL/kg/h) Inorganic fluorides from volatile agents (enflurane)
Insensible losses due to mechanical ven- c. Interstitial
tilaton Drugs
b. Reduced effective extracellular volume Autoimmune
Extravasation of fluid out of the vascular Viruses
compartment (the third space effect), Acute bacterial pyelonephritis
especially intra-abdominal and intratho- Malignant infiltration
racic operations d. Intrarenal obstruction
Heart failure Casts
Liver failure Crystals
c. Systemic vasodilation e. Renal artery occlusion
(1) Effect of anesthesia Aortic clamping during vascular surgery
Peripheral vasodilation f. Renal vein occlusion
Myocardial depression g. Intrarenal vascular
Renal efferent arteriolar vasodila- h. Atheroembolic
tion (especially in patients tak- Following release of the aortic clamp
ing angiotensin-converting enzyme Vascular surgery
inhibitor or angiotensin receptor Endoluminal procedures
blocker) Stent placement
Increase in antidiuretic hormone Coronary artery bypass graft
(2) Sepsis i. Vasculitis
(3) Liver failure j. Thrombotic microangiopathies
3. Renal Pregnancy
a. Ischemic acute tubular necrosis Hemolytic uremic syndrome
Renal ischemia Scleroderma
Inflammation Acelerated hypertension
Gut ischemia Glomerulonephritis
Impaired visceral perfusion 4. Postrenal
Portal endotoxemia (open abdominal aortic a. Urinary tract obstruction
aneurysm surgery) Ureters
Ischemia-reperfusion injury (open abdominal Bladder neck
and thoracic vascular surgery) Urethra
hemorrhage, or inadequate resuscitation can be easily 9.2 for patients who develop postoperative AKI compared
identified and corrected with proper monitoring. It has with those who do not.20 The mortality rate for contrast
become less common because of the new guidelines for mediainduced AKI is reported to be 34% compared with
volume resuscitation that promote early, goal-directed 7% for patients who do not develop AKI after exposure to
therapy for patients with sepsis and the systemic inflam- contrast media.21 In the ICU, the mortality rate of AKI is
matory response syndrome (SIRS).2 10% to 15% if it is an isolated organ failure, and increases
to as high as 40% to 90% if it is associated with other
organ failure.22
EPIDEMIOLOGY
AKI is a common problem in hospitalized patients and
is associated with a high morbidity and mortality rate.3 What Is the Pathophysiology
It affects 5% of all hospitalized patients and between
10% and 25% of all critically ill patients.4,5 Acute tubular
of Acute Kidney Injury?
necrosis (ATN) after surgery accounts for 20% to 25% of
all cases of hospital-acquired AKI, many with prerenal Prerenal azotemiareversible renal insufficiency due to
causes.6 Despite the significant advances in perioperative renal hypoperfusionand ATN, or a combination of both,
and intraoperative patient care, AKI is still a significant are among the most common causes of AKI among
complication that increases the postoperative mortality surgical patients.1,23
rate.2 The perioperative incidence varies according to the
etiology and definition of AKI, as well as the type of
surgery. AKI is common after cardiac surgery, occurring ACUTE TUBULAR NECROSIS
in 7.7% to 42% of patients with previously normal renal
function, once again depending on how AKI is defined.79 ATN results from a variety of synergistic combinations
Patients undergoing liver and cardiac transplantation also of ischemic and nephrotoxic insults rather than from
carry a higher risk for AKI.8 The incidence of severe any single insult.1,24 It is characterized by a complex
postoperative AKI requiring renal replacement therapy interaction between the injured tubular epithelial cells
(RRT) is reported between 0.6% and 2.7%.7,10 In the ICU, and the vascular smooth muscle and endothelial cells,
35% to 50% of cases of AKI are attributed to sepsis.1114 most likely genetically determined by the capacity of
Patients with preexisting renal impairment, hyperten- each group of these cells to regenerate and recover.2527
sion, cardiac disease, peripheral vascular disease, diabetes Pioneering studies of ATN pathophysiology in the early
mellitus, jaundice, and advanced age have a significantly 1940s indicated a uniform pattern of disturbed renal
higher risk for developing AKI. Preoperative renal dys- function, which was characterized by gross tubular
function was the single, most consistent predictor of dysfunction and extreme diminution in renal blood
postoperative renal failure.15 The nature of the surgi- flow (RBF), coupled with anatomic evidence of tubular
cal procedure also significantly affects the occurrence necrosis and regeneration.28 Recently, new data emerged
of postoperative AKI, with the highest incidence after that supported the existence of discrete phases of ischemic
CABG and vascular surgeries involving the aorta. For ATN: Prerenal, initiation, extension, maintenance, and
patients undergoing cardiac surgery, factors such as ad- repair (see Fig. 31.1).2729
vanced age, emergency surgery, low ejection fraction,
requirement for intra-aortic balloon pump, redo cardiac Prerenal Phase
surgery, diabetes mellitus, mitral valve surgery, duration
of cardiopulmonary bypass (CPB), and preoperative renal Renal hypoperfusion plays a role in the pathogenesis of
disease were independently associated with AKI.16 The prerenal azotemia and ATN. In experimental studies, AKI
development of postoperative renal dysfunction is also progressed from renal vasoconstriction and intact tubular
accompanied by a higher incidence of gastrointestinal function (prerenal azotemia) to established AKI with
bleeding, respiratory infections, and sepsis.17 Strategies tubular dysfunction if the renal ischemia was prolonged.30
directed toward the early recognition of patients with an Moreover, early intervention with fluid resuscitation was
increased risk for postoperative AKI, and use of known shown to prevent the progression from prerenal azotemia
preventive therapies in the perioperative and postopera- to established AKI in some instances.
tive periods, are of great importance, considering the high Although a decrease in RBF with diminished oxygen
prevalence of AKI and its detrimental effect on morbidity and substrate delivery to the tubule cells is an important
and mortality among this group of patients. ischemic factor, the relative increase in oxygen demand by
the tubules also plays a role in renal ischemia.30 Although
the kidney receives 20% to 25% of cardiac output, most of
MORTALITY that flow is directed to the renal cortex, with only a small
fraction reaching the vasa recta of the renal medulla.23
The mortality rate for patients with postoperative AKI is This area of low blood flow influences the concentration
reported between 14% and 19%.7,10,18 Patients requiring of urine but, ironically, places the medulla at risk of
RRT have a mortality rate as high as 28% to 69%.7,10,19 The further ischemic damage. Unlike the renal cortex that has
odds ratio for death in the first 30 days postoperatively is a partial pressure of O2 of approximately 50 mm Hg, the
446 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Normal epithelium
Ischemia-
Integrin
reperfusion
Na/K-ATPase
Loss of brush border
and cell polarity
Pre
ren Normal
al Differentiation and epithelium
re establishment
of polarity
Necrosis and apoptosis
Initia
Dedifferentiation
Ex of viable cells
ten very
sio
n Reco
GFR
Maintenance
Days
Onset
39 mo to
Oliguric/anuric Early Late reach normal
Renal blood flow
FIGURE 31.1 Cellular events, renal blood flow and GFR in different phases of ischemic acute
kidney injury. GFR, glomerular filtration rate. (Adapted from the following references: Cochrane
Injuries Group Albumin Reviewers: human albumin administration in critically ill patients:
systematic review of randomized controlled trials. Brit Med J 1998;317:235; Choi PT, Yip G,
Quinonez LG, et al. Crystalloids vs. colloids in fluid resuscitation: a systematic review. Crit Care Med
1999;27:200; Schortgen F, Lacherade JC, Bruneel F, et al. Effects of hydroxyethylstarch and gelatin
on renal function in severe sepsis: a multicentre randomized study. Lancet 2001;357:911.)
outer medulla has a partial pressure of oxygen in the range end-organ perfusion, including that of the kidneys, are
of 10 to 20 mm Hg.31 Perturbations in medullary blood inadequate.32
flow and oxygen delivery can, therefore, lead to anoxic
injury if they exceed the critical thresholds when they Initiation Phase
are not reversed by timely therapy. Cell injury and ATN
can result.23 The precise threshold for this injury and its After the initial insults, parenchymal injury develops,
clinical measurement are unclear. Persistent inadequate and kidney function may deteriorate. Experimental and
blood flow to the kidneys and progression to AKI following human studies indicate that tubular epithelial cells can
successful resuscitation and restored global hemodynamic suffer one of three distinct fates after ischemic AKI: Cell
parameters in severe sepsis and septic shock clearly death (apoptosis), sublethal cell injury, or escape from
indicate that the current means for measurements of injury.26
C H A P T E R 31/ A C U T E P E R I O P E R AT I V E K I D N E Y I N J U R Y 447
Apoptosis occurring in both distal and proximal such as the induction of heat shock protein (HSP) in
tubular cells is emerging as the major mechanism of tubular cells, are activated early and in parallel with
early and late cell death in ischemic and nephrotoxic mechanisms of renal injury. If they prevail, the injury
forms of AKI. Necrosis of the tubular cells is restricted to will be diminished at this stage, with complete recovery
the highly susceptible, outer medullary regions following to follow.43,44
severe renal injury.33 However, most cells remain viable,
either entirely escaping injury or undergoing sublethal Extension Phase
injury and subsequent recovery. Severe reduction in RBF
induces characteristic, rapidly occurring, and duration- Extension of microvascular endothelial injury resulting in
dependent effects in numerous renal cell types, although the perpetuation of hypoxia following the initial ischemic
these effects are seen most prominently in proximal event and exaggeration of inflammatory cascade are the
tubular cells.3436 hallmarks of this stage. Regional alterations in RBF per-
In sublethal tubule cellular injury, alterations in sist after the initial ischemic event and are of greater
the apical cytoskeleton result in loss of the brush magnitude in the outer medulla than in the outer cortex
border and cell polarity, and produce disruption of tight or inner medulla. Medullary blood flow remains severely
and adherens junctions.26,34 Loss of basolateral Na+ /K+ reduced to levels ranging from 10% to 50% of normal in
ATPase impairs proximal tubular sodium reabsorption, both human28 and animal models of ischemic AKI.45 The
with a consequent increase in the fractional excretion mechanisms underlying this persistent alteration of renal
of sodium.37 Redistribution of integrins to the apical perfusion following ischemic injury are complex and not
membrane results in detachment of viable cells from the completely understood. An imbalance between the medi-
basement membrane and promotes tubular obstruction.26 ators of renal vasoconstriction and renal vasodilatation,
Loss of surface membrane, tubular obstruction resulting as well as congestion of the renal microcirculation sec-
from sloughing of viable and dead tubular cells into ondary to endothelial injury and dysfunction, are among
the lumen, and cast formation decrease the glomerular the major proposed mechanisms.27,29
filtration rate (GFR).30,38 Obstruction of the tubular lumen Endothelial cells undergo an array of complex
eliminates glomerular filtration within that nephron by changes that alters their function and interaction with
increasing intratubular pressures to levels inconsistent multiple other cell types. Endothelial cell swelling, al-
with filtration.34 tered cell-to-cell attachment, and reduced endothelial cell
Unfortunately, human studies using forced diuresis basement membrane attachment increase permeability
with furosemide or mannitol fail to demonstrate an impact and interstitial edema.29,34 Thrombin activation and gen-
on the survival and renal recovery in patients with AKI, eration lead to an activated coagulation cascade, and
indicating that obstruction alone accounts only for minor the amplification of inflammatory cascade includes white
dysfunction.39,40 Tubular backleak of glomerular filtrate blood cell activation and cytokine generation.27 Altered
into the circulation results from denudation of the tubular vascular reactivity, produced by reactive, oxygen species-
basement membrane, intratubular obstruction, loss of cell mediated decreases in NO-mediated vasodilatation and
polarity, and disruption of tight and adherens junctions, endothelin-mediated vasoconstriction, potentiate the in-
further contributing to the decline of GFR in AKI.30 tense vasoconstriction and contribute to abnormal blood
Several important mechanisms underlie changes in flow.29 These changes are most prominent in the peritubu-
the tubular cell structure. A profound reduction in in- lar capillaries of the corticomedullary junction and outer
tracellular adenosine triphosphate (ATP) content occurs medulla, and further exacerbate medullary congestion,
early after ischemic renal injury, leading to several criti- hypoperfusion, and injury during the extension phase.27
cal metabolic consequences. A rapid degradation of ATP More widespread tubular cell death, desquamation,
results in the accumulation of adenine nucleotides and and luminal obstruction continue in the medulla. Injured
hypoxanthine, contributing to the generation of reactive endothelial and epithelial cells amplify the inflammatory
oxygen molecules that cause renal tubule cell injury by ox- cascade and produce multiple inflammatory cytokines and
idation of proteins, peroxidation of lipids, damage to DNA, chemokines. Endothelial activation enhances endothelial-
and induction of apoptosis.26 A rise in free intracellular leukocyte interactions and leads to leukocyte infiltration,
calcium causes protease and phospholipase activation, as well as further inflammatory and cytotoxic injury.27
with resultant cytoskeletal degradation.30 Ischemic acti- In a recent study, CD4+ T cells, working through both
vation of nitric oxide (NO) synthase and NO generation in interferon- (IFN- ) and costimulatory molecules, were
tubule cells further promotes cellular damage through ox- implicated as an important modulator of AKI.46 Com-
idant injury, as well as protein nitrosylation.41 Even with plement activation may also contribute to neutrophil
resolution of the initial ischemic insult, a steady decline recruitment and tissue damage.47 This activated inflam-
in RBF continues, caused by hemodynamic and vascu- matory cascade exacerbates hypoperfusion in the renal
lar abnormalities that are incompletely understood.28,42 microcirculation and promotes further release of cyto-
Tubuloglomerular feedback may be partly responsible for toxic inflammatory cytokines and reactive oxygen species.
the renal vasoconstriction observed in AKI, although its
role is not completely clear.26,42 Maintenance Phase
A complex cascade of inflammatory mechanisms is
initiated in this phase and will become fully evident if the During this phase, GFR is maintained at its nadir
injury is not alleviated. Intrarenal protective mechanisms, while parenchymal injury is established. RBF begins
448 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
to normalize, although it may remain slightly below intervention early in the course of AKI. Perhaps more im-
normal 12 to 20 weeks after the onset of ATN in spite portantly will be the development of strategies to identify
of clinical renal recovery.28 Tubular cellular injury with AKI susceptibility among individuals exposed to known
ongoing apoptosis coexists with regeneration during the ischemic and nephrotoxic injuries and to prevent injury
maintenance phase, the duration and severity of which altogether.
may be determined by the balance between cell survival
and death.44 Repair of both epithelial and endothelial cells
is critical to overall recovery, and measures to accelerate
the endogenous regeneration processes may be effective How Is Acute Kidney Injury
during this phase.
Although resident stem cells with tubulogenic capac- Clinically Assessed and
ity have been described in the kidneys after ischemic Diagnosed?
AKI,48 proliferation of the native tubule epithelial cells
seem to represent the primary healing source for re-
generation of the tubule epithelium.49,50 The surviving
renal tubule cells recapitulate phases and processes that GENERAL CONSIDERATIONS
are very similar to those during normal kidney devel-
opment.44,51 Viable cells dedifferentiate, proliferate, and Until approximately 60% of the renal mass is lost,
migrate across the basement membrane to reestablish patients are in a stage of decreased renal reserve. They
epithelial continuity. This sequence is followed by the generally have no signs or symptoms of renal dysfunction
process of redifferentiation and reestablishment of normal and present no specific problems for intensivists or
epithelial polarity and transport functions.44,51 Although anesthesiologists. The clinical diagnosis of AKI can be
the beneficial effect of bone marrowderived, mesenchy- described as a sudden deterioration in renal function
mal stem cells in ATN recovery has been demonstrated, with concomitant electrolyte, acidbase, and fluid balance
most of the evidence suggests that this protective effect is disturbance. Two issues are important in the clinical
mediated through powerful anti-inflammatory and an- approach to the patient with AKI:
tiapoptotic mechanisms, with decreased renal tubular
1. Treating any life-threatening features and
injury, rather than by differentiation.52,53
2. Identifying any cause of AKI that warrants specific
treatment
Recovery Phase
Any patient with low UO should be assessed for
Improvement in GFR and reestablishment of tubular possible prerenal causes of AKI. UO is very specific but far
integrity, with full differentiation and polarization of re- less sensitive to define acute renal failure (ARF). Severe
generated epithelial cells, are observed in this phase. The ARF can exist despite normal UO (i.e., nonoliguric),
mechanisms whereby most of the tubular cells escape but changes in UO can occur long before biochemical
cell death and recover completely after ischemic AKI are changes are apparent. Prerenal forms of ARF nearly
still under investigation. Induction of HSP is a highly always present with oliguria (<400 mL per day), although
conserved, innate cellular response and is found to be ac- very rarely nonoliguric forms are possible. Hypotension,
tivated after ischemic AKI, likely promoting cell survival shock, and respiratory failure should immediately be
by inhibiting apoptosis.43 The importance of inducing pro- assessed in the initial approach and treated if present.
tein involved in cell cycle control has recently emerged.54 Hyperkalemia is another feature that requires urgent
Engagement of the cell cycle is an important determinant treatment. Dehydration secondary to gastrointestinal
of whether cells survive the injury itself. Coordinated cell losses, pneumonia, bowel obstruction, and new onset
cycle control, initially manifested as cell cycle inhibition, is impairment of functional capacity in the elderly patient
necessary for optimum recovery from ATN. Hence, the im- are often the initial diagnoses. A definitive diagnosis of
portance of cell cycle inhibitors, p21 and 14-3-3r, induced AKI is made only when laboratory parameters become
by cellular stress from the initial ischemic or nephrotoxic available. The clinician should then ask a series of
insult, and their combined activities to check the cell cycle questions: Is this AKI? Are there any clues from the
at G1 and G2, can be appreciated to coordinate the cell history? Is the etiology prerenal? Could this be an
cycle and permit cell survival.54 obstructive process? Is intrinsic renal disease probable?
In conclusion, after the initial functional recovery What will the initial evaluation reveal?
from an episode of clinical AKI, a significant proportion of Anemia, hypocalcemia, and hyperphosphatemia are
patients exhibit persistent or progressive deterioration in not good indicators of renal failure chronicity. Renal ultra-
renal function. Animal studies, and some early morpho- sonography may reveal small kidneys, often with cortical
logic studies in human ATN, indicate that postischemic scarring and possible cyst formation, in chronic renal
kidneys demonstrate various sequelae of acute injury, in- disease. Normal sized kidneys, however, do not necessar-
cluding reduction in renal microvasculature, interstitial ily rule out current renal disease. A targeted history will
fibrosis, tubular hypercellularity and atrophy, and per- often be necessary. The history should include queries
sistent inflammation.28,5558 Future research efforts will concerning a recent throat infection (possible poststrep-
be directed toward a better understanding of the mech- tococcal glomerulonephritis). Drug history should include
anisms described and the development of a means for current and recent medications as well as over-the-counter
C H A P T E R 31/ A C U T E P E R I O P E R AT I V E K I D N E Y I N J U R Y 449
TABLE 31.2 Urine Analysis Differences between Prerenal Acute Kidney Injury and Acute
Tubular Necrosis
or azotemia, or both, are ruled out and CRI is not present. water. Renal vasoconstriction in sepsis seems to be due,
The disorder is increasingly recognized in the context of at least in part, to the ability of the tumor necrosis
multiple organ failure, especially in critically ill patients; factor alpha (TNF-) to release endothelin.76 Tubular
only a minority of cases in ICUs occurs without failure of sodium reabsorption is impaired secondary to tubular
another organ.11,12,64 damage. This aberration results in urine sodium values
ATN after surgery accounts for 20% to 25% of all cases >40 mMol per L and a fractional excretion of filtered
of hospital-acquired AKI, many of them with prerenal sodium >1%. The differences of urine tests in ATN and
causes.6,65 Groups at risk include patients with preex- prerenal azotemia are summarized in Table 31.2.
isting renal impairment, hypertension, cardiac disease,
peripheral vascular disease, diabetes mellitus, jaundice,
and advanced age. Newer forms of postsurgical ATN, such POSTRENAL ACUTE KIDNEY
as that following liver or cardiac transplantation, re-
flect changes in types of surgical interventions.3 Contrast INJURY
mediainduced AKI is increasingly recognized in hospi-
Postrenal AKI occurs in the presence of bilateral obstruc-
talized patients. Acute radiocontrast nephropathy is the
tion distal to the renal parenchyma. The GFR decreases
third, most common cause of ATN in patients admitted
with increased intratubular pressure. The most frequent
to the hospital, and up to 7% need temporary dialysis or
causes of AKI are obstruction of the urinary bladder neck
progress to end-stage renal disease.3 The occurrence of ra-
due to prostate enlargement and ureteral obstruction with
diocontrast nephropathy is associated with an increased
retroperitoneal fibrosis or pelvic tumors. Urinary system
risk of death and leads to an extended hospital stay and,
stones are also commonly related to AKI. The incidence of
consequently, increased health care costs.66
postrenal AKI is 2% to 10%. AKI can also be produced by
Several classes of antibacterial, antifungal, antiviral,
prostatic disease in men, with the examination revealing a
and antineoplastic agents are nephrotoxic. Additionally,
palpable bladder, a hypertrophied prostate, or both. Ure-
there are environmental agents and recreational drugs
thral catheterization yields a significant residual volume
that can cause AKI.6772 Sepsis is one of the most common
in these patients. In women, cervical carcinoma can lead
reasons for ATN in critically ill patients. AKI occurs in
to obstructive AKI.
approximately 19% of patients with moderate sepsis, 23%
Ultrasonography is the best tool for detecting obstruc-
with severe sepsis, and 51% with septic shock when blood
tion and may also yield information about the size of the
cultures are positive.73,74 The combination of AKI and
kidneys. One must keep in mind that hydronephrosis is
sepsis is associated with a mortality of 70%, as compared
rarely missed with ultrasound. Obstruction of the urinary
with a mortality of 45% among patients with AKI alone.
catheter during surgery or while the patient is in the ICU
The hemodynamic hallmark of sepsis is generalized
is a common cause of decreased UO and is relatively easy
arterial vasodilatation with an associated decrease in
to assess as long as one considers this possibility. The
systemic vascular resistance. These changes result in
first step to rule out obstruction is to irrigate the urinary
arterial underfilling, activation of the neurohumoral axis,
bladder catheter with sterile normal saline.
and an increase in cardiac output secondary to a decreased
cardiac afterload. Activation of the sympathetic nervous
system and the reninangiotensinaldosterone axis, the
nonosmotic release of vasopressin, and an increase in How Can Acute Kidney Injury
cardiac output are essential in maintaining the integrity
of the arterial circulation in patients with severe sepsis Be Prevented?
and septic shock. Unfortunately, these hemodynamic
changes can lead to AKI.75 Early in sepsis-related AKI, Although we should attempt to prevent AKI in all patients,
the predominant pathogenetic factor appears to be renal special focus should be utilized for those patients at espe-
vasoconstriction with intact tubular function, as shown cially high risk of developing AKI. At this time, no specific
by the increased reabsorption of tubular sodium and biomarkers define the early phase of AKI, but numerous
C H A P T E R 31/ A C U T E P E R I O P E R AT I V E K I D N E Y I N J U R Y 451
strategies are available for the primary prevention of ARF alone on the incidence of AKI has been studied under
under different clinical conditions. Optimal intravenous different conditions, including CABG, rhabdomyolysis,
hydration, maintenance of normal cardiac output, avoid- and vascular and biliary tract surgery. A reduction in the
ance of nephrotoxic agents, and following serum levels incidence of AKI with mannitol as compared to hydration
of drugs such as tacrolimus, cyclosporine aminoglyco- was not observed.8689 Mannitol is not associated with
sides, and vancomycin all seem to be important measures any risk reduction compared with saline alone for the
to prevent AKI.77 These measures are of special signif- prevention of contrast-induced AKI; moreover, it may
icance when RBF may already be compromised, as in even be harmful.90
elderly patients or those with heart failure, liver disease,
various renal infections, renal arterial stenosis, sepsis, or Sodium Bicarbonate
septic shock. Most antihypertensive drugs, including ACE
inhibitors and angiotensin receptor blockers, as well as In high-risk patients, a sodium bicarbonate infusion be-
nonsteroidal anti-inflammatory agents, impair RBF au- fore contrast media exposure can decrease the develop-
toregulation. These agents should be used with caution in ment of AKI. The recommended protocol is 154 mEq
the setting of preexisting renal dysfunction.43 per L of sodium bicarbonate in free water, as a bolus
of 3 mL/kg/hour for 1 hour, before the intravenous con-
trast agent, followed by an infusion of 1 mL/kg/hour for
INTRAVENOUS HYDRATION 6 hours after the procedure. In a randomized controlled
study, hydration with sodium bicarbonate before contrast
Optimal fluid resuscitation is the mainstay of AKI pre- exposure was more effective than hydration with sodium
vention. Whether one should use crystalloids or colloids chloride for prophylaxis of contrast mediainduced renal
remains somewhat unclear.7880 Although crystalloid failure.91
is generally considered appropriate, hydroxyethylstarch
may carry a higher risk for ARF compared to gelatin.81 In- Loop Diuretics
travenous hydration decreases the risk of AKI in postsurgi-
cal patients, especially that related to intravenous contrast Diuretic agents continue to be used in patients with AKI or
agents. The risk of developing contrast mediainduced for the prevention of AKI despite the lack of evidence sup-
AKI may be decreased by using 0.9% saline for hydration porting any benefit. The rationale for using loop diuretics
compared with 0.45% saline in dextrose. This effect is is related to the assumption that they decrease oxygen
even more pronounced in women, patients with diabetes, consumption in tubular cells by inhibiting transcellular
and patients who received a great volume of contrast sodium transport, thereby preventing or limiting ischemic
agent.80 A large, randomized, controlled study showed cell injury; loop diuretics may also vasodilate the cortical
that fluid resuscitation with saline or albumin has sim- vascular network. The increased tubular flow may reduce
ilar results in critically ill patients.82 Although it is well intratubular obstruction and backleak of filtrate, thereby
known that in traumatic rhabdomyolysis, early and ag- protecting the kidneys from AKI.
gressive fluid resuscitation has dramatic benefits,83 overly However, no evidence of improved survival with
aggressive resuscitation with intravenous fluids can lead diuretics in patients at risk for AKI and no decreased
to pulmonary edema, especially in patients with decreased incidence of AKI or need for dialysis has been associated
UO and diminished renal function. with diuretic use.92 Loop diuretics may convert oliguric
Optimization of hemodynamic status, which leads to AKI to its nonoliguric sister-state. Although the mortality
normal RBF, is one of the most essential measures for of nonoliguric AKI is lower than oliguric AKI,93,94 even
preventing AKI in sepsis. Unfortunately, the endpoints when diuretics converted oliguric AKI to a nonoliguric
of resuscitation under these conditions are not clearly state, no benefit on the severity of AKI or mortality
defined. Whereas supranormal cardiac output levels and was noted.93,94 Diuretic use that converts oliguric to
normal values for mixed venous oxygen saturation had nonoliguric AKI may delay not only the recognition of
no effect on mortality or on the incidence and severity of AKI, but subsequent consultation by a nephrologist and
AKI,84 early resuscitation to keep central venous oxygen the initiation of RRT.
saturation 70% or higher resulted in decreased mortality Diuretic use can possibly increase the risk of death
and less severe organ dysfunction in patients with severe or nonrecovery of renal function.95 Although creatinine
sepsis and in septic shock.85 The rates of fluid adminis- clearance decreases postoperatively after vascular and
tration have not been compared, but most of the studies thoracoabdominal operations, furosemide does not im-
reported fluids at a rate of 1 mL per kg for 6 to 12 hours prove creatinine clearance in these groups of patients.96
before and after using an intravenous contrast agent.77 Diuretics seem to worsen outcomes in ATN induced by
contrast media.90 In patients with CRI who underwent
cardiac angiography, furosemide increased the risk for
DRUG THERAPY AKI.97 A recent, large, prospective multicenter study noted
that diuretics are commonly prescribed in critically ill pa-
Mannitol tients with AKI, and their use is not associated with a high
mortality rate; however, they have no beneficial effect in
The benefits of mannitol for the prevention of AKI are patients with AKI.98 In the absence of compelling contra-
controversial. The effect of mannitol versus hydration dictory data from randomized, blinded clinical trials, the
452 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
widespread use of diuretics in critically ill patients with vasoconstriction, vasopressor use is of concern, as it
ARF is to be discouraged. may cause further vasoconstriction. In healthy humans,
norepinephrine decreases RBF;119 however, its effect on
Dopamine RBF depends upon a complex interaction on different
vascular beds and on the underlying patient condition.
Dopamine provides no benefit to AKI patients.99 Low-dose Norepinephrine increases blood pressure by stimulating
dopamine (1 to 3 g/kg/minute) has a renal vasodilator 1 receptors and augments cardiac output by stimulating
effect and increases RBF, diuresis, and natriuresis in an- 1 receptors. An excessive rise in systemic vascular resis-
imals and humans.100,101 Prospective, controlled studies tance can have a negative impact on cardiac output due
and meta-analysis revealed that dopamine does not re- to increased afterload. The net effect on renal vascular
duce mortality or promote renal recovery in patients with resistance is:119
AKI.97,99,100,102 Increase in systemic blood pressure with decreased
Dopamine induces diuresis without improving the renal sympathetic tone through a baroreceptor response
creatinine clearance.103 It does not decrease the require- that results in vasodilatation
ment for RRT and ICU care, nor does it decrease the length Autoregulatory vasoconstriction, owing to a rise in renal
of hospital stay or mortality.48 It can depress the respira- perfusion pressure
tory drive and trigger tachyarrhythmias, thereby leading Direct 1 -mediated renal vasoconstriction, which is of
to myocardial ischemia.100,101 In addition, dopamine can minor importance
induce intestinal ischemia secondary to shunting of blood
away from the bowel mucosa.104 The drug may also sup- Therefore, in a patient with sepsis who exhibits
press the release of all pituitary hormones except for systemic vasodilatation and impaired renal autoregula-
cortisol,105 which may be of significance in surgical pa- tion, norepinephrine can be expected to increase RBF if
tients. adequate volume resuscitation has been administered. In-
deed, it has been shown that norepinephrine increases UO
and GFR.120
N-Acetylcysteine
Vasopressin can be used safely to restore blood pres-
N-acetylcysteine scavenges reactive oxygen metabolites sure without compromising renal function in patients
and inhibits the synthesis of some proteins and cy- with septic shock. In general, norepinephrine appears
tokines, which may be harmful to the kidney.106 It superior to dopamine; in patients refractory to nore-
affects the tubular handling of creatinine without chang- pinephrine, early use of vasopressin is recommended.120
ing the GFR.43 Together with optimal volume hydration,
it is superior to hydration alone in patients at high Other Pharmacologic, Hormonal, and
risk for contrast-induced AKI.43,107,108 In several meta- Coagulation Factors in Acute Kidney Injury
analyses, N-acetylcysteine decreased the risk of contrast-
induced AKI by 50% in this group of patients,109112 Tumor Necrosis Factor-
although it did not show improved survival or a de-
TNF- is a cytokine that plays a role in the host response
creased need for RRT. N-acetylcysteine may decrease
to infection; it may also have specific renal effects.
the incidence of radiocontrast-induced nephropathy in
TNF- infusion decreases the GFR in animals.121,122 In
high-risk patients.113115 Although the risk of utilizing
animal models, anti-TNF antibodies are protective against
N-acetylcysteine seems to be minimal, it should be in ad-
the morbidity and mortality from sepsis.123 Passive
dition to hydration rather than in its place. Furthermore,
immunization to TNF- prevents renal cortical damage
isoosmolar radiocontrast agents, which have less risk for
during endotoxemia in an animal model.124 The use
contrast-induced AKI, should be used if possible.
of TNF-soluble receptor to neutralize TNF- protects
mice against lipopolysaccharide-induced renal failure.125
Calcium Channel Blockers Unfortunately, anti-TNF- antibodies and soluble TNF
receptor fusion proteins did not show any decreased
Calcium channel blockers used for renal protection in
mortality in patients with sepsis.123 Although TNF-
post-transplant patients with ATN are controversial, al-
indirectly affects the kidneys by inducing hypotension and
though one study did suggest that the prophylactic ad-
releasing inflammatory mediators into the circulation,
ministration of calcium-channel blockers may protect
several lines of evidence point to direct TNF-mediated
against delayed graft failure in patients who underwent
renal damage in sepsis.123 Despite the success of anti-TNF
renal transplantation.116 Calcium-channel blockers ad-
therapies in animal models, the beneficial effects of these
ministered after cardiac surgery for the prevention of
strategies in humans are marginal.
ARF seem to be promising, but more studies are required
before these agents can be used on a routine basis.117,118
Platelet-Activating Factor
Vasopressors The platelet-activating factor (PAF) has vasoactive,
platelet-aggregating, and proinflammatory properties.
If blood pressure remains inappropriately low after Serum and urinary concentrations of PAF are elevated in
adequate volume resuscitation, vasopressors are indi- patients with sepsis, and the levels correlate with the sever-
cated. In septic patients who already have intrarenal ity of AKI.126 The intrarenal infusion of PAF in rats results
C H A P T E R 31/ A C U T E P E R I O P E R AT I V E K I D N E Y I N J U R Y 453
in renal vasoconstriction and a decrease in GFR.127,128 The recruitment of circulating leukocytes into a tis-
Although experimental studies are encouraging, there is sue is directed by specific adhesive interactions between
no firm conclusion on the value of PAF antagonism in the leukocyte and the vascular endothelium. Renal inter-
septic AKI. cellular adhesion molecule-1 messenger RNA is increased
after ischemia-reperfusion in the mouse.138 Leukocytes
infiltrate the kidneys during sepsis, which results in renal
Nitric Oxide
dysfunction. At this time, however, there are no clinical
NO is the metabolic product of L-arginine and is produced studies being conducted with this agent.
by three major NO synthases: Endothelial NOS, neuronal
NOS, and inducible NOS. Endothelial NOS is expressed
constitutively in renal endothelial cells and plays a Coagulation Factors
significant role in vascular relaxation and inhibition of Disseminated intravascular coagulation is a frequent com-
leukocyte adhesion and platelet aggregation. Sublethal plication of sepsis and is associated with a poor prognosis.
endothelial injury and endothelial dysfunction, resulting It is characterized by the increased activation of the coag-
in an impaired release of NO produced by endothelial ulation cascade and increased intravascular formation of
NOS, have been described in ischemic kidneys.129,130 fibrin clots and endothelial damage. Impaired tissue blood
Nonselective NOS inhibition increases blood pressure supply contributes to organ dysfunction, including AKI.
and systemic vascular resistance and decreased cardiac The tissue factor forms a complex with factor VIIa, which
output in patients with septic shock.131133 However, no activates factor IX and X, initiating thrombin generation.
beneficial effects on renal function have been reported. The specific blockade of tissue factorfactor VIIa complex
can prevent renal injury in septic animals, but a benefit of
Endothelin-1 the tissue factor pathway inhibitors could not be shown
in patients with severe sepsis or septic shock.139
Endothelin-1 is a peptide with potent vasoconstrictor ef-
A double-blind, placebo-controlled, large random-
fects on the renal microcirculation, thereby reducing RBF
ized study revealed that antithrombin had no effect
and GFR. Infusion of endotoxins greatly increases plasma
on mortality and was associated with an increased
endothelin-1 levels in animal models. Endothelin-1 lev-
risk of hemorrhage.140 Protein C is activated by the
els are markedly elevated and correlate with morbidity
thrombinthrombomodulin complex on endothelial cells.
and mortality in sepsis.134 Nonselective endothelin antag-
Activated protein C inhibits thrombin generation by
onism increases the risk of contrast-induced nephropathy
inactivating factor Va and factor VIIIa. Besides its ef-
in patients with chronic renal failure undergoing coro-
fects on coagulation, activated protein C has direct
nary angiography. This negative effect may be related to
anti-inflammatory properties, including impairment of
an intrarenal steal phenomenon, because of a predomi-
leukocyte adhesion to the endothelium by binding selec-
nant increase of the cortical blood flow with a worsening
tions and inhibition of the production of inflammatory
of medullary ischemia.135 Further studies are required for
cytokines by monocytes. It stimulates the fibrinolytic
the endothelin-1 with endothelin antagonism.
response by inhibiting plasminogen-activator inhibitor
type1.
Prostaglandins, Thromboxanes, and Leukotrienes Reduced levels of protein C are found in patients
The metabolism of arachidonic acid by cyclooxygenase with sepsis and are associated with increased mortality.
results in the generation of prostaglandins (PGs) and In a randomized, placebo-controlled trial in patients with
thromboxanes, whereas lipoxygenase yields leukotrienes. severe sepsis, activated protein C significantly reduced
Both PGE2 and PGI2 induce renal vasodilatation and na- mortality,141 as well as the number of organ failures. For
triuresis, whereas thromboxane A2 , leukotrienes, and PG these reasons, including the decreased incidence of ARF,
F2 and PG H2 are potent renal vasoconstrictors. In patients activated protein C is recommended in these patients.
with sepsis, cyclooxygenase inhibition with ibuprofen re-
duced the synthesis of thromboxane and prostacyclin, but Growth Factors
had no effect on the development of AKI.136
Growth factors are required for recovery from AKI and
play a very important role in the regeneration and
Atrial Natriuretic Peptide restoration of the tubular epithelium. Epidermal growth
Atrial natriuretic peptide, a hormone produced in the factor and insulin-like growth factor I (IGF-I) are known
cardiac atria, increases GFR through vasodilatation of the mitogens for tubular epithelial cells. The expression of
afferent arterioles and vasoconstriction of the efferent epidermal growth factor and IGF-I and their receptors
arterioles. It inhibits the reabsorption of sodium and is increased after experimental renal injury. Infusion
redistributes renal medullary blood flow, thereby resulting of these growth factors in AKI may accelerate renal
in an improved supply and reduced demand of oxygen in epithelial regeneration and speed the time to recovery
the tubules. In patients with AKI after cardiac surgery, of renal function. In a multicenter, randomized, placebo-
atrial natriuretic peptide infusion improved RBF and controlled trial, IGF-I did not reduce the time to renal
GFR.137 However, there is no convincing evidence to recovery or mortality rates in 72 critically ill patients
support the use of natriuretic peptides for the treatment with AKI.142 In patients undergoing suprarenal aortic or
or prevention of AKI. renal artery surgery, no differences in serum creatinine
454 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
at time of discharge, length of ICU and hospital stay, or GFR criteria Urine output criteria
incidence of dialysis and mortality were observed between
IGF-I- and placebo-treated patients.143 Growth factors so
Increased SCreat 1.5 or UO < 0.5 mL/kg/h
far have not shown any beneficial effect on AKI. Risk
GFR decrease >25% 6h
High
Increased SCreat 2 or UO < 0.5 mL/kg/h sensitivity
Injury
PERIOPERATIVE ACUTE GFR decrease >50% 12 h
a
GFR decrease >75% 24 h or
Oliguri
The incidence of perioperative renal failure changes Failure
OR SCreat 4 mg/dL Anuria 12 h
according to the surgical procedure and the definition Acute rise 5 mg/dL
of AKI. AKI is still a very important complication that High
increases mortality and morbidity rates despite significant specificity
advances in perioperative and intraoperative care over the Loss
Persistent ARF = complete loss
last decade. Mortality, ICU, and hospital length of stay of kidney function >4 wk
were significantly increased in patients with AKI (both End stage kidney disease
ESKD
renal failure and renal dysfunction) compared with those (>3 mo)
who did not suffer these complications. Patients with renal
dysfunction were three times more likely to be discharged FIGURE 31.2 The RIFLE criteria for acute renal injury.
to an extended care facility compared with those without Proposed classification scheme for acute renal failure (ARF). The
renal dysfunction. classification system includes separate criteria for creatinine and
urine output (UO). A patient can fulfill the criteria through
Causes changes in serum creatinine (SCreat) or changes in UO, or both.
The criteria that lead to the worst possible classification should
Renal dysfunction in the surgical patient is usually be used. GFR, glomerular filtration rate; OR, operating room;
multifactorial. The most common cause is ATN resulting ARF, acute renal failure. (From: Bellomo R, Ronco C, Kellum JA,
from hypoxic nephron damage in the medullary region et al, and the ADQI Work Group: Acute renal failure-definition,
of the kidneys secondary to hypotension, hypovolemia, outcome measures, animal models, fluid therapy and
and/or dehydration. Prerenal azotemia also is a common information technology needs: the Second International
clinical problem predisposing to AKI. The common Consensus Coference of the Acute Dialysis Quality Initiative
risk factors for postoperative AKI include preexisting [ADQI] Group. Crit Care 2004;8: R204.)
renal insufficiency, type 1 diabetes mellitus, patient age
more than 65 years, major vascular surgery (including
coronary artery bypass and major aortic procedures),
69% for those individuals who required RRT.7,19 Patients
CPB longer than 3 hours, and recent exposure to
suffering from chronic ischemic states (i.e., renal artery
nephrotoxic agents such as radio-contrast dye, bile
stenosis, volume depletion, diabetes mellitus, or a recent
pigments, aminoglycoside antibiotics, and NSAIDs.15
acute ischemic event such as hemorrhage or exposure to
Patients with severe arteriosclerosis have reduced renal
radiocontrast agents) are more likely to suffer AKI when
perfusion, and an age-related decline in nephron mass also
subsequently exposed to an intraoperative ischemic insult.
is a risk factor for AKI.10,144 Finally, embolized material
Unfortunately, the patient population undergoing cardiac
following the release of an aortic clamp may lead to
or aortic surgery is characterized by precisely these risk
AKI, as may gut ischemia and impaired visceral perfusion
factors.
occurring during abdominal aortic aneurysm surgery.
Endotoxemia can activate other vasoactive compounds
that may then lead to AKI.145
efferent arteriolar vasoconstriction, responsible for the renin activity is also elevated as a result of a decrease
development of glomerular filtration pressure, is also nec- in circulating blood volume. Perioperative fluid losses
essary. The patient receiving chronic ACE inhibitors or an- include blood and intravascular fluid losses, insensible
giotensin II receptor antagonist treatment and undergoing losses, and losses secondary to third spacing, all of which
anesthesia and surgery may develop a significant decrease must be properly measured to the fullest extent possible,
in perfusion pressure, with decreased urine production. and recorded and replaced.
4. de Mendonca A, Vincent JL, Suter PM, et al. Acute renal 26. Devarajan P. Update on mechanisms of ischemic acute
failure in the ICU: Risk factors and outcome evaluated by kidney injury. J Am Soc Nephrol. 2006;l17:1503.
the sofa score. Intensive Care Med. 2000;26:915. 27. Molitoris BA, Sutton TA. Endothelial injury and dysfunc-
5. Clermont G, Acker CG, Angus DC, et al Renal failure in tion: Role in the extension phase of acute renal failure.
the UCU: Comparison of the impact of acute renal failure Kidney Int. 2004;66:496.
and end-stage renal disease on ICU outcomes. Kidney Int. 28. Bull GJ, Joekes AM, Lowe KG. Acute tubular necrosis of
2002;62:986. the kidney following abortion. Lancet. 1956;28:186.
6. Carmichael P, Carmichael AR. Acute renal failure in the 29. Sutton TA, Fisher CJ, Molitoris BA. Microvascular endothe-
surgical setting. Aust N Z J Med. 2003;73:144. lial injury and dysfunction during ischemic acute renal
7. Mangano CM, Diamondstone LS, Ramsay JG, et al. Renal failure. Kidney Int. 2002;62:1539.
dysfunction after myocardial revascularization: Risk fac- 30. Schrier RW, Wang W, Poole B, et al. Acute renal failure:
tors, adverse outcomes, and hospital resource utilization. Definitions, diagnosis, pathogenesis, and therapy. J Clin
Ann Intern Med. 1998;128:194. Invest. 2004;114:5.
8. Sadovnikoff N. Perioperative acute renal failure. Int Anes- 31. Brezis M, Rosen S. Hypoxia of the renal medullaits
thesiol Clin. 2001;39:95. implications for disease. N Engl J Med. 1995;332:647.
9. Tuttle KR, Worrall NK, Dahlstrom LR, et al. Predictors of 32. Sakr Y, Dubois MJ, De Backer D, et al. Persistent
ARF after cardiac surgical procedures. Am J Kidney Dis. microcirculatory alterations are associated with organ
2003;41:76. failure and death in patients with septic shock. Crit Care
10. Conlon P, Stafford-Smith M, White W, et al. Acute renal Med. 2004;32:1825.
failure following cardiac surgery. Nephrol Dial Transplant. 33. Bonegio R, Lieberthal W. Role of apoptosis in the patho-
1999;14:1158. genesis of acute renal failure. Curr Opin Nephrol Hypertens.
11. Liano F, Junco E, Pascual J, et al. The Madrid Acute Renal 2002;11:301.
Failure Study Group. The spectrum of acute renal failure 34. Molitoris BA. Actin cytoskeleton in ischemic acute renal
in the intensive care unit compared with that seen in other failure. Kidney Int. 2004;66:871.
settings. The Madrid Acute Renal Failure Study Group. 35. Kwon O, Phillips CL, Molitoris BA. Ischemia induces
Kidney Int Suppl. 1998;66:S16. alterations in actin filaments in renal vascular smooth
12. Cole L, Bellomo R, Silvester W, et al. A prospective, muscle cells. Am J Physiol Renal Physiol. 2002;282:F1012.
multicenter study of the epidemiology, management, and 36. Kuhne W, Besselmann M, Noll T, et al. Disintegration
outcome of severe acute renal failure in a closed ICU of cytoskeletal structure of actin filaments in energy-
system. Am J Respir Crit Care Med. 2000;162:191. depleted endothelial cells. Am J Physiol Heart Circ Physiol.
13. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure 1993;264:H1599.
in critically ill patients: A multinational, multicenter study. 37. Woroniecki R, Ferdinand JR, Morrow JS, et al. Dissociation
JAMA. 2005;294:813. of spectrin-ankyrin complex as a basis for loss of Na-K-
14. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of ATPase polarity after ischemia. Am J Physiol Renal Physiol.
acute renal failure in the intensive care unit: The Picard 2003;284:F358.
experience. Kidney Int. 2004;66:1613. 38. Bonventre JV, Weinberg JM. Recent advances in the
15. Novis BK, Roizen MF, Aronson S, et al. Association of pathophysiology of ischemic acute renal failure. J Am Soc
preoperative risk factors with postoperative acute renal Nephrol. 2003;14:2199.
failure. Anesth Analg. 1994;78:143. 39. Ho KM, Sheridan DJ. Meta-analysis of furosemide to
16. Bove T, Calabro MG, Landoni G, et al. The incidence prevent or treat acute renal failure. Br Med J. 2006;333:420.
and risk of acute renal failure after cardiac surgery. 40. Girbes AR. Prevention of acute renal failure: Role of vaso-
J Cardiothorac Vasc Anesth. 2004;18:442. active drugs, mannitol and diuretics. Int J Artif Organs.
17. Antonucci F, Bertolissi M, Calo L. Plasma endothelin and 2004;27:1049.
renal function during infrarenal aortic crossclamping and 41. Goligorsky MS. Whispers and shouts in the pathogenesis
nifedipine infusion. Lancet. 1990;336(8728):1449. of acute renal ischaemia. Nephrol Dial Transplant. 2005;
18. Chertow GM, Soroko SH, Paganini EP, et al. Mortality after 20:261.
acute renal failure: Models for prognostic stratification and 42. Lameire N, Van Biesen W, Vanholder R. Acute renal failure.
risk adjustment. Kidney Int. 2006;70:1120. Lancet. 2005;365:417.
19. Braams R, Vossen V, Lisman BAM, et al. Outcome in 43. Aufricht C. Heat-shock protein 70: Molecular supertool?
patients requiring renal replacement therapy after surgery Pediatr Nephrol. 2005;20:707.
for ruptured and non-ruptured aneurysm of the abdominal 44. Devarajan P, Mishra J, Supavekin S, et al. Gene expression
aorta. Eur J Vasc Endovasc Surg. 1999;18:323. in early ischemic renal injury: Clues towards pathogenesis,
20. Kashyap VS, Cambria RP, Davison JK, et al. Renal biomarker discovery, and novel therapeutics. Mol Genet
failure after thoracoabdominal aortic surgery. J Vasc Surg. Metab. 2003;80:365.
1997;26:949. Discussion 955. 45. Lieberthal W. Biology of ischemic and toxic renal tubular
21. Levy EM, Viscoli CM, Horwitz RI. The effect of acute cell injury: Role of nitric oxide and the inflammatory
renal failure on mortality. A cohort analysis. JAMA. 1996; response. Curr Opin Nephrol Hypertens. 1998;7:289.
275:1489. 46. Friedewald JJ, Rabb H. Inflammatory cells in ischemic
22. Venkataraman R. Prevention of acute renal failure. Crit acute renal failure. Kidney Int. 2004;66:486.
Care Clin. 2005;21:281. 47. Thurman JM, Lucia MS, Ljubanovic D, et al. Acute tubular
23. Blantz RC. Pathophysiology of pre-renal azotemia. Kidney necrosis is characterized by activation of the alternative
Int. 1998;53:512. pathway of complement. Kidney Int. 2005;67:524.
24. Zager RA. Endotoxemia, renal hypoperfusion, and fever: 48. Haller H, de Groot K, Bahlmann F, et al. Stem cells
Interactive risk factors for aminoglycoside and sepsis- and progenitor cells in renal disease. Kidney Int. 2005;68:
associated acute renal failure. Am J Kidney Dis. 1992;20:223. 1932.
25. Bonventre JV, Zuk A. Ischemic acute renal failure: An 49. Maeshima A, Sakurai H, Nigam SK. Adult kidney tubular
inflammatory disease? Kidney Int. 2004;66:480. cell population showing phenotypic plasticity, tubulogenic
458 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
capacity, and integration capability into developing kidney. 70. Evenepoel P. Acute toxic renal failure. Best Pract Res Clin
J Am Soc Nephrol. 2006;17:188. Anaesthesiol. 2004;18:37.
50. Duffield JS, Bonventre JV. Kidney tubular epithelium is 71. Bach PH. Acute renal failure associated with occupational
restored without replacement with bone marrow-derived and environmental settings. In: Molitoris BA, Finn WF, eds.
cells during repair after ischemic injury. Kidney Int. 2005; Acute renal failure: A companion to Brenner and Rectors the
68:1956. kidney, 1st ed. Philadelphia: WB Saunders; 2001:414.
51. Bonventre JV. Dedifferentiation and proliferation of surviv- 72. Turney JH. Acute renal failure associated with recreational
ing epithelial cells in acute renal failure. J Am Soc Nephrol. drug use. In: Molitoris BA, Finn WF, eds. Acute renal failure:
2003;14(Suppl 1):S55. A companion to Brenner and Rectors the kidney. 1st ed.
52. Krause D, Cantley LG. Bone marrow plasticity revisited: Philadelphia: WB Saunders; 2001:397.
Protection or differentiation in the kidney tubule? J Clin 73. Riedemann NC, Guo RF, Ward PA. The enigma of sepsis.
Invest. 2005;115:1705. J Clin Invest. 2003;112:460.
53. Togel F, Hu Z, Weiss K, et al. Administered mesenchymal 74. Rangel-Frausto MS, Pittet D, Costigan M, et al. The natural
stem cells protect against ischemic acute renal failure history of the systemic inflammatory response syndrome
through differentiation-independent mechanisms. Am J (SIRS): A prospective study. JAMA. 1995;273:117.
Physiol Renal Physiol. 2005;289:F31. 75. Schrier RW, Wang W. Acute renal failure and sepsis. N Engl
54. Price PM, Megyesi J, Safirstein RL. Cell cycle regulation: J Med. 2004;351:159.
Repair and regeneration in acute renal failure. Semin 76. Kon V, Badr KF. Biological actions and pathophysiologic
Nephrol. 2003;23:449. significance of endothelin in the kidney. Kidney Int. 1991;
55. Hall JW, Johnson WJ, Maher FT, et al. Immediate and 40:1.
long-term prognosis in acute renal failure. Ann Intern Med. 77. Kellum JA, Leblanc BM, Gibney CN, et al. Primary
1970;73:515. prevention of acute renal failure in the critically ill. Curr
56. Sikorski EM, Hock T, Hill-Kapturczak N, et al. The story so Opin Crit Care. 2005;11:537.
far: Molecular regulation of the heme oxygenase-1 gene in 78. Cochrane Injuries Group Albumin Reviewers. Human al-
renal injury. Am J Physiol Renal Physiol. 2004;286:F425. bumin administration in critically ill patients: System-
57. Basile DP, Donohoe D, Roethe K, et al. Renal ischemic atic review of randomized controlled trials. Br Med J.
injury results in permanent damage to peritubular capillar- 1998;317:235.
ies and influences long-term function. Am J Physiol Renal 79. Schierhout G, Roberts I. Fluid resuscitation with colloid or
Physiol. 2001;281:F887. crystalloid solutions in critically ill patients: A systematic
58. Gueler F, Gwinner W, Schwarz A, et al. Long-term effects review of randomized trials. Br Med J. 1998;316:961.
of acute ischemia and reperfusion injury. Kidney Int. 80. Choi PT, Yip G, Quinonez LG, et al. Crystalloids vs. colloids
2004;66:523. in fluid resuscitation: A systematic review. Crit Care Med.
59. Firth JD. Medical treatment of acute tubular necrosis. Q J 1999;27:200.
Med. 1998;91:321. 81. Schortgen F, Lacherade JC, Bruneel F, et al. Effects
60. Brater DC. Anti-inflammatory agents and renal function. of hydroxyethylstarch and gelatin on renal function in
Semin Arthritis Rheum. 2002;32(Suppl 1):33. severe sepsis: A multicentre randomized study. Lancet.
61. Gambaro G, Perazella MA. Adverse renal effects of anti- 2001;357:911.
inflammatory agents: Evaluation of selective and nonse- 82. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of con-
lective cyclooxygenase inhibitors. J Intern Med. 2003;253: trast media-associated nephropathy: randomized compar-
643. ison of 2 hydration regimens in 1620 patients undergoing
62. Morgan DB, Carver ME, Payne RB. Plasma creatinine and coronary angioplasty. Arch Intern Med. 2002;162:329.
urea: Creatinine ratio in patients with raised plasma urea. 83. Better OS, Rubinstein I. Management of shock and acute re-
Br Med J. 1977;2:929. nal failure in casualties suffering from the crush syndrome.
63. Patak RV, Lifschitz MD, Stein JH. Acute renal failure: Clin- Ren Fail. 1997;19:647.
ical aspects and pathophysiology. Cardiovasc Med. 1979; 84. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-
4:19. oriented hemodynamic therapy in critically ill patients.
64. Breen D, Bihari D. Acute renal failure as a part of multiple SvO2 Collaborative Group. N Engl J Med. 1995;333:
organ failure: the slippery slope of critical illness. Kidney 1025.
Int Suppl. 1998;66:S25. 85. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed
65. Deman A, Hoste E, Van Biesen W, et al. Insuffisance renale therapy in the treatment of severe sepsis and septic shock.
aigue postoperatoire: Epidemiologie, causes, pronostic N Engl J Med. 2001;345:1368.
et traitement. In: Lesavre P, Drueke T, Legendre C, 86. Ip-Yam PC, Murphy S, Baines M. Renal function and
et al. eds. Actualites nephrologiques jean hamburger. Paris: proteinuria after cardiopulmonary bypass: the effects of
Flammarion; 2004:227. temperature and mannitol. Anesth Analg. 1994;78:842.
66. Gruber SJ, Shapiro CJ. Nephropathy induced by contrast 87. Homsi E, Barreiro MF, Orlando JM, et al. Prophylaxis of
medium. N Engl J Med. 2003;348:2257. acute renal failure in patients with rhabdomyolysis. Ren
67. Dillon JJ. Nephrotoxicity from antibacterial, antifungal, and Fail. 1997;19:283.
antiviral drugs. In: Molitoris BA, Finn WF, eds. Acute renal 88. Beall AC, Holman MR, Morris GC. Mannitol-induced
failure: A companion to Brenner and Rectors the kidney, 1st osmotic diuresis during vascular surgery. Arch Surg. 1963;
ed. Philadelphia: WB Saunders; 2001:349364. 86:34.
68. Agraharkar M, Guba SC, Safirstein RL. Acute renal failure 89. Gubern JM, Sancho JJ, Simo J, et al. A randomized trial
associated with cancer chemotherapy. In: Molitoris BA, on the effect of mannitol on postoperative renal function in
Finn WF, eds. Acute renal failure: A companion to Brenner patients with obstructive jaundice. Surgery. 1988;103:39.
and Rectors the kidney. 1st ed. Philadelphia: WB Saunders; 90. Solomon R, Werner C, Mann D, et al. Effects of saline,
2001:365. mannitol, and furosemide to prevent acute decreases in
69. Boccalandro F, Anderson HV. Contrast-induced nephropa- renal function induced by radiocontrast agents. N Engl J
thy: Back to basics. J Invasive Cardiol. 2003;15:317. Med. 1994;331:1416.
C H A P T E R 31/ A C U T E P E R I O P E R AT I V E K I D N E Y I N J U R Y 459
91. Merten GJ, Burgess WP, Gray LV, et al. Prevention of 111. Alonso A, Lau J, Jaber BL, et al. Prevention of radiocontrast
contrast-induced nephropathy with sodium bicarbonate: nephropathy with N-acetylcysteine in patients with chronic
A randomized controlled trial. JAMA. 2004;291:2328. kidney disease: A meta-analysis of randomized, controlled
92. Kellum JA. The use of diuretics and dopamine in acute trials. Am J Kidney Dis. 2004;43:1.
renal failure: A systematic review of the evidence. Crit Care. 112. Pannu N, Manns B, Lee H, et al. Systematic review of the
1997;1:53. impact of N-acetylcysteine on contrast nephropathy. Kidney
93. Shilliday IR, Quinn KJ, Allison ME. Loop diuretics in the Int. 2004;65:1336.
management of acute renal failure: A prospective, double- 113. Brophy DF. Role of N-acetylcysteine in the prevention
blind, placebo-controlled, randomized study. Nephrol Dial of radiocontrast induced nephropathy. Ann Pharmacother.
Transplant. 1997;12:2592. 2002;36:1466.
94. Anderson RJ, Linas SL, Berns AS, et al. Nonoliguric acute 114. Marenzi G, Assanelli E, Marana I, et al. N-acetylcysteine
renal failure. N Engl J Med. 1977;296:1134. and contrast-induced nephropathy in primary angioplasty.
95. Ravindra LM, Maria TP, Sharon S, et al. for PICARD N Engl J Med. 2006;354:2773.
Study Group. Diuretics, mortality and nonrecovery of renal 115. Boccalandro F, Amhad M, Smalling RW, et al. Oral acetyl-
function in acute renal failure. JAMA 2002;288:2547. cysteine does not protect renal function from moderate to
96. Hager B. Effect of postoperative intravenous loop diuretic high doses of intravenous radiographic contrast. Catheter
on renal function after major surgery. Schwiez Med Cardiovasc Interv. 2003;58:336.
Wochenschr. 1996;126:666. 116. Lameire NH, De Vriese AS, Vanholder R. Prevention and
97. Lassnigg A, Donner E, Grubhofer G, et al. Lack of nondialytic treatment of acute renal failure. Curr Opin Crit
renoprotective effects of dopamine and furosemide during Care. 2003;9:481.
cardiac surgery. J Am Soc Nephrol. 2000;11:97. 117. Lumlertgul D, Wongmekiat O, Sirivanichai C, et al.
98. Uchino S, Doig GS, Bellomo R, et al. for The Beginning Intrarenal infusion of gallopamil in acute renal failure:
and Ending Supportive Therapy for the Kidney (B.E.S.T. A preliminary report. Drugs. 1991;42(Suppl 1):44.
Kidney) Investigators. Diuretics and mortality in acute renal 118. Piper SN, Kumle B, Maleck WH, et al. Diltiazem may
failure. Crit Care Med 2004;32:1669. preserve renal tubular integrity after cardiac surgery. Can J
99. Bellomo R, Chapman M, Finfer S, et al. Australian and New Anaesth. 2003;50:285.
Zealand Intensive Care Society (ANZICS) Clinical Trials 119. Schetz M. Vasopressors and the kidney. Blood Purif .
Group. Australian and New Zealand Intensive Care Society 2002;20:243.
Clinical Trials Group. Low-dose dopamine in patients with 120. Krismer AC, Wenzel V, Mayr VD, et al. Arginine vasopressin
early renal dysfunction: A placebo-controlled randomised during cardiopulmonary resuscitation and vasodilatory
trial. Australian and New Zealand Intensive Care Society shock: Current experience and future perspectives. Curr
(ANZICS) Clinical Trials Group. Lancet. 2000;356:2139. Opin Crit Care. 2001;7:157.
100. Denton MD, Chertow GM, Brady HR. Renal-dose 121. Bertani T, Abbate M, Zoja C, et al. Tumor necrosis factor
dopamine for the treatment of acute renal failure: Sci- induces glomerular damage in the rabbit. Am J Pathol.
entific rationale, experimental studies and clinical trials. 1989;134:419.
Kidney Int. 1996;50:4. 122. Van der Veen AH, Seynhaeve AL, Breurs J, et al. In vivo
101. Burton CJ, Tomson CR. Can the use of low-dose dopamine isolated kidney perfusion with tumour necrosis factor
for treatment of acute renal failure be justified? Postgrad alpha (TNF-alpha) in tumour-bearing rats. Br J Cancer.
Med J. 1999;75:269. 1999;79:433.
102. Kellum JA, Decker M. Use of dopamine in acute renal 123. Reinhart K, Karzai W. Anti-tumor necrosis factor therapy
failure: A meta-analysis. Crit Care Med. 2001;29:1526. in sepsis: Update on clinical trials and lessons learned. Crit
103. Duke GJ, Briedis JH, Weaver RA. Renal support in critically Care Med. 2001;29(Suppl):S121.
ill patients: Low-dose dopamine or low-dose dobutamine? 124. Fiedler VB, Loof I, Sander E, et al. Monoclonal antibody to
Crit Care Med. 1994;22:1919. tumor necrosis factoralpha prevents lethal endotoxin sep-
104. Segal JM, Phang PT, Walley KR. Low-dose dopamine sis in adult rhesus monkeys. J Lab Clin Med. 1992;120:574.
hastens onset of gut ischemia in a porcine model of 125. Knotek M, Rogachev B, Wang W, et al. Endotoxemic renal
hemorrhagic shock. J Appl Physiol. 1992;73:1159. failure in mice: Role of tumor necrosis factor independent
105. Van den Berghe G, de Zegher F. Anterior pituitary function of inducible nitric oxide synthase. Kidney Int. 2001;59:
during critical illness and dopamine treatment. Crit Care 2243.
Med. 1996;24:1580. 126. Mariano F, Guida G, Donati D, et al. Production of platelet-
106. Safirstein R, Andrade L, Vieira JM. Acetylcysteine and activating factor in patients with sepsis-associated acute
nephrotoxic effects of radiographic contrast agents-a new renal failure. Nephrol Dial Transplant. 1999;14:1150.
use for an old drug. N Engl J Med. 2000;343:210. 127. Wang J, Dunn MJ. Platelet-activating factor mediates
107. Briguori C, Manganelli F, Scarpato P, et al. Acetylcysteine endotoxin induced acute renal insufficiency in rats. Am
and contrast agent-associated nephrotoxicity. J Am Coll J Physiol. 1987;253:F1283.
Cardiol. 2002;40:298. 128. Tolins JP, Vercellotti GM, Wilkowske M, et al. Role of
108. Durham JD, Caputo C, Dokko J, et al. A randomized platelet activating factor in endotoxemic acute renal failure
controlled trial of N-acetylcysteine to prevent contrast in the male rat. J Lab Clin Med. 1989;113:316.
nephropathy in cardiac angiography. Kidney Int. 2002; 129. Conger J, Robinette J, Villar A, et al. Increased nitric oxide
62:2202. synthase activity despite lack of response to endothelium
109. Birck R, Krzossok S, Markowetz F, et al. Acetylcysteine for dependent vasodilators in postischemic acute renal failure
prevention of contrast nephropathy: Meta-analysis. Lancet. in rats. J Clin Invest. 1995;96:631.
2003;362:598. 130. Noiri E, Peresleni T, Miller F, et al. In vivo targeting of
110. Isenbarger DW, Kent SM, OMalley PG. Meta-analysis of inducible NO synthase with oligodeoxynucleotides protects
randomized clinical trials on the usefulness of acetylcys- rat kidney against ischemia. J Clin Invest. 1996;97:2377.
teine for prevention of contrast nephropathy. Am J Cardiol. 131. Grover R, Zaccardelli D, Colice G, et al. Glaxo Wellcome In-
2003;15:1454. ternational Septic Shock Study Group. An open-label dose
460 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
escalation study of the nitric oxide synthase inhibitor, N(G)- 148. Borghi B, Casati A, Iuorio S, et al. Frequency of hypotension
methyl-L-arginine hydrochloride (546C88), in patients with and bradycardia during general anesthesia, epidural anes-
septic shock. Glaxo Wellcome International Septic Shock thesia, or integrated epidural-general anesthesia for total
Study Group. Crit Care Med. 1999;27:913. hip replacement. J Clin Anesth. 2002;14:102.
132. Avontuur JA, Tutein Nolthenius RP, van Bodegom JW, 149. Emmett RS, Cyna AM, Andrew M, et al. Techniques for pre-
et al. Prolonged inhibition of nitric oxide synthesis in severe venting hypotension during spinal anesthesia for caesarean
septic shock: A clinical study. Crit Care Med. 1998;26:660. section. Cochrane Database Syst Rev 2002;(3):CD002251.
133. Petros A, Lamb G, Leone A, et al. Effects of a nitric oxide 150. Inman SR, Stowe NT, Albanese J, et al. Contrasting
synthase inhibitor in humans with septic shock. Cardiovasc effects of vecuronium and succinylcholine on the renal
Res. 1994;28:34. microcirculation. Anesth Analg. 1994;78:682.
134. Weitzberg E, Lundberg JM, Rudehill A. Elevated plasma 151. Wilson WC, Aronson S. Oliguria, A sign of renal success
levels of endothelin in patients with sepsis syndrome. Circ or impending renal failure? Anesthesiol Clin North America.
Shock. 1991;33:222. 2001;19:841.
135. Wang A, Holcslaw T, Bashore TM, et al. Exacerbation 152. Sural S, Sharma RK, Singhal M, et al. Etiology, prognosis,
of radiocontrast nephrotoxicity by endothelin receptor and outcome of post-operative acute renal failure. Ren Fail.
antagonism. Kidney Int. 2000;57:1675. 2000;22:87.
136. Bernard GR, Wheeler AP, Russell JA, et al. The Ibuprofen in 153. Donohoe JF, Venkatachalam MA, Bernard DB, et al.
Sepsis Study Group. The effects of ibuprofen on the physi- Tubular leakage and obstruction after renal ischemia:
ology and survival of patients with sepsis. The Ibuprofen in Structural-functional correlations. Kidney Int. 1978;13:208.
Sepsis Study Group. N Engl J Med. 1997;336:912. 154. Lowe MB. Effects of nephrotoxins on ischemia in experi-
137. Sward K, Valson F, Ricksten SE. Long-term infusion of mental haemoglobinuria. J Pathol Bacteriol. 1966;92:319.
atrial natriuretic peptide (ANP) improves renal blood flow 155. Ramsay JG. The respiratory, renal, and hepatic systems:
and glomerular filtration rate in clinical acute renal failure. effects of cardiac surgery and cardiopulmonary bypass. In:
Acta Anaesthesiol Scand. 2001;45:536. Mora CT, ed. Cardiopulmonary bypass. New York: Springer-
138. Kelly KJ, Williams WW Jr, Colvin RB, et al. Intercellular Verlag; 1995:147168.
adhesion molecule-1-deficient mice are protected against 156. Diegeler A, Matin M, Falk V, et al. Indication and
ischemic renal injury. J Clin Invest 1996;97:1056. patient selection in minimally invasive and off-pump
139. Warren HS, Suffredini AF, Eichacker PQ, et al. Risks and coronary artery bypass grafting. Eur J Cardiothorac Surg.
benefits of activated protein C treatment for severe sepsis. 1999;16(Suppl 1):S79.
N Engl J Med. 2002;347:1027. 157. Gamoso MG, Bute BP, Landolfo KP, et al. Off-pump versus
140. Warren BL, Eid A, Singer P, et al. Kyber-Sept Trial on-pump coronary artery bypass surgery and postoperative
Study Group: Caring for the critically ill patient. High-dose renal dysfunction. Anesth Analg. 2000;91:1080.
antithrombin III in severe sepsis: A randomized controlled 158. Toraman F, Evrenkaya S, Yuce M, et al. Highly positive
trial. JAMA. 2001;286:1869. intraoperative fluid balance during cardiac surgery is
141. Bernard GR, Vincent JL, Laterre PF, et al. Recombinant associated with adverse outcome. Perfusion. 2004;19:85.
human protein C Worldwide Evaluation in Severe Sepsis 159. Elliott WJ, Weber RR, Nelson KS, et al. Renal and hemody-
(PROWESS) study group: Efficacy and safety of recombi- namic effects of intravenous fenoldopam versus nitroprus-
nant human activated protein C for severe sepsis. N Engl J side in severe hypertension. Circulation. 1990;81:970.
Med. 2001;344:699. 160. White WB, Halley SE. Comparative renal effects of intra-
142. Hirschberg R, Kopple J, Lipsett P, et al. Multicenter clinical venous administration of fenoldopam mesylate and sodium
trial of recombinant human insulin-like growth factor I in nitroprusside in patients with severe hypertension. Arch
patients with acute renal failure. Kidney Int. 1999;55:2423. Intern Med. 1989;149:870.
143. Franklin SC, Moulton M, Sicard GA, et al. Insulin-like 161. Ranucci M, Soro G, Barzaghi N, et al. Fenoldopam
growth factor I preserves renal function postoperatively. prophylaxis of postoperative acute renal failure in high-
Am J Physiol. 1997;272:F257. risk cardiac surgery patients. Ann Thorac Surg. 2004;78:
144. Chew ST, Newman MF, White WD, et al. Preliminary report 1332.
on the association of apolipoprotein E polymorphisms 162. Bove T, Landoni G, Calabro MG, et al. Renoprotective
with postoperative peak serum creatinine concentrations action of fenoldopam in high-risk patients undergoing
in cardiac surgical patients. Anesthesiology. 2000;93:325. cardiac surgery: A prospective, double-blind, randomized
145. Welborn MB, Oldenburg HS, Hess PJ, et al. The relationship clinical trial. Circulation. 2005;111:3230.
between visceral ischemia, proinflammatory cytokines, and 163. Gamulin Z, Forster A, Morel D, et al. Effects of infra-renal
organ injury in patients undergoing thoracoabdominal aortic cross-clamping on renal hemodynamics in humans.
aortic aneurysm repair. Crit Care Med. 2000;28:3191. Anesthesiology. 1984;61:394.
146. Bellomo R, Ronco C, Kellum JA, et al.The ADQI workgroup. 164. Pain JA, Cahill CJ, Bailey ME. Perioperative complications
Acute renal failure-definition, outcome measures, animal in obstructive jaundice: Therapeutic considerations. Br J
models, fluid therapy and information technology needs: Surg. 1985;72:942.
The Second International Consensus Coference of the 165. Parks RW, Rowlands BJ, Diamond T. Renal function in
Acute Dialysis Quality Initiative (ADQI) Group. Crit Care jaundiced patients: A prospective analysis. Int J Clin Pract.
2004;8:R204. 1988;52:461.
147. Godet G, Fleron MH, Vicaut E, et al. Risk factors for 166. Pain JA, Cahill CJ, Gilbert JM, et al. Prevention of
acute postoperative renal failure in thoracic or thoracoab- postoperative renal dysfunction in patients with obstructive
dominal aortic surgery: A prospective study. Anesth Analg. jaundice: A multicentre study of bile salts and lactulose. Br
1997;85:1227. J Surg. 1991;78:467.
CHAPTER SODIUM, POTASSIUM,
32
AND MAGNESIUM
A
dice and a pancreatic mass presented for
a Whipple procedure. Past medical his-
tory was significant for type 2 diabetes
mellitus, hypothyroidism, hyperlipidemia, SODIUM
and hypertension. Medications included hy-
Sodium is the primary extracellular cation and accounts
drochlorothiazide, levothyroxine, glyburide, and simvas-
for most of the extracellular osmolality (see Equation 1):
tatin. Preoperative laboratory studies were remarkable for
serum potassium of 3.0 mEq per dL and serum creatinine
of 1.4 mg per dL. Preoperative electrocardiogram (ECG) Plasma osmolality = 2 Na+ (mEq/L)
showed sinus rhythm with a right bundle branch block. + glucose (mg/dL)/18
The patient was taken to surgery, which was notable for + BUN (mg/dL)/2.8 (32.1)
a surgical duration of 4 hours and blood loss of 1.5 L.
She received 3 units of packed red blood cells and 4 L of As a result, changes in Na+ levels can cause significant
normal saline. She was also given 40 mEq of potassium changes in serum osmolality and subsequent movement
chloride. Urine output during the case totaled 350 mL. of water down the osmotic gradient. This movement of
Following extubation, she was transported to the surgical water, depending on the direction, causes cellular swelling
intensive care unit (SICU) where she was noted to have or dehydration with associated morbidity.
frequent premature ventricular contractions (PVCs). The
ECG showed no evidence of ischemia. Her postopera-
tive potassium level was 2.6 mEq per dL. She received
60 mEq of potassium over 3 hours but continued to have What Is the Difference between
PVCs. Repeat potassium level was 2.9 mEq per dL, and
magnesium was 1.4 mEq per dL. She received 4 g of mag-
Osmolality and Tonicity?
nesium sulfate over 2 hours and an additional 60 mEq
of potassium chloride. Her PVCs subsequently resolved, The concepts of osmolality and tonicity are often misun-
and serum potassium and magnesium levels increased to derstood. Osmolality, defined as the number of osmoles of
3.7 mEq per dL and 2.0 mEq per dL, respectively. The re- solute per kilogram of water, is determined by the number
mainder of her postoperative course was uneventful, and of osmotically active particles in solution, whether they
she was discharged home on postoperative day 7. be permeant or impermeant to cell membranes. Tonic-
ity refers to the number of osmotically active particles
Why Are the Major Cations TABLE 32.1 Approximate Electrolyte Concentrations
Important in Anesthesia? Extracellular Intracellular Fluid
Electrolyte Fluid (mEq/L) (mEq/L)
Sodium (Na+ ), potassium (K+ ), and magnesium (Mg2+ ) Sodium 140 10
are the major cations in the body (see Table 32.1). Potassium 4 150
Perioperative alterations in concentrations of these elec- Magnesium 2 40
trolytes are common and can cause significant morbidity
461
462 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
that are impermeant to the cell membrane. Urea is an dL increase in glucose, the serum Na+ level decreases
example of an osmotically active substance that is per- approximately 1.4 mEq per L.2
meant to cell membranes. High urea concentrations will
increase serum osmolality, but, because the urea moves
freely across cell membranes, it does not cause hyper-
tonicity and therefore does not cause movement of water
HYPOOSMOLAL
across the cell membrane. Therefore, while hypoosmo- HYPONATREMIA
lal hyponatremia is always hypotonic, it is possible to
The vast majority of cases of perioperative hyponatremia
have hypotonic hyponatremia with hyperosmolality (e.g.,
are hypoosmolal. This entity can be further classified on
elevated blood urea nitrogen [BUN]).
the basis of the presence of hypovolemia, euvolemia, or
hypervolemia.
The kidneys normally have a remarkable ability to TABLE 32.2 Causes of Elevated Antidiuretic Hormone
maintain normal osmolality over a wide range of fluid (ADH) Levels
intake. A typical daily solute load is 1,000 mOsm. In
the absence of ADH, maximally dilute urine of 50 to Appropriate Hypovolemia
100 mOsm per kg can be excreted, allowing excretion Hypertonicity
of up to 20 L of fluid per day. In the face of maximal Inappropriate Pain
ADH stimulation, urine concentration can be as high as (SIADH) Stress
1,200 mOsm per kg. This capability of increased urinary Nausea and vomiting
concentration allows excretion of <1 L of fluid per day. Catecholamines
In addition to decreased blood volume and increased Hypoxia
osmolality, a number of other factors increase ADH Hypercapnia
secretion, including pain, emotional stress, positive- Drugs
pressure ventilation, nausea and vomiting, hypoxia, and Chlorpropamide
hypercapnia. Several of these factors are often present Carbamazepine
in the perioperative period and can cause hyponatremia Clofibrate
as a result of the elevated ADH levels.612 Chung et al. Vincristine
reported a series of surgical patients in which 4.4% Nicotine
developed a plasma concentration below 130 mEq per Malignancy
L within a week of surgery. Most of these cases were mild Bronchogenic carcinoma
and were not associated with neurologic deterioration.13 Pancreatic carcinoma
Other investigators have reported more severe cases of Duodenal carcinoma
postoperative hyponatremia with fatal cerebral edema.14 Prostatic carcinoma
An increase in ADH levels in the absence of osmotic Pulmonary Disorders
or hemodynamic stimuli is considered SIADH and is Infection
a diagnosis of exclusion. In addition to the common Positive-pressure ventilation/PEEP
perioperative factors listed earlier, causes of SIADH CNS Disorders
include malignancy, pulmonary disease, central nervous Tumors
system (CNS) disorders, and medications (see Table 32.2). Infection
Other causes of euvolemic hyponatremia include Trauma
psychogenic polydipsia, heavy beer drinking, renal failure,
and diuretics, particularly thiazides. These diuretics block SIADH, syndrome of inappropriate antidiuretic hormone secretion;
Na+ reabsorption in the distal convoluted tubule and PEEP, positive end-expiratory pressure; CNS, central nervous system.
prevent formation of maximally dilute urine.
symptoms. For example, glycine absorption from prostatic
Hypervolemic irrigation fluid is associated with reversible blindness.
Most hyponatremia is hypotonic, causing movement
Hypervolemic, hypoosmolal hyponatremia occurs in the of water into the cells. The main symptoms result from
setting of edematous states such as congestive heart failure brain edema and include lethargy, confusion, coma,
(CHF), cirrhosis, nephrotic syndrome, and advanced renal seizures, and weakness. In severe cases, cerebral edema
failure. Total body water (TBW) and Na+ content are in- advances to brain herniation. In acute hyponatremia,
creased, but the increase in TBW is proportionally greater, serum Na+ <120 mEq per L is associated with clinical
leading to hyponatremia. Edema is typically present. In manifestations and adverse outcome.15 If the onset is
the case of CHF and cirrhosis, Na+ and water retention are chronic, lower levels of Na+ will be tolerated, due to loss
stimulated by decreased circulating blood volume. With of intracellular organic osmolytes (previously termed, id-
renal failure, excretion of Na+ and water are impaired. iogenic osmoles) by brain cells over several days. Although
these patients are less likely to be symptomatic, they are at
risk for demyelination if the Na+ is corrected too aggres-
What Are the Clinical sively (see Section, How Is Hypernatremia Treated?).
Manifestations of
Hyponatremia? How Is the Patient with
Hyponatremia Evaluated?
Clinical manifestations are largely dependent on the tonic-
ity of plasma and the rapidity with which the changes in
tonicity occur. Hyperosmolal, hypertonic hyponatremia The initial approach to the patient with hyponatremia is
causes a shift of water from cells to the extracellular space, to calculate the plasma osmolality by Equation 1:
leading to cellular dehydration. In addition to thirst, these
patients generally have symptoms related to brain dehy- Plasma osmolality = 2 Na+ (mEq/L)
dration, including lethargy, weakness, seizures, and coma. + glucose (mg/dL)/18
Hypertonicity from toxic solutes may cause additional + BUN (mg/dL)/2.8
464 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Low sodium
Serum osmolality
Pseudohyponatremia Hyperglycemia
i.e., hyperlipidemia, Mannitol
hyperproteinemia Toxins
hypocalcemia are also associated with the impaired action TABLE 32.3 Treatment of Diabetes Insipidus
of ADH.21
Central Antidiuretic Hormone
Diagnosis Diabetes Desmopressin (DDAVP): 520 g
The diagnosis of DI is based on a hypotonic urine in the Insipidus every 1224 h nasally; 24 g IV
presence of hypernatremia and elevated plasma osmolal- or SC every 1224 h
ity, reflecting the inability of the kidney to conserve water. Aqueous arginine vasopressin: 510
Urine osmolality typically is <150 mOsm per L, with urine units SC, IM, or IC every 28 h
specific gravity <1.005. Polyuria is also a prominent fea- Lysine vasopressin: 12 sprays
ture, and urine volume may exceed 1,000 mL per hour. nasally every 36 h
Patients undergoing surgery may have undiagnosed DI, Potentiate Antidiuretic Hormone
which is compensated as long as the patient has access Action
to water, but manifests itself perioperatively while the Chlorpropamide: 250500 mg daily,
patient is nil per os. orally (danger of hypoglycemia)
Carbamazepine: 400800 mg daily,
orally
Clofibrate: 500 mg every 6 h, orally
How Is Hypernatremia
Nephrogenic Restrict salt
Treated? Diabetes Abundant water intake
Insipidus Thiazide diuretics: Conventional
Treatment of hypernatremia involves three goals: (i) Re- doses
store intravascular volume; (ii) replace free water deficit;
and (iii) prevent further losses. Because intravascular wa- IV, intravenously; SC, subcutaneously; IM, intramuscularly; IC,
ter represents only approximately 7% of TBW, even fairly intracellularly.
large deficits are typically not associated with severe in- Modified from Zaloga GP, Kirby RR, Bernards WC, et al. Fluids and
travascular depletion. The exception involves severe cases electrolytes. In: Civetta JM, Taylor RW, Kirby RR, eds. Critical care,
of polyuria and/or late intervention. Restoration of in- 3rd ed. Philadelphia: Lippincott-Raven; 1997:421.
travascular volume deficits takes priority over replacing
free water, and isotonic fluids should be administered
action of 2 to 8 hours. This agent is useful if the DI is
until the intravascular space is replete.
likely to be temporary, as occurs in some neurosurgery
The free water deficit can be calculated as follows:
patients. Intravenous administration may have significant
vasopressor effects and should be done slowly.
Normal TBW normal Na+
Desmopressin (DDAVP) is a vasopressin analog with
= Current TBW current Na+ (32.5) minimal vasoconstrictor effects and a longer duration
Free water deficit = Normal TBW current TBW of action. The drug can be given subcutaneously or
= (0.6)(weight in kg)[1 (140/current Na+ )] (32.6) intravenously at a dose of 1 to 2 g every 12 hours.
For chronic therapy, intranasal DDAVP is administered
Whereas some patients, particularly those undergoing at a dose of 5 to 20 g every 12 to 24 hours. Patient
pituitary resection, can drink enough hypotonic fluids to responses may vary significantly, and dosing should be
prevent or restore their free water deficit, many patients adjusted on the basis of urine output and sodium levels.
will have nausea or altered mental status and will need Lysine vasopressin can also be administered intranasally
intravenous replacement. The deficit can be replaced with but has a shorter duration of action. Several drugs can
D5W, or if hyperglycemia is poorly controlled, 0.25% also potentiate the effects of ADH and may be used in
saline. Especially in patients with chronic hypernatremia partial DI (see Table 32.3).
(present >48 hours), correction of Na+ should generally
be no faster than 1 to 2 mEq/L/hour. Faster correction
may place the patient at risk for cerebral edema, because POTASSIUM
the concentration of nonelectrolyte osmoles is increased
in the brain during chronic hypernatremia to maintain Perioperative alterations in serum K+ are common and
intracellular osmolality and volume. can be associated with significant morbidity and mortal-
In most cases of hypotonic fluid loss, adjustment of ity. Most of the total body potassium is intracellular, with
the maintenance fluid concentration and rate is sufficient a concentration of 150 mEq per L. Serum K+ normally
to correct the problem. However, patients with DI need ranges from 3.5 to 5.0 mEq per L. Serum values are ap-
further management to decrease renal losses. In nephro- proximately 0.4 mEq per L higher than plasma values, due
genic DI, removal of the causative drug may restore the to release of K+ during clot formation in the spun serum
renal response to ADH. Patients with central DI need sample.
replacement with ADH or one of its analogs. Aqueous Potassium plays a crucial role in the generation of
vasopressin can be given subcutaneously or intramuscu- an action potential in excitable tissues. The threshold
larly in a dose of 5 to 10 units, providing a duration of for an action potential is determined by the ratio of
C H A P T E R 3 2 / S O D I U M , P O TA S S I U M , A N D M A G N E S I U M 467
intracellular to extracellular K+ concentration, which hypertension or CHF, and often present with chronic
is maintained by the Na-K-ATPase pump in the cell hypokalemia. By blocking reabsorption of K+ , the thi-
membrane. Alterations of the intracellular to extracellular azide and loop diuretics increase K+ loss, which can lead
K+ ratio have multiple electrophysiologic effects involving to hypokalemia.
not only resting membrane potential, but also action Hypokalemia is very common in the intraoperative
potential duration, refractory period, and spontaneous and postoperative periods. Decreases in effective arterial
depolarization (automaticity) as well. blood volume (either by decreased cardiac output or va-
sodilation) cause activation of the sympathetic nervous
system and the renin-angiotensin-aldosterone system, re-
sulting in increases in catecholamine and aldosterone
How Are Potassium Levels levels.25 Increased catecholamines shift K+ intracellularly
by activation of -adrenergic receptors, whereas aldos-
Regulated? terone stimulates Na+ reabsorption at the expense of K+
secretion. Renal K+ loss can be further exacerbated by
Regulation of serum K+ levels can occur by either hypomagnesemia.
transcellular shift of K+ or by alterations in excretion of
potassium. There are a number of factors that stimulate
transcellular shifts, including acidbase status, insulin,
and catecholamines. The classic teaching on K+ shifts What Are the Consequences
with acidbase disorders has been that, for every 0.1
unit change in pH, the K+ level changes in the opposite of Hypokalemia?
direction by 0.6 mEq per L.22 However, the response
varies significantly, and K+ shifts caused by infusion of Clinical manifestations of hypokalemia are primarily
organic acids such as lactic or -hydroxybutyric acid cardiac and neuromuscular in nature. At a K+ level
are minimal. Insulin plays a critical role in K+ uptake <3 mEq per L, patients may complain of weakness or
in cells, and hyperkalemia stimulates insulin release fatigue, and may develop an ileus. These patients are
by the pancreas. The hyperkalemia seen with diabetic also more sensitive to neuromuscular blocking agents.
ketoacidosis may be due to insulin deficiency rather More severe hypokalemia (K+ <2.5 mEq per L) can
than acidosis. -Adrenergic agonists stimulate cellular cause paralysis and respiratory failure, and at levels below
potassium uptake, whereas -adrenergic agonists increase 2 mEq per L, rhabdomyolysis can occur.
serum K+ levels by promoting hepatic K+ release.23 Hypokalemia has multiple effects on the electrical ac-
tivity of the heart, including hyperpolarization, increased
duration of the action potential and refractory period,
and increased automaticity. These changes have long
What Are the Causes been thought to predispose to arrhythmias due to reentry
and increased automaticity, including ventricular ectopic
of Hypokalemia? beats, ventricular tachycardia, and ventricular fibrilla-
tion.26 However, the effect of preoperative hypokalemia
Hypokalemia, defined as a serum K+ level <3.5 mEq per on perioperative arrhythmias is somewhat controver-
L, is common in the perioperative period and can result sial. Two studies, one of which included patients with
from either K+ depletion or intracellular K+ shift. significant cardiovascular disease, were unable to demon-
strate an association between preoperative hypokalemia
and intraoperative arrhythmias, suggesting that patients
POTASSIUM DEPLETION with chronic hypokalemia can be safely anesthetized.27,28
However, a large, observational study of patients who
The typical North American diet includes 50 to 100 mEq underwent cardiac surgery showed a significant associa-
per day of K+ . Approximately 90% of the daily K+ load is tion between preoperative hypokalemia and perioperative
eliminated in urine, 10% is excreted in the stool, and arrhythmias.29
a small amount is lost in sweat. In the presence of The discrepancy between these studies may be due
decreased K+ intake, renal excretion decreases markedly to the chronicity of the hypokalemia. With chronic
after several days.24 hypokalemia, intracellular K+ levels change as well,
In the kidneys, K+ is freely filtered, and most is moving the ratio of extracellular to intracellular K+ back
reabsorbed in the proximal tubule. As a result, K+ balance toward normal, and diminishing the effect on the electrical
is primarily regulated by the amount of excretion in the activity of the heart. This change likely explains the
distal tubule. Potassium excretion is regulated primarily safety of anesthetizing patients with chronic hypokalemia.
by aldosterone, which stimulates tubular cells to reabsorb Patients undergoing cardiac surgery are more likely to
Na+ and secrete K+ into the tubular lumen. Increased Na+ have superimposed acute changes in serum K+ , with
and water delivery to the distal tubule will also stimulate more significant membrane effects than those undergoing
Na+ reabsorption and K+ secretion. noncardiac surgery. Our clinical experience suggests that
Perioperatively, a number of factors can cause acute perioperative hypokalemia, often associated with
hypokalemia. Many patients routinely take diuretics for major operations and large fluid shifts, is associated
468 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
with ventricular ectopy and atrial fibrillation, and that (40 mEq in 100 mL) can be administered through a cen-
replacement of K+ with or without magnesium will often tral catheter. Hypomagnesemia will prevent correction of
terminate these arrhythmias. hypokalemia and should be treated. Even if serum Mg2+
Electrocardiographic changes may become apparent levels are normal, replacement should be considered.
when the serum K+ is <3.0 mEq per L. These changes
include ST-segment depression, T-wave flattening, and
appearance of a U wave following the T wave. The U wave
may be quite prominent and mistaken for a T wave, but What Are the Causes
the QT interval is not typically prolonged.
Hypokalemia also potentiates the effect of digitalis of Hyperkalemia?
and may lead to digitalis toxicity. This toxicity may
manifest as any type of arrhythmia, including supraven- Hyperkalemia, defined as a serum potassium level
tricular tachycardia, ventricular tachycardia, heart block, >5.0 mEq per L, is less common in the perioperative
and bradyarrhythmias. Coexisting hypercalcemia or hy- period than hypokalemia. However, because severe hy-
pomagnesemia exacerbates the toxicity. perkalemia may be life threatening, it must be rapidly
detected and treated.
Three potential causes of hyperkalemia should be
considered: Decreased excretion, increased intake, and
cellular release of K+ . As was discussed earlier, K+ balance
How Is Hypokalemia is regulated by the kidneys, and decreased excretion is
Diagnosed and Treated? typically a result of renal failure. Certain medications
may also impair renal excretion, including K+ -sparing
diuretics (e.g., spironolactone, triamterene), angiotensin-
DIAGNOSIS converting enzyme (ACE) inhibitors, angiotensin receptor
blocking agents, nonsteroidal anti-inflammatory agents,
Hypokalemia is often suspected on the basis of the pa- cyclosporine, tacrolimus, and heparin.
tients history, with diuretic use being the main cause. Pa- Hyperkalemia due to increased potassium intake
tients with ongoing GI losses are also at risk. The presence is typically iatrogenic, either from rapid administration
of neuromuscular or cardiac effects may not be present of K+ replacement or from K+ -containing medications.
in mild hypokalemia, and perioperative hypokalemia is Administration of stored red blood cells is often cited as
often detected through routine laboratory monitoring. a potential cause of hyperkalemia. The cells release K+
over time; by 21 days, the K+ concentration may be 25
to 30 mEq per L of plasma.33 However, because each
TREATMENT unit of packed red blood cells contains approximately
100 mL of plasma, hyperkalemia is primarily a concern
Treatment of hypokalemia depends on the underlying only with rapid transfusion of large amounts of old blood.
cause. If intracellular shift of K+ is the cause, treatment Following transfusion, the cells begin to take up K+ , and
may simply involve correcting the cause, such as alkalosis hypokalemia may ensue.
or excessive -adrenergic stimulation. However, most Cellular release of K+ occurs in the setting of cellular
patients with hypokalemia will require K+ replacement. ischemia or death. Examples include bowel ischemia or
Perioperative K+ replacement usually involves chlo- necrosis, hemolysis, rhabdomyolysis, and tumor lysis. Of
ride salt. The route of replacement is determined by the particular concern to the anesthesiologist is the release
urgency of replacement and whether the patient is taking of large amounts of K+ following the administration of
medications by mouth, as well as whether peripheral or succinylcholine to patients with neuromuscular disease,
central venous access is available. Several oral replace- prolonged immobilization, or burn injury. In these
ment formulations are available. Complete repletion will conditions, proliferation of extrajunctional receptors leads
take some time, as a decrease in the serum K+ level of to the exaggerated response. An excellent review on this
1 mEq per L represents a total body deficit of approxi- topic has been recently published.34
mately 400 mEq.30
Intravenous K+ replacement must be undertaken with
great caution. Because the intravascular pool of K+ is ap-
proximately 12 to 20 mEq, rapid administration of K+ can What Are the Consequences
cause significant acute elevations in serum potassium. In
fact, multiple case reports of iatrogenic cardiac arrest due of Hyperkalemia?
to rapid K+ administration have been published.31,32 For
this reason, K+ is generally given at a rate of 10 mEq Hyperkalemia can cause severe arrhythmias, conduction
per hour. If faster administration is deemed necessary, abnormalities, and cardiac arrest and should be treated
the rate may be increased up to 40 mEq per hour in the aggressively. Characteristic ECG changes include peaked
presence of continuous ECG monitoring. Because K+ is ir- T waves, followed by flattening of the P wave, prolonged
ritating to peripheral veins, the concentration should also PR interval, and widened QRS complex, ultimately giving
be limited to 40 mEq per L. More concentrated solutions the ECG a sine wave pattern. The specific K+ levels
C H A P T E R 3 2 / S O D I U M , P O TA S S I U M , A N D M A G N E S I U M 469
TREATMENT
Severe Acute, symptomatic hyperkalemia is life threatening
hyperkalemia and must be treated aggressively. Even in the absence
of ECG changes, a serum potassium >6.0 mEq per
L probably warrants at least frequent monitoring and
telemetry, if not active treatment. Treatments fall into
three classes: (i) stabilization of the cardiac membranes;
(ii) intracellular shift of K+ ; and (iii) K+ removal (see
Table 32.4).
Calcium administration is the mainstay of initial
FIGURE 32.2 Effects of hyperkalemia as manifested on treatment. Calcium salts antagonize the adverse effects of
electrocardiogram. Note the peaked T waves with hyperkalemia on cardiac conduction and can be admin-
mild-to-moderate hyperkalemia and loss of P wave with QRS istered as either calcium chloride or calcium gluconate.
widening and progression to a sine wave appearance with severe Calcium chloride is a venous irritant and should be given
hyperkalemia. centrally at a dose of 15 to 50 mg per kg. Calcium glu-
conate can be given peripherally but has only one third
the calcium of calcium chloride, thereby requiring 50 to
associated with each ECG pattern vary,35 but serum K+ 150 mg per kg doses. The stabilizing effects of the calcium
exceeding 10 mEq per L can be expected to cause asystole salts are immediate and last 30 to 60 minutes.37
or ventricular fibrillation. Figure 32.2 shows characteristic Intracellular shift of K+ can be achieved in several
ECG changes with different degrees of hyperkalemia. ways. Insulin will increase cellular uptake. A dose of
Neuromuscular effects of K+ may also be seen, includ- 0.1 units per kg will decrease serum K+ within 15 to
ing weakness, paralysis, and respiratory arrest.36 How- 30 minutes, with a duration of 4 to 6 hours.37 Dextrose is
ever, cardiac toxicity typically precedes these symptoms. given concomitantly to avoid hypoglycemia.
-Agonists will also drive K+ intracellularly and lower
serum K+ in most patients. Albuterol may be delivered by
inhalation or intravenously. The K+ levels fall faster with
How Is Hyperkalemia the intravenous administration, but one animal study
Diagnosed and Treated? showed an early rise in serum K+ .38 However, a small,
transient rise in serum K+ may also occur with the
inhalational route.39 Nebulized albuterol, 10 to 20 mg, will
DIAGNOSIS lower serum K+ by 0.6 to 1.0 mEq per L. The decrease
occurs within minutes and lasts at least 2 hours.39,40
The diagnosis of hyperkalemia is typically made by routine Albuterol appears to act synergistically with insulin to
serum K+ measurement. In patients with continuous ECG lower serum potassium. Because some patients will not
respond to -agonists, they should not be relied on as the averages 30% to 50% of ingested Mg2+ , occurs mainly in
sole treatment. the small intestine and is inversely proportional to intake.
Alkalinization of the blood with sodium bicarbonate Magnesium elimination occurs primarily through
historically has been used to drive K+ intracellularly, the kidneys through filtration and reabsorption. If Mg2+
but its effectiveness has recently been questioned. One intake is low, excretion can decrease to <1 mEq per day.46
study showed no change in serum K+ 60 minutes after Unlike K+ , there does not appear to be any significant
administration of sodium bicarbonate to hyperkalemic, hormonal control of Mg2+ handled in the kidneys. The
nonacidotic patients.41 A follow-up study showed a threshold for renal excretion of Mg2+ occurs at a level very
moderate decline in K+ only after 4 hours.42 There is some close to normal serum levels. As a result, acute increases
suggestion that sodium bicarbonate is more effective in in serum magnesium will result in rapid elimination.
lowering serum K+ in acidotic patients.43 On the basis of
current data, it is reasonable to use sodium bicarbonate as
an adjunct, particularly in acidotic patients, but it should
not be relied on as a sole measure. How Is Hypomagnesemia
Although the above measures decrease serum K+ ,
none of them actually removes K+ from the body. Loop
Defined, and What Are Its
diuretics are excellent kaliuretics and can be used in Causes?
patients with renal function. However, most patients with
hyperkalemia have some degree of renal impairment. Hypomagnesemia, defined as a serum magnesium level
Sodium polystyrene sulfate, a K+ -binding resin, <1.8 mg per dL, can be caused by decreased intake,
has been a mainstay of treatment for K+ removal in increased losses, or transcellular redistribution. Dietary
patients with hyperkalemia. Each gram of resin binds causes include malnutrition, alcoholism, and parenteral
approximately 0.5 mEq of K+ . The main site of action nutrition with inadequate Mg2+ administration. Increased
is the colon, and the resin may be administered either loss of Mg2+ may occur through the GI tract or the kidneys.
orally or by enema, at a dose of 0.5 to 1.0 g per kg GI losses result from malabsorption, diarrhea, vomiting,
every 2 to 4 hours. When given orally, sorbitol is usually nasogastric suction, and fistulas. Secretions of the lower
added to avoid constipation. However, there are several GI tract contain higher levels of Mg2+ than those of the
reports of intestinal necrosis associated with the resin upper GI tract (10 to 14 mEq per L vs. 1 to 2 mEq per L).
in sorbitol administered orally or by nasogastric tube.44 Renal losses are a common cause of hypomagne-
Some investigators have questioned whether the use of semia, particularly in the perioperative period. Multiple
sodium polystyrene sulfate adds any significant benefit drugs that can increase renal excretion include diuretics
to the use of laxatives alone to enhance intestinal K+ (especially loop diuretics), cardiac glycosides, aminogly-
elimination.45 cosides,47 cyclosporine,48 and amphotericin B.49 Sodium
In refractory cases of hyperkalemia, dialysis may be loading and osmotic diuresis from mannitol or glucose
necessary. Hemodialysis is most effective and can remove also increase renal excretion, as does an elevated blood
25 to 30 mEq of K+ per hour. High flow, continuous renal alcohol level.
replacement therapy may be used in less emergent cases. Cellular uptake of Mg2+ may cause hypomagnesemia
Peritoneal dialysis can remove 10 to 15 mEq of K+ per without a change in total body magnesium. This trans-
hour. cellular shift may be seen in anabolic states, after
the administration of catecholamines or insulin, or
following parathyroidectomy for hyperparathyroidism.
MAGNESIUM Magnesium tissue precipitation or chelation may be seen
in pancreatitis or rhabdomyolysis. Large doses of citrate
Magnesium is a divalent cation that plays a role in multiple associated with massive blood transfusion chelate Mg2+ ,
cellular functions throughout the body. As a cofactor for and calcium and may decrease both cations.50
multiple enzymes, Mg2+ plays a role in oxidative phospho-
rylation, glycolysis, DNA transcription, protein synthesis,
nerve conduction, and ion transport. Most of the total
body Mg2+ (60% to 70%) is in bone, whereas 30% to 40% What Are the Consequences
is intracellular. Extracellular Mg2+ makes up only 1% of
total body Mg2+ . Normal serum Mg2+ ranges from 1.5 to of Hypomagnesemia?
1.9 mEq per L (1.8 to 2.4 mg per dL), of which approx-
imately 30% is protein-bound, primarily to albumin. Of Patients with hypomagnesemia are often asymptomatic,
the remaining 70%, most is present in the ionized form, and the disorder may go undetected. Whang and Ryder
whereas approximately 15% of the total serum magnesium measured 1,033 serum samples submitted for electrolyte
is complexed with anions. Most laboratories measure only determinations. Roughly half of the samples had low
total serum magnesium, which may not reflect the levels Mg2+ levels, but levels were ordered in only 10% of the
of the physiologically active Mg2+ . hypomagnesemic samples, suggesting that abnormalities
Magnesium balance is primarily maintained through were not suspected.51
dietary intake and renal excretion. The average Western Signs and symptoms of hypomagnesemia com-
diet contains 150 to 300 mg per day. Absorption, which monly involve the neuromuscular and cardiovascular
C H A P T E R 3 2 / S O D I U M , P O TA S S I U M , A N D M A G N E S I U M 471
systems. Neuromuscular effects include weakness, muscle Mg2+ administration. Because renal excretion is the
fasciculation, tremor, focal or generalized tetany, and al- primary route of Mg2+ , it is impaired in patients with
tered mental status. Hypocalcemia and hypokalemia often a glomerular filtration rate <30 mL per minute.62 Many
accompany hypomagnesemia and may exacerbate some antacids and cathartics contain magnesium and should
of the effects. be used with great caution in patients with renal failure.
Magnesium cardiovascular effects are of significant Iatrogenic hypermagnesemia can occur with overag-
concern perioperatively. Patients undergoing major opera- gressive Mg2+ repletion. However, it is most often encoun-
tions with large fluid shifts often develop hypomagnesemia, tered during treatment of preeclampsia and eclampsia,
likely through a combination of dilution and adrenergic where Mg2+ is standard therapy. Magnesium readily
stimulation. Catecholamines appear to cause hypomagne- crosses the placenta and may cause neonatal hyper-
semia by increased cellular uptake,5254 which may explain magnesemia. Other causes include lithium ingestion and
why the disorder is present in various stress states. hypothyroidism.
Perioperative hypomagnesemia is often associated
with both atrial and ventricular arrhythmias. However, as
is the case with hypokalemia, the role of hypomagnesemia
as a cause of arrhythmias has been questioned.55 Because
hypokalemia often accompanies hypomagnesemia, it may What Are the Consequences
be unclear which disorder is responsible for the arrhyth- of Hypermagnesemia?
mia. However, several reports of magnesium-associated
arrhythmias that resolved with Mg2+ replacement support
The most significant perioperative effects of hypermag-
the role of hypomagnesemia as the etiology.5658 Further-
nesemia involve the neuromuscular and cardiovascular
more, Mg2+ administration appears to reduce the risk of
systems. Magnesium excess inhibits acetylcholine release
atrial fibrillation following cardiac surgery.59,60 In light of
at the neuromuscular junction, diminishes the effect of
the available evidence, it seems reasonable to keep Mg2+ in
acetylcholine at the end plate, and decreases the excitabil-
the high normal range in patients at risk for perioperative
ity of the muscle fiber membrane. Magnesium potentiates
arrhythmias.
the effects of neuromuscular blocking agents and may pre-
Hypokalemia often accompanies hypomagnesemia
vent complete reversal by acetylcholinesterases. At levels
and may be refractory to treatment without replenishment
of 4 to 7 mEq per L, deep tendon reflexes are diminished
of Mg2+ .61 The mechanism appears to be related to renal
or absent, and at >9 mEq per L, muscle weakness may
K+ wasting. If patients do not respond appropriately
progress to paralysis and respiratory arrest.
to K+ replacement, magnesium administration should
Cardiovascular effects of hypermagnesemia include
be considered. Because serum Mg2+ may not reflect
vascular relaxation that can manifest as hypotension. Con-
intracellular levels, even patients with normal serum Mg2+
duction abnormalities occur with serum levels >5 mEq
levels may benefit from replacement.
per L. Electrocardiographic changes include prolonga-
tion of the PR, QRS, and ST intervals. When serum Mg2+
reaches 10 to 15 mEq per L, bradycardia, complete heart
block, and cardiac arrest occur.
How Is Hypomagnesemia
Treated?
Acute hypomagnesemia is typically treated with intra- How Is Hypermagnesemia
venous replacement, particularly in the perioperative pe-
riod. We routinely use magnesium sulfate, which contains
Treated?
8 mEq of magnesium per gram. If patients have signifi-
cant signs or symptoms, 2 g of magnesium sulfate infused Because most cases of hypermagnesemia are iatrogenic,
over 10 minutes will rapidly increase Mg2+ . However, due the first step in treatment is to stop Mg2+ administration.
to the relatively low renal threshold for excretion, much In life threatening situations, calcium chloride (1 g) or
of this dose will be lost through the kidneys. Therefore, gluconate (2 to 3 g) administered over 5 to 10 minutes
serial Mg2+ levels should be monitored, and an infusion will antagonize the toxic effects of Mg2+ , but the effect
of magnesium sulfate, 1 g per hour for 4 to 6 hours, may is transient. In patients with normal renal function, loop
be necessary. diuretics will facilitate Mg2+ excretion. Patients with renal
failure typically require dialysis.
2. Chronic changes in serum Na+ are compensated by 12. Berns AS, Anderson RJ, McKonald KM. Effect of hypercap-
nic acidosis on renal water excretion in the dog. Kidney Int.
changes in the concentration of organic osmolytes.
1979;15:116.
3. Chronic Na+ abnormalities should be corrected 13. Chung HS, Kluge R, Schrier RW, et al. Postoperative hypona-
slowly to prevent rapid movement of water across tremia: A prospective study. Arch Intern Med. 1986;146:333.
cell membranes. 14. Arieff AI. Hyponatremia, convulsions, respiratory arrest, and
4. Patients with chronic hypokalemia can generally be permanent brain damage after elective surgery in healthy
safely anesthetized. women. N Engl J Med. 1986;314:1529.
5. Acute hypokalemia may predispose to perioperative 15. Arieff AI, Llach F, Massry SG. Neurological manifestations
arrhythmias. and morbidity in hyponatremia. Medicine. 1976;55:121.
6. Hypokalemia is difficult to correct in the presence of 16. Mori T, Katayama Y, Kawamata T, et al. Improved efficiency
coexisting Mg2+ depletion. of hypervolemic therapy with inhibition of natriuresis by
fludrocortisone in patients with aneurysmal subarachnoid
7. Hyperkalemia can be life threatening and should be
hemorrhage. J Neurosurg. 1999;91:947.
treated aggressively. 17. Moro N, Katayama Y, Kojima J, et al. Prophylactic man-
8. Calcium salts will rapidly stabilize the membranes agement of excessive natriuresis with hydrocortisone for
of excitable cells and allow time for other measures efficient hypervolemic therapy after subarachnoid hemor-
to decrease serum K+ . rhage. Stroke. 2003;34:2807.
9. Insulin (combined with glucose to prevent hypo- 18. Ayus JC, Krothapalli RK, Arieff AI. Treatment of symp-
glycemia) will drive K+ into cells. Albuterol and tomatic hyponatremia and its relation to brain damage.
bicarbonate are often effective but are less reliable. A prospective study. N Engl J Med. 1987;317:1190.
10. Hypomagnesemia is common following large oper- 19. Arieff AI. Treatment of symptomatic hyponatremia: Neither
ations with major fluid shifts. haste nor waste. Crit Care Med. 1991;19:748.
20. Kleeman CR, Rubini ME, Lamdin E, et al. Studies in alcohol
11. Hypomagnesemia may predispose to perioperative
diuresis:II. The evaluation of ethyl alcohol as an inhibitor of
arrhythmias. the neurohypophysis. J Clin Invest. 1955;34:448.
12. Magnesium replacement should be given slowly (1 to 21. Berl T, Schrier RW. Disorders of water metabolism. In:
2 g per hour), because the renal threshold for Mg2+ Schrier RW, ed. Renal and electrolyte disorders, 5th ed.
excretion is just above the normal serum level. Philadelphia: Lippincott-Raven; 1997:1.
13. The toxic effects of hypermagnesemia can be antag- 22. Edwards R, Winnie AP, Ramamurthy S. Acute hypocapneic
onized by calcium salts. hypokalemia: An iatrogenic anesthetic complication. Anesth
Analg. 1977;56:782.
23. Jauchem JR, Vick RL. Phenylephrineinduced hyper-
REFERENCES kalemia: Role of the liver. Proc Soc Exp Biol Med. 1979;162:
1. Sterns RH, Spital A, Clark EC. Disorders of water balance. 207.
In: Kokko JP, Tannen RL, eds. Fluids and electrolytes, 3rd ed. 24. Womersley RA, Darragh JH. Potassium and sodium restric-
Philadelphia: WB Saunders; 1996:65. tion in the normal human. J Clin Invest. 1955;34:456.
2. Roscoe JM, Halperin ML, Rolleston FS, et al. Hyperglycemia- 25. Schrier RW. Body fluid volume regulation in health and dis-
induced hyponatremia: Metabolic considerations in calcula- ease: A unifying hypothesis. Ann Intern Med. 1990;113:155.
tion of serum sodium depression. Can Med Assoc J. 1975; 26. Helfant RH. Hypokalemia and arrhythmias. Am J Med. 1986;
112:452. 80:13.
3. Isotani E, Suzuki R, Tomita K, et al. Alterations in plasma 27. Vitez TS, Soper LE, Wong KC, et al. Chronic hypokalemia
concentrations of natriuretic peptides and antidiuretic hor- and intraoperative dysrhythmias. Anesthesiology. 1985;63:
mone after subarachnoid hemorrhage. Stroke. 1994;25:2198. 130.
4. Berendes E, Walter M, Cullen P, et al. Secretion of brain na- 28. Hirsch IA, Tomlinson DL, Slogoff S, et al. The overstated risk
triuretic peptide in patients with aneurysmal subarachnoid of preoperative hypokalemia. Anesth Analg. 1988;67:131.
hemorrhage. Lancet. 1997;349:245. 29. Wahr JA, Parks R, Boisvert D, et al. Preoperative serum
5. Tomida M, Muraki M, Uemura K, et al. Plasma concentra- potassium levels and perioperative outcomes in cardiac
tions of brain natriuretic peptide in patients with subarach- surgery patients. JAMA. 1999;281:2203.
noid hemorrhage. Stroke. 1998;29:1584. 30. Sterns RH, Cox M, Fieg PU, et al. Internal potassium balance
6. Cochrane JPS, Forsling ML, Gow NM, et al. Arginine and the control of the plasma potassium concentration.
vasopressin release following surgical operations. Br J Surg. Medicine. 1981;60:339.
1981;68:209. 31. Patient Safety Alert. Medication error prevention: Potassium
7. Deutsch S, Goldberg M, Dripps RD. Postoperative hypona- chloride. Int J Qual Health Care. 2001;13:155.
tremia with the inappropriate release of antidiuretic hor- 32. Wetherton AR, Corey TS, Buchino JJ, et al. Fatal intravenous
mones. Anesthesiology. 1966;27:250. injection of potassium in hospitalized patients. Am J Forensic
8. Thomas TH, Morgan DB. Post-surgical hyponatremia: The Med Pathol. 2003;24:128.
role of intravenous fluids and arginine vasopressin. Br J Surg. 33. Marshall M. Potassium intoxication from blood and plasma
1979;66:540. transfusions. Anaesthesia. 1962;17:145.
9. Haas M, Glick SM. Radioimmunoassayable plasma vaso- 34. Martyn JA, Richtsfeld M. Succinylcholine-induced hyper-
pressin associated with surgery. Arch Surg. 1978;113:597. kalemia in acquired pathologic states: Etiologic factors and
10. Rowe JW, Shelton RL, Helderman JH, et al. Influence of molecular mechanisms. Anesthesiology. 2006;104:158.
emetic reflex on vasopressin release in man. Kidney Int. 35. Ettinger PO, Regan TJ, Oldewurtel HA. Hyperkalemia,
1979;16:729. cardiac conduction, and the electrocardiogram: A review.
11. Anderson RJ, Pluss RG, Berus AS, et al. Mechanism of effect Am Heart J. 1974;88:360.
of hypoxia on renal water excretion in the dog. J Clin Invest. 36. Freeman SJ, Fale AD. Muscular paralysis and ventilatory
1978;62:769. failure caused by hyperkalemia. Br J Anaesth. 1993;70:226.
C H A P T E R 3 2 / S O D I U M , P O TA S S I U M , A N D M A G N E S I U M 473
37. Evans KJ, Greenberg A. Hyperkalemia: A review. J Intensive 50. Killen DA, Grogen EL, Gower RE, et al. Response of canine
Care Med. 2005;20:272. plasma ionized calcium and magnesium to the rapid infusion
38. DuPlooy WJ, Hay L, Kahler CP, et al. The dose-related of acid-citrate-dextrose (ACD) solution. Surgery. 1971;70:736.
hyper- and hypokalaemic effects of salbutamol and its 51. Whang R, Ryder KW. Frequency of hypomagnesemia and
arrhythmogenic potential. Br J Pharmacol. 1994;111:73. hypermagnesemia. Requested vs routine. JAMA. 1990;263:
39. Mandelberg A, Krupnik Z, Houri S, et al. Salbutamol 3063.
metered-dose inhaler with spacer for hyperkalemia-How 52. Whyte KF, Addis GJ, Whitesmith R, et al. Adrenergic control
fast? How safe? Chest. 1999;115:617. of plasma magnesium in man. Clin Sci (Lond). 1987;72:135.
40. Allon M, Dunlay R, Copkney C. Nebulized albuterol for acute 53. Ryzen E, Servis KL, Rude RK. Effect of intravenous
hyperkalemia in patients on hemodialysis. Ann Intern Med. epinephrine on serum magnesium and free intracellular red
1989;110:426. blood cell magnesium concentrations measured by nuclear
41. Blumberg A, Weidmann P, Shaw S, et al. Effect of various magnetic resonance. J Am Coll Nutr. 1990;9:114.
therapeutic approaches on plasma potassium and major reg- 54. Elliott D, Rizack M. Epinephrine and adrenocorticotropic
ulating factors in terminal renal failure. Am J Med. 1988;85: hormone-stimulated magnesium accumulation in adipocytes
507. and their plasma membranes. J Biol Chem. 1974;249:
42. Blumberg A, Weidmann P, Ferrari P. Effect of prolonged 3985.
bicarbonate administration on plasma potassium in terminal 55. Surawicz B. Is hypomagnesemia or magnesium deficiency
renal failure. Kidney Int. 1992;41:369. arrhythmogenic? J Am Coll Cardiol. 1989;14:1093.
43. Schwarz KC, Cohen BD, Lubash GD, et al. Severe acidosis 56. Moran J, Gallagher J, Peake S, et al. Parenteral magnesium
and hyperpotassemia treated with sodium bicarbonate sulfate versus amiodarone in the therapy of atrial tach-
infusion. Circulation. 1959;19:215. yarrhythmias: A prospective, randomized study. Crit Care
44. Rashid A, Hamilton SR. Necrosis of the gastrointestinal Med. 1995;23:1816.
tract in uremic patients as a result of sodium polystryrene 57. Singh RB, Singh VP, Bajpai HS. Refractory cardiac arrhyth-
sulfonate (Kayexelate) in sorbitol: An underrecognized mia due to hypomagnesmia. Acta Cardiol. 1975;30:499.
condition. Am J Surg Pathol. 1997;21:60. 58. Levine SR, Crowley TJ, Hai HA. Hypomagnesemia and ven-
45. Gruy-Kapral C, Emmett M, Santa Ana CA, et al. Effect tricular tachycardia: A complication of ulcerative colitis and
of single dose resin-cathartic therapy on serum potassium parenteral hyperalimentation in a nondigitalized noncardiac
concentration in patients with end-stage renal disease. J Am patient. Chest. 1982;81:244.
Soc Nephrol. 1998;9:1924. 59. Miller S, Crystal E, Garfinkle M, et al. Effects of magnesium
46. Barnes BA, Cope O, Harrison T. Magnesium conservation on atrial fibrillation after cardiac surgery: A meta-analysis.
in human beings on a low magnesium diet. J Clin Invest. Heart. 2005;91:618.
1958;37:430. 60. Alghamdi AA, Al-Radi OO, Latter DA. Intravenous magne-
47. Zaloga GP, Chernow B, Pock A, et al. Hypomagnesemia is sium for prevention of atrial fibrillation after coronary artery
a common complication of aminoglycoside therapy. Surg bypass surgery: A systematic review and meta-analysis.
Gynecol Obstet. 1984;158:561. J Card Surg. 2005;20:293.
48. Barton CH, Vaziri ND, Martin DC, et al. Hypomagnesemia 61. Whang R, Flink EB, Dyckner T, et al. Magnesium depletion
and renal magnesium wasting in renal transplant recipients as a cause of refractory potassium repletion. Arch Intern Med.
receiving cyclosporine. Am J Med. 1987;83:693. 1985;145:1686.
49. Barton CH, Vaziri ND, Cesario T. Renal magnesium wasting 62. Popovtzer MM, Schainuck LI, Massry SG, et al. Divalent ion
associated with amphotericin B therapy. Am J Med. 1984; excretion in chronic kidney disease: Relation to degree of
77:471. renal insufficiency. Clin Sci. 1970;28:297.
CHAPTER TURP SYNDROME
CASE SUMMARY and minimally invasive therapies have reduced the num-
ber of surgical interventions performed, TURP continues
73-year-old man with symptomatic be- to be necessary and is performed more than 100,000 times
A
nign prostatic hypertrophy is scheduled annually in the United States (and 25,000 times annually
for transurethral resection of the prostate in the United Kingdom).
(TURP). His past medical history is sig- Surgery performed through the urethra has a known
nificant for hypertension and a myocardial profile of risks. These include hemorrhage, perforation,
infarction 3 years ago, followed by a three- nerve injury, and later development of urethral strictures.
vessel, coronary artery graft. The blood pressure is now The TURP syndrome has traditionally been associated
well controlled with atenolol. He has occasional exertion- most notably with the acute onset of hypervolemia,
related chest pain that responds to a single dose of
profound hyponatremia, neurologic excitability, seizures,
nitroglycerin or rest. Transthoracic echocardiography re-
hemodynamic changes, renal failure, blindness, and, occa-
vealed left ventricular wall hypokinesis and an ejection
sionally, even death. This constellation of clinical findings
fraction of 42% with no other abnormalities.
stem from the surgical use and rapid intravascular ab-
Anesthesia is accomplished with a subarachnoid block
sorption of electrolyte-free irrigation fluid.
using 12.5 mg bupivicaine with epinephrine 1:200,000 that
Presenting signs and symptoms will depend both on
achieves a dense block to T8. Small ephedrine boluses are
the absorbed volume and the type of irrigant used. In
used to support the blood pressure until surgical resection
begins. The patients intraoperative course is uneventful general, more than 3 L of electrolyte-free, irrigating fluid
and surgery, using a bipolar resectoscope with normal needs to be absorbed to result in a severe form of this
saline as irrigant, is completed in 75 minutes. syndrome, whereas between 1 and 3 L is required to elicit a
The level of anesthetic block on admission to the mild syndrome.2 The amount of normal saline absorbed
postanesthesia care unit (PACU) has receded to T10. A total that will elicit symptoms in this aged patient group is
of 1,600 mL of normal saline had been infused, and esti- not known.
mated blood loss was 50 mL. Initially, the patient does well, An answer to the question posed at the end of the case
but over the ensuing hour, he develops progressive dyspnea synopsis requires, first, that there is agreement on what
and a lowering of his blood pressure. Physical examination constitutes the TURP syndrome. Historically, the syn-
reveals bilateral crackles, and chest radiograph demon- drome is considered to represent a severe clinical situation
strates diffuse fluffy infiltrates bilaterally. The patients with a large variety of symptoms affecting the circulatory
electrolytes are normal; specifically, the measured serum and neurologic systems. Milder forms clearly exist.3,4 The
sodium is 143 mEq per L ([Na+ ] = 140 mEq per L, preop- diagnosis is usually made during or after surgery with a
eratively). He denies vision changes and shows no signs of monopolar resectoscope. Evidence of acute serum elec-
neurologic excitability. He is treated with a diuretic for the trolyte dilution, such as hyponatremia or an acute serum
presumed congestive heart failure, diureses 800 mL, and concentration rise of a compound contained only in the
is discharged to the ward 2 hours later. Without evidence irrigation fluid, such as glycine or ethanol, must exist.
of electrolyte changes or neurologic symptoms, would this However, newer surgical instruments and techniques that
still be considered an example of the TURP syndrome? are touted to reduce the volume of absorbed irrigant or
that allow use of physiologic isoosmolar solutions for
irrigation will change this traditional presentation, elimi-
INTRODUCTION nating the possibility of acute hyponatremia.
These techniques can be categorized as follows:
Benign prostatic hypertrophy is a condition that ulti-
mately affects more than 50% of elderly men.1 The promi- Interstitial thermal ablativetransurethral microwave
nent clinical features of the disease arise from progressive thermotherapy, transurethral needle ablation, and in-
bladder outlet obstruction. Although improved medical terstitial laser
474
C HAPTE R 33/TU R P SYN DROM E 475
Minimally invasivetransurethral vaporization of the sorbitol, mannitol, and, with the bipolar resectoscopes,
prostate, photoselective vaporization of the prostate physiologic saline solutions.1
with KTP laser, and Holmium-YAG laser resection of In the 1950s, the incidence of the TURP syn-
the prostate drome was as high as 10%.10 With modern surgical
Transurethral vaporization-resection of the prostate and techniques and improved clinical awareness, this inci-
Techniques that utilize a bipolar resectoscope dence has been reduced to no more than 4% among
patients undergoing TURP,4 although occasionally the
Whichever of these newer treatment options ultimately
incidence is still reported much higher.11 These data
prevail, it seems certain that variability to the presenta-
show that, despite the advances that have been made,
tion of TURP syndrome will narrow and that the frequency
the TURP syndrome continues to occur with signifi-
of its occurrence will diminish, or perhaps even disap-
cant frequency. The diagnosis remains difficult to make
pear, as use of these techniques increase.1 In the future,
because the syndrome can develop over multiple path-
it will be the sequelae of acute hypervolemia rather
ways and lacks a stereotypical presentation (see Fig. 33.1,
than acute hyponatremia, hyperglycinemia, hypoosmo-
Table 33.2).
lality, and neurologic signs and symptoms that seal the
diagnosis.
Until these newer surgical interventions for TURP are
in broader use, however, there will remain an interest
in recognizing the complications from TURP associated Why Does Intravascular
with use of the monopolar resectoscope, still considered Volume Expansion Occur?
by many as the gold standard instrument for TURP.57
This chapter will focus on the pathophysiologic basis of
Absorption requires, as a first step, a path for entry of fluid
the TURP syndrome and discuss the prevailing opinion
into the body. Irrigation fluid gains direct intravascular
on the best anesthetic management practices.
access when the prostatic venous plexus is opened
during resection near the prostatic capsule. Once the
venous plexus is opened, whether recognized or not,
further absorption is possible for the remainder of the
How Is Transurethral procedure. Fluid can also be absorbed indirectly following
Resection of the Prostate an injury to the prostatic capsule or if the bladder neck
is divided, whereupon irrigation fluid extravasates into
Accomplished? the retroperitoneum or peritoneum, and intravascular
absorption occurs only gradually.12,13
Surgical features of TURP vary (see Table 33.1).4,8,9 The After fluid absorption by any routethe second step
speed of prostatic resection averages approximately 0.6 g on the path to developing the TURP syndromeis mea-
per minute, even when the newer bipolar resectoscopes surable in 34% (transurethral vaporization of the prostate)
are employed.9 The irrigation used ranges from distilled to 46% (TURP) of procedures.12,14 Extravasation of fluid
water to a variety of nonhemolytic glucose, urea, glycine, may occur in 4% to 20% of patients having TURP.4,9 The
TABLE 33.1 Average and Extreme Statistics Reported with Transurethral Resection
of the Prostate
a Trepanier CA, Lessard MR, Brochu J, et al. Another feature of TURP syndrome: Hyperglycaemia and lactic acidosis
Anesthesiology. 1979;50:355.
c Hjertberg H, Petterson B. The use of a bladder pressure warning device during transurethral prostatic resection
Dis. 1990;XVI:73.
TURP, transurethral resection of the prostate.
476 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
FIGURE 33.1 The variety of mechanisms and pathways that lead to transurethral resection of the
prostate (TURP) syndrome. The triggering event is the entry of irrigation solution into the
intravascular compartment (A) which raises intravascular volume (B) with its sequelae and
decreases (C) and/or increases (D) solute concentration. IV, intravenous.
force driving absorption by extravasation is the irrigation to occur, the irrigation fluid pressure must exceed the
fluid pressure in the bladder. Intravesicular pressure may receiving compartment pressure. Hence, intravascular ab-
be related to the height of the irrigation bag above the sorption can occur when intravesicular pressure exceeds
prostatic sinuses and explains why surgeons have been venous pressure, or approximately 15 cm H2 O (1.5 kPa).
admonished to raise the irrigation bag no more than For extravasation to occur through a perforated capsule or
60 cm above the height of the bladder.15 For absorption bladder neck, intravesicular pressure needs only to exceed
TABLE 33.2 Checklist Used to Define and Score Symptoms in the TURP Syndrome. (Number and Severity of
Symptoms Showed a Statistically Significant Increase as More Irrigating Fluid was Absorbed.)
Severity Score
Symptom 1 2 3
Circulatory
Chest pain Duration <5 min Duration >5 min Repeated attacks
Bradycardia HR drop 1020 bpm HR drop >20 bpm Repeated drops
Hypertension SAP up 1020 mm Hg SAP up >30 mm Hg Score (2) for 15 min
Hypotension SAP down 3050 mm Hg SAP down >50 mm Hg Repeated drops >50 mm Hg
Poor urine output Diuretics are needed Repeated use Diuretics ineffective
Neurologic
Blurred vision Duration <10 min Duration >10 min Transient blindness
Nausea Duration <5 min Duration 5120 min Intense or >120 min
Vomiting Single instance Repeatedly, <60 min Repeatedly, >60 min
Uneasiness Slight Moderate Intense
Confusion Duration <5 min Duration 560 min Duration >60 min
Tiredness Patient says so Objectively exhausted Exhausted for >120 min
Consciousness Mildly depressed Somnolent <60 min Needs ventilator
Headache Mild Severe <60 min Severe >60 min
TURP, transurethral resection of the prostate; HR, heart rate; SAP, systolic arterial pressure.
From: Hahn RG, Shemais H, Essen P. Glycine 1.0% versus glycine 1.5% as irrigating fluid during transurethral resection of the prostate. Br J
Urol. 1997;79:394.
C HAPTE R 33/TU R P SYN DROM E 477
the intra-abdominal pressure of approximately 5 cm H2 O and after TURP,27 suggesting that intravascular volume
(0.5 kPa).16,17 changes independent of osmolality can play an important
The mode of resectoscope use also plays a role in role in the morbidity and mortality associated with the
defining the intravesicular pressure. When the resecto- TURP syndrome.
scope is operated with continuous flow, as is favored
by many surgeons, less pressure builds in the bladder
than when resection is performed using intermittent flow.
However, use of a continuous flow resectoscope is no DETECTION
guarantee that intravesicular pressure will remain low;
Early detection of fluid absorption opens the possibility to
an obstruction to outflow caused by blood clots or tissue
arrest further absorption and avoid the TURP syndrome,
chips will make intravesicular pressure rise. Therefore,
either by stopping surgery or by keeping the fluid pressure
although the height of the irrigation bag defines the
low by applying a bladder pressure device. The earliest
maximum possible intravesicular pressure, no correla-
possible treatment of fluid absorption can also be started.
tion between bag height and fluid absorption has been
Classically, an awake patient undergoing TURP with re-
consistently demonstrated.18,19
gional anesthesia serves as his own miners canary for
The risk of having fluid absorption during TURP
early detection of absorption. The severity and number
increases slightly with increasing operating time, weight
of complaints worsen as the absorbed volume increases.
of prostate resected, and blood loss12 (see Fig. 33.2).
Complaints associated with increasing intravascular ab-
Surprisingly, the experience of the surgeon has not been
sorption include feeling bad, nausea, hypotension, and
validated to be of importance.4,20 Smoking is the only
chest pain. Extravasated fluid in the retroperitoneal or
patient factor associated with increased risk for high
peritoneal space provokes shoulder, back and chest pain,
volume absorption.21 Moreover, antidiuretic hormone
abdominal discomfort and distension, and shortness of
produced by the stress of surgery22 and increased renin
breath. Altered mental status, irritability, bradycardia,
and aldosterone secretion23 may also contribute to volume
and hypotension are more serious and are typically later
expansion by promoting water retention.
signs of increasing absorption. These complaints and ob-
Plasma volume expansion from absorbed irrigant
servations may emerge subtly or dramatically. Many mild
can occur rapidly (3.3 L in 20 minutes)24 during TURP
forms of the TURP syndrome probably go unrecognized.
and likely contributes to the hypertension and (reflex)
Moderate-sized absorptions typically only give rise to
bradycardia often seen.25 Absorption rates can reach
200 mL per minute.12 A patient with poor left ventricular symptoms 30 to 60 minutes after surgery.28 If patients
function may develop pulmonary edema when challenged are to be protected from the TURP syndrome, effective
with such an acute circulatory volume overload.26 A report and early means of quantifying absorption in patients un-
of five patients with severe TURP syndrome (two deaths, der general anesthesia and in awake, but asymptomatic,
two seizures, and one ventricular arrhythmia) found patients is required.
no significant variations in serum osmolalities before The TURP syndrome has been reported to occur as
quickly as 15 minutes from surgical start.24,29 Reliable
detection of such a rapid onset would be best achieved,
6,000 ideally, with a continuous method for monitoring ab-
sorption. If only intermittent monitoring methods are
5,000 available, sampling intervals should preferably not exceed
15 minutes during resection.
Fluid absorption (mL)
4,000
Serum Sodium
3,000
r = 0.36
A common method for detection of acute volume gain
2,000 of an electrolyte-free solution is serum sodium mea-
surement.24 A decreased serum sodium concentration is
1,000 indicative of the absorption of sodium-free irrigation fluid.
This method has the advantage of being widely available.
0 Its disadvantages are that it is cumbersome to obtain
0 1,000 2,000 3,000 4,000 serial blood samples free of intravenous fluid contam-
Blood loss (mL) ination, and a delay between collecting a sample and
receiving the laboratory results may result. Furthermore,
FIGURE 33.2 Poor relationship between blood loss and fluid because of its slow intravascular absorption, it may take 2
absorption in 817 patients undergoing TURP. The risk of having to 4 hours following extravasation of fluid before the de-
fluid absorption is increased in prolonged and bloody operations creased serum sodium concentration can be measured.30
(data from Ekengren J, Hahn RG. Blood loss during At present, bedside testing is possible (i-STAT, Abbott
transurethral resection of the prostate as measured by the Laboratories. Abbott Park, Illinois), and blood samples
HemoCue photometer. Scand J Urol Nephrol. 1993;27:501.), but can be analyzed for serum sodium at intervals less than
the relationship is not very strong (unpublished data). every 5 minutes, if desired. This approach will be use-
less for detecting absorption if an isotonic electrolyte
478 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
irrigation solution is used, because no change in serum compromised or delayed by increasing the proportions of
sodium would be expected, regardless of the volume of extravasated fluid.
fluid absorbed.
Miscellaneous Techniques
Breath Alcohol Levels
Also described but never widely adopted has been
Measurement of breath alcohol levels is an inexpensive monitoring the trends of the central venous pressure,37
and a simple method of monitoring irrigant absorption. plasma electrolyte concentrations (e.g., magnesium and
Glycine 1.5% mixed with ethanol 1% is an irrigation solu- calcium),38 irrigation solute concentration (glycine39,40
tion (Baxter Healthcare Ltd, Thetford, Norfolk, England) and sorbitol41 ), and transthoracic impedance.42
made for this purpose, and is available in some coun-
tries.31 Ethanol has also been added to mannitol and Prevention
sorbitol solutions32 and can be utilized in isotonic elec-
trolyte solutions. The technique does require the use of The principal strategies for reducing the incidence and
a breathalyzer to intermittently (every 5 to 10 minutes) severity of the TURP syndrome focus on reducing the
measure breath alcohol levels. Absorption has occurred three factors generally believed to be the most significant
when the breathalyzer detects any alcohol in the exhaled variables influencing absorption: (i) The magnitude of the
breath. This technique identifies intravascular absorption intravesicular fluid pressure; (ii) the number of prostatic
exceeding approximately 150 mL over 10 minutes. An venous and surgical perforation openings for fluid to gain
estimate of the volume absorbed intravascularly can be entrance; and (iii) the duration of high-pressure fluid
made from the breathalyzer value by using a published exposure to open veins or the retroperitoneal space.
nomogram.33 When fluid containing ethanol 1% is ex- Intravesicular pressure is generally lowest when a
travasated, a 15- to 20-minute delay can be expected trocar, suprapubic drainage catheter, or a continuous flow
before the breathalyzer measures any tracer.30 Although resectoscope is in use. Limiting irrigation fluid height
2% ethanol is more sensitive for detecting absorption, its further defines the maximum intraoperative bladder
use increases the possibility of intoxication when larger pressure that can occur. Restricting surgery time to 1 hour
volumes are absorbed. Ethanol, in any concentration, may and leaving a rim of tissue on the capsule until near the
not be the preferred absorption tracer if the patient is a end of the procedure where it can either be left (if signs
recovering alcoholic. To avoid all potential toxicity and of TURP syndrome are evident) or removed all at once
make the monitoring fully automatic, recent clinical trials have been advocated as ways to reduce the time that a
have added tracer doses of nitrous oxide to irrigating fluid large number of prostatic sinuses are open and capable of
and closely monitored the patients breath.34 absorbing fluid.10,43
Other efforts have attempted to limit the absorption
Volumetric Fluid Balance of irrigation fluid or blood loss through pharmacologic
means rather than further limiting absorption by redesign-
Calculation of volumetric fluid balance31 is another ing instruments or procedural techniques. Vasopressin,
method of assessing fluid absorption. Operating room a potent vasoconstrictor, injected into the prostate be-
staff track the volume of irrigant used and compare this fore TURP has been advocated as a possible way to
to the volume recovered from the surgical field. The limit absorption of the irrigant.44 These authors reported
difference represents the maximum volume absorbed. that insignificant absorption occurred among a series of
The technique is simple but often inaccurate because 36 patients who, after transrectal injection of vasopressin
of the incomplete recovery of irrigant, discrepancies in (10 units) into the prostate, underwent TURP, with water
labeled and actual irrigant bag volumes, and inaccurate as the irrigation fluid. Premedication for 4 weeks with
calibrations in the collection container. Estimates of blood chlormadinone acetate, an orally administered antiandro-
and urine losses must also be added to the collected gen, or a 3-month course of finasteride, a 5-reductase
volume, which further compromise the accuracy of this inhibitor that lowers dihydrotestosterone and angiogen-
method. The volumetric balance measures the sum of esis, have also been reported to significantly reduce the
intravascular absorption and extravasation and does not blood loss per gram of resected tissue.7
make a distinction between them.
had increased. After 30 to 35 minutes, when the rate of The TURP syndrome may develop and progress over
irrigant absorption slowed, flow from the plasma to the several hours, depending on the volume and distribution
interstitium increased to an average of 75 mL per minute. of fluid absorption between the intravascular spaces and
Three patients then became suddenly hypotensive (sys- intra- and retroperitoneal spaces. The rate of progression
tolic arterial blood pressure of 80 mm Hg or less) and, may be hastened by the release of prostatic endotoxins.
of these, two became hypotensive again after the proce- Despite routine antibiotic prophylaxis and negative pre-
dure. Three other patients suddenly became hypotensive operative urine cultures, endotoxemia can develop in up
within the first postoperative hour. A similar effect has to 45% of patients during TURP.50 Although usually man-
been demonstrated in animal studies45 (see Fig. 33.3). ifesting postoperatively, signs of sepsis have presented
There are several paths by which the acute fluid intraoperatively.50,51 The recommendation is to treat pa-
absorption and hypervolemia that initiate the syndrome tients with preoperative bacteria for several days before
can progress to a prolonged and potentially profound undertaking elective TURP.
hypotension. Severe hyponatremia by itself does not Bleeding and red blood cell destruction are additional
account for the hypotension.46,47 However, hypervolemia, sources of volume and oxygen-carrying capacity losses
which produces a dilution of serum electrolytes and that may further stress the heart. Absorption of distilled
proteins that manifests as hyponatremia, hypocalcemia, water during TURP can cause acute hypoosmolality
and low serum osmolality, combined with hypertension, with massive hemolysis.52 The hemoglobinemia that
may lead to a dramatic water flux along osmotic and follows such hemolysis, coupled with hypotension, can
hydrostatic pressure gradients out of the intravascular cause acute renal failure, and even death.53,54 The same
space.28 Together, these factors may be sufficient to impair mechanism may also cause permanent renal damage
cardiac function. Moreover, studies in mice and pigs in response to extravasation of distilled water.55 In
indicate that a severe volume overload damages the heart general, extravasation has a higher tendency than direct
by causing severe interstitial dilatation, hypoxic changes, intravascular absorption to cause arterial hypotension.2
and disruption of the histoskeleton.45,48,49 Any cardiac
compromise will only be further aggravated by other
conditions that can accompany symptomatic fluid gain in
a patient: High sympathetic blockade induced by regional ELECTROLYTE-FREE
anesthesia, liberation of cellular potassium, or acidosis. IRRIGATION SOLUTIONS
Pulmonary edema and hypovolemic shock develop when
fluid translocates into the lungs and other tissues. Distilled Water
Distilled water was the original irrigation used for
Glycine 1.5% Mannitol 5%
TURP.54 Without ionic solutes, distilled water provided
the best viewing field and resectoscope performance.
160
Electrolyte-free water did not disperse the electric cutting
current of the resectoscope and produced the least optical
Cardiac output (percent of baseline)
140
refraction. Its osmolality of zero reduced turbidity and
120 maintained a clear visual field by hemolyzing red cells
that had bled into the bladder during resection.
100 Water is said to have several additional advantages
over nonhemolytic solutions and may still enjoy a
80 resurgence of popularity if the claims that new techniques
reduce absorption are substantiated. Without dissolved
60 solutes, water provides more stable hemodynamics by
rapidly diffusing out of the intravascular compartment
40 Infusion into the interstitial compartment and, by the same
mechanism, produces less hyponatremia.56,57 The clinical
20 validity of these benefits is dubious, as the consequences
0 30 60 90 120 of diffusion of free water into brain and lungs, for example,
Time (min) are undesirable and serious. Hence, if distilled water
must be used, it should be restricted for use with smaller
FIGURE 33.3 Hypokinetic circulation after prolonged resections or with techniques known not to cause massive
administration of irrigating fluid. Cardiac output during and absorption. However, where procedural costs are a
after a 90-minute intravenous infusion of 100 mL/kg/hour of significant concern, water remains the least expensive and
either glycine 1.5% or mannitol 5% in 17 pigs (mean values). perhaps the only choice.57 When water is to be used, the
After the infusions, the mean arterial pressure decreased by considerable risks to the patient need to be acknowledged
20%. (Data derived from: Sandfeldt L, Riddez L, Rajs J, et al. and awareness of clinical events must not wane.58
High-dose intravenous infusion of urological irrigating fluids Complications arising from intravascular absorption
containing glycine and mannitol in the pig. J Surg that included intravascular hemolysis, renal failure and
Res. 2001;95:114.) death led to the introduction and wide acceptance of new
nonhemolyzing irrigation solutions. Glycine, sorbitol, and
480 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
mannitol are electrically nonconductive, but osmotically derangements of hyponatremia, hypoosmolality, hyper-
active solutes, that were added to irrigation fluids and ammonemia, and hyperglycinemia. The choices made for
reduced the occurrence of significant hemolysis and death anesthetic interventions can further impair the neurologic
by more than 50%.59,60 status and complicate the diagnosis of TURP syndrome.
The superimposition of centrally active drugs, such as
Nonhemolytic Irrigation Solutions propofol, benzodiazepines and narcotics,24 or a spinal
anesthetic causing hypotension that leads to nausea and
The nonhemolytic irrigation solutions (and calculated vomiting, can confuse the clinical picture and delay the
osmolalities) most often employed with the tradi- diagnosis.
tional monopolar resectoscopes include glycine 1.5%
(200 mOsm per L), sorbitol 3% (165 mOsm per L or
170 mOsm per kg), sorbitol 2.5% with 0.54% manni-
tol (167 mOsm per L), sorbitol 5% and mannitol 5% HYPONATREMIA
(275 mOsm per L), and dextrose 5% (252 mOsm per L).
The osmolality reported on the irrigant bag liners is cal- Profound hyponatremia with TURP has been impli-
culated and assumes there are no interactions between cated as the cause of visual aberrations, encephalopathy,
solute particles. Because these interactions do occur, a so- pulmonary edema, cardiovascular collapse, seizure, and
lutions measured osmolarity will be slightly lower (10 to death.61,62 The rate of irrigant absorption ranges from 10
20 mOsm per L) than the calculated osmolarity value. to 30 mL per minute of resection time.63 However, the
These osmolarities are generally hypotonic compared intravascular absorption rate alone can exceed 200 mL
with lactated Ringers solution (273 mOsm per L) and per minute.12 The incidence of serum sodium concentra-
0.9% normal saline (308 mOsm per L) that are used with tion <125 mEq per L following TURP may reach 10%43
bipolar resectoscopes. Although the osmotic activity of the with a mortality of symptomatic hyponatremia (headache,
irrigation solutions is reported as osmolarity, a measure of nausea, vomiting) of approximately 40%.64 The intravas-
the osmoles per liter of solution, it is conventional to refer cular absorption of irrigant from the perivesicular and
to serum osmotic activity as osmolarity, a measure of the retroperitoneal spaces may continue for up to 24 hours
osmoles per kilogram of solvent (water). When the concen- after surgery.13 The TURP syndrome may develop at any
tration of solute is very low, as is the case for electrolytes time while absorption continues.43,6567
in serum, osmolarity and osmolality become nearly the Dilutional hyponatremia may be aggravated by two
same, and the terms are often used interchangeably. other mechanisms. Electrolytes diffuse from the extracel-
lular fluid space to any accumulation of extravasated
nonelectrolyte fluid.2 If the patient is given diuretic
medication, additional electrolytes will be lost in the
What Are the Plasma Solute urine.23,38,46,68,69 Diuretics have been implicated as factors
for the rapid onset of hyponatremia.68 When used rou-
Effects Associated with the tinely or to treat hypervolemia following TURP, diuretics
Transurethral Resection of the may worsen hypoosmolality and provoke hypotension42
(see Fig. 33.4). Diuretics acting on the ascending loop of
Prostate Syndrome? Henle have an onset within 2 to 5 minutes. They inhibit
chloride uptake, causing urinary sodium loss, and pro-
Solute changes occurring during and after TURP may mote salt-wasting after TURP.15,23,38 Mannitol also wastes
alter the patients condition and neurologic function inde- sodium during the first 12 hours following TURP.70 There-
pendently of the volume-related effects (see Table 33.3). fore, if loop or other salt-wasting diuretics are used to treat
Most notably, central nervous system (CNS) symptoms as- hypervolemia, strong consideration should be given to a
sociated with TURP have been attributed to the metabolic concomitant infusion of salineeven in the presence of
TABLE 33.3 Changes in Neurologic Function during TURP Can be Triggered by One or More of
the Listed Factors
1.02 Infusion
1 HYPOOSMOLALITY
0.98 The magnitude, onset, and rate of correction of hy-
ponatremia are the most common aspects of the TURP
0.96
syndrome discussed in the literature. A strong case can
0.94
be made for the argument that the crucial physiologic de-
rangement of CNS function during TURP syndrome is not
0.92 from hyponatremia per se, but rather from acute hypoos-
0 20 40 60 80 100 molality. Acute deviations from normal serum osmolality
of 285 mOsm per kg will drive water into the space with
Time (min)
higher osmolality.
FIGURE 33.4 Rationale for the treatment of fluid absorption.
The fractional changes in blood hemoglobin and serum sodium Central Nervous System Effects
during and after infusing 1.15 L of glycine 1.5% intravenously in
10 volunteers over 20 minutes (mean, SD). At 100 minutes, they Serum sodium concentration should not be expected to
had voided 1.06 L which contained 31 mmol of sodium. Because have much influence on the CNS when one considers
hemoglobin was restored and the infused fluid excreted, the that the bloodbrain barrier, with an effective pore
appropriate treatment of the residual hyponatremia is by size of 8 A, is essentially impermeable to sodium ions,
replacing the lost sodium rather than by giving furosemide. but freely permeable to water. Furthermore, cerebral
edema is not caused by lowered serum colloid oncotic
pressure, but by lowered osmolality77 (see Fig. 33.5).
Symptoms of water intoxication in rabbits, induced by the
near-normal serum sodium concentrationsduring the administration of vasopressin and 2.5% glucose solution,
first 12 postoperative hours.15,23 were reversed by administering osmotically active agents
Hyponatremia during TURP is common, the degree such as urea and mannitol without correcting the serum
of which is related to the volume of irrigant absorbed sodium concentration.76
and the route of uptake. When absorbed by the direct Theoretically, a comparison of neuronal membrane
intravenous route, 1 L of fluid decreases serum sodium equilibrium potentials at normal and hyponatremic condi-
by 7 to 10 mEq per L, whereas extravasation is followed tions is possible with the Nernst equation (see Equation 1),
by a more modest reduction which peaks after several
hours.33 In the literature, the serum sodium concentration ENa = RT/F ln[Na]o /[Na]i (33.1)
decreased from 6 to 54 mEq per L, with an incidence
ranging from 7% to 26%.19,70 Decreases from normal in which ENa denotes the membrane equilibrium poten-
preoperative levels to 113 mEq per L and 104 mEq per tial, R is the universal gas constant, T is the absolute
L after just 15 minutes of resection using nonhemolytic temperature in Kelvin, F is the Faraday constant, and
irrigants have been reported.24,29 [Na]o and [Na]i represent sodium ion concentrations out-
Severe hyponatremia has been associated with hemo- side and inside the cell, respectively.
lysis and renal failure, cardiovascular and electrocardio- The Nernst equation predicts that the reduction in
gram (ECG) changes, respiratory compromise, rhabdomy- extracellular sodium concentration seen with a profound
olysis, seizure, coma, and death.38,47,62,71 73 However, TURP syndrome should only minimally alter neuronal
many patients experiencing similarly profound hypona- excitability. Replacing the [Na+ ]o value of 145 mmol per
tremic exposures have survived without any symptoms or L with 100 mmol per L in the Nernst equation only lowers
signs of water intoxication.39,41,70,74,75 For example, five the calculated transmembrane equilibrium potential of
patients had decreases in serum sodium concentration, 67 mV to 57 mV. Therefore, hyponatremia should not
from 34 mEq per L to 54 mEq per L, during TURP but substantially contribute to neuronal excitability when
had neither the TURP syndrome nor significant changes examined independently from serum osmolality, even
in serum osmolality.70 When 3% mannitol irrigation was when these changes are of the magnitude typically
used on one patient with TURP, serum sodium concen- associated with severe TURP syndrome.
tration decreased from 133 mEq per L to 99 mEq per L.74 One method of demonstrating the independent neu-
Osmolality was measured postoperatively at 290 mOsm rophysiologic effects of serum sodium concentration and
per kg, but was calculated at 239 mOsm per kg. This osmolality has been to measure field potentials from pre-
difference was attributed to the osmotic effect of manni- pared brain slices. Field potentials can be triggered or
tol not accounted for by the calculation. Another patient arise spontaneously and represent the voltage generated
survived a serum sodium change from a preoperative by the synchronous discharge of many neurons aligned
level of 136 mEq per L to 76 mEq per L postoperatively in parallel. Hypotonic saline (Na+ of 123 mmol per L and
482 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Mannitol 5%
with the largest serum sodium concentration decreases (by
Mannitol 3%
34 mOsm per kg to 54 mOsmol per kg) had no changes in
295 serum osmolality and no signs of the TURP syndrome.
285
How Are Hyponatremia
and Hypoosmolality Prevented
275 and Treated?
In theory, pretreatment with hypertonic saline may de-
220 crease the degree of dilutional hyponatremia and hypo-
osmolality when electrolyte-free fluid absorption occurs.79
Treatment for derangements of serum electrolyte and
180 osmolality should be based on an understanding of
Intracerebral pressure
(percent of baseline)
develop.8284 This entity is not well understood. Although or more are warranted, but only until symptoms abate.86
it is still referred to as central pontine myelinolysis, Treatment with diuretics along with hypertonic saline in-
the terminology changed when demyelination following fusions may be conducted following moderate to massive
a sudden elevation of serum osmolality was found in amounts of fluid absorption, provided that the circulatory
extrapontine areas, as well as in the pons. Osmotic situation is stable.
stress causing acute shrinkage of neuronal cells and
the release of myelinotoxic substances that result in
loss of oligodendrocytes and permanent demyelination
has been suggested as the cause.85,86 Therefore, cerebral What Is the Role of
demyelination appears less a disease of overly rapid serum
sodium correction and more as a manifestation of an acute
Hyperglycinemia?
hyperosmotic challenge.
Clinical presentation can vary, but signs of ODS in-
clude seizures, pseudobulbar palsy, dysphagia, dysarthria, HYPERGLYCINEMIA
vertigo, paraparesis or quadriparesis, spasticity, confu-
sion, or coma, typically presenting several days to a Glycine is a major inhibitory neurotransmitter such as
week after the osmotic stress is encountered.86 Mag- -aminobutyric acid (GABA) in the spinal cord and
netic resonance imaging may demonstrate the progression midbrain.90 It probably also has a significant role in
of lesions for several weeks. When ODS is suspected, higher cortical neurotransmission.90,91 That glycine may
plasmapharesis has been proposed as a therapeutic op- play a role in post-TURP encephalopathy and seizure is
tion, presumably to remove the speculated myelinotoxic suggested by nonketotic hyperglycinemia or glycine en-
substances.85 cephalopathy (GE). This disease is a heritable affliction
characterized by a defect in the glycine cleavage en-
zyme system,92 disturbed electrophysiologic function,93
intractable seizures, lethargy, spasticity, mental retarda-
TREATMENT tion, and death within the first few months of life.94 These
The need for treatment of TURP-associated acute hy- patients have plasma glycine levels up to 10 times greater
ponatremia must be assessed in an orderly manner. than those of normal infants (mean patient value range
Ideally, serum sodium concentration or exhaled ethanol 266 to 2, 027 mol per L; normal infant level = 209 mol
concentration should be reported together with osmolal- per L).94,95
ity in patients undergoing TURP using a nonhemolytic, Glycine may promote encephalopathy and seizure in
electrolyte-free irrigation (e.g., glycine, mannitol, or sor- GE92,96 through N-methyl-D-aspartate (NMDA), an exci-
bitol), because the measured serum sodium concentration tatory neurotransmitter. Magnesium can block NMDA-
or ethanol values may not reflect serum osmolality.27,70,74 receptor activity which is markedly potentiated by
Treatment of an asymptomatic and hyponatremic patient glycine.97 Therefore, along with its role as an inhibitory
with hypertonic therapy without knowing the serum os- transmitter, glycine may facilitate excitatory transmission
molality risks provoking ODS.87,88 Cautious surveillance in the brain through an allosteric activation of the NMDA
with repeated hematologic studies (including serum os- receptor.96
molality) to confirm recovery is recommended. High plasma glycine levels may be harmless, but
Treatment is always indicated in a symptomatic pa- high brain glycine levels can be fatal.95 Glycine may
tient. Although there is a growing consensus, the answer to reach high levels in the brain during TURP performed
how that treatment is most safely accomplished remains with 1.5% glycine irrigation. Toxic manifestations of
unsettled. Overly cautious and slow treatment for symp- glycine become manifest in humans (nausea, vomiting,
tomatic acute hyponatremia (0.7 mmol/L/hour)67,72 has headache, malaise, and weakness) at an infusion rate
been associated with a higher morbidity and mortal- of 3.5 mg/kg/minute.98 A prospective trial of 30 patients
ity than rapid correction (1.0 mmol/L/hour).68 Experts undergoing TURP found that 92% had an elevated serum
generally believed that safely correcting chronic hypona- level of glycine.99 Furthermore, serum glycine after TURP
tremia must be done far more slowly than the correction has been reported at levels greater than 14, 300 mol per
for an acute hyponatremia.86 L.39,43,100 This concentration is 17 times greater than that
The presence of symptoms has been described as in children dying from GE and more than 65 times that
the single, most important factor for determining mor- in normal adults (normal adult level = 219 mol per L).95
bidity and mortality from hyponatremia.64 The safest Furthermore, men older than age 40 have 29% higher
treatment of hyponatremia and hypoosmolality may be cerebrospinal fluid levels of free glycine than their younger
symptomatic.89 Osmolality should be monitored and cor- counterparts.101
rected aggressively, only until symptoms substantially
resolve; then correction should be continued slowly. To Visual Disturbances
assure the least chance of causing ODS, it is preferred
that the rate of serum sodium correction not exceed Visual disturbances in TURP syndrome vary in severity
0.5 mEq/L/hour or 8 mEq/L/day and that the acute cor- from blurred vision102104 to complete blindness.61,105,106
rection not exceed a serum target of 120 mEq per L. If the Some patients present with sluggish or fixed and di-
symptoms are severe, initial rates of up to 2 mEq/L/hour lated pupils,62,103,105 109 and total loss of light/dark
484 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
discrimination.105,106,109 Vision returns to normal within after the injection of similar volumes of retroperitoneal
24 to 48 hours as glycine levels approach normal val- water or lactated Ringers solution. This study did not
ues.43,99,105 This change is predictable because the half-life investigate whether the kidneys would eventually recover
of glycine is approximately 85 minutes;41 therefore, reas- from the apparent toxic insult. Hyperoxaluria has also
surance that unimpaired vision is expected to return is been proposed as an indirect route whereby glycine,
reasonable. Interestingly, several investigators have re- through its metabolism into oxalate and glycolate, could
marked on the unusual calmness of their patients facing cause renal failure in susceptible patients.118,119
what would seem to be an extremely frightening compli-
cation.105,109 Atropine106 or hyponatremia and cerebral Cardiac Effects
edema from overhydration may contribute to these visual
disturbances.61,62,104,108,110 Patients with cortical blind- Glycine (and water) almost exclusively is the irrigant
ness, on the other hand, lose all visual sensation (light associated with the TURP syndrome. Animal model in-
perception and the blink reflex), but retain the pupil- vestigations may explain why. Glycine solutions likely
lary responses to light and accommodation.103 Although injure cardiomyocytes through two mechanisms: Osmotic
it is difficult to separate the effects of serum sodium swelling and by a direct dose-dependent cardiotoxic ef-
concentration from those of other retinal transmitters, fect.120,121 This cardiotoxicity has been demonstrated in
sodium appears to play only a minor role in the visual several live mouse models where survival after intravenous
disturbances.111 infusions of glycine 1.5% was compared against survival
Glycine is a major inhibitory neurotransmitter in the with other solutions.122,123 Glycine-containing solutions
retina, just as it is in the spinal cord and brainstem.90,91 reached only a 66% or 33% survival rate compared with
It is found in the inner plexiform layer, amacrine, sorbitol-mannitol infusion or mannitol 5% infusion, re-
and bipolar cells of the human retina and causes spectively. These studies suggested that glycine promotes
hyperpolarization of ganglion cells.111 The sensitivity of bradycardia and death, independent of hyponatremia or
oscillatory potentials of the electroretinogram102,112 and hypoosmolality. The unfavorable effects of glycine on the
visual evoked potentials to glycine in the absence of large cardiovascular system have been described in mice, rats,
osmolality changes has been demonstrated.102 Infusions rabbits, sheep, pigs, and dogs. Glycine administration to
of glycine in volunteers correlate the acute deterioration of humans yielded slightly more hemodynamic effects after
visual acuity with delays of the signal transmission in the glycine 1.5% compared to sorbitol-mannitol and mannitol
retinal-cortical pathways.113 Therefore, glycine appears to 3%.124 A study using blinded fluid bags during 394 TURPs
affect the retinal physiologic condition independent of the demonstrated more irrigation-associated symptoms after
cerebral edema caused by hypoosmolality. absorption of glycine 1.5% than after mannitol 3%, in
Serum glycine levels in patients with visual changes particular with respect to neurologic symptoms such as
have been documented over a wide range. Such re- nausea, vomiting, and dizziness.125
ports have led to speculation on the existence of a
serum glycine concentration threshold for symptomatic
visual impairment (>4,000 mol per L)102 and blindness Encephalopathy and Seizures
(>13,734 mol per L).105
Glycine has not been reported to cause lasting impair- Glycine may be involved with TURP encephalopathy
ment of vision following TURP. However, in sheep, serum and seizure through its positive action on the NMDA
glycine concentrations in excess of 5,000 mol per L dam- receptor-channel system, as it is in GE.92,126 Seizures
aged the eye neurologically.114 Sustained (7 hours) eleva- following TURP associated with hyponatremia and hypo-
tion of vitreous glycine, despite a falling plasma level have osmolality are likely to be resistant to benzodiazepine
been observed after a single intravenous administration115 and anticonvulsant therapy and, in fact, such treatment
with plasma glycine (7,000 mol per L to 21,000 mol per may provoke apnea. Theoretically, an NMDA-receptor
L), comparable to maximal values reported in men under- antagonist126 or glycine antagonist92 may a better choice.
going TURP (24,800 mol per L).40,41,100,102,105 Sheep may Magnesium exerts a negative control on the NMDA
be more susceptible to the effects of high serum glycine, receptor.96,126 A serum magnesium level lowered by dilu-
because their plasma-to-cerebrospinal fluid concentration tion may make individuals more susceptible to seizures.
gradient is 16:1 compared with 50:1 for humans.115 In one Magnesium levels may be dramatically lowered follow-
study of seven sheep that received intravenous glycine, ing TURP in patients who have been treated with a loop
only the animal with the highest glycine level in the cere- diuretic.38 Therefore, a trial of magnesium therapy for
brospinal fluid died.116 seizures in patients in whom a glycine irrigant was used
during TURP deserves consideration, especially if mea-
sured osmolality is near normal.
Kidney Function
Poor urinary excretion after TURP is often associated with
arterial hypotension,2 but glycine may also exert toxic HYPERAMMONEMIA
effects on the kidney.117 A study in rats found histologic
evidence of glycine toxicity in their kidneys 6 hours after The portal bed and kidneys can metabolize glycine that
either the intravenous or intraperitoneal administration of gains intravascular access127 along two pathways. The pri-
large doses of 1.5% glycine solution. No toxicity was found mary pathway utilized by the liver and kidneys is oxidative
C HAPTE R 33/TU R P SYN DROM E 485
cleavage by glycine synthase,128 which leads to the for- Metabolism of the large sorbitol load likely produced the
mation of ammonia. Other potentially toxic products of hyperglycemia and mild acidosis. Nausea and vomiting
glycine metabolism include methylene tetrahydrofolate, may have been a result of hypervolemia and the mild
serine, and glyoxylic acid.40 The normal brain contains sedation from droperidol given to treat the nausea. It
a similar glycine cleavage enzyme system that causes appears that the enormous dose of sorbitol and the
oxidative deamination of glycine into carbon dioxide, a hyponatremia were reasonably well tolerated, whereas the
one-carbon fragment, and ammonia.95 large absorbed volume was predominantly responsible for
Absorption of glycine irrigation during TURP leads to the clinical picture.
elevations of serum ammonia, because patients undergo- Norris et al. reported a series of patients with TURP
ing retropubic resections without glycine do not develop in whom several TURP syndromes developed when using
hyperammonemia.100 Hyperammonemia to levels exceed- sorbitol-mannitol for irrigation.27 Fatal cases of fluid
ing 150 mol per L (normal 11 to 35 mol per L) has been absorption during hysteroscopy have been reported when
implicated in contributing to visual disturbances, muscle using sorbitol 3%.136,137
weakness, and encephalopathy.129 Serum ammonia levels Experimental studies show that infusion of fructose
following TURP with glycine 1.5% and in association with elicits lactic acidosis in a dose-dependent way; intra-
TURP syndrome have been reported at 500 mol per L venous infusion in healthy volunteers at a rate exceeding
and higher.130,131 Volunteers given 1 L of glycine 2.2% 1 g/kg/hour increased the plasma lactate from <1 mmol
developed blood ammonia levels ranging from 49 mol per L to 7 mmol per L in 90 minutes, which resulted in
per L to 354 mol per L, the elevation of which correlated a moderately severe metabolic acidosis.138 On assuming
with the development of mental symptoms.3 However, a that half the infused sorbitol is metabolized to fructose,
marked elevation of blood ammonia in response to glycine this rate of infusion would correspond to absorption of
probably occurs only in approximately 10% to 20% of the 3 L of sorbitol 5% per hour, which is feasible but very
population, which suggests a gene-linked effect.99,132 rare during surgery. Intolerance to fructose is a rare but
Factors limiting the elevation of plasma ammonia deadly condition.139,140
concentration when glycine irrigants are used include
hepatic arginine levels. Arginine acts in the liver to pre-
vent hepatic release of ammonia and accelerate ammonia MANNITOL
removal.127 The time necessary to deplete endogenous
Mannitol is a hexitol sugar and an isomer of glucose. It is
arginine stores needed for the rapid conversion of am-
commercially available in 3% and isotonic 5% concentra-
monia to urea may be as little as 12 hours, which
tions. It does not undergo metabolism before excretion
approximately equals the typical preoperative fast time.133
by the kidney. Its serum half-life with normal renal
The prophylactic administration of intravenous L-arginine
function is 100 minutes. Mild TURP syndrome (brady-
markedly lowered the rise in blood ammonia concentra-
cardia, hypotension) has been described with absorption
tion in fasting patients receiving intravenous glycine. The
of mannitol 3% (165 mOsm per L), likely because of
infusion of L-arginine with, or at the conclusion of, glycine
its hypoosmolarity.28,141 Studies in animals,120,123 volun-
administration prevented further increases in blood am-
teers,3,124 and patients125 show that mannitol solution is
monia concentration and accelerated the return of levels
better tolerated than glycine, but mannitol 5% expands the
to normal.127,133 Doses between 4 g (20 mmol) infused
plasma volume more than glycine 1.5%.3,45 This irrigating
over 3 minutes and 38 g (180 mmol) infused over 120
fluid should be used with caution in patients with poor
minutes were recommended. No toxicity was noted with
kidney function due to a markedly prolonged half-life of
either of these regimens.133
mannitol in these cases.
rate of 50 mL per kg over 1 hour in 20 volunteers.144 They myomectomy, and adhesiolysis are procedures at partic-
reported tiredness, problems to think, and abdominal ularly high risk of excessive fluid absorption. Absorption
pain in more than half the patients receiving normal in excess of 2 L occurs in up to 6% of procedures. As
saline, whereas less frequently after lactated Ringers so- a rule, fluid absorption is always possible when the pro-
lution.144 Adverse effects after major surgery are also cedures employ distending fluids. Fluid can gain direct
more common after fluid therapy based on normal saline intravascular access or enter the peritoneal cavity by way
than after lactated Ringers solution.145 Most evidence of the fallopian tubes or as a consequence of uterine per-
therefore suggests that the adverse effects from the rapid foration.129,153 Consistent with the experience gained with
absorption of normal saline during TURP would be an TURP, the volume of fluid absorbed has been linked to
issue, and that lactated Ringers solution could be a better the intrauterine pressure. When distending pressures are
irrigation solution than normal saline. The known draw- kept below 70 mm Hg, absorption is limited.
backs of normal saline is that the fluid induces a moderate Women seem particularly susceptible to acute os-
metabolic acidosis and is less readily excreted than lac- motic stress, including ODS, during correction of hy-
tated Ringers solution.144146 The latter effect is probably ponatremia. This process seems likely because of sex
due to a regulatory effect of the chloride ion on glomerular differences in cellular ion pump capacity,64 whereby cells
filtration.147 Moreover, normal saline expands the plasma in a womans hormonal milieu do not osmoregulate as
volume by 10% to 15% more than the same amount of quickly as that in a man. In fact, premenopausal women
lactated Ringers solution,146 which potentially aggravates have a 25-fold increased risk of brain damage or death
the cardiovascular strain from volume overload. from hyponatremic encephalopathy compared to men.154
This sensitivity to osmotic stress suggests that a more
conservative correction rate for hyponatremia and hypo-
osmolality in women is warranted than for treatment of
What Are Other Features of the TURP syndrome.
ventilation under general anesthesia.32 Factors believed As long as irrigation fluid under pressure is used to aid
to be responsible for this increased absorption were low surgical visualization, there can be no guarantee that the
central venous pressure after spinal anesthesia coupled TURP syndrome, or its equivalent in nonprostatic surgery,
with the negative pressures associated with spontaneous will be avoided.58,156 In the future, as more surgeons
breathing. adopt methods that utilize nonhemolytic and electrolyte-
containing irrigation solutions, TURP syndrome will
transform into a condition caused by hypervolemia alone.
GENERAL ANESTHESIA
Patients who will not tolerate lying still or are unable to co-
operate may require general anesthesia with a laryngeal KEY POINTS
mask airway or tracheal intubation. Outcome compar-
1. TURP syndrome is caused by disturbance of intravas-
isons of general anesthesia to regional anesthesia have
cular volume and/or serum osmolality.
been conducted.155 Increased intraoperative hypotension,
2. Four questions to which you should know the answer
heat loss, postoperative hypertension, dysrhythmias, and
before starting a TURP:
postoperative pain in the general anesthesia group have
a. What is the irrigation fluid?
been reported. However, insignificant differences in in-
(1) What is the solute?
traoperative blood loss, postoperative behavior, mem-
(2) What is the solutions calculated osmolality?
ory, major morbidity, mortality, or long-term outcome
b. What is the bag height over the prostate?
occurred.
c. What type of resectoscope is being used?
(1) Monopolar
(2) Interstitial thermal ablative
(3) Minimally invasive
What Are the Take (4) Transurethral vaporization-resection of the
Home Messages? prostate
(5) Bipolar
d. In what mode is the resectoscope being used?
The TURP syndrome stems from rapid and complex
changes in intravascular volume, solute, and electrolytes. (1) Continuous flow
Diagnosis and treatment are challenging, because aberra- (2) Intermittent flow
tions of solute and volume can occur simultaneously and 3. The mechanisms precipitating TURP syndrome can
may suggest opposing diagnoses and treatments. occur in any operation where irrigation or distension
Before reflexively treating severe hyponatremia with fluid is used within the bodynot just during TURP.
hypertonic saline, an anesthesia provider should make an 4. Treat symptomatic (mental status changes, seizures,
effort to exclude hypervolemia with a near-normal osmo- hypotension) patients aggressively; treat asymptom-
lality.70 Symptomatic cardiovascular or pulmonary com- atic aberrant values (hyponatremia, hyperglycemia)
promise requires aggressive intervention. After adequate very slowly, if at all.
pulmonary gas exchange and hemostasis are established, 5. For patients with seizure after TURP with a glycine ir-
administration of blood, positive inotropic agents, cal- rigation fluid, consider a trial of magnesium therapy,
cium,38,107 magnesium,38 diuretics, or augmentation of especially if measured osmolality is near normal.
intravascular volume may be needed. 6. Sodium loss is an important source of hyponatremia
Although monitoring of serum sodium or exhaled a few hours after a TURP during which absorption of
ethanol concentration during TURP is common practice an electrolyte-free irrigating fluid has occurred.
and is effective for assessing intravascular absorption,
it is slow to reveal extravasation. It may be beneficial Acknowledgment
to monitor serum osmolality as well. Avoidance of
glycine-containing irrigants should reduce the risk of The author wishes to thank Ms. Kelly Spaulding for her
the TURP syndrome, because glycine has significant editorial assistance in the preparation of this chapter.
CNS and cardiotoxic effects. Hypoosmolality appears to
be the principal culprit contributing to the neurologic REFERENCES
and hypovolemic changes. Supportive care remains the
mainstay of management for the renal, pulmonary, and 1. Issa MM, Young MR, Bullock AR, et al. Dilutional hypona-
cardiovascular complications of the TURP syndrome. tremia of TURP syndrome: A historical event in the 21st
The decreasing incidence of the TURP syndrome over century. Urology. 1994;64:298.
2. Olsson J, Nilsson A, Hahn RG. Symptoms of the
the last 40 years demonstrates that prevention, detection,
transurethral resection syndrome using glycine as the
and treatment have improved. Despite this progress, many
irrigant. J Urol. 1995;154:123.
questions remain. Defining when acute hyponatremia 3. Hahn RG, Stalberg HP, Gustafsson SA. Intravenous infu-
becomes chronic eludes us, and the roles of hyponatremia sion of irrigating fluids containing glycine or mannitol with
and hypoosmolality in the neurologic manifestations of and without ethanol. J Urol. 1989;142:1102.
TURP syndrome and how to avoid ODS when correcting 4. Collins JW, MacDermott S, Bradbrook RA, et al. Is
them are yet to be described. using ethanol glycine irrigating fluid monitoring and good
488 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
surgical practice enough to prevent harmful absorption 25. Hahn RG. Fluid and electrolyte dynamics during develop-
during transurethral resection of the prostate? BJU Int. ment of TURP syndrome. Br J Urol. 1990;66:79.
2006;97:1247. 26. Sellevold O, Breivik H, Tveter K. Changes in oncotic pres-
5. Ozmen S, KoSar A, Soyupek S, et al. The selection of the sure, osmolality and electrolytes following transurethral
regional anaesthesia in the transurethral of the prostate resection of the prostate using glycine as irrigating solution.
(TURP) operation. Int Urol Nephrol. 2003;35:507. Scand J Urol Nephrol. 1983;17:31.
6. Tam PC. Transurethral resection of the prostate: Reaf- 27. Norris HT, Aasheim GM, Sherrard DJ, et al. Symptoma-
firming the gold standard (editorial). Hong Kong Med J. tology, pathophysiology and treatment of the transurethral
2005;11:76. resection of the prostate syndrome. Br J Urol. 1978;45:420.
7. Lynch M, Anson K. Time to rebrand transurethral resection 28. Hahn RG. Fluid absorption in endoscopic surgery. Br J
of the prostate? Curr Opin Urol. 2006;16:20. Anaesth. 2006;96:8.
8. Gravenstein D. Transurethral resection of the prostate 29. Hurlbert BJ, Wingard DW. Water intoxication after 15
(TURP) syndrome: A review of the pathophysiology and minutes of transurethral resection of the prostate. Anes-
management. Anesth Analg. 1997;84:438. thesiology. 1979;50:355.
9. Rassweiler J, Teber D, Kuntz R, et al. Complications of 30. Hahn RG. Ethanol monitoring of extravascular absorption
transurethral resection of the prostate (TURP)incidence, of irrigating fluid. Br J Urol. 1993;72:766.
management, and prevention. Eur Urol. 2006;5:969. 31. Hahn RG. Early detection of the TUR syndrome by marking
10. Harrison RH III, Boren JS, Robinson JR. Dilutional the irrigating fluid with 1% ethanol. Acta Anaesthesiol
hyponatremic shock: Another concept of the transurethral Scand. 1989;33:146.
prostatic reaction. J Urol. 1956;75:95. 32. Gehring H, Nahm W, Baerwald J, et al. Irrigation fluid
11. Balzarro M, Ficarra V, Bartoloni A, et al. The pathophysi- absorption during transurethral resection of the prostate:
ology, diagnosis and therapy of the transurethral resection Spinal vs. General anaesthesia. Acta Anaesthesiol Scand.
of the prostate syndrome. Urol Int. 2001;66:121. 1999;43:458.
12. Hahn RG, Ekengren JC. Patterns of irrigating fluid ab- 33. Hahn RG. Ethanol monitoring of irrigating fluid absorp-
sorption during transurethral resection of the prostate as tion. Eur J Anaesthesiol. 1996;13:102.
indicated by ethanol. J Urol. 1993;149:502. 34. Hahn RG. Nitrous oxide as a marker for irrigating fluid
absorption. An experimental study in the pig. Scand J Urol
13. Oester A, Madsen PO. Determination of absorption of irri-
Nephrol. 2003;37:281.
gating fluid during transurethral resection of the prostate
35. Lyon RP, St. Lezin M, Thomas C, et al. Monitoring of
by means of radioisotopes. J Urol. 1969;102:714.
body weight during transurethral resection of the prostate:
14. Gray RA, Moores AH, Hehir M, et al. Transurethral
Preliminary report. J Endourol. 1994;8:161.
vaporization of the prostate and irrigating fluid absorption.
36. Shipstone DP, Inman RD, Beacock CJ, et al. Validation
Anaesthesia. 2003;58:787.
of the ethanol breath test and on-table weighing to
15. Madsen PO, Naber KG. The importance of the pressure
measure irrigating fluid absorption during transurethral
in the prostatic fossa and absorption of irrigating fluid
prostatectomy. BJU Int. 2002;90:872.
during the transurethral resection of the prostate. J Urol.
37. Mebust WK, Brady TW, Valk WL. Observations on cardiac
1973;109:446.
output, blood volume, central venous pressure, fluid and
16. Hulten J. Prevention of irrigating fluid absorption during
electrolyte changes in patients undergoing transurethral
transurethral resection of the prostate. Scand J Urol Nephrol
prostatectomy. J Urol. 1970;103:632.
Suppl. 1984;82:1.
38. Malone PR, Davies JH, Stanfield NJ, et al. Metabolic
17. Hulten JO, Sundstrom GS. Extravascular absorption of
consequences of forced diuresis following prostatectomy.
irrigating fluid during TURP. The role of transmural Br J Urol. 1986;58:406.
bladder pressure as the driving pressure gradient. Br J 39. Norlen H, Allgen LG, Vinnars E, et al. Glycine solution as
Urol. 1990;65:39. an irrigating agent during transurethral prostatic resection.
18. van Renen RG, Reymann U. Comparison of the effect of two Scand J Urol Nephrol. 1986;20:19.
heights of glycine irrigation solution on serum sodium and 40. Zucker JR, Bull AP. Independent plasma levels of sodium
osmolality during transurethral resection of the prostate. and glycine during transurethral resection of the prostate.
Aust N Z J Surg. 1997;67:874. Can Anaesth Soc J. 1984;31:307.
19. Hahn RG, Ekengren J. Absorption of irrigating fluid and 41. Norlen H, Allgen LG, Wicksell B. Sorbitol concentrations
height of the fluid bag during transurethral resection of the in plasma in connection with transurethral resection of the
prostate. Br J Urol. 1993;8083:72. prostate using sorbitol solution as an irrigating fluid. Scand
20. Hjertberg H, Jorfeldt L, Schelin S. Use of ethanol as marker J Urol Nephrol. 1986;20:9.
substance to increase patient safety during transurethral 42. Casthley P, Ramanathan S, Chalon J, et al. Decreases in
prostatic resection: Screening investigation of irrigating electric thoracic impedance during transurethral resection
fluid absorption in four hospitals and comparison of expe- of the prostatean index of early water intoxication. J Urol.
rienced and inexperienced urologists. Urology. 1991;38:423. 1981;125:347.
21. Hahn RG. Smoking increases the risk of large scale fluid 43. Ghanem AN, Ward JP. Osmotic and metabolic sequelae of
absorption during transurethral prostatic resection. J Urol. volumetric overload in relation to the TURP syndrome. Br
2001;166:162. J Urol. 1990;66:71.
22. Chung HM, Kluge R, Schrier RW, et al. Postopera- 44. Sharma DP, Harvey AB. Does intraprostatic vasopressin
tive hyponatremiaa prospective study. Arch Intern Med. prevent the transurethral resection syndrome? BJU Int.
1986;146:333. 2000;86:223.
23. Donatucci CF, Deshon GE Jr, Wade CE, et al. Furosemide- 45. Sandfeldt L, Riddez L, Rajs J, et al. High-dose intravenous
induced disturbances of renal function in patients under- infusion of urological irrigating fluids containing glycine
going TURP. Urology. 1990;35:295. and mannitol in the pig. J Surg Res. 2001;95:114.
24. Aasheim GM. Hyponatraemia during transurethral surgery. 46. Berg G, Fedor EJ, Fisher B. Physiologic observations related
Can Anaesth Soc J. 1973;20:274. to transurethral resection reaction. J Urol. 1962;87:596.
C HAPTE R 33/TU R P SYN DROM E 489
47. Mitnick PD, Bell S. Rhabdomyolysis associated with severe 69. Crowley K, Clarkson K, Hannon V, et al. Diuretics after
hyponatremia after prostatic surgery. Am J Kidney Dis. transurethral prostatectomy: A double-blind controlled
1990;XVI:73. trial comparing furosemide and mannitol. Br J Anaesth.
48. Hahn RG, Zhang W, Rajs J. Pathology of the heart after 1990;65:337.
overhydration with glycine solution in the mouse. APMIS. 70. Desmond J. Serum osmolality and plasma electrolytes
1996;104:915. in patients who develop dilutional hyponatremia during
49. Hahn RG, Olsson J, Sotonyi P, et al. Rupture of the transurethral resection. Can J Surg. 1970;13:116.
myocardial histoskeleton and its relation to sudden death 71. Hulten JO, Hahn RG. Monitoring irrigating fluid absorption
after overhydration with glycine 1.5% in the mouse. APMIS. during transurethral resection of the prostate (TURP)a
2000;108:487. comparison between 1 and 2% ethanol as a tracer. Scand J
50. Sohn MH, Vogt C, Heinen G, et al. Fluid absorption and Urol Nephrol. 1989;23:103.
circulating endotoxins during transurethral resection of the 72. Arieff AI. Hyponatremia, convulsions, respiratory arrest
prostate. Br J Urol. 1993;72:605. and permanent brain damage after elective surgery in
51. Yamamoto H, Omote K, Sonoda H, et al. A case of healthy women. N Engl J Med. 1986;314:1529.
sepsis that developed during transurethral resection of the 73. Arieff AI, Ayus JC, Fraser CL. Hyponatremia and death or
prostate. J Anesth. 2002;16:242. permanent brain damage in healthy children. Br Med J.
52. Creevy CD. Hemolytic reactions during transurethral pro- 1992;304:1218.
static resection. J Urol. 1947;58:125. 74. Kirschenbaum MA. Severe mannitol-induced hypona-
53. Kolmert T, Norlen H. Transurethral resection of the tremia complicating transurethral prostatic resection.
prostate a review of 1111 cases. Int Urol Nephrol. J Urol. 1979;121:687.
1989;21:47. 75. Wright HK, Gann DS. Severe postoperative hyponatremia
54. Creevy CD, Reiser MP. The importance of hemolysis in without symptoms of water intoxication. Surg Gynecol
transurethral prostatic resection: Severe and fatal reac- Obstet. 1962;115:553.
tions associated with the use of distilled water. J Urol. 76. Katzman R. Effect of electrolyte disturbances on the central
1963;89:900. nervous system. Annu Rev Med. 1966;17:197.
55. Dorotta I, Basali A, Ritchey M, et al. Transurethral 77. Zornow MH, Todd MM, Moore SS. The acute cerebral
resection syndrome after bladder perforation. Anesth Analg. effects of changes in plasma osmolality and oncotic
2003;97:1536. pressure. Anesthesiology. 1987;67:936.
56. Akan H, Sargin S, Dalva I, et al. Effect of distilled water and 78. Rosen AS, Andrew RD. Osmotic effects upon excitability in
a mixture of sorbitol-mannitol irrigation fluids on fluid- rat neocortical slices. Neuroscience. 1990;38:579.
electrolyte balance in patients undergoing transurethral 79. Russell D. Painless loss of vision after transurethral
prostatectomy. Int Urol Nephrol. 1997;29:574. resection of the prostate. Anaesthesia. 1990;45:218.
57. Memon A, Buchholz NP, Salahuddin S. Water as an irrigant 80. Andrew RD. Seizure and acute osmotic change: Clinical
in transurethral resection of the prostate: A cost-effective and neurophysiological aspects. J Neurol Sci. 1991;101:7.
alternative. Arch Ital Urol Androl. 1999;LXXI:131. 81. Restuccia T, Gomez-Anson B, Guevara M, et al. Effects of di-
58. Beal JL, Freysz M, Berthelon G, et al. Consequences of fluid lutional hyponatremia on organic osmolytes and water con-
absorption during transurethral resection of the prostate tent in patients with cirrhosis. Hepatology. 2004;39:1613.
using distilled water or glycine 1.5 per cent. Can J Anaesth. 82. Sterns RH, Riggs JE, Schochet SS Jr. Osmotic demyeli-
1989;36:278. nation syndrome following correction of hyponatremia.
59. Estey EP, Mador DR, McPhee MS. A review of 1486 N Engl J Med. 1986;314:1535.
transurethral resections of the prostate in a teaching 83. Ayus JC, Krothapalli RK, Arieff AI. Treatment of symp-
hospital. Can J Surg. 1993;36:37. tomatic hyponatremia and its relation to brain damage:
60. Emmett JL, Gilbaugh JH, McLean P. Fluid absorption dur- A prospective study. N Engl J Med. 1987;317:1190.
ing transurethral resection: Comparison of mortality and 84. Reuss S, Burger K, Claus H, et al. Acute moderate
morbidity after irrigation with water and non-hemolytic hyponatremia and its rapid correction: Effects on striatal
solutions. J Urol. 1969;101:884. and pontine ultrastructure in an animal model of the TURP
61. Appelt GL, Benson GS, Corriere JN Jr. Transient blind- syndrome. Eur J Anaesthesiol. 2004;21:231.
nessunusual initial symptom of transurethral prostatic 85. Bibl D, Lampl C, Gabriel C, et al. Treatment of central
resection reaction. Urology. 1979;13:402. pontine myelinolysis with therapeutic plasmapharesis.
62. Henderson DJ, Middleton RG. Coma from hyponatremia Lancet (Res Lett). 1999;353:1155.
following transurethral resection of the prostate. Urology. 86. Adrogue HJ. Consequences of inadequate management of
1980;15:267. hyponatremia. Am J Nephrol. 2005;25:240.
63. Hagstrom RS. Studies on fluid absorption during 87. Illowsky BP, Laureno R. Encephalopathy and myeli-
transurethral prostatic resections. J Urol. 1955;73:852. nolysis after rapid correction of hyponatremia. Brain.
64. Arieff AI. Hyponatremia. Mt Sinai J Med. 1990;57:125. 1987;110:855.
65. Bird D, Slade N, Feneley RC. Intravascular complications 88. Brunner JE, Redmond JM, Haggar AM, et al. Central pon-
of transurethral resection of the prostate. Br J Urol. tinemyelinolysis and pontine lesions after rapid correction
1982;54:564. of hyponatremia: A prospective magnetic resonance imag-
66. Sunderrajan S, Bauer J, Vopat R, et al. Posttransurethral ing study. Ann Neurol. 1990;27:61.
prostatic resection hyponatremic syndrome: Case report 89. Sarnaik AP, Meert K, Hackbarth R, et al. Management of hy-
and review of the literature. Am J Kidney Dis. 1984;4:80. ponatremic seizures in children with hypertonic salinea
67. Weissman JD, Weissman BM. Pontine myelinolysis and safe and effective strategy. Crit Care Med. 1991;19:758.
delayed encephalopathy following the rapid correction of 90. Pycock CJ, Kerwin RW. The status of glycine as a
acute hyponatremia. Arch Neurol. 1989;46:926. supraspinal neurotransmitter. Life Sci. 1980;28:2679.
68. Ashraf N, Locksley R, Arieff AI. Thiazide-induced hypona- 91. Pourcho RG, Goebel DJ, Jojich L, et al. Immunocytochem-
tremia associated with death or neurological damage in ical evidence for involvement of glycine in sensory centers
outpatients. Am J Med. 1981;70:1163. of the rat brain. Neuroscience. 1992;46:643.
490 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
92. Kish SJ, Dixon LM, Burnham WM, et al. Brain neurotrans- 115. Wright N, Seggie J. Glycine toxicokineticsvitreous fluid
mitters in glycine encephalopathy. Ann Neurol. 1988;24:458. concentration and visual impairment. Clin Invest Med.
93. Markand OH, Garg BP, Brandt IK. Nonketotic hyper- 1992;15:159.
glycinemiaelectroencephalographic and evoked potential 116. Stalberg HP, Hahn RG, Hjelmqvist H, et al. Haemodynam-
abnormalities. Neurology. 1982;32:151. ics and fluid balance after intravenous infusion of 1.5%
94. Perry TL, Urquhart N, Hansen S, et al. Studies of the glycine glycine in sheep. Acta Anaesthesiol Scand. 1993;37:281.
cleavage enzyme system in brain from patients with glycine 117. Maatman TJ, Musselman P, Kwak YS, et al. Effect of glycine
encephalopathy. Pediatr Res. 1977;12:1192. on retroperitoneal and intraperitoneal organs in the rat
95. Perry TL, Urquhart N, MacLean J, et al. Nonketotic model. Prostate. 1991;19:323.
hyperglycinemiaglycine accumulation due to absence of 118. Fitzpatrick JM, Kasidas GP, Rose GA. Hyperoxaluria
glycine cleavage in brain. N Engl J Med. 1975;292:1269. following irrigation for transurethral prostatectomy. Br J
96. Johnson W, Ascher P. Glycine potentiates the NMDA Urol. 1981;53:250.
response in cultured mouse brain neurons. Nature. 119. Rose GA, Fitzpatrick JM, Kasidas GP. Fluid absorption dur-
1987;325:529. ing transurethral resection [Letter]. Br J Med. 1981;282:317.
97. Pfeiffer-Linn C, Glantz RM. An arthropod NMDA receptor. 120. Zhang W, Andersson BS, Hahn RG. Effect of irrigating flu-
Synapse. 1991;9:35. ids and prostate tissue extracts on isolated cardiomyocytes.
98. Doolan PD, Harper HA, Hutchin ME, et al. The renal tubular Urology. 1995;46:821.
response to amino acid loading. J Clin Invest. 1956;35: 121. Zhang WB, Hahn RG. Double toxicity of glycine solution
888. in the mouse. Br J Urol. 1996;77:203.
99. Hoekstra PT, Kahnoski R, McCannish MA, et al. 122. Olsson J, Hahn RG. Glycine toxicity after high-dose i.v.
Transurethral prostatic resection syndrome - a new perspec- Infusion of 1.5% glycine in the mouse. Br J Anaesth.
tive. Encephalopathy with associated hyperammonemia. 1999;82:250.
J Urol. 1983;130:704. 123. Olsson J, Hahn RG. Survival after high-dose intra-
100. Hamilton Stewart PA, Barlow IM. Metabolic effects of venous infusion of irrigating fluids in the mouse. Urology.
prostatectomy. J R Soc Med. 1989;82:725. 1996;47:689.
101. Ferraro TN, Hare TA. Free and conjugated amino acids 124. Nilsson A, Randmaa I, Hahn RG. Haemodynamic effects
in human CSF: Influence of age and sex. Brain Res. of irrigating fluids studied by Doppler ultrasonography in
1985;338:53. volunteers. Br J Urol. 1996;77:541.
102. Wang JM, Creel DJ, Wong KC. Transurethral resection 125. Hahn RG, Sandfeldt L, Nyman CR. Double-blind random-
of the prostate, serum glycine levels, and ocular evoked ized study of symptoms associated with absorption of
potentials. Anesthesiology. 1989;70:36. glycine 1.5% or mannitol 3% during transurethral resection
103. Ovassapian A, Joshi CW, Brunner GH. Visual distur- of the prostate. J Urol. 1998;160:397.
bancesan unusual symptom of transurethral prostatic 126. Schwarcz R, Meldrum B. Excitatory amino acid antagonists
resection reaction. Anesthesiology. 1982;57:332. provide a therapeutic approach to neurological disorders.
104. Defalque RJ, Miller DW. Visual disturbances during Lancet. 1985;2:140.
transurethral resection of the prostate. Can Anaesth Soc 127. Nathans D, Fahey JL, Ship AG. Sites of origin and removal
J. 1975;22:620. of blood ammonia formed during glycine infusioneffect
105. Kaiser R, Adragna MG, Weis FR Jr. Transient blindness of L-arginine. J Lab Clin Med. 1958;51:124.
following transurethral resection of the prostate in an 128. Lehninger AL. Principles of biochemistry. New York: Worth
achondroplastic dwarf. J Urol. 1985;133:685. Publishers; 1982:531.
106. Gooding JM, Holcomb MC. Transient blindness following 129. Mushambi MC, Williamson K. Anaesthetic considerations
intravenous administration of atropine. Anesth Analg. for hysteroscopic surgery. Best Pract Res Clin Anaesthesiol.
1977;56:872. 2002;16:35.
107. Charlton AJ. Cardiac arrest during transurethral prosta- 130. Roesch RP, Stoelting RK, Lingerman JE, et al. Ammo-
tectomy after absorption of 1.5% glycine. Anaesthesia. nia toxicity resulting from glycine absorption during a
1980;35:804. transurethral resection of the prostate. Anesthesiology.
108. Still JA, Modell JH. Acute water intoxication during 1983;58:577.
transurethral resection of the prostate, using glycine 131. Shepard RL, Kraus SE, Babayan RK, et al. The role of
solution for irrigation. Anesthesiology. 1973;38:98. ammonia toxicity in the post transurethral prostatectomy
109. Wong KC. Transurethral resection of the prostate syndrome. Br J Urol. 1987;60:349.
anesthetic implications. Proceedings of the 58th Congress of 132. Hahn RG, Sandfeldt L. Blood ammonia levels after intra-
the International Anesthesia Research Society. Reno, Nevada: venous infusion of glycine with and without ethanol. Scand
1984:191. J Urol Nephrol. 1999;33:222.
110. Hahn RG. Hallucination and visual disturbances dur- 133. Fahey JL. Toxicity and blood ammonia rise resulting
ing transurethral prostatic resection. Intensive Care Med. from intravenous amino-acid administration in manthe
1988;14:668. protective effect of L-arginine. J Clin Invest. 1957;36:1647.
111. Massey SC, Redburn DA. Transmitter circuits in vertebrate 134. Kim AH, Keltz MD, Arici A, et al. Dilutional hypona-
retina. Prog Neurobiol. 1987;28:55. tremia during hysteroscopic myomectomy with sorbitol-
112. Wachtmeister L. Further studies of the chemical sensitivity mannitol distention medium. J Am Assoc Gynecol Laparosc.
of the oscillatory potentials of the electroretinogram (ERG). 1995;2:237.
Acta Ophthalmol. 1980;58:712. 135. Trepanier CA, Lessard MR, Brochu J, et al. Another feature
113. Hahn R, Andersson T, Sikk M. Eye symptoms, visual evoked of TURP syndrome: Hyperglycaemia and lactic acidosis
potentials and EEG during intravenous infusion of glycine. caused by massive absorption of sorbitol. Br J Anaesth.
Acta Anaesthesiol Scand. 1995;39:214. 2001;87:316.
114. Seggie J, Wright N. Glycine inhibition of pupillary re- 136. Baggish MS, Brill AIO, Ronsenweig B, et al. Fatal acute
sponses to pulses of light in conscious sheep. J Ocul glycine and sorbitol toxicity during operative hysteroscopy.
Pharmacol. 1989;5:343. J Gynecol Surg. 1993;9:137.
C HAPTE R 33/TU R P SYN DROM E 491
137. Arieff AI, Ayus JC. Endometrial ablation complicated by 147. Wilcox CS. Regulation of renal blood flow by plasma
fatal hyponatremic encephalopathy. JAMA. 1993;270:1230. chloride. J Clin Invest. 1983;71:726.
138. Bergstrom J, Hultman E, Roch-Norlund AE. Lactic acid 148. Santi M, Pinelli G, Ricci P, et al. Evidence
accumulation in connection with fructose infusion. Acta that 2-phenylpyrazolo[4,3-c]-quinolin-3(5H)-one antago-
Med Scand. 1968;184:359. nizes pharmacological, electrophysiological, and biochem-
139. Wagner K, Wolf AS. Death following fructose and sorbitol ical effects of diazepam in rats. Neuropharmacology. 1985;
infusions (German). Anaesthesist. 1984;33:573. 24:99.
140. Schulte MJ, Lenz W. Fatal sorbitol infusion in patient with 149. Morag M, Myslobodsky M. Benzodiazepine antagonists
fructose-sorbitol intolerance. Lancet. 1977;2:8030. abolish electrophysiological effects of sodium valproate in
141. Logie JRC, Keenan RA, Whiting PH, et al. Fluid absorption the rat. Life Sci. 1982;30:1671.
during transurethral prostatectomy. Br J Urol. 1980;52: 150. Schafer DF, Pappas SC, Brody LE, et al. Visual evoked
526. potentials in a rabbit model of hepatic encephalopathy.
142. Grove JJ, Shinaman RC, Drover DR. Noncardiogenic I. Sequential changes and comparisons with drug-induced
pulmonary edema and venous air embolus as complications comas. Gastroenterology. 1984;86:540.
of operative hysteroscopy. J Clin Anesth. 2004;16:48. 151. vanBoven MJ, Singelyn F, Donnez J, et al. Dilutional
143. Hahn RG, Drobin D, Stahle L. Volume kinetics of Ringers hyponatremia associated with intrauterine endoscopic laser
solution in female volunteers. Br J Anaesth. 1997;78:144. surgery. Anesthesiology. 1989;71:449.
144. Williams EL, Hildebrand KL, McCormick SA, et al. The 152. Rosenberg MK. Hyponatremic encephalopathy after roller-
effect of intravenous lactated Ringers solution versus 0.9% ball endometrial ablation. Anesth Analg. 1995;80:1046.
sodium chloride solution on serum osmolality in human 153. Olsson J, Berglund L, Hahn RG. Irrigating fluid absorption
volunteers. Anesth Analg. 1999;88:999. from the intact uterus. Br J Obstet Gynaecol. 1996;103:558.
145. Wilkes NJ, Woolf R, Mutch M, et al. The effects of 154. Ayus JC, Wheeler JM, Arieff AI. Post-operative hypona-
balanced versus saline-based hetastarch and crystalloid traemic encephalopathy in menstruant women. Ann Intern
solutions on acid-base and electrolyte status and gastric Med. 1992;117:891.
mucosal perfusion in elderly surgical patients. Anesth Analg. 155. Agin C. Anesthesia for transurethral prostate surgery. Int
2001;93:811. Anesthesiol Clin. 1993;31:25.
146. Drobin D, Hahn RG. Kinetics of isotonic and hyper- 156. Hjertberg H, Petterson B. The use of a bladder pressure
tonic plasma volume expanders. Anesthesiology. 2002;96: warning device during transurethral prostatic resection de-
1371. creases absorption of irrigating fluid. Br J Urol. 1992;69:56.
E . H E M ATOLOG I C
A N D H E M O STAT I C
CHAPTER ACUTE ANEMIA
CASE SUMMARY
What Is Perioperative Anemia?
64-year-old, 72-kg man is seen at the local
A
emergency department for a 3-day history
The occurrence of anemia in the surgical population has
of hematemesis and dark stools. The pa-
long attracted the interest of the anesthesia, surgical, and
tient has a past medical history of heavy
alcohol intake and bleeding from esophageal critical care communities. Its presence has been identified
varices. On arrival, the patient is noted to be as an independent risk factor for infection, requirement
mildly dehydrated, confused, markedly tachycardic, and for transfusion,13 and increased perioperative morbidity
hypotensive. He is also complaining of new onset of chest and mortality.46 However, the lack of a consistent
tightness associated with vomiting. A 12-lead electrocar- and standard definition of what constitutes anemia has
diogram reveals marked ST depressions in the anterior limited the interpretation and comparison of available
lead and sinus tachycardia. During initial evaluation, ap- data, and complicates the reporting of definitive anemia
proximately 250 mL of bright red blood emesis was noted. prevalence in the perioperative setting. An attempt to
Hematocrit at this time was reported at 29%. A large bore establish a numeric definition of the lower limit of
intravenous site was accessed, and aggressive fluid resus- normal hemoglobin was recently reported.7 The values are
citation was begun by rapid infusion of 2 L of normal derived from two large, well documented databasesThe
saline and 500 mL of 5% albumin. Elective endotracheal Third US National Health (and Nutritional Examination
intubation was achieved for airway protection and im- Survey and the Scripps-Kaiser database) that allow the
pending endoscopic sclerotherapy. Hematocrit was noted exclusion of individuals who are not normal. This
to be 18% and platelet count 25,000 per mm3 following analysis concludes that a hemoglobin level below 13.7 g
fluid resuscitation. Borderline hypotension and subtle ST per dL, in a white man between 20 and 60 years of age,
changes persist in all anterior leads despite improving has approximately a 5% chance of being a normal value.
hypotension and tachycardia. The patient was trans- The corresponding value of hemoglobin in women of all
fused with three units of packed red blood cells (PRBCs) ages is 12.2 g per dL (see Table 34.1). These values differ
and one unit of platelets, with marked improvement in slightly from the widely accepted values proposed by the
hemodynamic, electrocardiogram, and laboratory values. World Health Organization (WHO) that were published
Following initial hemodynamic resuscitation, the patient in 1968.8 The WHO defines anemia as a hemoglobin
underwent successful therapeutic endoscopy for definite of <13 g per dL in adult males and <12 g per dL in
treatment of active variceal hemorrhage under general adult nonpregnant females. These small but significant
endotracheal anesthesia. Cardiology workup revealed un- numeric discrepancies confound the true prevalence of
diagnosed coronary artery disease, thereby explaining the anemia in the general population and suggest that a
ST changes in the presence of hypotension and severe single numeric value to diagnose clinical anemia is
acute anemia. difficult to establish. Therefore, a more individualized
492
CHAPTER 34/ACUTE ANEMIA 493
TABLE 34.1 Proposed Lower Limits of Normal Therefore, perioperative anemia needs to be eval-
Hemoglobin Concentration of the Blood for White uated on an individual basis, taking into account the
and Black Adults patients age, gender, medical condition, analysis of risk
factors, and the current surgical scenario. The functional
Group Hemoglobin (g/dL) consequences of anemia are far more relevant in clinical
anesthesia than the simplicity of a laboratory value.
White Men Acute anemia may result from a sudden drop in the
2059 y 13.7 hemoglobin or hematocrit due to hemolysis, acute hemod-
60 y 13.2 ilution, or acute blood loss. In all instances, the amount
White Women of plasma volume present in blood influences the hemat-
2049 y 12.2 ocrit and hemoglobin concentration. It is possible for an
50 y 12.2 individual to be severely anemic and yet have a normal
Black Men hematocrit. This often occurs in the presence of acute
2059 y 12.9 hemorrhage, where rapid loss of intravascular volume en-
60 y 12.7 tails the simultaneous loss of RBCs and plasma. Following
Black Women a sudden loss of 20% of total blood volume, 20 to 60 hours
2049 y 11.5 are required to restore a normal blood volume by endoge-
50 y 11.5 nous plasma replacement.11,12 The decrease in RBC mass,
or anemia, does not become clinically apparent until fluid
From: Beutler E, Waalen J. The definition of anemia: What is the
replacement therapy or endogenous compensatory mech-
lower limit of normal of the blood hemoglobin concentration?
Blood. 2006;107:1747. anisms reestablish normal blood volume. On the contrary,
it is also possible to have laboratory evidence of anemia
without clinical evidence of disease; this is due to the
and physiologic approach to the management of anemia ability of the body to compensate for decreased RBC mass
and hemoglobin levels in patient care is a reasonable goal. in chronic situations.
Preoperative
2,000 Hct
Postoperative
1,800 Hct
1,600
1,400
Patients
1,200
1,000
800
600
400
200
0
<20 2125 2630 3135 3640 4145 4650 >50
Hematocrits
The heart is exquisitely prone to injury in the pres- Normal RBC survival is approximately 120 days.
ence of anemia. In the resting individual, the myocardial Erythrocytes progress from blast precursors in the bone
muscle extracts 70% to 75% of the oxygen delivered marrow over a period of 5 days. They are then released
through the coronary circulation. Myocardial oxygen into the circulation in the form of reticulocytes (immature
extraction cannot increase above resting levels in the erythrocytes), maturing in 1 day and circulating for
presence of higher oxygen demand. Consequently, higher an average of 120 days before being destroyed by the
oxygen delivery must be achieved by increasing coronary reticuloendothelial system. All the by-products of the
blood flow. catabolism of RBCs remain available for reutilization in
Transient and asymptomatic changes in the electro- cases of physiologic destruction or hemolytic anemia, as
cardiogram, reversible changes in cognition, and fatigue contrasted with acute blood loss in which RBCs are lost
have been reported in normal individuals during induced as a whole.
isovolemic anemia with hemoglobin levels between 5 and Mechanisms responsible for acute anemia are limited
7 g per dL.1416 Moreover, in older patients (>85 years) to those related to excessive red cell destruction (hemolytic
and critically ill patients, the presence of anemia is asso- anemia), acute blood loss, and iatrogenic dilutional ane-
ciated with an increased number of blood transfusions, mia. There are disorders such as iron deficiency anemia,
diminished organ function, and increased mortality.10,17 which results from inadequate iron intake or chronic
bleeding, and deficiency of vitamin B12 and folic acid that
manifest as a chronic inability of the bone marrow to
increase red cell production beyond the rate of red cell
What Are the Causes destruction. However, because of the time that it takes for
and Differential Diagnosis chronic anemia to develop, compensatory mechanisms
such as increases in 2,3-diphosphoglycerate (2,3-DPG) in
of Anemia? the RBCs facilitate oxygen unloading in the tissue and
mask clinical symptoms. On the contrary, acute anemias
Assuming that the circulatory system is intact, a decrease such as hemolytic anemia, acute hemorrhage, and acute
in RBC mass, or anemia, results from an imbalance in hemodilution rapidly trigger a reflex sympathetic response
the mechanisms responsible for RBC production and that augments (cardiac output) CO and forces a blood flow
destruction. The degree to which the signs and symptoms redistribution in the body until more chronic and effective
of anemia are manifest depends upon the magnitude and compensatory mechanisms become established.
speed of the reduction in RBC mass, volume of fluid loss, if Acute anemia related to blood loss is common in
any, and the individuals ability to compensate. Table 34.2 the perioperative setting. The clinical response to de-
provides a classification of the causes of acute and chronic creased blood volume becomes more notorious as blood
anemia. loss becomes more severe. During severe hemorrhage,
CHAPTER 34/ACUTE ANEMIA 495
TABLE 34.2 Causes of Anemia in the presence of acute blood loss. Clinical signs of acute
blood loss correlate well with the percentage of total-body
Hemorrhagic Acute Blood Loss blood volume loss (see Table 34.3).18 A normal individual
External can rapidly lose 10% to 20% of blood volume without
Internal signs or symptoms of anemia. However, a rapid blood
Chronic Blood Loss loss of more than 50% of total blood volume presents with
clinical signs of severe hypovolemic shock and death.
Hemolytic Acute Hemolysis
In patients with slow, chronic blood loss, compensatory
(Inherited or Extravascular
mechanisms allow for more dramatic losses of RBCs with
Acquired) Intravascular
minimal clinical signs of anemia. Chronic reestablishment
Chronic Hemolysis
and maintenance of intravascular volume limit the
Membrane defects
cardiovascular response to hypovolemia. However, signs
Metabolic defects
of decreased total RBC mass may ensue.
Hemoglobin disorders
In general, clinical signs of blood loss anemia are
Environmental causes
tachycardia, orthostatic and resting hypotension, and ev-
Impaired Red Cell Maturation Disorders idence of reduced end organ perfusion, manifested by
Production Cytoplastic defects altered mental status and decreased urine output. In
Severe iron deficiency healthy subjects, symptoms produced by bleeding may
Thalassemia become apparent when the hematocrit falls below 35%.
Sideroblastic anemia These symptoms include mild fatigue, palpitations, dysp-
Nuclear defects nea, excessive sweating with exercise, and a requirement
Folate deficiency for an increased amount of sleep.11
Vitamin B12 deficiency In the case of acute anemia due to hemolysis, in-
Other marrow diseases travascular volume is intact and the clinical signs and
Hypoproliferative Disorders symptoms are directly related to anemia. Clinical signs of
Iron deficiency hemolytic anemia are pallor, jaundice, tachycardia, a new
Inflammation systolic ejection murmur, and dark colored urine. Symp-
Reduced erythropoietin toms include marked fatigue, dizziness, and palpitations.
Marrow damage
Extracorpuscular:
Immune
Transfusion reaction
Hemolytic disease of the newborn
Cold agglutinin disease
Cold hemolysin disease
Autoimmune hemolytic anemia
Drug-induced hemolytic anemia
Nonimmune
Malaria
Bacterial toxins
Excessive Toxic chemicals
red blood cell Splenomegaly
destruction Microangiopathy
Intracorpuscular:
Hereditary
Hemoglobinopathies
Thalassemia
Spherocytosis, eliptocytosis
Glucose-6-phosphate dyhydrogenase
deficiency
Acquired
Thermal injury
Lead poisoning
Paroxy smal nocturnal hemoglobinuria
Acute anemia
Birth bearing
Antepartum and postpartum bleeding
Acute blood loss Surgically induced
Trauma
Coagulopathy (diagnosed vs. undiagnosed)
Iatrogenic
Dilutional anemia
Acute normovolemic hemodilution
The rate at which the blood loss occurs heavily influences increase in heart rate. The latter is responsible for the
the magnitude of the compensatory mechanism in acute increase in CO associated with hypovolemic anemia. The
blood loss anemia. reduction in the blood viscosity seen at the postcapillary
With acute blood loss, an absolute decrease in venule with moderate normovolemic anemia results in
the RBC count will occur; however, hematocrit or increased venous return to the heart and increased CO.
hemoglobin will not immediately reflect the quantity of Early canine studies demonstrated that lowering blood
blood loss. With more subtle and chronic blood loss, viscosity with a colloid exchange-transfusion to a target
sufficient restoration of plasma volume ensues and a of half the normal hematocrit dramatically increased CO,
marked decrease in RBC count is seen in the face of compared to dogs whose effective hemoglobin was halved
minimal cardiovascular response due to maintenance of by exchange-transfusion with RBCs containing methe-
intravascular volume. moglobin.22 This explains the increase of total and local
CO also increases with increasing degrees of nor- blood flow as a compensatory response to preserve oxygen
movolemic anemia. In fact, an inverse relation has been delivery as long as normovolemia is preserved.
established between hemoglobin level and CO during nor- In the presence of low hemoglobin/hematocrit and
movolemic anemia.19,20 The main mechanisms responsi- stable hemodynamics, optimal oxygen delivery and uptake
ble for increases in CO during normovolemic anemia are are of primary concern. There is a unique relation between
reduced blood viscosity and increased sympathetic stimu- oxygen consumption (VO2 ), tissue blood flow, and the
lation.21 In this condition, the increase in CO is primarily extraction of oxygen from the blood (arterialvenous
the result of an increase in stroke volume rather than an oxygen difference). The Fick principle states that oxygen
CHAPTER 34/ACUTE ANEMIA 497
TABLE 34.3 Signs and Symptoms of Acute Blood Loss of Increasing Severity
Percentage of
Total Blood mL in
Volume 70-kg Man Clinical Signs
10 500 None; rare vasovagal syncope
20 1,000 Usually none at rest; slight orthostatic hypotension and tachycardia with activity
30 1,500 Anxiety and restlessness; flat neck veins; tachycardia and resting hypotension possible
40 2,000 Cold, clammy skin; rapid, thready pulse; air hunger possible; CVP, cardiac output, and
arterial blood pressure below normal at rest
50 2,500 Hypovolemic shock; diaphoresis; disorientation or decreased consciousness; death
consumption (VO2 ) is equal to the product of flow and the The second portion of the equation represents the amount
arteriovenous oxygen content difference (CaO2 CvO2 ). of oxygen dissolved in the plasma and not bound to
hemoglobin. After multiplying the partial pressure of
oxygen dissolved in plasma (PaO) by its solubility coeffi-
VO2 = TBF (CaO2 CvO2 ) cient (0.003), the amount of dissolved oxygen available for
extraction is relatively minor. It is clear that the primary
TBF = specific tissue blood flow or whole body CO. mechanisms to alter the arterial oxygen content are by
It is expected that to preserve the baseline oxygen changing the hemoglobin concentration and its saturation
consumption (VO2 ) of the body in the presence of reduced with oxygen.
blood oxygen content (arterial and venous), as in acute
anemia, CO must increase or mixed venous oxygen
content must decrease. Therefore, when flow limitations
or hypovolemia are present during anemia, increased What Are the Changes at the
oxygen extraction comes into play. This is measured by the Level of the Red Blood Cell
difference between the arterial and venous oxygen content
(CaO2 CvO2 ). As a rule, an increase in heart rate and a and Hemoglobin Molecule
fall in central venous oxygen saturation (SvO2 ) suggest in Response to Acute Anemia?
excessive hemodilution or hypovolemia.12
Arterial oxygen content (CaO2 ) is primarily dependent
Several cellular and molecular adjustments take place in
on the number of RBCs available for oxygen transport, as
the presence of acute anemia. The ultimate result of these
demonstrated by the equation for the content of oxygen
adjustments is the reduction of hemoglobins affinity for
in arterial blood:
oxygen by the rightward shift of the oxyhemoglobin dis-
sociation curve (see Fig. 34.3). This unique characteristic
CaO2 = (1.34 Hgb SaO2 ) + (0.003 PaO2 ) facilitates the unloading of oxygen from the hemoglobin
FIGURE 34.3 Oxyhemoglobin dissociation curve. Once the curve shifts to the right, the p50
increases, which means that a higher partial pressure of oxygen is required to saturate 50% of the
hemoglobin present.
498 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
molecule to peripheral tissues. The oxygen extraction of earliest references to support the 10 g per dL hemoglobin
anemic blood by peripheral tissues increases the concen- requirement appeared in a 1941 publication that stated
tration of deoxyhemoglobin in the RBC, which stimulates . . .when hemoglobin concentration is less than 8 to 10
the production of 2,3-diphosphoglycerate (2,3-DPG). The grams per 100 cubic centimeters of whole blood, it is
increased concentration of 2,3-DPG decreases the affin- wise to give a blood transfusion before operation.24
ity of hemoglobin for oxygen. Significant changes in This statement, in addition to nearly a half a century
hemoglobin affinity and 2,3-DPG concentration may occur of unsubstantiated clinical tradition, guided clinicians in
under extreme hemodilution (hematocrit <15) in acute the preoperative management of anemia. It was not until
circumstances. The shift in hemoglobins affinity for oxy- 1960 that Penden et al. concluded that in the absence of
gen is far more effective as a compensatory mechanism acute blood loss, blood transfusion should not be given
during chronic conditions of anemia than during acute to correct deficits in blood volume, and by doing so, they
normovolemic or blood loss anemia. demonstrated the risks of fluid overload in chronically ill
patients.25
With the advent of open-heart surgery, many clinical
studies reported successful management of patients with
What Influences the Rate hematocrits of 20% or less during cardiopulmonary by-
pass.26,27 However, recent data suggest that lower hema-
of Blood Regeneration after tocrits on bypass are associated with increased adverse
Blood Loss? outcomes during coronary artery bypass grafting.28 The
recognition that anemia does not necessarily equate to
hypovolemia lead to improvements in anesthetic manage-
The critical factors for mounting an effective erythro-
ment of patients with reduced hemoglobin levels without
poiesis are the presence of stimulus for blood production,
the use of blood transfusions. In addition, recent data
the functional ability of the bone marrow to respond to
suggest that critically ill patients can tolerate hemoglobin
the stimulus, and the presence of the basic elements that
levels of 7 g per dL, and support the association between
support the building process. The main drive for stimu-
transfusion and diminished organ function as well as
lating the erythropoietic process is a low hematocrit level.
between low hemoglobin levels, transfusion, and mortal-
Normal RBC turnover, or physiologic hematopoiesis, is
ity.17 These experiences, in addition to concerns related to
considered to be 1% of the total circulating red cell mass
transfusion-associated infections and the growing short-
per day, or 7% per week. This represents the normal abil-
age of RBCs available for transfusion, has led to a more
ity of the bone marrow to respond to the typical red cell
functional concept of anemia in the perioperative setting.
lifespan in a healthy individual.
Anemia continues to be a very common diagnosis in
In the presence of acute blood loss anemia, both RBCs
the critically ill patient (>95% of patients in some series)
and the products of RBC catabolism (especially iron) used
and, despite data regarding RBC transfusion thresholds
by the bone marrow for the regeneration of RBCs are
and risks of RBC transfusion, practice patterns have not
lost. This may limit the ability of the bone marrow to
changed dramatically, as reported by Corwin et al. in
mount an effective erythropoietic response. In the case of
2004.29
hemolytic anemia, all the products of RBC destruction
The decision to correct preoperative anemia in cur-
remain in the circulatory reticuloendothelial systems.
rent practice should take into account the patients specific
Despite the shorter lifespan of RBCs in hemolytic anemias,
needs. The anesthesiologist should evaluate the cause and
the bone marrow is generally able to mount an effective
duration of the anemia, intravascular fluid volume, risk
erythropoietic response in the presence of the stimulus
of surgical bleeding and, most importantly, the presence
and all circulating elements for RBC construction.
of comorbidities such as congestive heart failure, coro-
Large quantities of plasma proteins and hemoglobin
nary artery disease with or without previous myocardial
must be synthesized to replace the deficits resulting from
ischemia, peripheral or cerebrovascular disease, and pul-
acute blood loss. Plasma volume equilibration occurs
monary function. The need to improve oxygen delivery
within the first 20 to 40 hours of acute blood loss, and
by increasing the oxygen-carrying capacity of the blood
plasma protein regeneration is usually complete within a
should be the only indication for correction of anemia in
week of acute blood loss.11,23
the perioperative setting.
of the surgical procedure. The benefit and efficacy of TABLE 34.4 Surgical Procedures Commonly Associated
this blood transfusion strategy relies on the significant with Substantial Anticipated Blood Loss
release of endogenous erythropoietin (EPO) in response
to phlebotomy and blood collection, while still remaining Coronary artery bypass
within the normal range (4 to 26 g per mL). Patients Major vascular surgery
undergoing PAD may donate one unit of blood as often Primary hip replacement
as twice a week until 72 hours before surgery, as far Revision, hip replacement
in advance of surgery as possible (up to 42 days) to Total knee replacement
allow compensatory erythropoiesis to correct the induced Major spine surgery with instrumentation
anemia.30 The production of two units of blood in Selected neurologic procedures
excess of basal erythropoiesis occurs in response to Hepatic resections
three phlebotomies at weekly intervals and a threshold Radical prostatectomy
hematocrit for blood donation of 36%.31 Physiologic
replacement of the red cell mass by increasing cell
production is a gradual process. Patients subjected to
general population, upon the consent of the patient and
more aggressive phlebotomy (2 units weekly) respond by
physician who ordered the PAD, once it is determined that
producing 2 to 3 blood units due to endogenous EPO
the patient will not require a PRBC transfusion.
during the same time period. When recombinant human
Table 34.4 summarizes those surgical procedures
EPO therapy is instituted, the equivalent of 5 units of
commonly associated with substantial anticipated blood
blood may be produced in response to acute anemia
loss, and for which perioperative autologous blood
induced by PAD.32
collections are most beneficial. Traditionally, autologous
There are no firm age or weight minimums for the
blood has not been recommended for procedures that
preoperative collection of autologous blood, although
rarely (fewer than 10% of cases) require transfusion, such
blood collection personnel calculate the amount of blood
as cholecystectomy, herniorrhaphy, vaginal hysterectomy,
that may be collected on the basis of estimated blood
and uncomplicated obstetric deliveries.33 During the
volume and the beginning RBC mass for individual
mid-1990s, the protection afforded by autologous blood
patients. In small or anemic patients, the amount of
donation was considered limited and did not justify the
blood that can be safely collected may make PAD
increased cost because of increased allogeneic blood
impractical. Many pediatric patients and some adults with
testing and safety. Today, the cost of allogeneic blood
developmental delays or other neurologic conditions may
is rising as blood bank shortages occur and the role of
not be able to cooperate with the PAD process.
autologous blood transfusion is evolving. The high cost
of banked blood currently will become even more costly
in the future due to its shortage. Additionally, because of
the publics perception on the safety of elective allogeneic
When Is Preoperative blood transfusion, autologous blood donation plays an
Autologous Donation Useful? important role, even in elective surgeries associated with
a low risk of transfusion.
PAD is useful for patients undergoing a surgical procedure
for which there is a high likelihood of significant blood
loss, and for which blood is likely to be crossmatched and
possibly transfused. Many hospitals use the maximum Who Is Eligible to Provide
surgical blood order schedule. This schedule includes all a Preoperative Donation
surgical procedures at the particular institution and the
number of units of packed RBCs to be crossmatched of Autologous Blood?
for each procedure. The recommended crossmatch to
transfusion ratio (C:T ratio) is 2:1, which allows for Determining the eligibility of a patient for PAD is largely
crossmatching twice the number of units of blood that the responsibility of the primary physician caring for the
are expected to be transfused for the intended surgical patient. A minimal hematocrit of 33% before donation
procedure. In the case of autologous donation, the less has been suggested.34 Identifying comorbidities that pose
likely the possibility of transfusion, the more likely that additional health risks in the presence of acute induced
donated blood will not be used, and possibly wasted. anemia should be the primary goal of the predonation
The number of units of autologous blood obtained interview process. Underlying diseases and the patients
preoperatively is based on the number of units that ability to tolerate physiologic changes due to phlebotomy
would be crossmatched before surgery if allogeneic blood are the determining factors for PAD. Age alone is not
were being used following the maximum surgical blood considered a limitation to autologous blood donation.
order schedule. This approach is designed to allow the However, in the pediatric and some adult populations the
collection of enough autologous blood so that fewer than stress related to venipuncture and difficult venous access
10% of patients who were undergoing surgery would limit its utility. In small patients, the volume of blood
receive allogeneic blood transfusion. Many blood centers collected at each donation is reduced in proportion to
now allow autologous blood to be released for use in the body weight; in general, donations up to 10% of total blood
500 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
volume are well tolerated. Autologous donations can be the input received from stretch receptors located in the
safely obtained from patients with stable coronary artery aortic arch, carotid arteries, heart, and lungs. By increas-
disease, stable valvular heart disease, and congenital heart ing circulating levels of norepinephrine, epinephrine, and
disease.3537 angiotensin II, the sympathetic nervous system attempts
Individuals ordinarily excluded from donating blood to increase systemic vascular resistance and maintain in-
for the random blood donor pool, such as those with travascular volume and tissue perfusion. The physiologic
evidence of hepatitis B and human immunodeficiency response to acute anemia varies, depending on the acuity
virus (HIV) infection because of concerns for safety and the magnitude of the insult.
of both patients and personnel, are considered eligible In acute anemia due to hemolysis or iatrogenic
for autologous blood donation and transfusion at some dilution, changes in red cell mass occur without disturbing
institutions. Denying these patients the full and equal intravascular volume. In fact, CO increases because of
enjoyment of goods or services could be a violation of reduced blood viscosity and improved microcirculatory
the Americans with Disabilities Act.38 If an institution flow at the postcapillary venule. The main physiologic
decides to provide autologous blood services to HIV- response to moderate normovolemic anemia is therefore
infected individuals, standards require special labeling an increase in stroke volume without compensatory
of the units, including For Autologous Use Only and tachycardia. Normovolemic anemia becomes detrimental
biohazard labels.39,40 In addition, institutions are required when the RBC mass is incapable of maintaining optimal
by regulation to provide for the quarantine, handling, and oxygen delivery.
disposal of blood units not suitable for general use.41 Intraoperative recognition of anemia requires mon-
itoring, not only of vital signs and perfusion, but also
understanding and monitoring the surgical procedure.
Periodic measurements of all vital signs, according to
What Are the Risks Associated the American Society of Anesthesiologists (ASA) basic
monitoring standards,43 and routine examination of the
with the Preoperative patient and surgical field are helpful in assessing on-
Autologous Donation of Blood? going bleeding. With progressive bleeding, the skin of
the patient becomes cold to touch, and appears pale de-
One in 16,000 autologous blood donations are associated spite proper active warming measures. In patients with
with an adverse reaction such as lightheadedness, hy- hemoglobinopathies or other forms of hemolysis, diffuse
potension, bradycardia, loss of consciousness, or seizures bleeding from intravenous puncture sites and mucous
that are severe enough to require hospitalization.42 This is membranes can be noted. This should also raise the sus-
in contrast with 1 in 198,119 (0.0005%) during allogeneic picion of ongoing coagulopathy due to coagulation factor
blood donation. The risk for autologous blood donation deficiency or disseminated intravascular coagulation.
is by far greater than the risk associated with community The cardiovascular response depends on the amount
(random) blood donations by healthy individuals. of blood loss with respect to total blood volume. Only
A patient who has provided a PAD of one or more mild tachycardia may be present with <15% blood loss;
units is not completely spared the possibility of allogeneic hypotension presents with a blood loss between 15% to
blood transfusion. If the loss of blood during a surgical 30%. Narrow pulse pressure and marked hypotension
procedure exceeds the allowable blood loss and the quan- reflect 30% to 40% of total blood volume. Blood loss
tity of available patient-specific autologous blood, then the above 40% leads to hemorrhagic shock unless prompt
patient may require an allogenic blood transfusion. The resuscitative measures are taken. Congestive heart failure
discussion of preoperative autologous blood collection and signs of active ischemia should raise the suspicion for
and perioperative transfusion should include considera- severe acute anemia, especially in patients with known
tion of the relative risks of and informed consent for both coronary or peripheral vascular disease. A progressive
autologous and allogeneic blood transfusion. The trans- decrease in urine output should alert the anesthesiologist
fusion of autologous blood shares many of the same risks that intravascular volume deficit may be increasing.
associated with allogeneic blood transfusions, including Intraoperative hemoglobinuria or urobilinogen could
the possibility for bacterial contamination, clerical errors, indicate hemolysis.
and volume overload. Once intraoperative anemia is suspected, additional
laboratory studies should establish its magnitude and
metabolic consequence. Laboratory investigation of acute
anemia should not preclude prompt treatment of the
How Is Intraoperative Anemia actively bleeding, unstable patient. Fluid resuscitation
should be initiated while laboratory investigations take
Recognized and Evaluated? place. An isolated hematocrit level during acute hemor-
rhage may be misleading because of the simultaneous
Acute intraoperative anemia denotes a precipitous drop loss of cell mass and plasma. Only after aggressive fluid
in the red cell mass due to acute hemorrhage, dilution, resuscitation takes place and the intravascular volume is
or hemolysis. Anemia due to acute blood loss leads to restored does the hematocrit become useful. Acidbase
intraoperative hypovolemia and hypotension. The sympa- status usually aids in the assessment of adequate oxy-
thetic nervous system increases its activity in response to genation, ventilation, and peripheral perfusion. Metabolic
CHAPTER 34/ACUTE ANEMIA 501
acidosis develops during acute hypovolemic anemia, and Potential complications of EPO therapy include prob-
the presence of base deficit and lactic acid indicates sig- lems related to a relatively rapid increase in blood viscosity
nificant anaerobic metabolism. and vascular resistance, such as thrombotic vascular
events (myocardial infarction, angina pectoris, deep ve-
nous thrombosis, superficial phlebitis) and hypertension.
Despite these concerns, perioperative therapy with EPO
What Adjuvant Therapy May is felt to be safe and well tolerated in appropriate pa-
tients. Practitioners not accustomed to the routine use
Be Indicated to Promote and monitoring of patients who receive this treatment
Erythropoiesis? should consider hematology consultation for assistance
in the management of EPO therapy.
Select patients may be candidates to undergo therapy
with agents aimed at increasing the RBC mass, with
the primary goal of avoiding or reducing the need for
allogeneic RBC transfusion. This includes preoperative What Anesthetic Interventions
patients scheduled to undergo elective surgical procedures May Reduce Intraoperative
that commonly involve significant blood loss and the
need for red cell transfusion, and postoperative patients Blood Loss?
who have developed acute anemia in the perioperative
period but are normovolemic and able to tolerate the There are a considerable number of factors under the
anemia for several weeks as the RBC mass is restored. complete or partial control of the anesthesiologist that can
Adjuvant therapy typically includes exogenous EPO and reduce intraoperative blood loss or exposure of the patient
supplementation with iron and vitamins, particularly to allogeneic RBC transfusion. These include nonspecific
vitamins B12 and folate. interventions such as patient positioning, management of
EPO is a glycoprotein hormone that is the primary ventilation and arterial carbon dioxide levels, and mainte-
regulator of erythropoiesis. It is primarily produced in nance of normothermia unless some degree of hypother-
the kidney, but also in small amounts in the liver. mia is specifically indicated, as during cardiopulmonary
Circulating EPO levels are regulated by a classic negative bypass. More specific anesthetic interventions include the
feedback loop; there is an inverse relation between tissue use of controlled hypotension, regional anesthetic tech-
oxygenation and serum EPO levels. Two sources of EPO niques, and autologous blood transfusion, which includes
include a recombinant DNA technique that employs preoperatively collected autologous blood (PAD), the in-
mammalian expression vectors (epoietin ) and EPO traoperative salvage of shed blood, and the use of ANH.
obtained and purified from human urine. Both types of
EPO have identical amino acid structures but different,
although similar, oligosaccharide side chains.
The primary indications for EPO therapy have been POSITIONING
chronic anemia associated with chronic renal failure, HIV Positioning of the patient depends largely upon the
infection, and cancer. During the last 10 years, there has operative site and needs of the surgeon. When possible,
been increasing experience with EPO in perioperative the operative site can be elevated or positioned above
patients, in whom the hormone increases erythropoiesis the level of the heart, which helps to reduce venous
and RBC mass and may reduce the need for allogeneic pressure and blood pooling at the surgical site. However,
blood transfusion during or after surgery. Three specific this positioning does increase the relative risk of venous
strategies for the use of EPO in elective preoperative air embolism. In surgical procedures on the spine,
patients are: careful positioning of the patient to avoid pressure on
1. Increase or maintain RBC mass in patients who the abdominal wall and solid organs serves to reduce
undergo PAD of blood pressure on collateral venous plexuses in the spinal
2. Increase RBC mass in patients without PAD epidural space.44
3. Increase RBC mass before planned acute normo-
volemic hemodilution (ANH) in the operating room
EPO has been administered to perioperative patients MANAGEMENT
according to a wide range of dosing schemes. One goal of OF VENTILATION
current investigation is to identify the most cost-effective
dosing regimen for different perioperative uses. EPO may During positive-pressure mechanical ventilation, mean
be administered either intravenously or subcutaneously. intrathoracic pressure is increased during the inspiratory
Typical doses are in the range of 100 to 500 IU per kg, cycle and may be further increased by the addition of
administered one to three times a week during the 3 to positive end-expiratory pressure. Increased intrathoracic
4 weeks before elective surgery. It is essential to provide pressure decreases venous return and may contribute
the patient with adequate amounts of supplemental iron to peripheral pooling of blood. Spontaneous ventilation
and vitamins, or the response to the relatively costly EPO reduces mean intrathoracic pressure and facilitates ve-
treatment may be suboptimal. nous blood return from the body. Resistance to airflow
502 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
during expiration also influences venous return and may REGIONAL ANESTHETIC
impact blood loss during surgery. Effective treatment of
bronchospasm, optimal adjustment of the I:E (inspiratory TECHNIQUES
to expiratory) ratio, and avoiding obstruction of breath-
ing circuits and endotracheal tubes can indirectly reduce The effect of regional anesthesia, specifically central neu-
blood loss.45 Providing effective ventilation and avoidance raxial (subarachnoid and epidural) blockade, on hemo-
of hypercarbia serves to minimize sympathetic nervous dynamics and measured blood loss in a variety of
system activation and the associated increases in CO, different surgical procedures is complex. In general,
vascular resistance, and blood pressure. these blocks are associated with reductions in mean
arterial, peripheral venous, and central filling pressures
while CO is maintained near normal or even increased.
Studies supporting the notion of reduced blood loss
MAINTENANCE during regional compared to general anesthesia are
OF NORMOTHERMIA conflicting, although some investigations claim a 30%
to 40% reduction in blood loss.48 Strict comparisons
In homeothermic organisms, including humans, hemo- of regional and general anesthetic techniques are dif-
static mechanisms function most effectively within the ficult because of a number of confounding factors,
range of normothermia. Below approximately 35 C, there including mode of ventilation, use of muscle relax-
is progressive enzymatic inhibition of both platelets and ants, differences in operative site, and other coexisting
clotting factors through temperature-induced alterations morbidities.
in kinetic activities.46 Hypothermia also independently
reduces the circulating platelet count. Importantly, hem-
orrhage and hypothermia are two thirds of the lethal
triad, which includes acidosis, that characterize death AUTOLOGOUS BLOOD
from acute hemorrhagic shock.47 Interventions that main- TRANSFUSION
tain body temperature or reduce heat losses include
active and passive heating and humidification of respi- This includes the appropriate use of any preoperatively
ratory gases, increasing the ambient temperature in the collected autologous blood from the patient (PAD),
operating room, employing forced-air warming blankets intraoperatively salvaged and processed RBCs through the
and mattresses, minimizing exposure of the patient, and use of a cell saver suctioning and collection apparatus,
warming fluids and blood products before infusion. and ANH.
AS-1, ADSOL (adenine, dextrose, sorbitol, sodium chloride; Fenwal Division of Baxter Healthcare Corp, Deerfield, IL); AS-3, Nutricel (Cutter
Biological, Division of Miles Labs, Emeryville, CA); WBC, white blood cell; CPD, citrate phosphate dextrose; CPDA-1, citrate phosphate dextrose
adenine; PRBCs, packed red blood cells; CMV, cytomegalovirus; HLA, human leukocyte antigen.
504 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
are determined, a process that takes 5 to 15 minutes; liver disease, and perhaps patients with excessive bleeding
once these types are known, type-specific blood that who have been taking antiplatelet agents including
has not been crossmatched may be released by the blood aspirin.51
bank and transfused with little risk of transfusion reac-
tion.
In elective situations, autologous or directed-donor
homologous blood may be available. Autologous blood
FIBRINOLYTIC INHIBITORS:
includes blood the patient donated for him or her- EPSILON-AMINOCAPROIC
self preoperatively, the patients blood obtained intra-
operatively in the process of ANH or cell salvage, or
ACID AND TRANEXAMIC
blood obtained postoperatively from a variety of salvage ACID
devices.
Two drugs, epsilon-aminocaproic acid and tranexamic
One unit of PRBCs or whole blood may be admin-
acid are lysine analogs that bind to plasminogen and
istered in as little as several minutes to as long as 4
plasmin at the binding sites for fibrinogen and fib-
hours, depending upon the severity of anemia, extent of
rin, and therefore interfere with the fibrinolytic process.
coexisting hypovolemia, and flow rate limitations of vas-
They are known as antifibrinolytic or antiplasmin agents.
cular access catheters and fluid administration devices.
Tranexamic acid is 6 to 10 times more potent than
Factors to be considered in determining the end-point of
epsilon-aminocaproic acid. Both drugs are initially ad-
RBC transfusion include restoration of the hemoglobin
ministered with an intravenous loading dose followed
or hematocrit to the minimum acceptable level for the
by continuous infusion. Potential side effects of antifib-
individual patient, coexisting morbidities or conditions
rinolytic agents are hypotension with rapid intravenous
such as prematurity or advanced age, and the likeli-
administration and various complications related directly
hood of continued blood loss following completion of
to inhibition of the fibrinolytic system. Pulmonary em-
the transfusion.
bolism; renal, carotid, and cerebral artery thrombosis;
and deep venous thrombosis have been described in case
reports.51,52
TABLE 34.6 Complications of Treatment of Acute reflect effective intravascular red blood cell mass for
Anemia a matter of hours or a day or more.
4. Arterial oxygen content is primarily dependent upon
Effects of Fluid/Volume Administration RBC mass, as expressed in the equation:
CIRCULATORY OVERLOAD
Electrolyte abnormalities CaO2 = (1.34 Hgb in g/dL SaO2 )
Dilution of coagulation factors + (0.003 PaO2 )
Embolism of air or particles
Hypothermia 5. Anesthetic interventions that may reduce intra-
Reaction to anticoagulants in blood products (citrate operative blood loss include appropriate patient
toxicity) positioning, effective ventilation, maintenance of
Immunologic reactions normothermia, controlled hypotension, and ANH.
HEMOLYSIS OF RED BLOOD CELLS 6. Treatment of acute blood loss anemia must include
Febrile (nonhemolytic) transfusion reaction correction of hypovolemia and the reduction of
Allergic reaction intravascular RBC mass. At the point that organ and
Transfusion-related acute lung injury (TRALI) tissue oxygen delivery becomes critically impaired,
Anaphylactic reaction which varies among different types of patients,
Delayed hemolytic transfusion reaction transfusion of blood products containing RBCs is
Immune modulation often required.
GRAFT-VERSUS-HOST DISEASE 7. The treatment of acute anemia entails of a number of
Increased susceptibility to infection possible complications, including intravascular vol-
EFFECTS ON ORGAN/TISSUE GRAFT SURVIVAL ume overload, electrolyte disturbances, transmission
Increased susceptibility to malignancy recurrence of infectious disease, a range of effects upon the im-
Propagation of infectious disease mune system, and a variety of transfusion reactions.
INFECTIOUS
Bacterial REFERENCES
Viral
Parasitic 1. Keating EM, Meding JB, Faris PM, et al. Predictors of
transfusion risk in elective knee surgery. Clin Orthop. 1998;
357:50.
2. Goodnough LT, Vizmeg K, Sobecks R, et al. Prevalence
and classification of anemia in elective orthopedic surgery
patients: Implications for blood conservation programs. Vox
Sang. 1992;63:90.
What Are Potential 3. Benoist S, Panis Y, Pannegeon V, et al. Predictive factors
for perioperative blood transfusions in rectal resection for
Complications of Treatment cancer: A multivariate analysis of a group of 212 patients.
Surgery. 2001;129:433.
of Acute Anemia? 4. Dunne JR, Malone D, Tracy JK, et al. Perioperative anemia:
An independent risk factor for infection, mortality, and
There are a variety of ways to categorize the complications resource utilization in surgery. J Surg Res. 2002;102:
of treatment of acute anemia: Near or long-term, infec- 237.
5. Carson JL, Duff A, Poses RM, et al. Effect of anaemia and
tious or noninfectious, immunologic or nonimmunologic.
cardiovascular disease on surgical mortality and morbidity.
Table 34.6 lists the potential complications of treatment
Lancet. 1996;348:1055.
of acute anemia with a transfusion of one or more blood 6. Dougenis D, Patrinou V, Filos KS, et al. Blood use in lung
products. Discussion of these complications is beyond the resection for carcinoma: Perioperative elective anaemia does
scope of this chapter. not compromise the early outcome. Eur J Cardiothorac Surg.
2001;20:372.
7. Beutler E, Waalen J. The definition of anemia: What
is the lower limit of normal of the blood hemoglobin
concentration? Blood. 2006;107:1747.
8. Blanc B, Finch CA, Hallberg L, et al. Nutritional anaemias.
KEY POINTS Report of the WHO Scientific Group. WHO Tech Rep Ser.
1. Acute anemia has two broad causes: Blood loss and 1968;405:1.
hemolysis. 9. Spence RK, Cernaianu AC, Carson J, et al. Transfusion and
2. No single hemoglobin or hematocrit value defines surgery. Curr Probl Surg. 1993;30:1101.
anemia; a variety of functional characteristics and 10. Izaks GJ, Westendorp RG, Knook DL, et al. The definition of
anemia in older persons. JAMA. 1999;281:1714.
comorbidities must be considered in labeling a patient
11. Coleman DH, Stevens AR Jr, Dodge HT, et al. Rate of
anemic. blood regeneration after blood loss. AMA Arch Intern Med.
3. In acute blood loss anemia, isolated hemoglobin or 1953;92:341.
hematocrit readings can be deceiving and promote 12. Messmer K. Compensatory mechanisms for acute dilutional
a false sense of security. They may not accurately anemia. Bibl Haematol. 1981;47:31.
506 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
13. Shander A, Knight K, Thurer R, et al. Prevalence and 35. Britton LW, Eastlund DT, Dziuban SW, et al. Predonated
outcomes of anemia in surgery: A systematic review of the autologous blood use in elective cardiac surgery. Ann Thorac
literature. Am J Med. 2004;116:59S. Surg. 1989;47:529.
14. Leung JM, Weiskopf RB, Feiner J, et al. Electrocardiographic 36. Love TR, Hendren WG, OKeefe DD, et al. Transfusion of
ST-segment changes during acute, severe isovolemic hemod- predonated autologous blood in elective cardiac surgery.
ilution in humans. Anesthesiology. 2000;93:1004. Ann Thorac Surg. 1987;43:508.
15. Weiskopf RB, Kramer JH, Viele M, et al. Acute severe 37. Owings DV, Kruskall MS, Thurer RL, et al. Autologous blood
isovolemic anemia impairs cognitive function and memory donations prior to elective cardiac surgery: Safety and effect
in humans. Anesthesiology. 2000;92:1646. on subsequent blood use. JAMA. 1989;262:1963.
16. Toy P, Feiner J, Viele MK, et al. Fatigue during acute 38. Americans with Disabilities Act. (Pub. L. No. 101336, 104
isovolemic anemia in healthy, resting humans. Transfusion. Stat. 327 (codified at 42 U.S.C. 1210112213), 1990.
2000;40:457. 39. 21 CFR 606.121 Recognition and use of a standard for uni-
17. Vincent JL, Baron JF, Reinhart K, et al. Anemia and blood form blood and blood component container labels. www.fda
transfusion in critically ill patients. JAMA. 2002;288:1499. .gov/cber/gdlns/unilabbld.htm. Accessed May 29, 2007.
18. Hillman RS, Hershko C. Acute blood loss anemia. In: 40. Menitove JE, ed. Standards for blood banks and transfusion
Beutler E, Lichtman MA, Coller BS, et al. eds. Williams services, 18th ed. Bethesda: American Association of Blood
hematology, 6th ed. Columbus: McGraw-Hill; 2001:677. Banks; 1997:49.
19. Duke M, Abelmann WH. The hemodynamic response to 41. 21 CFR 606.40 Code of federal regulation title 21,
chronic anemia. Circulation. 1969;39:503. Vol 7. www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/
20. Chapler CK, Cain SM. The physiologic reserve in oxygen CFRSearch.cfm?fr=606.40. 2006.
carrying capacity: Studies in experimental hemodilution. 42. Popovsky MA, Whitaker B, Arnold NL. Severe outcomes of
Can J Physiol Pharmacol. 1985;64:7. allogenic and autologous blood donation: Frequency and
21. Spahn DR, Leone BJ, Reves JG, et al. Cardiovascular and characterization. Transfusion. 1995;35:734.
coronary physiology of acute isovolemic hemodilution: A re- 43. ASA. Standards for basic anesthetic monitoring. Approved
view of nonoxygen-carrying and oxygen-carrying solutions. by the House of Delegates on October 21, 1986, and last
Anesth Analg. 1994;78:1000. amended October 25, 2005. Available at: http://www.asahq
22. Murray JF, Escobar E, Rapaport E. Effects of blood viscosity .org/publicationsAndServices/standards/02.pdf. Accessed
on hemodynamic responses in acute normovolemic anemia. November 7, 2006.
Am J Physiol. 1969;216:638. 44. Park CK. The effect of patient positioning on intraabdominal
23. Ebert RV, Stead EA Jr, Gibson JG. Response of normal pressure and blood loss in spinal surgery. Anesth Analg. 2000;
subjects to acute blood loss. Arch Intern Med. 1941;68:578. 91:552.
24. Adams RC, Lundy JS. Anesthesia in cases of poor surgical 45. Bertrand D, Hannhart B, Laxenaire MC. Haemodynamic
risk: Some suggestions for decreasing the risk. Surg Gynecol effects of inspiratory and expiratory resistances in anaes-
Obstet. 1941;71:1011. thetic breathing systems during induced hypotension. Eur J
25. Penden JC Jr, Maxwell M, Ohin A, et al. A consideration of in- Anaesthesiol. 1985;2:353.
dications for preoperative transfusions based on analysis of 46. Rohrer MJ, Natale AM. Effect of hypothermia on the
blood volumes and circulating proteins in normal and mal- coagulation cascade. Crit Care Med. 1992;20:1402.
nourished patients with and without cancer. Ann Surg. 1960; 47. Dutton RP. Hypothermia and hemorrhage. In: Spiess BD,
151:303. Spence RK, Shander A, eds. Perioperative transfusion
26. Kirklin J. Cardiac surgery. New York: Churchill Livingstone; medicine, 2nd ed. Philadelphia: Lippincott Williams &
1993. Wilkins; 2006:481.
27. Martineau RJ. Pro: A hematocrit of 20% is adequate to wean 48. Modig J, Karlstrom G. Intra- and post-operative blood loss
a patient from cardiopulmonary bypass. J Cardiothorac Vasc and haemodynamics in total hip replacement when per-
Anesth. 1996;10:291. formed under lumbar epidural versus general anaesthesia.
28. DeFoe GR, Ross CS, Olmstead EM, et al. Lowest hematocrit Eur J Anaesthesiol. 1987;4:345.
on bypass and adverse outcomes associated with coronary 49. Shander A. Acute normovolemic transfusiona practical
artery bypass grafting. Ann Thorac Surg. 2001;71:769. approach. Transfus Alternatives Transfus Med. 1999;4:7.
29. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT study: 50. Cash JD, Gader A, De Costa J. The release of plasminogen
Anemia and blood transfusion in the critically illcurrent activator and factor VIII by LVP, AVP, DDAVP, AT III, and
clinical practice in the United States. Crit Care Med. 2004; OT in man. Br J Haematol. 1984;30:363.
32:39. 51. Levy JH. Pharmacological approaches to prevent or decrease
30. Goodnough LT. Autologous blood donation. Crit Care. bleeding in surgical patients. In: Spiess BD, Spence RK,
2004;8(Suppl 2):S49S52. Shander A, eds. Perioperative transfusion medicine, 2nd ed.
31. Kasper SM, Gerlich W, Buzello W. Preoperative red cell Philadelphia: Lippincott Williams & Wilkins; 2006:403.
production in patients undergoing weekly autologous blood 52. Tanaka K, Takao M, Yada I. Alterations in coagulation and
donation. Transfusion. 1997;37:1058. fibrinolysis associated with cardiopulmonary bypass during
32. Goodnough LT, Monk TG, Andriole GL. Erythropoietin open heart surgery. J Cardiothorac Anesth. 1989;3:181.
therapy. N Engl J Med. 1997;336:933. 53. Mangano DT, Tudor IC, Dietzel C, et al. The risk associated
33. Renner SW, Howanitz PJ, Bachner P. Preoperative autolo- with aprotinin in cardiac surgery. N Engl J Med. 2006;
gous blood donation in 612 hospitals. Arch Pathol Lab Med. 354:353.
1992;116:613. 54. Food and Drug Administration. Public health advisory
34. Cazzola M, Mercuriali F, Brugnara C. Use of recombinant aprotinin injection (marketed as Trasylol). Available
human erythropoietin outside the setting of uremia. Blood. at: http://www.fda.gov/cder/drug/advisory/aprotinin.htm. Ac-
1997;89(12):42484267. cessed March 25, 2006.
CHAPTER TRANSFUSION REACTIONS
CASE SUMMARY
What Is the Historical
72-year-old woman was scheduled to un-
Perspective of Complications
A
dergo elective primary arthroplasty of her
right hip for osteoarthritis. She had coro- and Reactions to Blood
nary artery disease and had an angioplasty
with a stent insertion in her proximal right
Transfusion?
coronary artery 4 months before her surgery.
Her medications included aspirin, clopidogrel, atenolol, A historical perspective on blood transfusions was pub-
and rosuvastatin. She weighed 55 kg, and was 156 cm tall. lished in 1998 by Rossi et al.1 William Harvey first
Clopidogrel was stopped 5 days before surgery. Preopera- described the circulatory system in 1628 in his book,
tive full blood count revealed a hemoglobin concentration Exercitatio Anatomica de Motu Cordis et Sanguinis in
of 10.5 g per dL, and her platelet count was 220 109 per L. Animalibus (An Anatomical Exercise on the Motion of
During surgery under general anesthesia, she was noted to the Heart and Blood in Animals). The first recorded
have diffuse bleeding and lost approximately 3 L of blood. blood transfusion in a human was performed by Jean
Four liters of lactated Ringers solution, 6 units of packed Baptiste Denis in Paris on June 15, 1667. Denis trans-
red blood cells (RBCs), 4 units of fresh frozen plasma fused the blood of a lamb to a 15-year-old boy. The
(FFP), and 4 platelet units were administered. She was ex- first reports of human-to-human blood transfusion began
tubated postoperatively and transferred to the orthopedic emerging in the early 19th century. These transfusions
ward. That evening, she became progressively dysphonic predated the recognition of different blood groups and
and hypotensive, and was transferred to the intensive care were confounded by difficulty with collection, storage,
unit. Her chest radiograph showed nonspecific pulmonary and administration of blood. Not surprisingly, the out-
infiltrates, and she had a low PaO2 /FIO2 index. The diag- comes of these bold treatments were indifferent at best.
nosis was transfusion-related acute lung injury (TRALI). It was not until Karl Landsteiner identified the A, B, and
Mechanical ventilation was initiated, and the patient re- C (later renamed O) blood groups in 1900 that we began
quired 4 days of positive-pressure ventilation. She suffered to understand the importance of immunohematology. Ot-
hemodynamic instability and required vasopressor ther- tenberg introduced compatibility testing in 1907. The Rh
apy. There was no evidence of myocardial infarction or system was discovered in 1939 and by the 1940s, col-
ischemia. She was discharged from the intensive care unit lection and storage of donor blood, along with reliable
after 6 days and spent another week convalescing in the compatibility testing, had rendered blood transfusion al-
hospital before being discharged home. most routine, and apparently safe. However, in 1943,
TRALI is one of many recognized complications Beeson described posttransfusion hepatitis and by 1962,
associated with blood transfusion. The complications it became apparent that there was an association be-
of transfusion are serious, can be fatal, and are costly. tween paid blood donors and posttransfusion hepatitis.
To minimize the incidence of transfusion reactions, In 1983, the first report of the acquired immune de-
the clinician needs to understand the pathophysiology ficiency syndrome (AIDS) transmission associated with
involved and appreciate the measures required in avoiding transfusion (in an infant) was published. Hepatitis and
unnecessary transfusion. AIDS are not alone, and transmission of bacterial and
507
508 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 35.3 Current Estimates of Risk of Transmission in the development of transfusion-associated hepatitis,
of Viral Diseases from Blood Transfusion although it appears that this hepatitis is self-limiting and
does not result in chronic liver disease.6 TTV has not been
Estimated Risk of conclusively linked to human disease.
Transmission Per Unit
Virus of Blood Transfused
TT virus 1:10 to 1:50
SEN virus 1:50 What Are the Special Concerns
Hepatitis B virus 1:31,000 to 1:82,000a Related to Cytomegalovirus
Hepatitis C virus 1:3,100,000
Human immunodeficiency 1:4,700,000 to 1:10,000,000
Transmission?
virus
CMV transmission in particular, is associated with trans-
a The risk for risk of clinical disease transmission for HBV is estimated fusion of leukocytes; hence, leukodepleted RBCs and
at 1:1,200,000. platelet units are less likely to be a vehicle for its trans-
Data from: British Columbia Provincial Blood Coordinating Office. mission. Blood products that do not contain cellular
Physicians Guide 2004: physicians guide for blood and blood prod- components do not transmit CMV; FFP and cryoprecipi-
uct utilization. Vancouver, B.C.: British Columbia Provincial Blood tate are safe in this regard. CMV infection is an undesirable
Coordinating Office; 2003 [cited 2006 May 3]. Available from:
outcome in immunocompromised patients, and patients
http://www.pbco.ca/images/stories/ic%202003%20final%
20with%20changes.pdf. Accessed May 31, 2007. Physicians
at greatest risk should receive only leukodepleted blood
guide 2004 blood and blood product utilization. Available at from CMV seronegative donors. These patients include
http://www.pbco.ca/ic.html. Accessed May 3, 2006. the following:
CMV seronegative pregnant women
of time between a donor becoming potentially infected CMV seronegative women undergoing intrauterine
and demonstrating nucleic acid testingdetectable HIV transfusion
nucleic acids is 13 days. Before this, the window for the CMV seronegative patients undergoing allogeneic bone
antibody testing alone was approximately 22 days. marrow transplantation
The estimated risks for transmission of viral disease Patients undergoing solid organ transplantation from a
per blood component transfused are calculated mathe- CMV seronegative donor
matically using input variables such as the virus incidence CMV seronegative patients with conditions that are
in the donor population, as well as the window of time of likely to require an allogeneic bone marrow transplan-
the current screening techniques to identify the disease. tation
The current estimates5 of transmission risk of some CMV seronegative patients with HIV infection
viral diseases from blood transfusion are listed in
Table 35.3.
chain through ingestion of contaminated meat products. immunologically mediated reaction, typically caused by
It is likely that the time from infection with this particular recipient antibodies reacting with donor RBC antigens.
prion to the onset of symptoms may be many years or Usually, it is the preformed anti-A or anti-B antibodies
even decades. Whereas CJD is not associated with blood in the recipient which react with donor RBCs and
transfusion, vCJD may well be. In at least one patient cause rapid destruction. Occasionally, anti-Rhesus or anti-
in the United Kingdom who died from vCJD, there is Kidd antibodies are involved. On rare occasions, donor
strong suggestive evidence that vCJD was transmitted antibodies can react with recipient antigens.
by transfusion of contaminated blood products;3 another Activation of the complement system is a major
report of preclinical vCJD was likely to have been component of an acute HTR (AHTR). Opsonized RBCs
transmitted by blood transfusion.2 are phagocytized by macrophages, a process mediated
As of yet, there is no screening test for vCJD, by complement receptors. Activated phagocytes produce
and diagnosis is usually confirmed at autopsy. Because inflammatory cytokines that, in turn, result in the
of the long latency period between infection and the clinical manifestations of AHTR.10 The appreciation
development of symptoms, it is difficult to estimate of the pathophysiology of AHTR should lead to the
accurately the number of donors contaminated with development of specific treatment strategies in the future.
vCJD and the exact risk of transmission and subsequent However, at present, treatment is largely supportive.
development of this disease. It is possible, although there Management of AHTR includes immediate cessation of
is no conclusive evidence, that universal leukodepletion the offending transfusion and supportive therapy of
may reduce the risk of transmission of vCJD. compromised organ systems. Severe AHTRs usually occur
soon after the initiation of transfusion of incompatible
blood. HTRs are variable in their clinical presentation;
they can range from innocuous to life-threatening or fatal
What About NonInfectious reactions. There is a classical triad of fever, pain, and
hematuria, which is seldom seen. HTRs are clinically
Transfusion Reactions? characterized by dyspnea, fever and chills, chest and
back pain, hypotension, and tachycardia. Other symptoms
Noninfectious transfusion reactions include those medi- include nausea, pain at the site of the intravenous
ated by immunologic mechanisms, as well as those that infusion site, flushing, and apprehension. Disseminated
affect the recipients immune system directly or indi- intravascular coagulation and acute renal failure are
rectly. Other nonimmunologic and noninfectious compli- major complications associated with a high mortality.
cations also occur, including mechanistic complications Indeed, AHTR is the most common cause of death
of transfusion. associated with transfusion. The successful management
of an HTR depends on its early recognition. Blood from
both the recipient and donor should be sent to the blood
bank to confirm whether incompatible transfusion has
What Are the Immunologic occurred. Other investigations to confirm the diagnosis
Complications of Blood include direct and indirect Coombs tests. Free hemoglobin
may be seen in the serum of clotted blood. Serum bilirubin
Transfusion? is elevated, usually within a few hours of the reaction, and
hemoglobinuria is invariably a feature. Anemia ensues
These complications include acute transfusion reactions from hemolysis. Subsequent disseminated intravascular
and delayed transfusion reactions, and can further be coagulation and acute renal failure are both related to the
classified as being either hemolytic or nonhemolytic. Ana- consequences of hemolysis.
phylactic reactions, TRALI, transfusion-associated graft- Since the recognition of the ABO and Rhesus blood
versus-host disease (TA-GVHD), and immunosuppression groups, and because of routine compatibility testing, these
are also all recognized complications of blood transfu- reactions are usually a consequence of an administra-
sion. Posttransfusion purpura (PTP) is a rare compli- tive error. The Serious Hazards of Transfusion (SHOT)
cation. Febrile, nonhemolytic transfusion reactions are scheme in the United Kingdom has published accumu-
frequent but relatively benign reactions to blood trans- lated data from 8 years of national surveillance; from this
fusion. Immunosuppression following blood transfusion comprehensive data set,10 it is evident that the most fre-
is an increasingly recognized phenomenon with serious quently occurring, adverse event in transfusion medicine
implications for blood product recipients. is the transfusion of incompatible blood. On the basis of
a denominator of 27 million blood components issued by
the United Kingdom blood services, it is estimated that the
risk of incompatible blood transfusion is 1:15,000. The risk
What Are Hemolytic of ABO-incompatible blood transfusion is 1:100,000, and
the risk of death from an incompatible blood transfusion
Transfusion Reactions? is approximately 1:1,500,000.11 This estimate is consistent
with those in other studies from other countries.
When incompatible blood is transfused to a patient, There are frequently complex system failures identi-
the consequence is referred to as an HTR. This is an fied in the post hoc analysis of incompatible transfusions.
512 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
The process of ordering, supplying, and administering of the offending trigger and supportive therapy, includ-
blood is a complex one, with many potential points where ing epinephrine, vasopressors, volume resuscitation, and
error can and does occur. These include incorrect patient antihistamine and corticosteroid treatment.
identification, errors in sample labeling, laboratory errors, If patients are known to be IgA-deficient and have
and, of course, the incorrect administration of blood prod- anti-IgA antibodies, they should receive washed RBCs and
ucts. Mislabeling of samples is frequent, and is estimated blood components from IgA-deficient donors whenever
to occur once per every 165 samples collected. Mislabeled transfusion is required.
and incorrectly collected samples are the most common Minor allergic reactions, including mild urticarial
sources of error. Sample collection errors are between reactions, pruritus, erythema, and mild respiratory symp-
1,000 and 10,000 times more likely to occur than viral toms occur with relative frequency and mandate the im-
transmission from blood transfusion.12 mediate cessation of transfusion and supportive therapy.
If the reactions resolve, and the cutaneous manifestations
respond to antihistamine therapy, transfusion may re-
sume, provided there are no signs or symptoms of a severe
What Are Delayed Transfusion anaphylactic reaction. Pretreatment with corticosteroids
and antihistamines may prevent these reactions.
Reactions?
Delayed hemolytic transfusion reactions (DHTRs) are
less severe than AHTRs. They occur in patients who What Is Transfusion-Related
have developed antibodies to RBCs following previous
transfusion or pregnancies. Usually the antibody presence Acute Lung Injury?
is weak and is not detected during standard crossmatching
procedures. Typically, DHTRs occur between 24 and TRALI is a clinical syndrome, which includes dyspnea,
72 hours following transfusion, and they may be clinically hypoxemia, and bilateral pulmonary infiltrates on radio-
so benign as to remain undetected. These reactions graphs, and occurs within 24 hours of blood transfusion,
tend to be hemolytic reactions. Although DHTRs may with no other apparent cause. Hypotension and fever are
be clinically significant, it appears that they are not common features. Typically, symptoms begin within 4 to
often fatal.10 DHTRs manifest with progressive anemia 6 hours from the start of transfusion, and the severity
secondary to hemolysis. There may be associated elevation of this syndrome can range from mild to severe.13 Virtu-
of bilirubin and hematuria. Fever and leukocytosis may ally all patients who develop TRALI require supplemental
mimic infection. If DHTR is suspected, the blood bank oxygen therapy, and up to 70% of patients with TRALI
should be notified, and the patient should be tested to will require mechanical ventilation.14 The pathogenesis
identify the culprit antibody. Future transfusions should of TRALI is almost certainly related to human leukocyte
be free of the corresponding antibody. antigen (HLA) antibodies in donor blood, although recip-
ient antibodies have also been implicated. It is thought
that these donor antibodies bind to antigens on recipi-
ent leukocytes, with subsequent monocyte activation and
What Is the Incidence of the increased intracellular generation of interleukin-1,
tumor necrosis factor , and tissue factor. The release
Anaphylaxis and Allergic of these cytokines probably results in the secondary acti-
Reactions, and What vation of neutrophils and endothelial cells, with the net
result of endothelial damage and capillary leak. The pul-
Mechanisms Are Involved? monary endothelial/epithelial interface is proposed as the
initial site of damage, which is postulated to stimulate the
Pure or primary anaphylactic reactions appear to be production of further proinflammatory mediators, with
uncommon after transfusion of RBCs and are mostly asso- recruitment of more inflammatory cells. The acute respi-
ciated with platelet and FFP transfusion.10 Anaphylactic ratory distress syndrome is the end result and the common
reactions were believed to result from a reaction between presenting feature following primarily extrapulmonary in-
donor immunoglobulin A (IgA) and anti-IgA antibodies, sults. The pathogenesis of TRALI has not yet been fully
which are produced by IgA-deficient recipients; however, elucidated, and donor blood, which contains HLA anti-
the SHOT report10 does not support this theory. Never- bodies, will not universally cause TRALI in recipients.
theless, it would seem prudent to administer IgA-deficient A popular theory is that patient factors may contribute
blood components to IgA-deficient patients. to developing TRALI. Hypoxia, recent surgery, cytokine
Most anaphylactic reactions remain unexplained. therapy, active infection or inflammation, massive trans-
Anaphylaxis usually occurs within 45 minutes of the be- fusion, and biologically active lipids in stored blood have
ginning of transfusion. Urticaria appears to be a common all been implicated as contributing factors; when a sec-
finding when anaphylaxis occurs following blood transfu- ond hit of HLA antibodies in donor blood is given to
sion. Other clinical signs of anaphylaxis include airway these patients, they go on to develop TRALI.15,16 The true
swelling and obstruction, bronchospasm, and cardiovas- incidence of TRALI is unknown, but it is almost certainly
cular collapse. Treatment includes immediate cessation underdiagnosed.
CHAPTER 35/TRANSFUSION REACTIONS 513
The proper treatment of TRALI depends on making of donor lymphocytes. Photochemical treatment has also
the diagnosis; it can be difficult to distinguish TRALI been suggested as an effective alternative to irradiation of
from other causes of respiratory compromise. Supportive blood products.17
treatment, including mechanical ventilation with lung
protective ventilation strategies, is accepted treatment.
Hypotension should be treated with judicious volume
resuscitation, and, occasionally, pressor agents will be What Is Posttransfusion
required. It is imperative to distinguish fluid overload and
pulmonary edema from cardiac failure from TRALI, as Purpura?
intravenous fluid resuscitation could be dangerous in the
former. One of the distinguishing features of TRALI is PTP is an unusual but serious complication of blood
that its prognosis is significantly better than that from transfusion thought to be associated with transfusion of
other forms of acute lung injury (ALI). Overall mortality blood containing platelet antigens to patients who do not
from TRALI is 6% to 10%, relatively low compared to ALI. have that antigen. Most cases occur following transfu-
Recovery usually occurs within a few days, and it appears sion of platelets from human platelet antigen (HPA)-1a
that long-term pulmonary complications such as fibrosis positive donors to HPA-1b homozygous recipients. Mul-
or structural damage are rare.13 tiparous patients are more likely to develop PTP as a
When TRALI is diagnosed or suspected, the culprit result of sensitization to HPA-1a antigens during prior
donor blood should ideally be tracked, and the donor pregnancies. PTP is caused by the reaction of preformed
traced and excluded from further contributions to the alloantibodies in the recipient with the platelet antigens
donor blood pool. Because multiparity is associated with in the donor blood. Clinically, the features of PTP in-
higher rates of HLA sensitization, initially only samples clude a sudden and precipitous fall of the platelet count in
from female (multiparous) donors should be tested.15 the recipient of a blood transfusion to fewer than 15,000
per L 3 to 12 days following transfusion. Patients show
signs of purpura, epistaxis, mucous membrane hemor-
rhage, and bleeding from the gastrointestinal and urinary
What Is Transfusion-Associated tracts. This bleeding is often life-threatening, and in-
tracranial hemorrhage can be fatal. Mortality approaches
Graft-Versus-Host Disease? 10%, and is usually due to intracranial bleeding. Intra-
venous globulin is used to treat this condition, and
TA-GVHD is a rare, but frequently fatal, complication diagnosis requires a high index of suspicion, along with
of blood transfusion. This disease is the result of en- serologic detection of HPA, typically HPA-1a in the vic-
graftment and the proliferation of donor lymphocytes in tim.18
transfusion recipients. Donor T cells become activated Initial diagnosis may be confused with heparin-
against alloantigens in the recipient, which in turn leads induced thrombocytopenia. Patients who have had PTP
to lymphocyte proliferation and the production of cy- should subsequently receive only antigen-negative trans-
tokines. TA-GVHD may occur following transfusion of fusions.
RBCs, platelet concentrates, FFP, or granulocytes. The
development of this condition is unusual because, ordi-
narily, donor lymphocytes are destroyed by the recipient
immune system. There are two circumstances when this What Are Febrile Nonhemolytic
fails: When the recipient is immunocompromised and
when there is specific HLA matching between donor and Transfusion Reactions?
recipient, as may occur with transfusion of components
transfused from relatives, or by chance. Immunodefi- Febrile reactions occur approximately once in every 300
cient states where this may occur include Hodgkins RBC transfusions and once in every 10 platelet trans-
and non-Hodgkins lymphoma, leukemia, myeloma, and fusions. Febrile reactions are attributed to nonspecific
other conditions associated with diminished, recipient cytokines present in the transfused blood, as well as to an-
cell-mediated immunity, including patients undergoing tibodies in the recipient, which react with non-RBC donor
bone marrow or stem cell transplants. antigens (usually on donor leukocytes). The typical febrile
The initial clinical presentation of this condition is reaction occurs within several hours of transfusion and
nonspecific and includes fever, hepatic failure, erythema, may be accompanied by rigors, nausea, vomiting, and hy-
and diarrhea. Pancytopenia ensues, usually approximately potension. Treatment requires recognition and is largely
2 weeks following the transfusion. Overwhelming sepsis is supportive. Acetaminophen is the mainstay therapy, but
the usual cause of death. Bone marrow biopsy can confirm corticosteroids may also be of benefit. Antihistamine ther-
the diagnosis, and aggressive immunosuppressive therapy apy has not been shown to be of any use, and nonsteroidal
may improve outcome, although survival is rare. anti-inflammatory drugs are not routinely recommended
Leukodepletion does not entirely prevent TA-GVHD, because of the potential compromise of renal function in
and it is traditionally recommended that individuals at patients who may be at significant risk for transfusion
risk receive blood products that have been exposed to reactions. Table 35.4 lists the noninfectious transfusion
-radiation, which destroys any potential for engraftment reactions.
514 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
transfusion-induced reaction, which occurs primarily in appears that this is not a major clinical concern, and there
those who receive multiple transfusions over prolonged is evidence that the significant coagulation and formation
periods of time. Patients with thalassemia, sickle cell of microaggregates is not a risk in the context of rapid
disease, and myelodysplasia are at greater risk of this blood transfusion.23
complication. Resuscitation should not be delayed by changing
Ringers lactate solutions for normal saline, but when
blood is being diluted with crystalloid and given slowly, it
would seem prudent to mix it with saline.
What Can Cause Hypotension
During Transfusion?
Is Autologous Cell Saved Blood
Hypotension associated with vasoactive substances is
considered a risk. Normal angiotensin-converting en- Free of Complications?
zyme (ACE) activity and normal pulmonary circulation
are required to metabolize bradykinin. Bradykinin is Complications associated with transfusion of salvaged
generated when plasma is frozen because of activation autologous blood have also been reported. These include
of the prekallikrein activator. This may occur in the rigors and pyrexia after the infusion of unwashed, sal-
donor plasma or after infusion into recipients. Negatively vaged, and citrated blood from surgical drains following
charged blood filters can also induce the generation of knee replacement surgery. The mechanism of these reac-
bradykinin. Hypotension has been reported in patients tions is not clear, but intravenous infusions of cytokines
receiving FFP while on cardiopulmonary bypass and in in the salvaged blood and acrylic monomers from bone ce-
patients on ACE inhibitors who receive blood through ment in the blood salvage system have been implicated.10
negatively charged filters. Bradykinin is thought to play Patientswhoreceiveautologouscellsavedbloodshould
a role in all of these circumstances. ACE inhibition re- be monitored closely, and adverse reactions dealt with and
duces the ability of these patients to metabolize the reported in the same way that they are reported in cases of
exogenous bradykinin. Prestorage leukoreduction should allogeneic blood transfusion. Similarly, predonated autol-
reduce bradykinin generation, as leukodepleting filters are ogous blood is not free from the mechanistic and metabolic
positively charged. complications of blood transfusion. Human error can occur
Occasionally, transfusion of blood and blood products whenautologousbloodisbeinggiven,andbacterialcontam-
may be associated with severe hypertensive reactions, and ination of autologous blood can also occur.
it is thought that pressor agents may be generated by
transfusion, although this has not been proven.
drugs are prescribed. Surgical techniques and anesthetic of O2 consumption and CO2 production by bacteria,
management can impact blood loss, and antifibrinolytic nucleic acid testing for bacterial DNA, and direct bacterial
agents can reduce perioperative transfusion requirements culture.27
in high-risk groups, although they may introduce throm-
botic complications. The use of intraoperative cell salvage
can reduce the need for transfusion and monitoring of co-
agulation function. Point-of-care testing and adherence What Conclusions Can We
to protocol-driven strategies may reduce unwarranted
transfusion of blood components. There are emerging Draw, and How Should We
indications for the use of recombinant factor VIIa to re- React when a Transfusion
duce blood loss during surgery; however, the enthusiasm
for the use of antifibrinolytic therapy, as well as for the Reaction Is Suspected?
off-license administration of factor VIIa should be tem-
pered against the potential for adverse thromboembolic If an adverse reaction is suspected during blood transfu-
events.24 sion, the transfusion should be immediately discontinued
Transfusion triggers should be selected for patients, and the hospital transfusion service should be notified.
taking into account the comorbidity. Restrictive blood Residual blood products should be returned to the blood
transfusion triggers in critically ill patients, other than bank and a blood culture should be drawn from the pa-
those with acute infarction or unstable angina, has been tient at a site remote from the implicated transfusion.
shown to probably be superior to liberal transfusion This will aid the diagnosis of sepsis and contribute to the
strategies.25 The search for safe, oxygen-carrying blood differential diagnosis of immediate transfusion reactions.
substitutes is ongoing, and despite more than 20 years of Many delayed reactions may go undetected and conse-
investigation, there is still no product readily available. quently unreported. If, for example, a patient develops
graft-versus-host-disease, they may conceivably die from
overwhelming infection, and the diagnosis of TA-GVHD
may never be made. It is essential, therefore, for clinicians
Does Leukoreduction Reduce dealing with blood transfusion to maintain a high index
of suspicion when recently transfused patients become
the Incidence of Transfusion seriously ill.
Reactions? There has been much publicity regarding the compli-
cations of blood transfusions in many countries around
Leukoreduction (WBC filtration) is advocated by many as the world. Most of the public concern has focused on
a means to reduce some of the complications of blood the transmission of infectious agents and the potential
transfusion, particularly HLA alloimmunization, CMV for transmission of disease. It is clear that with current
transmission, and recurrent febrile reactions. Several screening techniques and the high degree of vigilance and
blood agencies routinely provide prestorage leukodeple- standards that are expected of blood product providers,
tion. Prestorage leukodepletion is preferable to bedside the risk of disease transmission from contaminated blood
filtration because the delay between filtration and transfu- is now very low. Nevertheless, we need to remain cautious;
sion affords time for the breakdown of bradykinin, which as yet, we do not know the exact risks for newer and exotic
can cause hypotension in patients on ACE inhibitors. diseases such as SARS and vCJD. The World Health Orga-
Universal leukodepletion, although widely practiced, is nization recommends routine testing of all donated blood
expensive. Questions remain as to the benefit versus the for transfusion-transmissible infections, and collection of
cost of this practice.26 blood only from voluntary, unpaid blood donors who are
at low risk of carrying transmissible infections.
The immunologic complications of blood transfusion
are also not yet completely understood. As we develop our
How Can We Reduce the understanding of the risks of TRALI, immunosuppression,
and the effects of blood transfusion on major outcomes
Incidence of Bacterial from surgical procedures and therapies, we will modify
Contamination of Blood our use of these valuable resources.
The future may hold the availability of hemoglobin
Products? substitutes and genetically engineered blood factors,
which hopefully will reduce the number of complica-
Improved donor screening, meticulous asepsis when per- tions from blood transfusion. At present, however, blood
forming venipuncture, and removal of the first aliquot transfusion remains a key component of anesthesia care.
of donor blood can reduce this complication. Leukocyte It is the responsibility of everyone who initiates or admin-
reduction, as well as minimizing and optimizing storage isters blood transfusions to use all necessary measures
time are also important. Ruling out bacterial contami- and rationale in doing so. Multidisciplinary collaboration,
nation before transfusion has also been advocated and blood conservation strategies, and appropriate transfu-
includes visual inspection of components, direct staining sion triggers should be employed to reduce inappropriate
of donor blood, bacterial ribosomal assays, measurement transfusion.
518 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
It is very significant that, by far, the most frequent infusion of microaggregates, air embolism, hemol-
complication of, and the biggest risk to, patients who re- ysis, hypothermia, and impairment of hemostatic
ceive blood transfusion is that of incompatible blood or function. These complications are frequently associ-
blood products being administered. Blood banks and lab- ated with massive transfusion.
oratories that provide and test blood for incompatibility 13. Hypotension may be due to bradykinin generated
have very rigorous processes in place to prevent incompat- by negatively charged blood filters. Patients on
ible blood transfusion. The primary responsibility remains ACE inhibitors who receive blood through nega-
with the clinical personnel who collect and administer tively charged filters might become hypotensive.
blood to ensure that all samples are meticulously col- Prestorage leukoreduction should reduce bradykinin
lected and labeled. Absolute adherence to guidelines and generation, as leukodepleting filters are positively
protocols to prevent incompatible transfusions remain the charged.
mainstay of avoiding an essentially human set of errors. 14. Saline is the preferred carrier for blood products;
Safe bedside transfusion practice is a mandatory require- however, in clinical practice, rapid transfusion of
ment, and anesthesia practitioners have a very big role blood through intravenous sets that have been
to play in the safe administration of blood and blood primed with lactated Ringers solution is not likely
products. to cause problems.
15. Autologous blood, whether predonated or cell-
salvaged, is not free of risks to the patient.
16. Blood conservation is an important component of
KEY POINTS reducing the need for transfusion and the risks of
transfusion to patients.
1. Although apparently benign viral infection is poten-
17. Leukoreduction is increasingly practiced during
tially relatively common, serious viral illnesses are
collection of blood and, although expensive, appears
currently very rarely associated with transfusion of
to justify the cost by reducing the incidence of
blood and blood products.
transfusion reactions.
2. Special considerations should be given to higher-risk
18. Meticulous attention to sterility is essential, starting
patients, such as those with immunosuppression,
with collection of blood, and should be carried
parturients, multiparous women, and patients who
through to completion of the administration of blood
have undergone transplantation.
products.
3. Bacterial infections are the most common infective
19. Human error remains the most avoidable source of
complications, particularly following platelet trans-
transfusion reactions. Incompatible transfusions are
fusion.
invariably the result of human failures in transfusion
4. Bacterial sepsis is a very serious complication of
practice.
transfusion with a high mortality.
5. AHTRs are caused by incompatible transfusions.
These are life-threatening complications, with asso- REFERENCES
ciated high morbidity and mortality.
1. Rossi EC, Simon TL, Moss GS, et al. A history of transfu-
6. True anaphylaxis is rare and may be more common sion. In: Spiess BD, ed. Perioperative transfusion medicine.
in IgA-deficient patients. Baltimore: Lippincott Williams &Wilkins; 1998:3.
7. TRALI is an underdiagnosed complication of blood 2. Peden AH, Head MW, Ritchie DL, et al. Preclinical vCJD
transfusion, with serious implications to patients. It after blood transfusion in a PRNP codon 129 heterozygous
consumes many healthcare resources. patient. Lancet. 2004;364:477.
8. TA-GVHD is the result of engraftment and prolifera- 3. Llewelyn CA, Hewitt PE, Knight RS, et al. Possible transmis-
tion of donor lymphocytes in transfusion recipients, sion of variant Creutzfeldt-Jakob disease by blood transfu-
and survival is unlikely. sion. Lancet. 2004;363:417.
9. PTP can occur following transfusion of platelets 4. Sandler SG, Yu H, Rassai N. Risks of blood transfusion and
their prevention. Clin Adv Hematol Oncol. 2003;5:307.
from HPA-1a positive donors to HPA-1b homozy-
5. British Columbia Provincial Blood Coordinating Office.
gous recipients. Mortality is high (10%), usually due Physicians Guide 2004: physicians guide for blood and
to intracranial bleeding, and requires a high index blood product utilization. Vancouver, B.C.: British Columbia
of suspicion. Multiparous patients are more likely to Provincial Blood Coordinating Office; 2003 [cited 2006
develop PTP as a result of sensitization to HPA-1a May 3]. Available from: http://www.pbco.ca/images/stories/
antigens during prior pregnancies. ic%202003%20final%20with%20changes.pdf. Accessed May
10. Febrile, non-HTRs are common, but are largely 31, 2007.
benign when caused by nonspecific cytokines in 6. Sagir A, Kirschberg O, Heintges T, et al. SEN virus infection.
transfused blood. Rev Med Virol. 2004;14:141.
11. Immunosuppression is common, and may have far- 7. Schmidt M, Brixner V, Ruster B, et al. NAT screening
of blood donors for severe acute respiratory syndrome
reaching consequences in particular subgroups of
coronavirus can potentially prevent transfusion-associated
recipients of blood and blood products.
transmissions. Transfusion. 2004;44:470.
12. Mechanistic complications are associated with tech- 8. Kleinman S, Chan P, Robillard P. Risks associated with trans-
niques of transfusion and can have serious effects fusion of cellular blood components in Canada. Transfus Med
on patients. These include circulatory overload, Rev. 2003;17:120.
CHAPTER 35/TRANSFUSION REACTIONS 519
9. Davenport RD. Pathophysiology of hemolytic transfusion 18. Woelke C, Eichler P, Washington G, et al. Post-transfusion
reactions. Semin Hematol. 2005;42:165. purpura in a patient with HPA-1a and GPIa/IIa antibodies.
10. Stainsby D, Cohen H, Jones H, et al. Serious hazards of Transfus Med. 2005;16:69.
transfusion: annual report 2004. Manchester, U.K.: The 19. Spiess BD. Blood transfusion: The silent epidemic. Ann
Serious Hazards of Transfusion Steering Group; 2005 Thorac Surg. 2001;72:S1832.
[cited 2006 April 26]. Available from: http://www.shotuk.org/ 20. Spiess BD, Royston D, Levy JH, et al. Platelet transfusions
SHOTREPORT2004.pdf. Accessed May 31, 2007. during coronary artery bypass graft surgery are associated
11. Stainsby D. ABO incompatible transfusionsexperience with serious adverse outcomes. Transfusion. 2004;44:1143.
from the UK serious hazards of transfusion (SHOT) 21. Engoren MC, Habib RH, Zacharias A, et al. Effect of blood
scheme transfusions ABO incompatible. Transfus Clin Biol. transfusion on long-term survival after cardiac operation.
2005;12:385. Ann Thorac Surg. 2002;74:1180.
12. Dzik WH, Murphy MF, Andreu G, et al. An international 22. Johnston TD, Chen Y, Reed RL. Functional equivalence
study of the performance of sample collection from patients. of hypothermia to specific clotting factor deficiencies.
Vox Sang. 2003;85:40. J Trauma. 1994;37:413.
13. Kopko PM, Marshall CS, MacKenzie MR, et al. Transfusion- 23. Lorenzo M, Davis J, Negin S, et al. Can Ringers lactate be
related acute lung injury. Report of a clinical look-back used safely with blood transfusions? Am J Surg. 1998;175:
investigation. JAMA. 2002;17:1968. 308.
14. Moore SB. Transfusion-related acute lung injury (TRALI): 24. OConnell KA, Wood JJ, Wise RP, et al. Thromboembolic
Clinical presentation, treatment and prognosis. Crit Care adverse events after use of recombinant human coagulation
Med. 2006;34:S114. factor VIIa. JAMA. 2006;18:293.
15. Curtis BR, McFarland JG. Mechanisms of transfusion- 25. Herbert PC, Wells G, Blajchman MA, et al. A multicenter ran-
related acute lung injury (TRALI): Anti-leukocyte antibodies. domized, controlled clinical trial of transfusion requirements
Crit Care Med. 2006;34:S118. in critical care. N Engl J Med. 1999;340:409.
16. MacLennan S. Transfusion related acute lung injury (TRALI). 26. Dzik WH. Leukoreduction of blood components. Curr Opin
Information document INF/MED/CM/026/01. Available at: Hematol. 2002;9:521.
http://www.blood.co.uk/hospitals/guidelines/pdf/TRALI.pdf. 27. Hillyer CD, Joesphson CD, Blajchman MA, et al. Bacterial
Accessed April 26 2006. contamination of blood components: Risks, strategies, and
17. Grass JA, Wafa T, Reames A, et al. Prevention of transfusion- regulation. Joint ASH and AABB educational session in
associated graft-versus-host disease by photochemical treat- transfusion medicine. Hematology (American Society of
ment. Blood. 1999;93:3140. Hematology Education Program Book). 2003:575.
CHAPTER BLEEDING COMPLICATIONS
A
rhagia was scheduled for a total abdominal administration of pharmacologic agents. The Section
hysterectomy. Her history was otherwise un- How Should the Clinician Approach the Diagnosis
remarkable; she was taking no medications, and Treatment of Bleeding Diatheses? presents an
and had not had prior surgical procedures. approach to the diagnosis and treatment of bleeding dis-
Hemoglobin was 11.1 g per dL. A general orders. Disorders of the hemostatic mechanism may also
anesthetic was administered and proceeded uneventfully. result in hypercoagulable states (see Chapter 38).
At the conclusion of the hysterectomy, the surgeon noted
difficulty achieving a dry field and ultimately closed the
wound with a Jackson-Pratt drain in place. The nursing
notes report three dressing changes during the patients How Does the Normal
stay in the postanesthesia care unit, and blood tinging of
the urine was noted. However, vital signs remained stable, Hemostatic Mechanism
and the patient was transferred to the surgical ward. At Function?
10:00 PM that evening, a house officer was summoned to
evaluate the patient for relative oliguria and mild hypoten-
sion (systolic pressures in the 90s). A crystalloid bolus was In this chapter, only an abbreviated discussion follows.
administered. At the time of the first evaluation of vital A detailed description of the coagulation mechanism
signs following the nursing shift change, the patient was can be found in another publication.1 The nomenclature
found in a state of agonal respiration with no palpable (numerals and common names) and the half-lives of the
pulses. Resuscitation was attempted but was unsuccess- clotting factors are presented in Table 36.1.
ful. A blood specimen drawn earlier by the house officer
revealed a hematocrit of 15%. Factor analysis performed
on blood drawn during the attempted resuscitation led THE COAGULATION
to a suspected postmortem diagnosis of von Willebrand MECHANISM
disease (vWD), the presence of which was confirmed in
first-degree relatives. Although the classical, dual cascade model of coagulation
with its intrinsic and extrinsic pathways (see Fig. 36.1)
probably provides a reasonable model of coagulation, as
INTRODUCTION it is evaluated in vitro by the activated partial thrombo-
plastin time (aPTT) and prothrombin time (PT) determi-
Abnormalities of hemostasis that result in clinical coag- nations respectively, that model provides an inadequate
ulopathies are common in the operating room and the representation of in vivo coagulation. It suggests impor-
intensive care unit (ICU). Understanding the etiology and tant roles for factors XII and XI, congenital deficiencies
treatment of these abnormalities requires a knowledge of of which cause relatively little clinical disturbance of co-
the normal coagulation mechanism, which is discussed in agulation. In addition, it fails to explain why a patient
greater detail in Chapter 38, but a brief review is presented with hemophilia who lacks an intrinsic pathway factor
here (see Section How Does the Normal Hemostatic (hemophilia A, factor VIII; hemophilia B, factor IX) can-
Mechanism Function?). Abnormalities of hemostasis not achieve hemostasis through the unaffected extrinsic
may be either congenital (see Section What Are the Im- pathway. A description of the current understanding of
portant Congenital Abnormalities of Hemostasis?) the three stages of the hemostatic process, which has
520
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 521
been thoroughly defined and described by Hoffman, are to yield a complex of TF and activated (indicated by
summarized in the following text and in Figure 36.2.2 a lower case a) factor VII (Fig. 36.2B). The TF-VIIa
complex then activates factors IX and X (Fig. 36.2C). Ac-
Activation tivated factor X (Xa) then activates factor V (Fig. 36.2D),
leading to the formation of the prothrombinase com-
Activation of the coagulation process begins when dis- plex (Xa and Va) on the phospholipid surface provided
ruption of the vascular endothelium exposes blood to by TF. The prothrombinase complex catalyzes the con-
tissue factor (TF) (Fig. 36.2A). TF activates factor VII version of prothrombin (factor II to thrombin (FIIa)
(Fig. 36.2E). This initial formation of thrombin serves
Contact activation Injury or inflammation to advance the coagulation process to the more efficient
exposes tissue factor amplification phase that follows. Note that the TF-VIIa
XII XIIa
complex-mediated activation of Xa is a self-limited one
that generates only small amounts of thrombin. It is
XI XIa TF regulated in a negative feedback manner by the factor Xa-
VIIa VII
mediated generation of tissue factor pathway inhibitor
Intrinsic
pathway
IX IXa (TFPI).3 Teleologically, this pathway probably serves to
VIIIa Extrinsic
prevent extensive and spontaneous clot formation in the
pathway extravascular space. It provides the explanation for why
(Platelet surface) IXa-VIIIa TF-VIIa patients with hemophilia (A or B) cannot simply rely on
the extrinsic pathway (TF-VIIa mediated formation of Xa)
X X
Xa to generate thrombin. The thrombin yield of that pathway
Va is too modest to provide the thrombin burst necessary for
the eventual consolidation of the platelet plug by fibrin
Xa-Va
(Fig. 36.2I).
Common
pathway (Thrombin) II IIa (Prothrombin)
Amplification
(Fibrinogen) I Ia (Fibrin) Whereas it was the phospholipid surface provided by
XIIa membrane-bound TF that initiated the coagulation pro-
Cross-linked fibrin cess, it is now the phospholipid surface provided by
activated platelets that serves to perpetuate it. The breach
FIGURE 36.1 The intrinsic and extrinsic pathways of in the vascular tree that began the activation process also
coagulation. Factors in the inactive form are within shaded exposed platelets to collagen, to which they become bound
squares. The shaded ovals represent phospholipid surfaces by von Willebrand factor (vWF) through the glycoprotein
(tissue factor [TF] or platelets). (GP) Ib-IX-V receptor complex on the platelet surface (see
Fig. 36.3). The thrombin just generated by the TF-bound
522 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
A
TF
Fibrin Fibrinogen
B
TF I II IIa
Prothrombinase
VII VIIa complex Xa-Va
C
TF-VIIa
IX IXa X Xa
X Xa IX IXa
H
XI XIa
D IXa-VIIIa
TF-VIIa Va Tenase
Xa IIa IIa complex
V Va vWF-VIII:C
Prothrombinase
E G
complex TF-VIIa ctive platelet
Xa-Va Ina VIIIa
Va
F
II IIa
(Thrombin)
IIa
FIGURE 36.2 The coagulation mechanism. See text for a detailed explanation. TF, extravascular
membrane-bound tissue factor; vWF-VIII:C, circulating factor VIII bound to its carrier protein, the
von Willebrand factor. (Reproduced with permission from Drummond JC, Petrovitch CT.
Hemotherapy and hemostasis. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical anesthesia, 5th
ed. Philadelphia: Lippincott Williams & Wilkins; 2006:208.)
GPIIb/IIIa
GPIIb/IIIa Fibrinogen
GPIb GPIIb/IIIa
GPIb
vWF
Endothelial cells
vWF
Subendothelium
FIGURE 36.3 Platelet adhesion and aggregation. When the endothelium is denuded, von
Willebrand factor (vWF) binds to collagen in the subendothelial layer. Platelets adhere through their
glycoprotein 1b (GPIb)-IX-V receptors to vWF. Platelets aggregate to one another by cross-linking
through fibrinogen (or vWF, not shown) between GPIIb-IIIa receptors expressed on the platelet
surface during the process of platelet activation. (Reproduced with permission from Drummond JC,
Petrovitch CT. Hemotherapy and hemostasis. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical
anesthesia, 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2006:208.)
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 523
prothrombinase complex supports the amplification of the is impaired. In high concentrations, FDPs impair platelet
coagulation process in four ways, which are as follows: function, inhibit thrombin, and prevent the cross-linking
of fibrin strands,4 and thereby lead to bleeding. Excess
1. First, it further activates the adjacent platelets
plasmin can also lead to a bleeding diathesis. This is in
(Fig. 36.2F). That activation results in platelet sur-
large part because plasmin degrades fibrin. However, be-
face changes, most notably the appearance of the
cause it is a serine protease, plasmin can also degrade
GPIIb-IIIa receptor, and the release of the contents
other coagulation factors, including fibrinogen, factors V,
of platelet granules. Among those contents are adeno-
VIII, XIII, and vWF.5 In addition, plasmin can also digest
sine diphosphate (ADP), a powerful platelet activator
the platelet receptor, GPIb. Accordingly, circulating plas-
and proaggregant that rapidly recruits other platelets
min inhibits platelet function and disrupts coagulation in
to the growing platelet mass, and FV.
several ways. Circulating plasmin is normally inhibited
2. Thrombins second effect promotes the activation of
immediately by various antiplasmins, the most important
FV to FVa (Fig. 36.2G).
of which is 2 -antiplasmin. Antiplasmin capacity may be
3. Third, thrombin releases circulating FVIII from its
exceeded in conditions in which the fibrinolytic system
carrier molecule (vWF) and activates it (Fig. 36.2G).
produces large quantities of plasmin (primary fibrinoly-
4. Fourth, thrombin activates factor XI (see Fig. 36.2H).
sis, DIC) and may contribute to the bleeding diathesis that
Factor XIa in turn activates factor IX (Fig. 36.2H),
occurs in these conditions.
further adding to the pool of factor IXa that first formed
during the activation phase above (Fig. 36.2C). The
net result of this amplification stage is the availability
of activated platelets and activated factors V, VIII, NATIVE CONTROL
and IX.
OF COAGULATION
Propagation There are numerous anticoagulant mechanisms that hold
the normal coagulation mechanism in check. These are
The platelet then provides the phospholipid surface on mentioned only briefly here because abnormalities of
which two coagulation factor complexes form to lead to these mechanisms are more likely to manifest themselves
the explosive generation of thrombin. First, factors VIIIa as hypercoagulable states rather than bleeding diatheses.
and IXa form the tenase complex, which activates fac- The first line of defense is composed of substances secreted
tor X (Fig. 36.2H). The resultant Xa forms an additional by normal endothelium that inhibit platelet aggrega-
prothrombinase complex (Xa-Va), and large amounts of tion (prostacyclin, nitric oxide) and coagulation (heparan
thrombin are elaborated (Fig. 36.2I). Thrombin (IIa) cat- sulphate, ADPases, thrombomodulin) and promote fibri-
alyzes the formation of fibrin, which acts to cross-link the nolysis (tissue plasminogen activator [tPA]). Circulating
platelets to stabilize the friable platelet plug (Fig. 36.3). inhibitors include protein C and antithrombin III (ATIII).
Thrombin also activates the thrombin activatable fibrinol- Protein C (with protein S as a cofactor), which is activated
ysis inhibitor and factor XIII (fibrin stabilizing factor), at sites of coagulation by a complex of thrombomodulin
both of which serve to stabilize the fibrin clot. Fibrin and thrombin, inhibits factors Va and VIIIa. ATIII is a cir-
monomers initially aggregate relatively loosely to form culating serine protease inhibitor that binds to thrombin
clot composed of soluble fibrin (fibrin S), held together and thereby inactivates it. It also binds and inactivates,
loosely by hydrogen bonds. Fibrin stabilizing factor (not although less avidly, each of the activated clotting factors
shown in figure) mediates the formation of covalent pep- of the classical intrinsic coagulation cascade: Factors
tide bonds between the fibrin monomers to produce a XIIa, XIa, IXa, and Xa.6
stable (insoluble) fibrin clot.
Fibrinolysis
What Are the Important
Fibrinolysis leads to the dissolution of fibrin clots. Fibri- Congenital Abnormalities
nolysis, which requires the production of plasmin from
plasminogen, serves to eventually remodel fibrin clots and of Hemostasis?
recanalize vessels that have been occluded by thrombo-
sis. The degradation products yielded by plasmin action Patient history is invaluable in the identification of disor-
on fibrin are called fibrin degradation products (FDPs)or ders of hemostasis. Abnormalities of primary hemostasis,
fibrin split products. Under normal circumstances, FDPs which are usually the result of reduced platelet number or
are removed from the blood by the liver, kidney, and retic- function, will be revealed by evidence of skin and mucosal
uloendothelial system. However, if FDPs are produced bleeding, including easy bruising, petechiae, prolonged
at a rate that exceeds their normal clearance capacity, bleeding from minor skin lacerations, recurrent epistaxis,
they will accumulate. This occurs most often when the and menorrhagia. Coagulation abnormalities are associ-
fibrinolytic system is excessively active (e.g., disseminated ated with deep bleeding events including hemarthroses
intravascular coagulation [DIC]), or when liver function or hematomas after blunt trauma.
524 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
that the results of common coagulation tests (platelet classified as mild, moderate, and severe. With mild dis-
count, aPTT, PT) may be normal in the patient with vWD. ease, factor levels are 5% to 30% of normal, and abnormal
Although the half-life of VIII:C is diminished in vWD, bleeding usually occurs only following trauma. With mod-
there is usually sufficient factor VIII:C to yield a normal erate disease, factor levels are 1% to 5% of normal, and
aPTT in basal conditions. spontaneous bleeding occasionally occurs. Most patients
The established treatments for vWD are DDAVP and with hemophilia have the severe form of the disease. Fac-
factor concentrates.9 DDAVP, which promotes release of tor VIII:C levels are <1% of normal, and spontaneous
vWF, is effective first therapy for the majority (approxi- bleeding episodes are frequent. Disease severity typically
mately 80%) of patients with vWD, including those with correlates with the level of clotting factor activity. As with
type I and type IIA disease. However, the identification of vWD, patients with hemophilia should avoid other agents
subtype IIB (described previously) is important because that interfere with hemostasis, for example, heparins,
DDAVP will cause thrombocytopenia in these patients.9 aspirin, and other platelet-inhibiting agents.
DDAVP (also discussed in Section How Should the
Clinician Approach the Diagnosis and Treatment of Diagnosis and Treatment of Hemophilia A
Bleeding Diatheses?), given intravenously in a dose
History will typically reveal the x-linked recessive pattern
of 0.3 g per kg, increases factor VIII:C and vWF two-
of disease inheritance. The diagnosis is confirmed by a
to fivefold in most patients. Its effect is maximal after 30
prolonged aPTT (with a normal PT and BT) and factor
minutes, and increased levels persist for 6 to 8 hours.12 For
assays demonstrating a deficiency of factor VIII coagulant
patients in whom the response to DDAVP is inadequate,
activity with normal levels of vWF, factor IX, and factor XI.
factor concentrates containing vWF and factor VIII will
Hemophilia A is treated with plasma-derived concentrates
be appropriate;9 virally inactivated concentrates are avail-
that have been treated by viral attenuation procedures or
able. Antifibrinolytic agents, epsilon-aminocaproic acid
with recombinant factor VIII (rFVIII).14 Before elective
(EACA) and tranexamic acid (TXA), are sometimes used
surgery, factor supplementation should be managed by
in combination with DDAVP to manage these patients
a hematologist. Factor replacement is typically chosen
during the perioperative period,13 and may be given in-
to achieve a target procoagulant activity. A procoagulant
travenously, or orally. They have also been administered
level of 25% is a common target for achieving control of a
topically as mouthwashes in patients with vWD under-
spontaneous bleeding episode. The necessary replacement
going dental extractions. Oral contraceptives (estrogens)
must be calculated on the basis of the patients plasma
have been used to treat patients with vWD and men-
volume (40 mL of plasma per kilogram of body weight).
orrhagia or who are undergoing elective surgery.13 The
One unit of procoagulant activity is defined as the amount
mechanism of action of the estrogens is not understood,
of procoagulant activity present in 1 mL of plasma with
although an effect on vWF synthesis is suspected. An-
100% of the normal level. For an 80-kg patient (plasma
tiplatelet drugs should be avoided.
volume 3,200 mL), 800 units of factor VIII:C (25%
3,200 mL) would be required. For elective surgery, the
The Hemophilias target level of factor VIII:C activity is typically 50% to
100% of normal. Many patients with hemophilia develop
Hemophilia A and B are sex-linked recessive disorders, inhibitors to factor VIII:C. The presence of the inhibitor
which occur almost exclusively in males. Hemophilia A increases the amount of factor VIII:C that must be
is the result of a deficient or functionally defective fac- administered to manage a given hemostatic challenge.
tor VIII:C. Hemophilia B (Christmas disease) and C are Recombinant activated factor VIIa (see the following text)
caused by a deficiency or abnormality of factors IX and may be required for the patient with inhibitors.
XI, respectively.14 Hemophilia C is an autosomal reces- DDAVP may also be effective in mild hemophilia,
sive disorder that occurs almost exclusively in Ashkenazi and is thought to cause the release of factor VIII:C
Jews.14 The relative frequencies of the three hemophilias from liver endothelial cells. There is a large variation in
are: Factor VIII:C (85%); factor IX (14%); and factor XI patient response to DDAVP. It is most effective in patients
(1%). Inherited deficiencies of factors II, VII, V, and XI with factor VIII:C levels >5%.15,16 As with vWD (above),
also occur but are rare.14 Patients with hemophilia most DDAVP is administered intravenously in a dose of 0.3 g
commonly experience deep tissue bleeding, hemarthroses, per kg, in 50 mL of saline, over 15 to 30 minutes. It
and hematuria. Approximately 50% of the operations causes a prompt increase in factor VIII:C. Tachyphylaxis,
performed in patients with hemophilia are orthopedic however, limits its usefulness.
procedures required for treatment of the arthritic conse- The antifibrinolytics, EACA and TXA, have been
quences of hemarthroses. widely used before dental procedures. However, they are
contraindicated in bleeding episodes involving joints or
Hemophilia A the urinary tract because they inhibit the clearance of
clots from those spaces.
Factor VIII is a large macromolecule with two compo-
nents: Coagulant factor VIII (VIII:C) and vWF. The VIII:C
molecule circulates bound to and protected by vWF. Pa- Hemophilia B
tients with hemophilia A have normal levels of vWF but Like hemophilia A, factor IX deficiency is also an
reduced or defective factor VIII:C. Hemophilia A occurs x-linked recessive disorder. It occurs in approximately
in approximately 1 in 10,000 males. Hemophilia A is 1 in 25,000 males14 and produces a bleeding diathesis
526 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
of Hemostasis?
suppression. Infiltration of the bone marrow by cancer
For organizational purposes, bleeding disorders can be cells or replacement by fibrosis will also result in inade-
classified according to which of the three hemostatic quate platelet production.
processes are involved: Primary hemostasis (platelet
disorders); coagulation (clotting factor disorders); and Nonimmunologically Mediated Consumption
fibrinolysis (production of inhibitors, e.g., FDPs). Some This type of consumption occurs with the extensive acti-
disorders involve more than one process. Coagulation vation of coagulation with or without the occurrence of
tests may also focus on determining whether the clinical DIC. After extensive tissue damage, for example, burns or
problem involves primary hemostasis (decreased platelet massive crush injuries, with the associated denuding of
count, increased BT, etc.), coagulation (prolonged PT and the vascular endothelium, the normal process of hemosta-
aPTT, decreased factor levels, etc.), fibrinolysis (increased sis activates platelets, leading to their consumption and to
FDPs, increased D-dimer), or a combination of the thrombocytopenia. Similarly, the interaction of platelets
three. with nonendothelialized structures, such as large vascular
grafts or with native vessels during any extensive vasculitis
(e.g., toxemia of pregnancy), can also lead to a transient
ACQUIRED DISORDERS thrombocytopenia. The many conditions that cause DIC
OF PLATELETS (discussed in the following text) will also cause platelets
to be consumed or destroyed faster than they can be
The clinical conditions that cause an isolated disorder produced.
of primary hemostasis typically involve abnormalities of
either platelet number or function.8 Immunologically Mediated Consumption
This pattern of consumption can be caused by various
Thrombocytopenia drugs (heparin, quinidine, cephalosporins), autoimmune
disorders (thrombotic thrombocytopenic purpura, sys-
Platelets are derived from megakaryocytes in the bone
temic lupus erythematosis, rheumatoid arthritis), and
marrow in response to thrombopoietin, which is synthe-
alloimmunization resulting from previous blood transfu-
sized by the liver. The causes of thrombocytopenia may
sions or pregnancy. Heparin-induced thrombocytopenia
be categorized (see Table 36.2) as inadequate production
is discussed in the subsequent text.
by the bone marrow, increased peripheral consump-
tion or destruction (nonimmune mediated), increased
peripheral destruction (immune mediated), dilution, and Dilution of Platelets
sequestration. Massive transfusion may be associated with dilutional
thrombocytopenia (see subsection Massive Transfusion).
Decreased Bone Marrow Production
Platelets are derived from megakaryocytes in response to Sequestration
thrombopoietin, which is synthesized by the liver. Phys- Normally, approximately one third of platelets are se-
ical and chemical agents (radiation and chemotherapy), questered in the spleen. With splenic enlargement, more
various drugs (thiazide diuretics, sulfonamides, diphenyl- are sequestered, and thrombocytopenia may result. This
hydantoin, alcohol), infectious agents (hepatitis B, tuber- may occur with splenomegaly of any cause, including
culosis [TB], overwhelming sepsis), and chronic disease cirrhosis of the liver, although in that condition, decreased
states (uremia, liver disease) can all cause bone marrow production also contributes to thrombocytopenia.
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 527
Disorders of Platelet Function Caffeine, aminophylline, and theophylline are also in-
hibitors of phosphodiesterase and similarly produce mild,
Uremia reversible platelet inhibition.
Adenosine Diphosphate Receptor Antagonists. Ac-
Platelet dysfunction occurs commonly in patients with
tivation of the ADP receptor leads to expression of the
uremia. It is attributed to the accumulation of acids
IIb-IIIa receptor on the platelet surface. Ticlopidine and
that are thought to interfere with the platelets ability to
clopidogrel, both used primarily for stroke prophylaxis,
expose the PF3 phospholipid surface. These compounds
block the ADP receptor noncompetitively and irreversibly,
can be dialyzed, and dialysis frequently improves the
and thereby inhibit ADP-induced platelet aggregation.
hemostatic defect. An abnormality in the interaction of
Ticlopidine has been withdrawn from the market be-
vWF with platelet receptors is also suspected. DDAVP
cause of the occurrence of neutropenia and thrombotic
rapidly improves platelet adhesiveness in patients with
thrombocytopenic purpura.
uremia;17 the mechanism is not known with certainty.
Glycoprotein IIb-IIIa Receptor Antagonists. The
However, DDAVP has been shown to cause increased
GPIIb-IIIa site, by which fibrinogen crosslinks platelets,
expression of GPIb-IX in platelets.18 Erythropoietin and
is the final common pathway for platelet aggregation. The
conjugated estrogens have also been observed to cause
IIb-IIIa antagonists (abciximab, tirofiban, eptifibatide),
gradual improvement of the hemostatic defect associated which cause reversible inhibition of this cross-linking,
with uremia. The mechanisms of these effects are similarly have been used principally in the management of acute
unidentified. Cryoprecipitate will also improve the platelet coronary syndromes. These agents all require intravenous
dysfunction of uremia but, given the efficacy of DDAVP, administration. The half-lives are approximately 12 hours
the associated risks are not justified. Life-threatening for abciximab and 2.5 hours for tirofiban and eptifi-
bleeding in the patient with uremia should be managed batide.20 However, abciximab has a relatively high affinity
by the administration of platelet concentrates. for the IIb-IIIa receptor, and platelet dysfunction may be
longer (approximately 48 hours) than would be inferred
Antiplatelet Agents from the half-life. All these agents may cause throm-
bocytopenia (incidence: Abciximab 2.5%; tirofiban and
Numerous platelet-inhibiting medications are adminis-
eptifibatide 0.5%).21,22 These agents also cause prolonga-
tered to reduce the risk of myocardial infarction (MI),
tion of the activated clotting time (ACT).20
stroke, and other thromboembolic complications. They
induce platelet dysfunction by several mechanisms, in-
Herbal Medications and Vitamins
cluding inhibition of cyclooxygenase (COX), inhibition of
phosphodiesterase, ADP receptor antagonism, and block- Several herbal medications, including ginseng, gingko
ade of the GPIIb-IIIa receptor. biloba, garlic, and ginger (for mnemonic purposes, the
Cyclooxygenase Inhibitors. Aspirin is the proto- Gs) may cause inhibition of platelet function. The actual
type. Aspirin produces irreversible inhibition of platelet risks are not well defined. Nonetheless, they should be
COX, and therefore prevents the synthesis of thromboxane discontinued before surgery, and in particular, before
A2 , a potent platelet proaggregant and vasoconstrictor. In neurologic, cardiac, and cosmetic surgical procedures.
moderate doses, there is selective sparing of the synthesis Vitamin E is also a platelet inhibitor and should similarly
of prostacyclin (antiaggregant, vasodilator), which results be withheld.23,24 (See also Chapter 65.)
in shifting the balance substantially in favor of platelet in-
hibition. All of the nonsteroidal anti-inflammatory agents Other Conditions
(e.g., ibuprofen, indomethacin, phenylbutazone) similarly Myeloproliferative and myelodysplastic syndromes are as-
inhibit COX. However, their inhibition is promptly re- sociated with intrinsic defects of both platelet morphology
versible upon drug clearance. The recently introduced and function. The platelet dysfunction that occurs in con-
COX-2 inhibitors selectively inhibit the COX-2 isoform junction with other complex hemostatic disorders (liver
(responsible for generating the mediators of pain and disease, fibrinolytic states including DIC) is discussed in
inflammation) while sparing the COX-1 isoform (respon- the following text.
sible for many of the adverse effects of COX inhibitors
including gastric damage, decreased renal blood flow
and inhibition of platelet thromboxane A2 ). Accordingly, ACQUIRED DISORDERS
platelet function should not be impaired. However, prob- OF CLOTTING FACTORS
ably because COX-2 inhibitors reduce prostacyclin gener-
ation by vascular endothelial cells, they appear to tilt the (INCLUDING
natural balance toward platelet aggregation,19 which may ANTICOAGULANT THERAPY)
explain in part the increased rate of myocardial ischemic
events in patients taking specific COX-2 inhibitors. Vitamin K Deficiency
Phosphodiesterase Inhibitors. Cyclic adenosine
monophosphate (AMP) inhibits platelet aggregation, and Synthesis of clotting factors II, VII, IX and X, as well as
levels of cyclic AMP are increased by the inhibition of protein C and protein S by the liver, requires the presence
phosphodiesterase. Dipyridamole, used for stroke, and of vitamin K. Vitamin K is required for the carboxylation
cilostazol appear to act primarily by this mechanism. of these factors. Without the carboxyl group, these factors
528 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
are unable to adhere to phospholipid surfaces during urgency, fresh frozen plasma (FFP) is commonly adminis-
the coagulation process. When vitamin K deficiency tered. However, prothrombin complex concentrate, which
occurs, the K-dependent factors are depleted in an order contains factors II, VII, IX, X, has been reported to be
determined by their half-lives. Factor VII has the shortest more effective than FFP26,27 because FFP administration
half-life and is depleted first, followed by factors IX and frequently fails to achieve adequate levels of factor IX26
X, and finally factor II. and furthermore, some patients cannot tolerate the requi-
Vitamin K refers to a group of vitamins.24 Vitamin site volume, that is, approximately 15 mL per kg of FFP.
K1 (phylloquinone) is found in leafy green vegetables. If FFP or concentrates are administered, and sustained
Vitamin K2 (menaquinone) is synthesized by the normal reversal is desired, vitamin K should also be administered
intestinal flora, and it is therefore uncommon for pa- because of the short (6 hours) half-life of factor VII. rFVIIa
tients to develop vitamin K deficiency solely because of (discussed in the following text) has also been reported to
dietary deficiency. However, it may occur commonly in achieve rapid normalization of INR.28
patients who are receiving parenteral nutrition without Heparin Therapy. Heparin inhibits coagulation
vitamin K supplementation and who are being treated principally through its interaction with ATIII. Heparin,
concurrently with broad-spectrum antibiotics that alter in binding to ATIII, causes a conformational change
or destroy the gut flora. Vitamin K deficiency can de- that greatly increases ATIIIs thrombin inhibitory activ-
velop in as little as 1 week. Newborns, who have a sterile ity. ATIII also inhibits several activated factors including,
gut at birth, have been noted to develop vitamin K de- in addition to IIa (thrombin), Xa, IXa, XIa, and XIIa. It is
ficiency. Because vitamin K is a fat-soluble vitamin, it most active against thrombin and Xa. Heparin similarly
requires bile salts for absorption from the jejunum. Pa- increases the activity of a second circulating antithrom-
tients with biliary obstruction, pancreatic insufficiency, bin, heparin cofactor II, which inhibits thrombin but not
malabsorption syndromes, gastrointestinal (GI) obstruc- the other activated factors. Its contribution to the clinical
tion, or rapid GI transit can develop vitamin K deficiency effects of heparin is uncertain. Heparin resistance can
because of inadequate absorption. occur in patients who are deficient in ATIII on either a
hereditary or an acquired basis. The latter may occur in
Diagnosis and Treatment of Vitamin K Deficiency patients on sustained heparin therapy or in the presence
of depletion by a consumptive coagulopathy. Heparin re-
Vitamin K deficiency will cause prolongation of the PT. sponsiveness can be restored by the administration of
This occurs because factor VII is depleted first. With more ATIII concentrates29,30 or FFP.
severe deficiency, as levels of factors IX and X decrease,
the aPTT will also be increased. Platelet count will remain a. LOW MOLECULAR WEIGHT HEPARIN (LMWH): Low
normal. Vitamin K may be administered orally, intramus- molecular weight fractions of heparin are supplanting
cularly, or intravenously. Urgent treatment of vitamin K subcutaneous unfractionated heparin and coumadin
deficiency is best accomplished by the intramuscular or for DVT prophylaxis and treatment.31 There are sev-
intravenous administration of vitamin K (Aquamephy- eral available agents, including certoparin, dalteparin,
ton), usually in doses of 1 to 5 mg. Vitamin K should be danaparoid, enoxaparin, reviparin, and tinzaparin.
administered slowly to avoid hypotension. Improvement These agents do not appear to differ in terms of effi-
of the coagulation abnormality will begin to be apparent cacy.32 Enoxaparin is used most widely in the United
within 6 to 8 hours. States. LMWHs also act through ATIII but have greater
Warfarin Therapy. Warfarin produces its antico- activity against factor Xa than thrombin (IIa). The
agulant effect by competing with vitamin K for the ratio of Xa/IIa activity varies among the agents (enoxa-
carboxylation-binding sites (see the preceding text), and parin 3.8:1; tinzaparin 1.9:1).33 As a consequence, the
thereby causing depletion of factors II, VII, IX, X, protein effect of these agents on standard coagulation tests
C, and protein S. As with vitamin K deficiency, factor VII will vary (minimal for enoxaparin34 ), as will the effect
is the first factor to be depleted. Thereafter, factors IX and of protamine neutralization, which is very incomplete
X are depleted, and then factor II. As a result, initially only for enoxaparin. Coagulation testing is usually not
the PT will be prolonged. With greater doses, the aPTT required or performed. If laboratory testing is deemed
will become prolonged as well. necessary, the anti-Xa level is the appropriate test.
Warfarin therapy (most commonly for deep vein LMWH causes less platelet inhibition and is associated
thrombosis [DVT], pulmonary embolus [PE], atrial fibril- with a lesser incidence of heparin-induced thrombo-
lation, prosthetic cardiac valves, and protein S or protein cytopenia.35 There has been considerable variation in
C deficiency) is adjusted according to the International the dosage regimens employed. In Europe, it has been
Normalized Ratio (INR) (see tests of the hemostatic mech- common to begin prophylactic administration 12 to
anism). Bleeding is the principal untoward effect. Rapid 24 hours preoperatively. Postprocedure administration
reversal (12 to 24 hours) of warfarin effect25 can be accom- is more common in North America, but there is little
plished by the intravenous, slow administration of 5 mg of evidence to suggest the superiority of either practice.32
vitamin K. Smaller doses, 0.5 to 3 mg, should be used in sit- Twice daily dosing with enoxaparin has been common
uations of lesser urgency or when the objective is to reduce in North America. However, once-daily regimens are
rather than normalize INR. The INR should be rechecked usually sufficient and may actually be safer in some
at 6-hour intervals. Vitamin K administration may have to circumstances. Because of the relatively long half-life
be repeated at 12-hour intervals. In situations of greater of enoxaparin, twice daily dosing poses a problem with
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 529
respect to removal of epidural catheters because there not inhibited, and it may be appropriate to administer
is no anticoagulant nadir. platelet inhibitors simultaneously.
b. HEPARIN-INDUCED THROMBOCYTOPENIA/THROMBOSIS
Direct Thrombin Inhibitors. The DTIs are now
(HITT): The clinical manifestations of heparin-
induced thrombocytopenia/thrombosis (HITT) are used most commonly for CPB when there are con-
most commonly thrombotic and thromboembolic traindications to heparin. The available DTIs include
events (DVT, PE, limb or acral ischemia, MI, stroke), hirudin, argatroban, lepirudin, and bivalirudin. Hirudin
rather than a bleeding diathesis.36 Accordingly, this occurs naturally (in the saliva of the medicinal leech),
chapter will not provide a detailed description. As and the others are synthetic. Bivalirudin has been used
many as 5% of patients who receive heparin therapy for most frequently during CPB, and effective suppression of
5 days will develop thrombocytopenia that results from hemostatic activation has been reported.44 There are dis-
the development of antibodies (usually immunoglobu- advantages, however. The first is that there is no antidote,
lin G [IgG]) directed against platelet factor 4-heparin and termination of effect is therefore largely dependent on
complexes. HITT appears to be dose-related and is renal elimination. The exception is bivalirudin, which is
more common with bovine than porcine heparin. On- in part cleared by proteolysis by thrombin. In the patient
set usually occurs after several days in the heparin with renal failure or in urgent situations, elimination can
naive patient but can occur much more quickly (10 be accomplished by dialysis or hemofiltration.45 Accurate
to 12 hours) in those exposed within the preceding monitoring of the anticoagulant cannot be accomplished
100 days.37 LMWH-associated HITT occurs at a much with the common coagulation tests. Although the DTIs
lesser frequency and requires longer periods of expo- will prolong the ACT, as well as the TT, aPTT, and PT,
sure.38 Treatment requires withdrawal of heparin and they do not do so in a reliable dose-related manner. The
institution of an alternate anticoagulant (e.g., a direct ecarin clotting time (ECT) (see the following text) is prob-
thrombin inhibitor [DTI] such as hirudin, argatroban, ably the preferred method of monitoring.46 However, the
lepirudin, bivalirudin or a heparinoid such as dana- ECT is not widely available in North America, and the
paroid), but not a LMWH. Warfarin is contraindicated use of DTIs during CPB has been reported using either
because the inhibition of proteins C and S by warfarin protocol-driven administration43,47 or dosing to maintain
in the face of ongoing platelet clumping may aggra- kaolin-ACT values not less than 450 seconds.48
vate thrombosis. Platelets may similarly contribute to Ximelagatran is a DTI that can be taken orally. It has
thrombosis and should not be administered unless undergone several clinical trials in the prevention of recur-
thrombocytopenia is extreme. rent venous thromboembolism with favorable results.49 It
c. HEPARIN IN CARDIOPULMONARY BYPASS: A comprehen- has a relatively short half-life (approximately 4 hours)
sive review of the use and monitoring of heparin and is largely eliminated by the kidneys. Because of pre-
therapy in cardiopulmonary bypass (CPB) is beyond dictable bioavailability, coagulation monitoring is usually
the scope of this chapter, and extensive reviews are not employed. DDAVP, rFVIIa and FFP have been used
available elsewhere.39 In brief, the common practice is in the management of bleeding complications.50 Xime-
to administer sufficient heparin to maintain ACT >480 lagatran has recently (February 2006) been withdrawn
to 500 seconds for the duration of bypass. Although from the market worldwide because of serious liver injury
there is no universal agreement, it appears that there is (http://www.astrazeneca.com/pressrelease/5217.aspx).
greater hazard in allowing ACT to be on the low side Inhibitors of Activated Factor X. These agents
than in maintaining more complete heparinization.40 (fondaparinux, idraparinux) act through antithrombin.49
Platelet activation is less apparent when longer ACTs The prototype is fondaparinux, which is used most com-
are maintained. Whether this is a function of direct monly as alternative for DVT prophylaxis.31,51 Its advan-
inhibition of platelets, which are subject to contact tages include very predictable uptake (after once-daily
activation by the CPB circuit, binding of vWF or the subcutaneous administration) and kinetics that make
result of reduced formation of thrombin and inhibition monitoring and dosage adjustment unnecessary.49,52
of platelets by fibrin breakdown products (FDPs) is not However, these agents have long half-lives (fondaparinux,
apparent to the authors of this chapter. Protamine is 17 hours; idraparinux, 80 hours49 ), and there is no rever-
used to reverse heparins effect. Many clinicians em- sal agent. Excretion is through the kidneys. Therapeutic
ploy a milliliter for a milliliter technique. However, doses do not cause changes in PT, aPTT, or ACT.49 rFVIIa
titration of protamine against ACT is ideal to avoid has been used in the treatment of overdose.53
the excessive administration of protamine, which has
inherent anticoagulant effects including platelet inhi-
bition, stimulation of tPA release from endothelium, ACQUIRED COMBINED
and inhibition of fibrinogen cleavage by thrombin.41
Various alternatives have been used for the patient DISORDERS OF PLATELETS
with HIT who requires CPB. Plasmapheresis before AND CLOTTING FACTORS
surgery with subsequent use of heparin has been re-
ported.42 Nonheparin anticoagulants, including the Massive Transfusion
defibrinogenating agent Ancrod (from the venom of the
Malaysian pit viper) and, more commonly DTIs, have The rapid transfusion of large volumes of stored blood can
been employed.43 The contact activation of platelets is have numerous physiologic consequences (see Table 36.3).
530 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
load is limited, and the citrate in the CPD solution should TABLE 36.4 Hemostatic Abnormalities in Liver Disease
be metabolized by the liver to bicarbonate, making any
transfusion-related acidosis self-correcting. Clinically, in Thrombocytopenia
the injured patient who is hypotensive, poorly perfused, Decreased production
and has inadequate tissue oxygenation, it will be difficult Hypersplenism
to determine what portion of the metabolic acidosis is Increased consumption (low grade DIC)
due to rapid transfusion and what portion is due to RBC- Impaired platelet function
generated lactic acid.62 Because acidosis contributes to Decreased FDP clearance
the coagulation dysfunction associated with the shock Decreased synthesis of clotting factors
state, the appropriate course is to perform periodic blood Vitamin K deficiency (diet, malabsorption)
gas analysis and administer bicarbonate if indicated. Decreased hepatocyte function
Increased factor consumption
Citrate Intoxication Decreased clearance of activated factors
Decreased synthesis of inhibitors (protein C,
Contemporary additive solutions contain citrate, which
Protein S)
achieves anticoagulation by chelation of ionized cal-
Increased fibrinolysis
cium. When large volumes of stored blood (>1 blood
Decreased clearance of tPA
volume) are administered rapidly, the citrate can bind
Decreased synthesis of 2 -antiplasmin
and thereby cause a temporary reduction in ionized cal-
Decreased synthesis of PAI-1
cium levels. Citrate is normally metabolized efficiently by
the liver, and reduction of ionized calcium levels should DIC, disseminated intravascular coagulation; FDP, fibrin degradation
not occur until transfusion exceeds 1 mL/kg/minute, product; tPA, tissue plasminogen activator; PAI-1, plasminogen
that is, approximately 1 unit of blood per 5 minutes activator inhibitor-1.
in an average-sized adult. Note that the increasingly
common additive solution preservatives have a much impaired or when DIC (see the following text) complicates
smaller citrate content than citrate-phosphate-dextrose- the coagulation disturbance. Alcohol can also contribute
adenine blood, and that most of the citrate administered to platelet dysfunction by direct inhibition of the synthesis
during massive transfusion is in the FFP rather than of ADP, adenosine triphosphate (ATP), and thrombox-
the PRBCs. Impaired liver function or perfusion will ane A2 . Accordingly, a normal platelet count does not
lower the rate threshold for developing citrate intoxi- provide assurance of intact primary hemostasis in a pa-
cation. In the absence of significant liver dysfunction, tient with advanced liver disease. DDAVP may improve
citrate-related impairment of coagulation is unlikely. platelet function, but transfusion of platelet concentrates
Note also that critical cardiac consequences occur be- may still be necessary.
fore hypocalcemia invokes significant implications for
coagulation. Impaired Coagulation
With impaired hepatic function, factor production de-
creases and consumption increases. All of the clotting
ACQUIRED COMBINED factors, with the probable exception of factor VIII, are
DISORDERS OF PLATELETS synthesized in the liver. As with vitamin K deficiency,
hepatic disease first results in a deficiency of factor VII be-
AND CLOTTING FACTORS cause it has the shortest half-life. Thereafter, deficiencies
WITH INCREASED in factors II, IX, and X will develop. Dietary deficiency
of vitamin K (common in alcoholics) and diminished
FIBRINOLYSIS hepatic secretion of bile constituents (leading to malab-
sorption) will exaggerate these deficiencies. If vitamin K
Liver Disease deficiency is the cause of impaired coagulation rather
Chronic liver disease is associated with derangements than hepatic damage, parenteral vitamin K may be help-
of all three phases of hemostasis: Primary hemostasis, ful in restoring factor levels of II, VII, IX, and X. Further
coagulation, and fibrinolysis63 (see Table 36.4). deterioration of hepatic function will result in decreas-
ing levels of factors, I, V, XI, and XII. Impaired hepatic
function can also lead to a thrombotic tendency, which
Impaired Primary Hemostasis in turn leads to increased consumption of clotting fac-
Liver disease can contribute to both thrombocytopenia tors. This occurs because of impairment of two processes
and impaired platelet function. Decreased thrombopoi- that normally serve to inhibit coagulation. First, synthe-
etin secretion by the liver leads to decreased platelet sis of the native anticoagulants, ATIII, protein C, and
production. Hypersplenism may also contribute to throm- protein S may be impaired; second, hepatic clearance
bocytopenia. Platelet dysfunction can occur when liver of activated clotting factors from the circulation may be
disease is sufficiently advanced that clearance of FDPs, reduced, resulting in persistent activation of the coagula-
which coat the platelet surface and impair aggregation, is tion cascade.
532 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
reactions can also liberate TF-equivalent material into the antibiotic therapy. The obstetric conditions are fre-
circulation and incite DIC. Certain malignancies, most quently self-limited, although evacuation of the uterus
notably promyelocytic leukemia and adenocarcinomas, or hysterectomy may be warranted. Hypovolemia, aci-
are associated with DIC. However, with malignancy- dosis, and hypoxemia should be corrected to prevent
associated DIC, thrombotic manifestations are more their contribution to the DIC process. When bleeding
likely to appear first, whereas with the others mentioned is or may become life threatening, the consumptive
above, the hemorrhagic diathesis is often the first clinical coagulopathy must be treated. Platelets will be re-
manifestation. quired for thrombocytopenia, for example, <50,000
A few general conditions such as acidosis, shock, per mm3 . FFP will replace clotting factor deficiencies.
and hypoxia are associated with DIC. Shock promotes Fibrinogen levels should be raised to >100 mg per dL.
coagulation because one of the control mechanisms When hypofibrinogenemia is severe (<50 mg per dL),
(rapid blood flow) is compromised. Clearance of activated cryoprecipitate may be required. Six units of cryopre-
clotting factors is reduced when blood flow is decreased. cipitate will increase fibrinogen levels by approximately
Acidosis and hypoxia may contribute to both tissue and 50 mg per dL in a 70-kg patient.70
endothelial damage.
The clinical manifestations of DIC are a consequence Heparin has been advocated. However, the con-
of both thrombosis and bleeding. Bleeding is a more temporary practice is to restrict its use to only those
common clinical presentation in patients with acute, situations in which thrombosis is clinically problematic,
fulminant DIC. Petechiae, ecchymoses, epistaxis, gin- principally DIC associated with malignancies. There is
gival/mucosal bleeding, hematuria, and bleeding from no proven benefit in situations where bleeding is the
wounds and puncture sites may be evident. With the predominant manifestation. Antifibrinolytics have been
chronic forms of DIC, thrombotic manifestations are considered; however, their use in the face of widespread
more likely. Organs with the greatest blood flow (e.g., thrombosis is potentially disastrous, and they should not
kidney and brain) typically sustain the greatest damage. be used. ATIII concentrates have been administered in
Pulmonary function may deteriorate as a consequence of the hopes that its administration will serve to slow the
microthrombus accumulation. runaway coagulation process. However, a beneficial ef-
fect on outcome from DIC has not been confirmed,69
a. DIAGNOSIS: There is no absolutely consistent constel-
and therefore its use should be viewed as experimental.
lation of laboratory findings among routine tests.68
An insufficiency in the protein C endogenous coagulation
Increased PT and aPTT, thrombocytopenia, decreased
inhibition system is thought to contribute to the pro-
fibrinogen level, and the presence of FDPs and
thrombotic state in DIC. Activated protein C (drotrecogin
D-dimer may all be noted. The peripheral smear
alfa) has been shown to decrease mortality and organ
may reveal schistocytes (fragmented RBCs reflecting
failure in patients with severe sepsis and a high risk of
the microangiopathy that occurs as a consequence
death (as defined by an APACHE II [Acute Physiology
of widespread fibrin deposition). Thrombocytopenia
and Chronic Health Evaluation] score of 25 and failure
(<100,00 per L) is not always evident early in the
of more than one organ system).71,72 Improvement was
process, but true DIC without sequential reduction
also evident among patients with sepsis with overt DIC.73
in platelet count is very unlikely. PT and aPTT may
Its use should be considered in any sustained episode
remain normal in spite of decreasing factor levels be-
of DIC.73
cause of the presence of high levels of activated factors,
including thrombin and Xa. Fibrinogen level may not
be decreased, that is, <100 mg per dL, initially. Fib-
rinogen is an acute phase reactant that increases in
response to stress, and the early consumption of fib- How Should the Clinician
rinogen may simply reduce its levels to normal. FDPs
are a sensitive measure of fibrinolytic activity, although
Approach the Diagnosis
they not specific for DIC. D-dimer (which is a break- and Treatment of Bleeding
down product of the crosslinked fibrin in a mature clot) Diatheses?
is somewhat more specific for DIC, but not entirely so,
and should be measured when DIC is suspected.
Various other laboratory assays have been em-
ployed to support a diagnosis of DIC,68 but should LABORATORY EVALUATION
probably not be considered as part of the anesthe-
siologists routine. They include levels of prothrombin OF THE HEMOSTATIC
fragments F1+F2 (a marker of prothrombin conversion MECHANISM
to thrombin-increased), thrombin-ATIII complexes (in-
creased), ATIII (decreased), 2 -antiplasmin (decreased Evaluation of Primary Hemostasis
by binding to excess plasmin), protein C (decreased),
plasminogen (decreased), and factor VIII (decreased in Only brief descriptions of the most common tests are pro-
DIC but normal with hepatic failure without DIC). vided herein. Detailed and more comprehensive descrip-
b. TREATMENT: Treatment should focus on management tions (including the platelet function analyzer, platelet
of the underlying condition. Septicemia will require aggregometry) are provided elsewhere.1
534 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Platelet Count be prolonged until the fibrinogen level is below 100 mg per
Normal platelet counts range between 150,000 and dL. A prolonged PT does not define the exact hemostatic
440,000 per mm3 . Counts below 150,000 per mm3 are defect. However, if the aPTT (see the following text) is
defined as thrombocytopenia. Spontaneous bleeding is normal, then a prolonged PT is most likely to represent a
unlikely in patients with platelet counts >10,000 to 20,000 deficiency or abnormality of factor VII. Because factor VII
per mm3 .74 With counts from 40,000 to 70,000 per mm3 , has the shortest half-life of the clotting factors synthesized
bleeding induced by surgery may be severe. in the liver, factor VII is the clotting factor that first
becomes deficient with liver disease, vitamin K deficiency
or warfarin therapy. Prolongation of the PT may also be
Bleeding Time due to deficiencies of multiple factors. However, when
A prolongation of the BT may be due to thrombocytopenia, multiple factor deficiencies coexist, the PTT or aPTT (see
platelet dysfunction (adhesion, aggregation), and vascular the following text) will usually be prolonged as well.
abnormalities. BT is prolonged in patients with many
conditions that cause platelet dysfunction (e.g., use International Normalized Ratio
of aspirin, uremia). However, prolonged BT has been
Another difficulty with the PT test is that many types of
observed with numerous disorders that are not associated
thromboplastin reagent are used. This results in a wide
with platelet dysfunction, such as vitamin K deficiency
variation in normal values, which makes comparison of
of the newborn, amyloidosis, congenital heart disease,
PT results between laboratories difficult. The INR was
and the presence of factor VIII inhibitors.75 Whether the
introduced to circumvent this difficulty.76 Each throm-
BT test represents a specific measure of in vivo platelet
boplastin is compared with an internationally accepted,
function is debated. In spite of the correlation of BTs
standard thromboplastin and assigned an International
with conditions known to influence platelet function,
Sensitivity Index (ISI). If the test thromboplastin is equiv-
and in spite of BT quite reliably becoming progressively
alent to the international standard, it will have an ISI
prolonged as platelet count falls below 80,000 per L,
index of 1. Once the ISI number has been determined, PT
there are no convincing data to confirm that BT is
test times obtained with that reagent are normalized and
a reliable predictor of the bleeding that will occur in
reported as an INR.77
association with surgical procedures.
of coagulation). Fresh whole blood is added to a test tube per dL may occur. Because of this increase, in spite of
that contains a particulate surface activator of factors XII rapid fibrinogen consumption during a hypercoagulable
and XI. The time to clot formation is measured. Partial state such as DIC, the fibrinogen level may still appear to
thromboplastin or a platelet phospholipid substitute is not be normal.
added. Coagulation is therefore dependent upon adequate Laboratory Evaluation of Fibrinolysis
amounts of platelet phospholipid being present in the
1. FIBRIN DEGRADATION PRODUCTS AND D-DIMER: The
blood sample. The automated ACT is widely used to
FDP test identifies the breakdown products of fibrin
monitor heparin therapy in the operating room. Normal
(crosslinked or uncrosslinked), and fibrinogen itself.
values are in the range of 90 to 120 seconds.78
The D-dimer assay is specific for breakdown products
of crosslinked fibrin. FDPs will be increased in any state
Thrombin Time of accelerated fibrinolysis including advanced liver dis-
Thrombin time (TT) is a measure of the ability of ease, CPB, administration of exogenous thrombolytics,
thrombin to convert fibrinogen to fibrin. This test, for example, streptokinase, and DIC. D-dimer is specific
which is performed by adding exogenous thrombin to to conditions in which extensive lysis of the crosslinked
citrated plasma, bypasses all the preceding reactions. fibrin of mature thrombus occurrs, in particular DIC
The TT may be prolonged by conditions that affect but also DVT and PE.
either the substrate, fibrinogen, or the action of the 2. THE THROMBOELASTOGRAM: Thromboelastography
enzyme, thrombin. The TT is prolonged when there provides a measure of the mechanical properties of
is an inadequate amount of fibrinogen (<100 mg per evolving clot as a function of time.82 A principal ad-
dL) or when the fibrinogen molecules that are present vantage of this test is that the processes it measures
are abnormal (dysfibrinogenemia), as in advanced liver require the integrated action of all the elements of the
disease. Thrombins enzymatic function can be hindered hemostatic process: Platelet aggregation, coagulation,
by inhibitors such as heparin (complexed to ATIII), and fibrinolysis. A specimen of blood is placed in a ro-
FDPs or by inhibitors that may be seen in patients tating cuvette into which a piston is lowered. As clot
with plasma cell myeloma and other immunoproliferative formation begins, the pistons rotation varies as a func-
conditions.80 The normal TT is <30 seconds. tion of the adherence of the evolving fibrin clot to the
piston. The rotation of the piston results in a to-and-fro
Reptilase Time excursion of a stylus. The amplitude of that oscillation
is proportional to the speed of piston rotation.
Reptilase time can be used to differentiate between the
effects of heparin and FDPs. Reptilase, which is derived Figure 36.4 depicts a normal thromboelastogram
from a snake venom, converts fibrinogen to fibrin. The (TEG). Several parameters are derived from the TEG.
action of reptilase is unaffected by heparin but is inhibited The most commonly used and their interpretation are
by FDPs. A prolonged TT and a normal reptilase time as follows.83 R, the reaction time, is the interval until
suggest the presence of heparin. Prolongation of both TT initial clot formation. It requires thrombin formation, and
and reptilase time will occur in the presence of FDPs or prolongation is usually indicative of an intrinsic pathway
when fibrinogen level is low. The normal reptilase time is factor deficiency. K, the clot formation time, is the interval
14 to 21 seconds. required after R for the TEG to achieve a width of
20 mm. Prolongation occurs with deficiencies of thrombin
Ecarin Clotting Time formation or generation of fibrin from fibrinogen. The
angle, like K, is a measure of the speed of clot formation.
The DTIs (hirudin, argatroban, bivalirudin, lepirudin) are
A decrease of the angle has similar significance to a
used most often in patients with HITT. At low DTI concen-
prolongation of K. MA, the maximum amplitude, is a
trations, TT and ACT provide reasonable correlations with
measure of the strength of the fully formed clot. It reflects
DTI concentrations; however, with the concentrations re-
quired for CPB, the correlation becomes poor and the risk
of underdosing and overdosing with these agentsfor
which there are no antagonistsbecomes significant. The a
ECT provides better correlation and can be used for moni-
MA A60
toring in this context.81 The test employs the venom of the 20 min
R
saw-scaled (aka sawtooth) viper (Echis carinatus). A met- 60 min
R+K
alloprotease in the venom converts normal prothrombin
to a form that is inhibited by the DTIs in a dose-dependent
F
manner.
FIGURE 36.4 The normal thromboelastogram and the variables
Fibrinogen Level commonly derived from it. See text for explanation. MA,
Normal values are between 160 and 350 mg per dL. Below maximum amplitude; A, amplitude. (From Kang Y, Lewis JH,
100 mg per dL, fibrinogen may be inadequate to produce a Navalgund A, et al. Epsilon-aminocaproic acid for treatment of
clot. Fibrinogen is rapidly depleted during DIC. A marked fibrinolysis during liver transplantation. Anesthesiology.
increase in fibrinogen may occur in response to stress 1987;66:766, with permission.)
such as surgery and trauma. Levels in excess of 700 mg
536 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Normal Thrombocytopenia Hypercoagulation PT, aPTT, and occasionally BT. When a complicated
Hemophilia Fibrinolysis disruption of the hemostatic mechanism is suspected,
further testsincluding fibrinogen, TT, and assays for
FDP and the D-dimermay be ordered. Because the
coagulation defects that appear are revealed most of-
ten as abnormal values of PT and/or aPTT, Figure 36.6
provides an algorithm for the evaluation of those abnor-
malities.
a aPTT prolongation is more likely to occur with LMWHs with lower Xa/IIa effect ratios, eg., tinzaparin, than with greater ratios, e.g., enoxaparin.
b Bleeding time may also be prolonged in association with a marked aPTT increase.
c DIC may be distinguished by the presence of D-dimers.
aPTT, activated partial thromboplastin time; PT, prothrombin time; TT, thrombin time; FDPs, fibrin degradation products; , increased; ,
decreased; N, normal; HITT, heparin-induced thrombocytopenia/thrombosis; ASA, aspirin; NSAIDs, nonsteroidal anti-inflammatory drugs;
vWD, von Willebrand disease; vWF, von Willebrand factor; LMWH, low molecular weight heparin; DIC, disseminated intravascular coagulation.
Reproduced with permission from Petrovitch CT, Drummond JC. Coagulopathic states. In: ODonnell JM, Nacul FE, eds. Surgical intensive
care medicine. Boston/Dordrecht/London: Kluwer Academic Publishers; 2001:495.
deficiencies (such as liver disease, vitamin K deficiency, the aPTT as well. Similarly, the development of liver
the administration of warfarin, or the coagulopathy disease will lead to deficiencies of factor VII first and
associated with massive transfusion or DIC). Heparin initially prolong only the PT. With further deterioration
therapy or congenital disorders of hemostasis are more of liver function, both the PT and the aPTT will be pro-
probable causes. longed. Liver disease can also lead to thrombocytopenia
and platelet dysfunction. Acquired coagulation factor in-
Prolonged Prothrombin Time (Normal Platelet Count hibitors are rare but can occur in patients with lymphoma
and Activated Prothrombin Time) or collagen vascular disease.
Differential diagnosis includes vitamin K deficiency, war-
farin administration, early liver dysfunction, factor VII
deficiency, and acquired coagulation factor inhibitors.
Prolonged Prothrombin Time and Activated
Because, among the vitamin K-dependent factors, factor Prothrombin Time (Normal Platelet Count)
VII has the shortest half-life, depletion of the vitamin Differential diagnosis includes vitamin K deficiency,
K-dependent factors will first prolong the PT and later warfarin, and heparin. Although liver disease can produce
538 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
PT aPTT
prolonged prolonged
Fbg, FII, FV, Fbg and FV Fbg, FII, FV Fbg but FII, TT TT and Fbg TT, Fbg TT, Fbg
and FVII normal; and FVII FV and FVII Fbg normal normal DD DD normal
normal FII and FVII normal
Liver APA or
Artifact, APA Vitamin K Fbg DIC or hyper- Fbg
insufficiency Heparin intrinsic factor
or Heparin deficiency deficiency deficiency fibrinolysis deficiency
or DIC
Liver Factor/
DIC Perform APA
insufficiency Inhibitor
likely likely tests assays
FIGURE 36.6 An approach to the evaluation of prolonged PT and/or aPTT. PT, prothrombin time;
aPTT, activated partial thromboplastin time; Fbg, fibrinogen; F, factor; TT, thrombin time; DD,
D-dimer; APA, antiphospholipid antibody (e.g., lupus anticoagulant, anticardiolipin and anti-B2 -GPI
antibodies); DIC, disseminated intravascular coagulation. (Modified from: Bombeli T, Spahn DR.
Updates in perioperative coagulation: Physiology and management of thromboembolism and
haemorrhage. Br J Anaesth. 2004;93:275.)
multiple factor deficiencies and this pattern, the platelet TREATMENT OF DISORDERS
count is usually decreased. FDPs will also be elevated (see
the following text). OF HEMOSTASIS
Treatment of individual disorders has been discussed in
Prolonged Prothrombin Time, Activated Prothrombin the preceding text. This section presents a general descrip-
Time, and Thrombin Time (Normal Platelet Count) tion of the use of blood components and pharmacologic
agents in the treatment of bleeding disorders.
Differential diagnosis includes heparin, DTIs or FDPs, hy-
pofibrinogenemia, and dysfibrinogenemia. Simultaneous
prolongation of the TT makes the diagnosis of simple Fresh Frozen Plasma
vitamin K deficiency or warfarin therapy unlikely. TT is
sensitive to minute levels of heparin. Addition of pro- The indications for FFP listed in Table 36.7 represent
tamine or reptilase time will identify heparin. FDPs may a consolidation of recommendations offered by the
be elevated with fibrinolytic therapy, DIC, or liver disease. American Society of Anesthesiologists in 1996 and the
DIC and liver disease usually result in thrombocytope-
nia as well. A normal platelet count makes heparin or
extensive fibrinolysis more likely. TABLE 36.7 Indications for the Administration of Fresh
Frozen Plasma
Prolonged Prothrombin Time, Activated Prothrombin
Correction of single coagulation factor deficiencies for
Time, Thrombin Time (Decreased Platelet Count) which specific concentrates are not available
Differential diagnosis includes DIC, dilution by massive (principally factor V)
transfusion, liver disease, and heparin therapy. FDPs and Correction of multiple coagulation factor deficiencies,
D-dimer are elevated in DIC and allow differentiation e.g., DIC, with evidence of microvascular bleeding and
from dilutional effects and excess heparin. Heparin causes PT and/or aPTT >1.5 times normal
thrombocytopenia only when prolonged exposure results Urgent reversal of warfarin therapya
in HIT. FDPs, but not D-dimer, are elevated in severe liver Correction of microvascular bleeding during massive
disease. transfusion (>1 blood volume) when PT/aPTT cannot
The interpretation of coagulation tests may be made be obtained in a timely manner
more difficult by the fact that patients who develop a
bleeding diathesis in the perioperative period may have a Prothrombin complex concentrate (II, VII, IX, X) is an alternative
more than one bleeding disorder, for example, DIC and that has been reported to be more effective than FFP.25
coagulopathy due to massive transfusion, and may also DIC, disseminated intravascular coagulation; PT, prothrombin time;
have a surgical cause for bleeding. aPTT, activated partial thromboplastin time.
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 539
British Committee for Standards in Haematology in patient in the face of similar platelet counts. This occurs
2004.88,89 Although there is actually little formal proof because some platelets are more effective than others.
of the efficacy of FFP,90 its use to restore coagulation When thrombocytopenia results from peripheral destruc-
factor levels is inevitably valid in many clinical situations. tion of platelets, the bone marrow continues to produce
normal, young, large platelets that are hemostatically very
Platelets effective. A patient with these platelets may have more
effective primary hemostasis than a patient with the same
The indications for platelet administration presented in platelet count but whose platelets were produced by a less
Table 36.8 represent a consolidation of recommendations active, less healthy bone marrow.
offered by the American Society of Anesthesiologists One platelet unit will typically increase platelet
in 1996 and the British Committee for Standards in count by 5,000 to 10,000 per L. However, the increase
Haematology in 2003.88,91 The guidelines are general, must be verified by platelet count, especially in patients
and there is a substantial latitude (and requirement) for who may have been alloimmunized by frequent platelet
clinician judgment. administration.
The threshold for platelet administration that will
most often be relevant to anesthesiologists will lie be- Cryoprecipitate
tween 50,000 and 100,000 per L. The threshold within
that range at which platelets are administered should Cryoprecipitate contains factor VIII, the vWF, fibrinogen,
be based on the likelihood of the intended procedure fibronectin, and factor XIII. Virally inactivated factor VIII
to cause bleeding, the hazard of bleeding should it oc- coagulation factor concentrates, some of which contain
cur, for example, intracranial neurosurgery greater than clinically effective concentrations of vWF (antihemophilic
peripheral orthopedics, and the presence or possibility of factor, e.g., Humate P [ZLB Behring L.L.C, King of
additional causes of coagulation disturbance, for example, Prussia, PA], Alphanate [GriJfols Inc., Los Angeles, CA])
recent administration of antiplatelet agents, CPB, DIC, are now available. As a result, hemophilia A and vWD
dilution due to large volume administration. Bleeding are usually treated (in consultation with a hematologist)
manifestations can vary substantially from patient to with those concentrates rather than cryoprecipitate. The
remaining indications for cryoprecipitate are presented in
Table 36.9.92
TABLE 36.8 Indications (Expressed as Current Patient
Platelet Count) for the Administration of Platelets
Recombinant Factor VIIa (rFVIIa,
Nonbleeding patients without 10,000/L NovoSeven, Novo Nordisk, Bagsvaerd,
other abnormalities of Denmark)
hemostasis74
Lumbar puncture, epidural 50,000/L rFVIIa (rFVIIa, NovoSeven, Novo Nordisk, Bagsvaerd,
anesthesia, central line Denmark) was developed for the treatment of patients
placement, endoscopy with with hemophilia A or B and inhibitors to exogenous factor
biopsy, liver biopsy or VIII or factor IX preparations. However, it is fast becom-
laparotomy in patients without ing the hemostatic agent of last (and sometimes earlier) re-
other abnormalities of sort in many clinical situations. Table 36.10 lists the many
hemostasis
Intended procedures in which 100,000/L
closed cavity bleeding might TABLE 36.9 Indications for the Administration
be especially hazardous, e.g., of Cryoprecipitate
neurosurgery
To maintain platelets during Not less than Prophylaxis before surgery or treatment of bleeding in
ongoing bleeding and 50,000/L patients with congenital dysfibrinogenemias
transfusion Microvascular bleeding when there is a disproportionate
To maintain platelets during DIC Not less than decrease in fibrinogen, e.g., DIC and very massive
with ongoing bleeding 50,000/L transfusiona , with fibrinogen <80100 mg/dL (or
Microvascular bleeding Clinician judgment assay result not available)
attributed to platelet Prophylaxis before surgery or treatment of bleeding in
dysfunction, e.g., uremiaa , hemophilia A and von Willebrand disease if
postcardiopulmonary bypass concentrates are unavailable or ineffective
or in association with massive Bleeding due to uremia that is unresponsive to
transfusion desmopression (DDAVP)
a Aftera trial of desmopressin, if permitted by the clinical situation. a Fresh frozen plasma is the first line component for the factor
DIC, disseminated intravascular coagulation. depletion associated with massive transfusion.
Levi MM, Vink R, de Jonge E. Management of bleeding disorders by DIC, disseminated intravascular coagulation; DDAVP, 1-deamino-
prohemostatic therapy. Int J Hematol. 2002;76(Suppl 2):139. 8-D-arginine vasopression.
540 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 36.10 Reported Off-Label Uses of Recombinant arbitrary, algorithm in place at the UCSD provides for the
Activated Factor VII (rFVIIa) administration of 60 g per kg for profuse bleeding un-
responsive to conventional therapy. Such dose is rounded
Excessive anticoagulation (coumadin, anti-Xa, DTIs) to the nearest 1,200 g, in recognition that the agent is
Thrombocytopenia supplied in vials of 1.2 mg. The half-life is approximately
Platelet dysfunction (uremia, aspirin, clopidogrel, IIb/IIIa 2.5 hours, and repeat dosing at 2-hour intervals may be
inhibitors, Glanzmann thrombasthenia, required. It is our opinion, based first on empiric, anec-
Bernard-Soulier syndrome, vWD) dotal clinical experience and second on the principle that
Hepatic dysfunction/failure additional doses can be readily administered if necessary,
Gastrointestinal hemorrhage that for uncontrolled bleeding in the operating room or
Intracerebral hemorrhage ICU, 20 g per kg is a reasonable first dose.
Surgery (trauma, post-CPB, spine, liver transplantation,
prostatectomy, vascular surgery, neurosurgery) Desmopressin
Obstetric hemorrhage (accreta, HELLP syndrome)
Postoperative bleeding Desmopressin (DDAVP) is a synthetic analog of the natu-
ral hormone, vasopressin. The actions of vasopressin are
Xa, activated factor X; DTI, direct thrombin inhibitor; vWD, von Wille- thought to mediated by two general classes of receptors:
brand disease; CPB, cardiopulmonary bypass; HELLP, hemolysis- V1 receptors, which mediate smooth muscle contraction
elevated liver enzymes-low platelets. in the peripheral vasculature; and V2 receptors, which
regulate water reabsorption in the collecting ducts of the
nephron. Desmopressin has activity at only the V2 recep-
causes of clinical bleeding, in addition to its labeled uses in tors. In fact, desmopressin is a more potent antidiuretic
hemophilia A and B and congenital factor VII deficiency, than vasopressin with more prolonged activity. The hemo-
for which effective use of rFVIIa has been described in static effects of desmopressin are thought to be mediated
the published literature.28,93 Of the reported uses, effi- by low-affinity, extra-renal V2 -like receptors.17 Desmo-
cacy has been confirmed by a prospective, randomized pressin causes release of coagulation factor VIII:C, vWF,
controlled trial for only intracerebral hemorrhage, open and tPA. Desmopressin is thought to release VIII:C from
prostatectomy, and post-CPB management.9496 Negative the sinusoidal liver endothelial cells and the vWF from
prospective trials have been reported in the context of endothelial cells. In mild hemophilia A, desmopressin can
elective liver resection and pelvic reconstruction.97,98 increase the circulating factor VIII:C concentration two-
The mechanism of action is not entirely certain. It is to sixfold. The effect is maximal after 30 minutes, with
clearly more than an augmentation of the native functions elevated levels persisting for 6 to 8 hours.12 Desmopressin
of factor VII. Were that the case, it would not be effective in also increases platelet adhesiveness and shortens the BT,
hemophilia. However, factor VIIa, whose preferred ligand perhaps by increasing platelet expression of GPIb/IX.18
is TF, also undergoes low-affinity binding to activated
platelets. In the concentrations achieved with typical Indications
rFVIIa dosing, the serum levels are several hundred times
those achieved physiologically and are probably sufficient Desmopressin has proven to be effective treatment for
to activate Xa on the platelet surface (Fig. 36.2H) and certain types of vWD and mild hemophilia (see the
preceding text). Desmopressin has been shown to reduce
achieve the thrombin burst necessary to support the
the BT in a variety of conditions associated with platelet
propagation phase of coagulation.2 Because rFVIIa is an
dysfunction. It produces rapid and temporary correction
active procoagulant only when it is in contact with TF
of prolonged BTs in patients with uremia following
or activated platelets, unwanted coagulation has been
intravenous or intranasal administration. In cirrhotic
relatively infrequent. However, a review of adverse events
patients, desmopressin increases the concentrations of
reported to the U. S. Food and Drug Administration (FDA)
larger vWF multimers and shortens prolonged BTs.
suggests that thrombotic complications, for example,
Desmopressin also decreases the prolonged BTs caused
thromboembolic stroke, MI, arterial thrombosis, and PE,
by many drugs including aspirin, nonsteroidal anti-
probably do occur when rFVIIa is administered to patients
inflammatory drugs, dextran, ticlopidine, and heparin.17
with active bleeding.99 The incidence of these events,
The prophylactic use of desmopressin in cardiac sur-
although probably low, is impossible to discern from
gical patients has been controversial. Because platelet
the available evidence. Moreover, it is the (nonevidence
dysfunction and thrombocytopenia are common in that
based) opinion of these authors that it should be viewed
setting, numerous studies have been performed. Those
as relatively contraindicated in clinical states in which TF
that have revealed decreased blood loss or blood prod-
may be circulating freely, that is, in most of the conditions
uct administration have involved principally patients
associated with DIC.
who were predisposed to blood loss (e.g., redo proce-
The appropriate dosing of this expensive agent
dures100103 ) and patients receiving aspirin.104
($1,020 for 1.2 mg at University of California, San Diego
[UCSD]) is not well defined. The dose used most often
in hemophilia has been 90 g per kg. However, doses Dosage Recommendations
as low as 20 g per kg have been effective in other cir- Desmopressin is commonly administered intravenously
cumstances in some reports.95 The current, somewhat in a dose of 0.3 g per kg. The effect of desmopressin is
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 541
rapid. Peak levels of factor VIII:C and vWF are achieved with certainty. However, thrombin is a serine protease
within 30 to 60 minutes, with the effect lasting for that can activate platelets through a protease-activated re-
several hours.17 Desmopressin administration may be ceptor on the platelet surface.113 Better preservation of the
repeated after 8 to 12 hours, although tachyphylaxis may GPIb receptor (which is necessary for initial platelet adhe-
occur and the interval should ideally be as long as is sion to vascular defects) has been reported during CPB in
clinically feasible.9 When used in cardiac surgery, the patients who received aprotinin.115 Aprotinin also appears
drug should be administered after termination of CPB. to reduce neutrophil activation and transmigration across
Water balance should be monitored. However, although capillary endothelium, perhaps through an effect on an
congestive cardiac failure and hyponatremia and seizures endothelial protease-activated receptor,116 and may there-
in children have been reported, clinically significant water fore also blunt the neutrophil-mediated component of the
retention is relatively uncommon. Desmopressin may be response to endothelial injury. All the factors in the coagu-
administered as a nasal spray and is available for home lation cascade, except V and VIII, are serine proteases and
use for patients with mild hemophilia and those with vWD are inhibited to some extent by aprotinin. Consistent with
(type I). Intravenous administration causes a more rapid this is the observation that thrombin formation during
rise in FVIII:C levels and is therefore preferable for acute CPB is reduced by aprotinin but not TXA.117
hemostatic challenges.
Use of Antifibrinolytics in Cardiac Surgery
Antifibrinolytics
Meta-analyses of the many studies of these agents per-
Antifibrinolytic agents have been used frequently in formed in the context of CPB confirm that, overall, blood
situations in which exaggerated fibrinolysis is suspected of loss and the administration of allogeneic blood is dimin-
contributing to intraoperative bleeding. The situations in ished by the use of all three agents.105,110,118,119 Concern
which favorable effects on blood loss and replacement has been expressed that antifibrinolysis might lead to an
have been reported include CPB procedures, hepatic increased rate of graft occlusion, MI, and renal failure.
transplantation, scoliosis surgery, total joint replacement, However, the meta-analyses have not borne out those
and prostate surgery.105112 The use of antifibrinolytic concerns.105,118,119 There does not appear to be a clear
mouthwashes in the context of dental procedures in consensus as to which of the three agents is most ap-
patients with hemophilia has been mentioned previously. propriate in the context of CPB. Various authors have
There are three commonly available antifibrinolytics: The argued that EACA and TXA are preferable to aprotinin be-
lysine analog, EACA and TXA, and the serine protease cause they are less expensive and have apparently similar
inhibitor, aprotinin. efficacy.120,121 However, the most recent of those meta-
analyses reported a trend toward a reduced incidence of
Epsilon-Aminocaproic Acid and Tranexamic Acid atrial fibrillation and a reduced incidence of stroke in
patients receiving aprotinin.119 These observations have
EACA and TXA bind to and produce a structural change served to create a bias in favor of aprotinin.
in both plasminogen and plasmin. That structural change The patterns of use of antifibrinolytic agents in
prevents the conversion of plasminogen to plasmin cardiac surgery vary substantially among institutions.
and also prevents plasmin from degrading fibrinogen Few appear to use these agents on a routine basis
and fibrin. The dual action of these agents results for all CPB procedures. Some reserve their use for
in two effects on the hemostatic mechanism. First, situations more likely to be associated with post-CPB
decreased synthesis of plasmin from plasminogen results bleeding, for example, redo procedures, circulatory arrest
in reduced fibrinolysis. The second effect of these drugs, procedures. Still others appear to reserve antifibrinolytics
the inactivation of plasmin, decreases the formation of
for refractory bleeding post CPB. The latter seems less
degradation products of fibrinogen and fibrin. These FDPs
logical because much of the activation of the hemostatic
have anticoagulant effects, including the inhibition of
mechanism occurs during CPB.
platelet aggregation and the inhibition of the cross-linking
A recent publication reporting increased incidences of
of fibrin strands, which are thereby avoided.
renal failure, MI, heart failure, stroke, and encephalopa-
thy in patients who receive aprotinin during cardiac
Aprotinin surgery122 has been met with some skepticism. The con-
Aprotinin produces its antifibrinolytic effect by a differ- cerns include the nonstandardized anesthetic and surgical
ent mechanism. It is an inhibitor of numerous serine management practices in the multinational patient cohort
protease enzymes including plasmin and kallikrein. The and, more importantly, the possibility that the results rep-
latter participates in the process of contact activation of resent the administration of aprotinin to the subset of
factor XII. As a consequence of its inhibition of plas- patients perceived to be at highest risk. The expression
min, aprotinin, like EACA and TXA, prevents degradation of concern has included, in our opinion, an unprece-
of fibrinogen and fibrin. As is the case with EACA and dented and remarkably rapid response by a Task Force of
TXA, the reduction in FDPs should improve both platelet the Board of Directors of the Society for Cardiovascular
and coagulation function. However, aprotinin is believed Anesthesia offering the opinion that the suggestion to
to have additional beneficial effects on the inflammatory immediately curtail all use of aprotinin is premature.123
response to CPB in general and on platelets in particu- There is a small but finite rate of allergic responses to
lar.113,114 The mechanism of these effects is not known aprotinin. Accordingly, a test dose has been recommended
542 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
for patients who have had prior exposure to aprotinin. vascular injury to serve as meeting places for
The potential for developing sensitivity has been used activated clotting factors.
as a rationale for avoiding administration of aprotinin 2. vWD is the most common congenital bleeding diathe-
in circumstances in which it is not clearly indicated, sis and often does not clinically manifest itself until
for example, first time coronary artery bypass grafting the time of the patients first surgical procedure. It
(CABG), so that it may be used safely in the event of a should be suspected in instances of unexplained or
redo procedure. The risk of an anaphylatic/anaphylactoid unusual postoperative bleeding.
response appears to decline very substantially with reex- 3. Although a dilutional coagulopathy is unlikely in the
posure intervals >6 months.124 setting of an isovolemic exchange of <1.5 blood
volumes, in the event of resuscitation from hypov-
Use of Antifibrinolytics in Liver Transplantation olemia or with the simultaneous occurrence of other
Accelerated fibrinolysis occurs commonly in patients un- conditions promoting consumption (tissue injury, sta-
dergoing hepatic transplantation. This is probably, in part, sis, acidosis), a coagulopathy may occur with much
the consequence of decreased clearance of activated clot- smaller exchange volumes.
ting factors by the diseased liver. More importantly, hep- 4. Advanced liver disease produces a complex and vari-
atic clearance ceases entirely during the anhepatic phase. able coagulation disturbance that includes abnormal-
In addition, with reperfusion of the donor liver, there is an ities of primary hemostasis, impaired coagulation
explosive release of tPA into the systemic circulation.125 (either accelerated or inhibited), and increased fib-
Aprotinin, EACA, and TXA have all been used and reported rinolysis, which in some patients will result in the
to reduce blood loss in hepatic transplantation.82,108,112 equivalent of a low grade DIC.
Some advocate the prophylactic administration to all pa- 5. The aPTT test is relatively insensitive to factor X
tients,108 whereas others administer these agents only in deficiency or inhibition. As a result, it does not provide
response to the demonstration, typically by thromboelas- a reliable measure of the clinical effect of LMWHs,
tography, of hyperfibrinolysis.82,126 which cause greater inhibition of factor X than factor
II (thrombin).
6. Because of factor VIIs short half-life, vitamin K defi-
Use of Antifibrinolytics in Orthopedic and Other ciency, warfarin administration, and liver dysfunction
Surgery will first result in isolated prolongation of the PT. As
There have been numerous reports of a reduction of trans- factor depletion becomes more severe, aPTT may also
fusion requirement in scoliosis and joint replacement be prolonged.
surgery.106,107,109,111 However, meta-analysis has not con- 7. rFVIIa has been reported to be an effective hemostatic
firmed the efficacy in other types of noncardiac surgery, agent in a very wide variety of situations involving
and the available data do not permit a determination of both platelet dysfunction and/or coagulation defects.
the relative efficacy of EACA, TXA, and aprotinin.105 Thrombotic complications are rare but have been re-
ported. The risk is theoretically greatest in situations
in which there is likely to be exposed or circulat-
ing tissue factor-like material or extensive damage to
CONCLUSIONS endothelial surfaces.
Preoperative evaluation must identify those patients 8. Desmopressin (DDAVP) causes release of vWF and fac-
whose inherited or acquired medical conditions or whose tor VIII:C and expression of platelet GPIb receptors.
current medications may influence the hemostatic pro- It may be effective in the treatment of bleeding as-
cess. With respect to medications, there are a rapidly sociated with vWD, mild hemophilia A, and platelet
increasing number of agents that are administered specif- dysfunction caused by uremia, liver disease, and as-
ically for the purpose of altering the hemostatic balance pirin.
(e.g., clopidogrel, tPA, LMWH). As the patient proceeds 9. The antifibrinolytics, EACA (Amicar), TXA, and apro-
through surgery and the postoperative period, the anes- tinin have all been shown to reduce blood product
thesiologist must determine whether bleeding is surgical administration in several surgical situations including
in nature or the result of a preexisting or evolving hemo- CPB, major joint replacement, scoliosis surgery, and
static defect that will require the transfusion of hemostatic liver transplantation. In terms of reduction of blood
blood componentsplatelets, FFP, or cryoprecipitateor loss, none of the three has been shown to be superior.
the administration of pharmacologic agents.
REFERENCES
1. Drummond JC, Petrovitch CT. Hemotherapy and hemosta-
KEY POINTS sis. In: Barash PG, Cullen BF, Stoelting RK, ed. Clinical
anesthesia, 5th ed. Philadelphia: Lippincott Williams &
1. Numerous control mechanisms serve to localize the
Wilkins; 2006:208.
coagulation process to sites of vascular injury. Chief 2. Hoffman M. A cell-based model of coagulation and the role
among them is the necessity for specific phospholipid of factor VIIa. Blood Rev. 2003;17(Suppl 1):S1.
surfaces surfaces (TF, activated platelets), which 3. Broze GJ Jr. The role of tissue factor pathway inhibitor in a
under normal circumstances occur only at sites of revised coagulation cascade. Semin Hematol. 1992;29:159.
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 543
4. Bick RL. Disseminated intravascular coagulation. Objective for rapid reversal of oral anticoagulation. Thromb Res.
criteria for diagnosis and management. Med Clin North Am. 2004;113:371.
1994;78:511. 28. Ghorashian S, Hunt BJ. Off-license use of recombinant
5. Nilsson IM. Coagulation and fibrinolysis. Scand J Gastroen- activated factor VII. Blood Rev. 2004;18:245.
terol Suppl. 1987;137:11. 29. Bucur SZ, Levy JH, Despotis GJ, et al. Uses of antithrombin
6. Mehta JL, Kitchens CS. Pharmacology of platelet-inhibitory III concentrate in congenital and acquired deficiency states.
drugs, anticoagulants, and thrombolytic agents. Cardiovasc Transfusion. 1998;38:481.
Clin. 1987;18:163. 30. Lemmer JH, Despotis GJ Jr. Antithrombin III concentrate
7. Nurden AT. Qualitative disorders of platelets and megakary- to treat heparin resistance in patients undergoing cardiac
ocytes. J Thromb Haemost. 2005;3:1773. surgery. J Thorac Cardiovasc Surg. 2002;123:213.
8. White GC II. Congenital and acquired platelet disorders: 31. Geerts WH, Pineo GF, Heit JA, et al. Prevention of
Current dilemmas and treatment strategies. Semin Hematol. venous thromboembolism: The seventh ACCP confer-
2006;43:S37. ence on antithrombotic and thrombolytic therapy. Chest.
9. Rodeghiero F, Castaman G. Treatment of von Willebrand 2004;126:338S.
disease. Semin Hematol. 2005;42:29. 32. Bombeli T, Spahn DR. Updates in perioperative coagu-
10. Mendolicchio GL, Ruggeri ZM. New perspectives on von lation: Physiology and management of thromboembolism
Willebrand factor functions in hemostasis and thrombosis. and haemorrhage. Br J Anaesth. 2004;93:275.
Semin Hematol. 2005;42:5. 33. Groce JB III. Treatment of deep vein thrombosis using low-
11. Ewenstein BM. von Willebrands disease. Annu Rev Med. molecular-weight heparins. Am J Manag Care. 2001;7:S510.
1997;48:525. 34. Boneu B, de Moerloose P. How and when to monitor a
12. Vischer UM, de Moerloose P. von Willebrand factor: From patient treated with low molecular weight heparin? Semin
cell biology to the clinical management of von Willebrands Thromb Hemost. 2001;27:519.
disease. Crit Rev Oncol Hematol. 1999;30:93. 35. Schwarz RP. The preclinical and clinical pharmacology of
13. Mannucci PM. Treatment of von Willebrands disease. Novastan (Argatroban). In: Pifarre R, ed. New anticoagu-
N Engl J Med. 2004;351:683. lants for the cardiovascular patient. Philadelphia: Hanley &
14. Lee JW. von Willebrand disease, hemophilia A and B, and Belfus, Inc.; 1997:231.
other factor deficiencies. Int Anesthesiol Clin. 2004;42:59. 36. Jang IJ, Hursting MJ. When heparins promote thrombosis:
Review of heparin-induced thrombocytopenia. Circulation.
15. Warrier AI, Lusher JM. DDAVP: A useful alternative to blood
2005;111:2671.
components in moderate hemophilia A and von Willebrand
37. Warkentin TE, Kelton JG. Temporal aspects of heparin-
disease. J Pediatr. 1983;102:228.
induced thrombocytopenia. N Engl J Med. 2001;344:1286.
16. Levi MM, Vink R, de Jonge E. Management of bleed-
38. Gruel Y, Pouplard C, Nguyen P, et al. Biological and
ing disorders by prohemostatic therapy. Int J Hematol.
clinical features of low-molecular-weight heparin-induced
2002;76(Suppl 2):139.
thrombocytopenia. Br J Haematol. 2003;121:786.
17. Lethagen S. Desmopressina haemostatic drug: State-of-
39. Despotis GJ, Gravlee G, Filos K, et al. Anticoagulation
the-art review. Eur J Anaesthesiol Suppl. 1997;14:1.
monitoring during cardiac surgery: A review of current and
18. Gordz S, Mrowietz C, Pindur G, et al. Effect of desmopressin
emerging techniques. Anesthesiology. 1999;91:1122.
(DDAVP) on platelet membrane glycoprotein expression
40. Despotis GJ, Joist JH, Hogue CW Jr, et al. More effective
in patients with von Willebrands disease. Clin Hemorheol
suppression of hemostatic system activation in patients
Microcirc. 2005;32:83.
undergoing cardiac surgery by heparin dosing based on
19. DeWitt DL. Cox-2-selective inhibitors: The new super
heparin blood concentrations rather than ACT. Thromb
aspirins. Mol Pharmacol. 1999;55:625. Haemost. 1996;76:902.
20. Kam PC, Egan MK. Platelet glycoprotein IIb/IIIa antago- 41. Body SC, Morse DS. Coagulation, transfusion and cardiac
nists: Pharmacology and clinical developments. Anesthesi- surgery. In Spiess BD, Counts RB, Gould SA, eds. Perioper-
ology. 2002;96:1237. ative transfusion medicine. Baltimore: Williams & Wilkins;
21. Merlini PA, Rossi M, Menozzi A, et al. Thrombocytopenia 1998:419.
caused by abciximab or tirofiban and its association with 42. Messmore H, Upadhyay G, Farid S, et al. Heparin-
clinical outcome in patients undergoing coronary stenting. induced thrombocytopenia and thrombosis in cardiovas-
Circulation. 2004;109:2203. cular surgery. In: Pifarrer R, ed. New anticoagulants for the
22. Nurden P, Clofent-Sanchez G, Jais C, et al. Delayed cardiovascular patient. Philadelphia: Hanley & Belfus Inc;
immunologic thrombocytopenia induced by abciximab. 1997:83.
Thromb Haemost. 2004;92:820. 43. Dyke CM, Smedira NG, Koster A, et al. A comparison of
23. Szuwart T, Brzoska T, Luger TA, et al. Vitamin E reduces bivalirudin to heparin with protamine reversal in patients
platelet adhesion to human endothelial cells in vitro. Am J undergoing cardiac surgery with cardiopulmonary bypass:
Hematol. 2000;65:1. The EVOLUTION-ON study. J Thorac Cardiovasc Surg.
24. Couris RR. Vitamins and minerals that affect hemostasis 2006;131:533.
and antithrombotic therapies. Thromb Res. 2005;117:25. 44. Koster A, Yeter R, Buz S, et al. Assessment of hemostatic
25. Makris M, Watson HG. The management of coumarin- activation during cardiopulmonary bypass for coronary
induced over-anticoagulation annotation. Br J Haematol. artery bypass grafting with bivalirudin: Results of a pilot
2001;114:271. study. J Thorac Cardiovasc Surg. 2005;129:1391.
26. Makris M, Greaves M, Phillips WS, et al. Emergency oral 45. Koster A, Chew D, Grundel M, et al. An assessment of
anticoagulant reversal: The relative efficacy of infusions different filter systems for extracorporeal elimination of
of fresh frozen plasma and clotting factor concentrate bivalirudin: An in vitro study. Anesth Analg. 2003;96:1316.
on correction of the coagulopathy. Thromb Haemost. 46. Weitz JI. New anticoagulants for treatment of venous
1997;77:477. thromboembolism. Circulation. 2004;110:I19.
27. Lubetsky A, Hoffman R, Zimlichman R, et al. Efficacy and 47. Mann MJ, Tseng E, Ratcliffe M, et al. Use of bivalirudin, a
safety of a prothrombin complex concentrate (Octaplex) direct thrombin inhibitor, and its reversal with modified
544 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
ultrafiltration during heart transplantation in a patient 69. Levi M. Current understanding of disseminated intravascu-
with heparin-induced thrombocytopenia. J Heart Lung lar coagulation. Br J Haematol. 2004;124:567.
Transplant. 2005;24:222. 70. Carey MJ, Rodgers GM. Disseminated intravascular coag-
48. Dyke CM, Koster A, Veale JJ, et al. Preemptive use of ulation: Clinical and laboratory aspects. Am J Hematol.
bivalirudin for urgent on-pump coronary artery bypass 1998;59:65.
grafting in patients with potential heparin-induced throm- 71. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and
bocytopenia. Ann Thorac Surg. 2005;80:299. safety of recombinant human activated protein C for severe
49. Bates SM, Weitz JI. New anticoagulants: Beyond heparin, sepsis. N Engl J Med. 2001;344:699.
low-molecular-weight heparin and warfarin. Br J Pharma- 72. Abraham E, Laterre PF, Garg R, et al. Drotrecogin alfa
col. 2005;144:1017. (activated) for adults with severe sepsis and a low risk of
50. Mohapatra R, Tran M, Gore JM, et al. A review of the oral death. N Engl J Med. 2005;353:1332.
direct thrombin inhibitor ximelagatran: Not yet the end of 73. Dempfle CE. Coagulopathy of sepsis. Thromb Haemost.
the warfarin era. Am Heart J. 2005;150:19. 2004;91:213.
51. Turpie AG, Bauer KA, Eriksson BI, et al. Superiority 74. Rebulla P, Finazzi G, Marangoni F, et al. The threshold
of fondaparinux over enoxaparin in preventing venous for prophylactic platelet transfusions in adults with acute
thromboembolism in major orthopedic surgery using myeloid leukemia. Gruppo Italiano Malattie Ematologiche
different efficacy end points. Chest. 2004;126:501. Maligne dellAdulto. N Engl J Med. 1997;337:1870.
52. Nutescu EA, Helgason CM. Evolving concepts in the 75. Rodgers RP, Levin J. A critical reappraisal of the bleeding
treatment of venous thromboembolism: The role of factor time. Semin Thromb Hemost. 1990;16:1.
Xa inhibitors. Pharmacotherapy. 2004;24:82S. 76. Poller L. The British system for anticoagulant control.
53. Samama MM, Gerotziafas GT, Elalamy I, et al. Bio- Thromb Diath Haemorrh. 1975;33:157.
chemistry and clinical pharmacology of new anticoagulant 77. Henriksen R. Instrumentation and quality control of
agents. Pathophysiol Haemost Thromb. 2002;32:218. hemostasis. In: Lotspiech-Steininger C, Steine-Martin E,
54. Wolberg AS, Meng ZH, Monroe DM III, et al. A systematic Koepke J, eds. Clinical hematology. Philadelphia: JB Lip-
evaluation of the effect of temperature on coagulation en- pincott; 1992:695.
zyme activity and platelet function. J Trauma. 2004;56:1221. 78. Ellison N, Silberstein LE. Hemostasis in the perioperative
55. McLoughlin TM, Greilich PE. Preexisting hemostatic de- period. In: Stoelting RK, Barash PG, Gallagher TJ, eds.
fects and beeding disorders. In: Lake CL, Moore RA, ed. Advances in anesthesia, 1st ed. Chicago: Year Book Medical
Blood: Hemostasis, transfusion, and alternatives in the peri- Publishers; 1986:67.
operative period. New York: Raven Press; 1995:25. 79. Freiberger JJ, Lumb PD. How to manage intraoperative
56. Ferrara A, MacArthur JD, Wright HK, et al. Hypothermia bleeding. In: RW V, ed. Perioperative problems/catastrophes.
and acidosis worsen coagulopathy in the patient requiring Philadelphia: JB Lippincott co; 1987:161.
massive transfusion. Am J Surg. 1990;160:515. 80. Triplett DA. Overview of hemostasis. In Menitove JE,
57. Wang HE, Callaway CW, Peitzman AB, et al. Admission McCarthy LJ, eds. Hemostatic disorders and the blood bank.
hypothermia and outcome after major trauma. Crit Care Arlington: American Association of Blood Banks; 1984:1.
Med. 2005;33:1296. 81. Casserly IP, Kereiakes DJ, Gray WA, et al. Point-of-care
58. Hiippala ST, Myllyla GJ, Vahtera EM. Hemostatic factors ecarin clotting time versus activated clotting time in
and replacement of major blood loss with plasma-poor red correlation with bivalirudin concentration. Thromb Res.
cell concentrates. Anesth Analg. 1995;81:360. 2004;113:115.
59. Ho AM, Larmakar MK, Dion PW. Are we giving enough 82. Kang Y, Lewis JH, Navalgund A, et al. Epsilon-
coagulation factors during major trauma resuscitation? Am aminocaproic acid for treatment of fibrinolysis during liver
J Surg. 2005;190:479. transplantation. Anesthesiology. 1987;66:766.
60. Forestner JE. Massive transfusion protocol for trauma. ASA 83. Traverso CI, Caprini JA, Arcelus JI. The normal thromboe-
Newsl. 2005;69:7. lastogram and its interpretation. Semin Thromb Hemost.
61. Johansson PI, Hansen MB, Sorensen H. Transfusion 1995;21(Suppl 4):7.
practice in massively bleeding patients: Time for a change? 84. Tuman KJ, Spiess BD, McCarthy RJ, et al. Effects of
Vox Sang. 2005;89:92. progressive blood loss on coagulation as measured by
62. Crosby ET. Perioperative haemotherapy: I. Indications for thrombelastography. Anesth Analg. 1987;66:856.
blood component transfusion. Can J Anaesth. 1992;39:695. 85. Kang W. Blood coagulation during liver, kidney, and
63. DeLoughery TG. Management of bleeding with uremia and pancreas transplantation. In Lake CL, Moore RA, eds.
liver disease. Curr Opin Hematol. 1999;6:329. Blood: Hemostasis, transfusion, and alternatives in the
64. Mammen EF. Coagulation defects in liver disease. Med Clin perioperative period. New York: Raven Press; 1995:529.
North Am. 1994;78:545. 86. Zuckerman L, Cohen E, Vagher JP, et al. Comparison
65. Kahl B, Schwartz B, Mosher D. Profound imbalance of pro- of thrombelastography with common coagulation tests.
fibrinolytic and anti-fibrinolytic factors (tissue plasminogen Thromb Haemost. 1981;46:752.
activator and plasminogen activator inhibitor type 1) and 87. Colon-Otero G, Cockerill KJ, Bowie EJ. How to diagnose
severe bleeding diathesis in a patient with cirrhosis: Cor- bleeding disorders. Postgrad Med. 1991;90:145.
rection by liver transplantation. Blood Coagul Fibrinolysis. 88. American Society of Anesthesiologists. Practice guidelines
2003;14:741. for blood component therapy: A report by the American So-
66. Staudinger T, Locker GJ, Frass M. Management of acquired ciety of Anesthesiologists Task Force on Blood Component
coagulation disorders in emergency and intensive-care Therapy. Anesthesiology. 1996;84:732.
medicine. Semin Thromb Hemost. 1996;22:93. 89. OShaughnessy DF, Atterbury C, Bolton Maggs P, et al.
67. Bakker CM, Knot EA, Stibbe J, et al. Disseminated intravas- Guidelines for the use of fresh-frozen plasma, cryoprecipi-
cular coagulation in liver cirrhosis. J Hepatol. 1992;15:330. tate and cryosupernatant. Br J Haematol. 2004;126:11.
68. Bick RL. Disseminated intravascular coagulation: Current 90. Stanworth SJ, Brunskill SJ, Hyde CJ, et al. Is fresh
concepts of etiology, pathophysiology, diagnosis, and treat- frozen plasma clinically effective? A systematic review of
ment. Hematol Oncol Clin North Am. 2003;17:149. randomized controlled trials. Br J Haematol. 2004;126:139.
C H A P T E R 3 6 / B L E E D I N G C O M P L I C AT I O N S 545
91. British Committee for Standards in Haematology, Blood 108. Porte RJ, Molenaar IQ, Begliomini B, et al. Aprotinin
Transfusion Task Force. Guidelines for the use of platelet and transfusion requirements in orthotopic liver trans-
transfusions. Br J Haematol. 2003;122:10. plantation: A multicentre randomised double-blind study.
92. Lane TA. UCSD Medical center blood bank handbook. Avail- EMSALT Study Group. Lancet. 2000;355:1303.
able at: http://health.ucsd.edu/labref/. Accessed October 2, 109. Shiga T, Wajima Z, Inoue T, et al. Aprotinin in major
2006. orthopedic surgery: A systematic review of randomized
93. Spahn DR, Tucci MA, Makris M. Is recombinant FVIIa controlled trials. Anesth Analg. 2005;101:1602.
the magic bullet in the treatment of major bleeding? Br J 110. Cid J, Lozano M. Tranexamic acid reduces allogeneic
Anaesth. 2005;94:553. red cell transfusions in patients undergoing total knee
94. Mayer SA, Brun NC, Begtrup K, et al. Recombinant arthroplasty: Results of a meta-analysis of randomized
activated factor VII for acute intracerebral hemorrhage. controlled trials. Transfusion. 2005;45:1302.
N Engl J Med. 2005;352:777. 111. Camarasa MA, Olle G, Serra-Prat M, et al. Efficacy of
95. Friederich PW, Henny CP, Messelink EJ, et al. Effect of aminocaproic, tranexamic acids in the control of bleeding
recombinant activated factor VII on perioperative blood during total knee replacement: A randomized clinical trial.
loss in patients undergoing retropubic prostatectomy: Br J Anaesth. 2006;96:576.
A double-blind placebo-controlled randomised trial. Lancet. 112. Ickx BE, van der Linden PJ, Melot C, et al. Comparison of
2003;361:201. the effects of aprotinin and tranexamic acid on blood loss
96. Diprose P, Herbertson MJ, OShaughnessy D, et al. Activated and red blood cell transfusion requirements during the late
recombinant factor VII after cardiopulmonary bypass re- stages of liver transplantation. Transfusion. 2006;46:595.
duces allogeneic transfusion in complex non-coronary car- 113. Cirino C, Napoli C, Bucci M, et al. Inflammation-
diac surgery: Randomized double-blind placebo-controlled coagulation network: Are serine protease receptors the
pilot study. Br J Anaesth. 2005;95:596. knot? Trends Pharmacol Sci. 2000;21:170.
97. Lodge JP, Jonas S, Oussoultzoglou E, et al. Recombinant 114. Landis CR, Haskard DO, Taylor KM. New antiinflammatory
coagulation factor VIIa in major liver resection: A ran- and platelet-preserving effects of aprotinin. Ann Thorac
domized, placebo-controlled, double-blind clinical trial. Surg. 2001;72:S1808.
Anesthesiology. 2005;102:269. 115. van Oeveren W, Harder MP, Roozendaal KJ, et al. Aprotinin
98. Raobaikady R, Redman J, Ball JA, et al. Use of activated protects platelets against the initial effect of cardiopul-
recombinant coagulation factor VII in patients undergoing monary bypass. J Thorac Cardiovasc Surg. 1990;99:788.
reconstruction surgery for traumatic fracture of pelvis 116. Landis RC, Asimakopoulos G, Poullis M, et al. The an-
or pelvis and acetabulum: A double-blind, randomized, tithrombotic and antiinflammatory mechanisms of action
placebo-controlled trial. Br J Anaesth. 2005;94:586. of aprotinin. Ann Thorac Surg. 2001;72:2169.
99. OConnell KA, Wood JJ, Wise RP, et al. Thromboembolic 117. Kuitunen A, Hiippala S, Vahtera E, et al. The effects of
adverse events after use of recombinant human coagulation aprotinin and tranexamic acid on thrombin generation and
factor VIIa. JAMA. 2006;295:293. fibrinolytic response after cardiac surgery. Acta Anaesthesiol
100. Cattaneo M, Harris AS, Stromberg U, et al. The effect of Scand. 2005;49:1272.
desmopressin on reducing blood loss in cardiac surgerya 118. Levi M, Cromheecke ME, de Jonge E, et al. Pharmacological
meta-analysis of double-blind, placebo-controlled trials. strategies to decrease excessive blood loss in cardiac
Thromb Haemost. 1995;74:1064. surgery: A meta-analysis of clinically relevant endpoints.
101. Salzman EW, Weinstein MJ, Weintraub RM, et al. Treat- Lancet. 1999;354:1940.
ment with desmopressin acetate to reduce blood loss after 119. Sedrakyan A, Treasure T, Elefteriades JA. Effect of apro-
cardiac surgery. A double-blind randomized trial. N Engl J tinin on clinical outcomes in coronary artery bypass graft
Med. 1986;314:1402. surgery: A systematic review and meta-analysis of random-
102. Czer LS, Bateman TM, Gray RJ, et al. Treatment of se- ized clinical trials. J Thorac Cardiovasc Surg. 2004;128:442.
vere platelet dysfunction and hemorrhage after cardiopul- 120. Munoz JJ, Birkmeyer NJ, Birkmeyer JD, et al. Is epsilon-
monary bypass: Reduction in blood product usage with aminocaproic acid as effective as aprotinin in reducing
desmopressin. J Am Coll Cardiol. 1987;9:1139. bleeding with cardiac surgery? A meta-analysis. Circulation.
103. Mongan PD, Hosking MP. The role of desmopressin acetate 1999;99:81.
in patients undergoing coronary artery bypass surgery. 121. Casati V, Guzzon D, Oppizzi M, et al. Hemostatic effects of
A controlled clinical trial with thromboelastographic risk aprotinin, tranexamic acid and epsilon-aminocaproic acid
stratification. Anesthesiology. 1992;77:38. in primary cardiac surgery. Ann Thorac Surg. 1999;68:2252.
104. Gratz I, Koehler J, Olsen D, et al. The effect of desmopressin 122. Mangano DT, Tudor IC, Dietzel C. The risk associated with
acetate on postoperative hemorrhage in patients receiving aprotinin in cardiac surgery. N Engl J Med. 2006;354:353.
aspirin therapy before coronary artery bypass operations. 123. Society of Cardiovascular Anesthesiologists. SCA Statement
J Thorac Cardiovasc Surg. 1992;104:1417. regarding the use of antifibrinolytic drugs. Available at:
105. Henry DA, Moxey AJ, Carless PA, et al. Anti-fibrinolytic www.scahq.org/sca3/aprotinin.shtml. Accessed August 23,
use for minimising perioperative allogeneic blood trans- 2006.
fusion. The Cochrane Collaboration. 2006. Available 124. Dietrich W, Spath P, Zuhlsdorf M, et al. Anaphylactic reac-
at: http://www.cochrane.org/reviews/en/ab001886.html. Ac- tions to aprotinin reexposure in cardiac surgery: Relation
cessed October 2, 2006. to antiaprotinin immunoglobulin G and E antibodies. Anes-
106. Khoshhal K, Mukhtar I, Clark P, et al. Efficacy of aprotinin thesiology. 2001;95:64.
in reducing blood loss in spinal fusion for idiopathic 125. Porte RJ, Bontempo FA, Knot EA, et al. Systemic effects
scoliosis. J Pediatr Orthop. 2003;23:661. of tissue plasminogen activator-associated fibrinolysis and
107. Lemay E, Guay J, Cote C, et al. Tranexamic acid reduces its relation to thrombin generation in orthotopic liver
the need for allogenic red blood cell transfusions in transplantation. Transplantation. 1989;47:978.
patients undergoing total hip replacement. Can J Anaesth. 126. Kufner RP. Antifibrinolytics and orthotopic liver transplan-
2004;51:31. tation. Transplant Proc. 1998;30:692.
CHAPTER SICKLE CELL ABNORMALITIES
37
AND ACUTE PORPHYRIA
CASE SUMMARY
What Is the Genetic Cause
38-kg, 16-year-old, African American man
of Sickle Cell Disease?
A
with a history of sickle cell disease (SCD) is
scheduled for a cholecystectomy. He has had
multiple hospitalizations for painful crisis, Hb consists of two chains and two non- chains. In the
acute chest syndrome (ACS), and priapism, case of normal adult Hb, these non- chains are chains.
and has often required transfusion therapy. The chain production is coded by two alleles located
He suffered a cerebrovascular accident (CVA) 1 year ago in the chromosome 11, whereas the chain production
that has left him with a right-sided hemiparesis and is controlled by four alleles paired on chromosome 16.
mild cognitive dysfunction. The patient has been on a Abnormal Hb states can result from underproduction of
transfusion protocol since the CVA. His hemoglobin (Hb) a globin chain or production of an abnormal amino acid
is 8.2 g per dL, and his liver enzymes are mildly elevated. sequence within a chain.
The patient was admitted the day before surgery and Underproduction of a given chain results in the group
received 10 mL per kg of packed red blood cells that raised of disorders known as thalassemia. -Thalassemia is the
his Hb to 10.1 g per dL and lowered his HbS to 31%. The underproduction of the chain. There are four types of
patient received intravenous hydration before and during -thalassemia, ranging in severity from mild to severe,
an uneventful general anesthetic for a laparoscopic chole- depending on how many of the four -globulin chains
cystectomy. In the postanesthesia care unit, he developed are underproduced. -Thalassemia is commonly seen in
shivering and pain, and subsequently became combat-
people from the Mediterranean Sea region. It results from
ive and agitated. His pulse oximeter oxygen saturation
an underproduction of chains. Heterozygotes have a
(SpaO2 ) dropped to 85%, and he developed a hacking, dry
mild anemia known as thalassemia minor. Homozygotes
cough. A chest radiograph demonstrated a new left lower
are known as thalassemia major, or Cooleys anemia,
lobar infiltrate.
and are usually transfusion-dependent. Patients may be
Treatment includes supplemental oxygen, intra-
genetically coded to have both a sickling disorder and one
venous fluids, a forced air warming blanket, and mor-
of the thalassemias.
phine. He is admitted to the intensive care unit for
treatment of ACS. He then developed respiratory fail- SCD is the most common of all hereditary disorders,
ure, which required 1 week of mechanical ventilation, and affecting up to 0.2% of the adult African American
is eventually discharged after a 4-week hospital course. population with SCD, 8% with sickle cell trait,1 and
approximately 50,000 children in the United States with
SCD.2,3 A single amino acid substitution at position 6
INTRODUCTION on the chain is responsible for the condition. SCD is
inherited as an autosomal recessive disorder following
Anesthesiologists are frequently called upon to care for a predictable Mendelian pattern. Therefore, heterozygote
patients with hemoglobinopathies. As in this case, the com- (HbAS) parents will have a 25% chance of producing
plications of the disease often occur in the postoperative either a normal (HbAA) or SCD (HbSS) and a 50% chance
period. Adequate preoperative preparation, intraoperative of producing another heterozygote (HbSA, trait) child.
and postoperative management may prevent many of the Normal adult red blood cells contain three types of
complications associated with this disease. Hb, HbA (2 ,2 ) and small quantities of HbA2 (2 ,2 ) and
546
C HAPTE R 37/SIC K LE C E LL ABNOR MALITI E S AN D AC UTE POR P HYR IA 547
TABLE 37.1 Genotype and Hemoglobin Electrophoresis Patterns in Various Hemoglobinopathy Syndromes
Hemoglobinopathy Neonatal
Syndrome Genotype Electrophoresis HbA % HbS % HbF % HbC %
Sickle cell trait HbSA FAS 5560 <50 12 0
Sickle + thalassemia HbS + FAS or FS 1530 5575 220 0
Hemoglobin SC disease HbSC FSC 0 4555 28 4555
Sickle 0 thalassemia HbS 0 FS 0 5080 230 0
Sickle cell disease HbSS FS 0 >80 220 0
HbA, hemoglobin A; HbS, sickle hemoglobin; HbF, fetal hemoglobin; HbC, hemoglobin C; HbS + , hemoglobin sickle + thalassemia; HbS 0 ,
hemoglobin sickle 0 thalassemia; 0 , thalassemias with absent production of globin; + , thalassemias with reduced but not absent
production of globin.
HbF (2 2 ). Patients with SCD have >50% HbS, with has been shown to be beneficial in some patients suffering
the remainder being HbF or HbA2 (see Table 37.1). They a vaso-occlusive crisis, including stroke and ACS.
contain no HbA unless they have a double heterozygous With time, these complex interactions between the
condition such as both SCD and -thalassemia. However, deformed and sticky red cells and the endothelium results
they will always have more than 50% HbS. Individuals in widespread chronic endothelial inflammation, injury,
with a combination of normal HbA and <50% HbS have and organ dysfunction.
sickle cell trait. Infants may have >70% HbF, which The HbS red cell is unstable and insoluble, result-
persists for up to 4 months of age when the fetal red ing in early red cell destruction, sickling, and endothelial
cells are replaced by hematopoiesis of adult red cells as damage. Red blood cells with HbSS begin sickling when
chain production replaces chain production. When the oxygen saturation falls below 85% (PaO2 of approxi-
fetal Hb is present and persists into adulthood, it can mately 40 to 50 mm Hg). Acidosis, hypoxia, intracellular
provide protection against sickling. HbC and thalassemia dehydration, and vascular stasis increase the sickling
+ also provide some protection against sickling, and process (see Table 37.2). Decreased cardiac output and
these patients usually have a milder clinical course. hypovolemia lead to increased transit time through the
TABLE 37.3 Age-Specific Clinical Complications Often Present in Patients with Sickle Cell Disease
Age Condition Clinical Presentation
Infant Dactylitis Painful vaso-occlusion, hand-foot syndrome
Splenic sequestration Anemia, hypovolemia, death
Child Aplastic crisis Bone marrow hypoplasia, severe anemia
Splenic infarction Prone to infection with encapsulated organisms
Infection Sepsis, osteomyelitis, Staphylococcus or Salmonella
Vaso-occlusive crisis Pain, usually of the extremities, may be abdominal
Reactive airway disease May lead to acute chest syndrome
Enuresis Inability to concentrate urine
Adolescent to adult Acute chest syndrome Chest pain, fever, cough, new infiltrate on CXR
Stroke Microinfarcts or hemorrhagic
Priapism May lead to impotence
Leg ulcers
Cholelithiasis Pigment gallstones due to hemolysis
Chronic renal insufficiency
Hematuria
Orbital infarction
hypoxic environment of the capillary bed, also increasing oxygen, exchange transfusion, hydration, aggressive respi-
the sickling process. The consequence of sickling is en- ratory support, and antibiotics (see Table 37.4). Inhaled
dothelial adhesion and occlusion of the microvasculature. NO may also have a unique role in treating ACS.8
Erythrocytes, leukocytes, platelets, vascular endothelium, Cerebral vascular accidents occur in approximately
NO, and the coagulation cascade are involved in the vas- 5% of subjects with SCD.9 Most CVAs in children are
cular injury that results from the sickled red blood cell. due to ischemic infarcts, whereas adults may also be
In fact, SCD should be considered as much a vascular hemorrhagic. It has recently been recognized that as
endothelial disease as a red blood cell disease. many as 15% to 25% of asymptomatic children have
radiographic evidence of silent cerebral infarcts. In many
centers, patients with SCD are screened with transcranial
Doppler, and those with elevated flow velocities in the
What Are the Clinical middle cerebral and terminal internal carotid arteries are
placed on hypertransfusion regimens. Individuals with a
Manifestations of Sickle Cell previously documented CVA are placed on transfusion
Disease? regimens, usually for a period of 10 years. It is only
recently that we are noting that patients coming off these
regimens are experiencing recurrence.
Sickle cell problems begin in infancy and culminate in
Subjects who have had either ACS or CVA should be
multiorgan damage in adulthood after years of endothelial
considered high-risk patients and may benefit from ag-
damage, microinfarcts and ischemic damage to end
gressive preoperative transfusion with a target HbS level.
organs (see Table 37.3). Because of immunologic deficits
Aside from these acute manifestations, the anesthesi-
and splenic dysfunction, these patients are at very high
ologist must be aware of the considerable risk for chronic
risk for overwhelming sepsis. One of the main reasons
end organ dysfunction. These patients may have sickle cell
for widespread postnatal screening for this disease is so chronic lung disease, cardiac dysfunction with high output
children can be placed on penicillin until the age of 6 congestive heart failure, and renal insufficiency with poor
to assure appropriate immunoprophylaxis. The clinical concentrating ability, retinopathy, and liver disease.
manifestations of SCD can be grouped into acute and
chronic. Typically, the acute problems have been grouped TABLE 37.4 Indications for Red Blood Cell Transfusion
as various crises: Splenic sequestration, hyperhemolytic, in Patients with Sickle Cell Disease
aplastic, and vaso-occlusive. The vaso-occlusive crisis can
be further broken down into: Painful, priapism, ACS, and Transient ischemic attack
stroke. It is these vaso-occlusive events that are of the Stroke
most concern during the perioperative period. Acute chest syndrome
ACS and stroke are the most concerning periopera- Aplastic crisis
tive complications. ACS may result from pneumonia, fat Splenic sequestration crisis
emboli, pulmonary vaso-occlusion, and sequestration. It Severe hemolytic anemia
occurs in 10% to 20% of postoperative patients,68 and may Priapism
rapidly progress to respiratory failure. The mortality rate Spinal cord infarct
from ACS is 2% to 12%. This, more than anything else, ac- Refractory vaso-occlusive crisis
counts for the very high perioperative mortality rate of 1% Pregnancy
in these patients. Treatment for ACS includes supplemental
C HAPTE R 37/SIC K LE C E LL ABNOR MALITI E S AN D AC UTE POR P HYR IA 549
The goal of the perioperative management is to of high concentrations of the more soluble HbF in the
mitigate those factors that may promote the acute crises first 4 months of life, the solubility test is inaccurate in
while managing the issues associated with any chronic detecting HbS in these infants. The normal infant has a
organ dysfunction. mixture of fetal Hb and HbA. The result from the neonatal
screening is reported as HbFA. When HbS is present
with only fetal Hb, the infant has SCD (or a multiple
heterozygote variant), and it is reported as HbFS. The
Should Patients Be Screened infant with sickle cell trait has all three: HbF, S, and A
(Table 37.1).
for Sickle Cell Disease?
Presently, newborn screening programs are available in
the United States; therefore, most newborns are tested What Is the Perioperative
at birth for multiple genetic abnormalities, including
electrophoresis. Unfortunately, the degree to which this
Management of Patients
information is communicated to and understood by the with Sickle Cell Disease?
parents is quite variable. In the older infant, child, or adult,
a more rapid solubility test, which promotes formation of
insoluble sickle Hb, is available. If the solubility test is PREOPERATIVE
positive, it will require a Hb electrophoresis to determine
the type of hemoglobinopathy.
PREPARATION
A few facts may be helpful in deciding whether a As noted in the preceding text, surgical morbidity and
preoperative screening program has any true merit. The mortality is increased in patients with SCD. Historically,
overall mortality rate for adults undergoing anesthesia mortality rates as high as 10% and morbidity rates as high
and surgery has reached a lofty, 6- level of 1 in 300,000. as 50% have been reported.10 More recent studies have
Mortality rates for children overall is considerably worse reported a mortality of approximately 1%.11 SCD causes
(approximately 1 in 80,000). The mortality rate for pa- such widespread organ damage, that it is difficult to give a
tients with SCD undergoing surgery is 1%. For this reason, single critical pathway for standardized preoperative care
we believe it is imperative to institute screening protocols which will apply to all patients. The patients hematologist
for SCD. Table 37.5 shows the various procedures fol- should be intimately involved in the preoperative prepara-
lowed at our institution to determine the sickle cell status tion. Collectively, the hematologist, anesthesiologist, and
on preoperative patients who are considered to be at risk. surgeon can assess the patients perioperative risk and
Because 95% of all patients with SCD would have had determine any special requirements.
some clinical manifestations by age 9, we do not screen In general, the patient should be admitted to the
beyond this age. If a child is found to have SCD, sickle hospital the day before surgery to permit assessment
HbC disease, or sickle -thalassemia, they should be seen by hematology, anesthesiology, and surgery. An elective
by their hematologist who will assist in the preparation surgical procedure should not proceed in a patient with
for surgery. SCD and an ongoing infection, such as urinary tract
infection or respiratory tract infection, because these
could lead to a painful crisis, ACS, or CVA. Intravenous
fluids should be administered while the patient is fasting.
How Do I Interpret a Neonatal Although the recommendation of an infusion rate one
and a half times the maintenance fluid requirement of
Hemoglobin Electrophoresis? a balanced salt solution is frequently made, there is no
evidence that excessive hydration is beneficial. The goal
Neonatal screening requires a Hb electrophoresis to of intravenous fluid therapy is to prevent dehydration
determine the presence of HbS. Because of the presence throughout the perioperative period.
The role of preoperative transfusion continues to be
TABLE 37.5 Methods to Determine Sickle the Presence somewhat controversial. Based on the concern for high
of Sickle Cell Abnormalities in Populations at Risk mortality and morbidity rates, hematologists and anes-
for This Disease thesiologists in the 1980s and 1990s provided aggressive
transfusion practices, with target Hb of 10 mg per dL
Newborn screen from the state laboratory and HbS concentration of 30% or less. This was con-
A report from the childs pediatrician sidered by many to be the standard of care for most
A reliable report from the parent, along with a patients until the Preoperative Transfusion in Sickle Cell
hemoglobin determination of 10 g/dL or greater Disease Study Group published their findings in 1995.12
A sickle cell preparation and, if positive, a hemoglobin This group determined that the complication rate, includ-
electrophoresis ing death and ACS, was equal in a group that underwent
A child whose age is 10 y with a negative history for aggressive transfusion with a target Hb of 10 mg per
sickle cell disease and a hemoglobin determination of dL and HbS of 30% compared to those who had a sim-
10 g/dL ple correction of anemia to an Hb of 10 mg per dL.
There has never been a randomized controlled study
550 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 37.6 Risk Assessment of the Patient and the Procedure in Patients with Sickle Cell Disease
Low-risk patients Infrequent and mild crisis: May be associated with sickle + thalassemia or persistent fetal
hemoglobin
No preexisting pulmonary, cardiovascular, or central nervous system complications of the
disease
High-risk patients Preexisting pulmonary, cardiovascular, or central nervous system complications of the
disease (by clinical and/or imaging and blood flow assessments)
Frequent and/or moderate to severe crisis
Very high-risk patients Recent or frequent ACS, CVA or TIA
Low-risk surgery Minor procedures (myringotomy and tubes)
High-risk surgery Intrathoracic, intra-abdominal, intracranial, large blood loss or fluid shifts, airway surgery
(including tonsillectomy and/or adenoidectomy), emergency procedures
ACS, acute chest syndrome; CVA, cerebrovascular accident; TIA, transient ischemic attack.
retrospective studies have supported the safe use of tourni- respiratory function, and pain control in the postoperative
quets in patients with SCD.20 One of the recommendations period. In addition to aggressive analgesia for the surgical
when using tourniquets include fully exsanguinating the incision, the patient may develop vaso-occlusive crisis
extremity to remove as many HbS red blood cells from or ACS, which also require pain management. These
the vascular bed as possible. The planned use of a tourni- patients may have been on opioids for vaso-occlusive
quet during extremity surgery may alter the preoperative crisis, and often have developed tolerance to these
transfusion goals of lowering the HbS concentration. medications. In these cases, a higher-than-usual dose is
frequently required to achieve adequate analgesia. Local
anesthetic injection, peripheral nerve blocks, neuraxial
POSTOPERATIVE blockade, intravenous narcotics, and nonsteroidal anti-
COMPLICATIONS inflammatory drugs (when not contraindicated) are useful
adjuvants to decrease postoperative pain in these opioid-
The postoperative period is the most vulnerable time
tolerant patients. Vaso-occlusive crisis, renal insufficiency,
for SCD complications to occur.10 It is important to
or CVA may complicate the postoperative course.
maintain adequate oxygenation, perfusion, hydration,
P A R T II PORPHYRIA
CASE SUMMARY incidence and prevalence of the specific types varies widely
among countries. Porphyria can present with cutaneous
33-year-old woman presents with acute ab- symptoms, systemic symptoms, or both. Of most concern
A
dominal pain, nausea, and vomiting. She for the anesthesiologist are the acute porphyrias: Acute
also complains of muscle pain and weak- intermittent porphyria; variegate porphyria; hereditary
ness in both arms. She has recently been coproporphyria; and plumboporphyria.
trying to lose weight with a calorie-reduced
diet. Her mother and sister have a history of
acute intermittent porphyria. The surgeon schedules her
for a laparoscopic appendectomy. What Are the Clinical
Manifestations of an Acute
Attack?
What Is Porphyria?
Acute attacks almost always start with severe abdominal
The porphyrias are a group of inherited disorders of heme pain, although the clinical presentation can be varied and
synthesis. The porphyrias can be classified according to episodic (see Table 37.10). The most common presenting
their site of enzymatic defect (see Table 37.8).21 The symptoms are abdominal pain, extremity pain or pares-
specific enzyme defects lead to the accumulation of thesia, vomiting, and constipation.23 Acute attacks may
porphyrins and heme precursors (see Table 37.9).22 The be fatal and can be treated with heme arginate.24,25 When
neuromuscular weakness and quadriplegia occurs, recov-
ery may be prolonged from weeks to months. During an
TABLE 37.8 Classification of the Types of Porphyria acute attack, blood, urine, and stool studies will demon-
strate the abnormal accumulation of porphyrins and heme
Porphyrias Inheritance synthesis substrates. DNA analysis and family history can
HEPATIC help identify affected individuals during the asymptomatic
Acute intermittent porphyria Autosomal dominant phases. An acute attack is rare before puberty and occurs
Porphyria cutanea tarda Autosomal dominant mostly after a precipitating event. Precipitating factors in-
Variegate porphyria Autosomal dominant clude hormonal fluctuations, fasting, dehydration, stress,
Hereditary coproporphyria Autosomal dominant infection, and medications.26 Of most importance to the
ALA dehydratase deficiency Autosomal recessive anesthesiologist is choosing an anesthetic that will not
induce an acute attack.
ERYTHROPOIETIC
Congenital erythropoietic Autosomal recessive
porphyria ANESTHETIC MANAGEMENT
Erythropoietic protoporphyria Autosomal dominant
A careful history is required to identify individuals with
ALA, -aminolevulinic acid. porphyria, including a detailed family history. Patients in
552 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
acute crisis may have impending respiratory failure due to 2. SCD leads to chronic endothelial damage throughout
progressive muscle weakness or increased aspiration risk the body, leading to multiple organ dysfunctions.
due to cranial nerve dysfunction. Fluid and electrolyte 3. Patients with SCD, sickle -thalassemia, and sickle
imbalance may be present during an acute attack. HbC disease who also have a history of recent or fre-
The choice of anesthetic technique and medications quent ACS, CVA, or transient ischemic attack (TIA)
must avoid those factors known to precipitate an acute are at very high risk for postoperative complications.
attack (see Table 37.11).27,28 Most information concerning 4. Preoperative transfusion to an Hb of 9 to 10 g
anesthetic use is based on historical and anecdotal per dL is recommended in all but the lowest risk
reports; even medications once considered safe may situations.
precipitate an attack. Regional anesthesia is a reasonable 5. Provide adequate oxygenation (50% inspired oxygen),
alternative, keeping in mind the decreased intravascular oxygen saturation >95%, hydration, normothermia,
volume and autonomic dysfunction that may be present Hb of 9 to 11 gm per dL, and pain control throughout
during an acute attack. General anesthesia can be safely the perioperative course.
administered if precipitating medications are avoided 6. Never transfuse to a Hb >11 gm per dL.
and care is taken to preserve fluid balance, provide 7. Postoperative complications include painful crisis,
carbohydrates, and decrease stress hormones through ACS, CVA, and renal insufficiency.
adequate perioperative pain control.
Unfortunately, the symptoms of acute porphyria may
resemble an acute surgical abdomen, thereby leading to
unnecessary abdominal exploration. When faced with a TABLE 37.11 Common Medications Categorized for Use
patient with porphyria, the internet can be a great resource in Porphyria
for researching the safety of a particular medication.
Generally Safe Considered Unsafe
Propofol Barbiturates, thiopental, etc.
KEY POINTS SICKLE CELL: Volatile agents (except Etomidate
halothane)
1. Children younger than 10 years should have a reliable Nitrous oxide Phenytoin
report of testing for sickle cell status. Opioids Carbamazepine
Anticholinergics Imipramine, nortriptyline,
amitriptyline, etc.
TABLE 37.10 Clinical Symptoms of Acute Porphyria Anticholinesterases Calcium channel blockers,
nifedipine, amlodipine,
Abdominal pain felodipine, nimopdipine,
Autonomic neuropathy etc.
Constipation or diarrhea Neuromuscular Alcohol, cannabis, tobacco,
Extremity pain or paresthesia blockers: and cocaine
Peripheral neuropathy Depolarizing and
Back or chest pain nondepolarizing
Loss of sensation Midazolam Ergot preparations
Psychiatric symptoms (hallucinations, paranoia, Bupivacaine Sulfonamide antibiotics:
depression) Sulfamethoxazole and
Cranial neuropathy trimethoprim
Seizures Chloral hydrate Metoclopramide
Coma Ketorolac, ibuprofen, Clindamycin
Basal ganglion, cerebellar or pyramidal tract symptoms etc.
C HAPTE R 37/SIC K LE C E LL ABNOR MALITI E S AN D AC UTE POR P HYR IA 553
38 Christine S. Rinder
Y
on a 30-year-old woman, 5 ft 3 in., and Genetic mutations that reduce the functional levels
180 lbs. She suffered from femoral deep of endogenous anticoagulantsantithrombin, protein C,
vein thrombosis at age 18 after starting and protein Shave long been known to predispose to
oral contraceptive pills. Her family history venous thromboembolism (VTE). However, new genetic
includes an aunt who died of pulmonary abnormalities are being discovered that contribute to hy-
embolism. She is not anticoagulated, and a total knee percoagulability. Indeed, with appropriate testing, causes
replacement surgery is planned; her orthopedic surgeon of thrombophilia are being detected in as many as 50%
sent her to you to find out, What are the chances that of VTE cases. However, genetic factors are only rarely the
anesthesia will give me a clot and Ill die like my aunt? sole contributor to VTE. Genetic sources of thrombophilia
create a lifelong hypercoagulable state that give rise to
clinical VTE only episodically. In most cases of VTE, hy-
percoagulabilityacquired or environmentalserves as
What Is the Influence a triggering factor. As anesthesiologists are often ac-
of Hypercoagulability tively involved in clinical events that create acquired
hypercoagulability, for example, pregnancy, surgery, and
in Thrombosis? malignancy, it is important to stay abreast of advances in
the understanding and treatment of thrombosis.
Thrombosis in either the venous or arterial circulation Historically, blood clotting has been viewed as a se-
may have catastrophic consequences. In the venous cir- ries of enzymatic reactions in which the participants,
culation, three vascular conditions occurring singly or, sequence, and regulation were independent of location. In-
more often, in concert, put patients at increased risk creasingly, recognition of the unique conditions operating
for thrombosis. These include (a) stasis, (b) injury to in the arterial circulation, including (a) the high velocity
vascular endothelium, and (c) hypercoagulability. Hyper- of blood flow, creating shear forces in vessels of smaller
coagulability may be thought of as a state of exaggerated diameter, (b) its complex rheology at vessel branch points,
activation of coagulation. A comparable triad that pre- and (c) its unique anticoagulant requirements (i.e., better
disposes to arterial thrombi includes flow limitation from efficacy of antiplatelet agents as opposed to antithrom-
atherosclerosis, plaque rupture as a uniquely arterial form bin agents), has mandated the development of an arterial
of endothelial injury, and arterial hypercoagulability, an clotting model distinct from that operating in the venous
entity just beginning to be understood. Fundamental to circulation.1 In concert with this improved understanding
the prevention and management of hypercoagulability of the physiology of arterial hemostasis has come an ap-
is a better understanding of thrombosis and its endog- preciation for how tightly regulated the controls on this
enous regulators. Accordingly, this review will begin by system must be maintained. Indeed, given the spectrum of
describing the most current model for the initiation and clinical complications associated with defects in arterial
propagation of clotting, together with the role of natural hemostasis, including catastrophic blood loss at one end
anticoagulants, the activity of which is critical to the pre- and infarction at the other, it is not surprising that mul-
vention of thrombosis. Using that model, the etiology of tiple interdependent factors keep a very tight rein on this
different hypercoagulable states will be explored. process. Accordingly, although there are conditions that
Sources of hypercoagulability can be divided into give rise to both venous and arterial hypercoagulability,
two major classes: (a) a congenital predisposition caused a subset of heritable and acquired conditions exist that
554
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 555
Va
VIIIa and IXa. When these active proteases are bound in
Platelet
combination with negatively charged phosphatidylserine,
Thrombin
ThrombinThrombin the resulting enzyme complex enhances the binding of
ThrombinThrombinThrombin
ThrombinThrombinThrombinThrombin
its substrate, factor X. The speed of X activation by this
Fibrinogen
complex is also increased by platelet binding; indeed,
when compared to the reaction speed of free proteases in
Fibrin
solution, assembly of the entire reaction on the platelet
membrane increases the catalytic efficiency of X activation
by approximately 13 million-fold.
FIGURE 38.2 The two-phase model of coagulation. In the Assembly of the prothrombinase complex is similarly
initiation phase of coagulation, tissue factor (TF) exposed on a dependent on the activated platelet surface for optimum
subendothelial fibroblast after vessel injury complexes with activity. Much of factor V released by the activated platelet
small amounts of factor VIIa present in the circulation. This (subsequently thrombin-activated) stays bound to the
complex then activates a small amount of factor X to Xa in the platelet surface. Like the Xase complex, the platelet-bound
presence of an activated platelet. The platelet-bound Xa converts prothrombinase complex (i.e., Xa-Va-phosphatidylserine)
a tiny amount of prothrombin to thrombin. This small amount enhances prothrombin activation to thrombin 300,000-
of thrombin then sparks the propagation phase of coagulation. fold faster than free Xa and Va acting on prothrombin
Thrombin activates factors XI, VIII, and V at or near the formed in solution. Platelet-bound Xa is the rate-limiting
activated platelet. Factor IX is activated by either factor XIa or reagent in prothrombin cleavage. The substrate for
the TF-VIIa complex. Factor IXa complexes with the factor VIIIa this enzyme complex, prothrombin, binds to GPIIb/IIIa
activated by thrombin, and on the platelet surface generates on both activated and inactivated platelets, potentially
factor Xa with remarkable kinetic efficiency. The platelet-bound providing a source for thrombin generation in both the
factor Xa complexes with factor Va, which converts initiation and propagation phases.
prothrombin to explosive amounts of thrombin. This thrombin Evidence that factor XI further amplifies the propaga-
in turn converts fibrinogen to fibrin, thereby sealing the vessel tion phase is growing.7 As noted in the preceding text, Xa
injury beneath. levels are rate limiting to the prothrombinase complex, par-
ticularly once the switch is made from extrinsic (FVIIa:TF
complex) to intrinsic (FVIIIa:FIXa complex) Xase. Small
therapy; however, they do not model the entire sequence amounts of IXa can be generated by the TFVIIa com-
of events necessary for effective hemostasis. They fail to plex, but its ability to sustain IXa generation is limited by
give information relevant to thrombin generation dur- its endogenous inhibitor, tissue factor pathway inhibitor
ing the propagation phase, which determines whether a (TFPI).7 To generate Xa in amounts sufficient to fuel the
persistent clot will form, or whether endogenous antico- propagation phase, an alternative and kinetically superior
agulants and fibrinolytic regulators are able to constrain source of IXa is needed. Factor XI is another zymogen
excess clot growth. activated by the minute amounts of thrombin generated
Thrombin generated during the initiation phase is during initiation, but this activation only occurs on the
a potent platelet activator, thereby providing both an activated platelet surface.7 Platelet-bound FXIa is ideally
activated platelet surface membrane and platelet-released located to activate FIX, which also binds to the platelet
factor V (which thrombin promptly converts to Va). Factor surface, helping to speed and localize this activity. Addi-
VIII, conveniently brought to the bleeding site by its tionally, binding to the platelet surface protects FXIa from
carrier, the von Willebrand factor (vWF), is also activated its inhibitor, protease nexin 2. Therefore, FXIa generation
by thrombina step that causes its release by vWF. on the activated platelet is key to providing sufficient FIXa
FVIIIa then complexes with the picomolar amounts of to maintain Xa generation through the catalytically more
factor IXa, also generated by the TFVIIa complex during efficient intrinsic Xase complex.
the initiation phase, forming the FVIIIaIXa complex, a In the venous circulation, the kinetic advantage of
pivotal point in the successful generation of a clot. coagulation cascade assembly on the activated platelet
The formation of the FVIIIaIXa complex on the surface is readily apparent; recent in vitro work has
platelet surface heralds the switch from FXa generation clarified the minimum platelet count (i.e., the dose of
by the TFVIIa complex to the intrinsic Xase pathway. platelets) necessary for the reaction sequence to proceed.8
This switch is of enormous kinetic advantage, with the When all coagulation zymogens are present, thrombin is
intrinsic Xase complex exhibiting approximately a 50-fold not generated unless platelets are present as a source
greater efficiency at Xa generation. The bleeding diathesis of phospholipid. Once platelets are added, thrombin
associated with hemophilia is testament to the hemostatic generation begins and grows with increasing numbers
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 557
of platelets, up to a threshold of 10,000 platelets per L. kinetics when bound to heparin or endogenous heparan
Increases in platelets beyond this number have no effect sulfate), protein C requires activation by thrombin-
on the efficiency of the reaction, suggesting that, as had stimulated endothelial cell thrombomodulin. Activated
been empirically observed in thrombocytopenic patients, protein C (APC) then acts in concert with its cofactor,
the platelet level must decrease to very low levels (i.e., protein S, to limit the rate of thrombin generation by
<10,000 per L) to increase the risk of venous bleeding. inactivating the essential procoagulant cofactors, FVa
This contrasts sharply with the arterial circulation where and FVIIIa.9 Again, the activated platelet membrane
the minimum platelet count needed to ensure hemostasis promotes clot success by protecting FVIIIa and FVa from
for operative procedures is at least five times that number. inactivation by APC.10
a All
numbers pertain to heterozygous state.
DVT, deep venous thrombosis.
Adapted with permission from: van der Meer FJ, Koster T, Vandenbroucke JP, et al. The leiden thrombophilia
study (LETS). Thromb Haemost. 1997; 78:631.
558 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
more than 40% of the VTE that developed did so in resistance in a single family in 1993, and subsequently
association with either surgery or pregnancy.12 Only found that among other VTE patients, their plasma often
11% of VTE were spontaneous, that is, had no known exhibited resistance to the normal anticoagulant effect of
precipitating factors. APC.15 Specifically, the addition of exogenous APC to their
Hereditary deficiencies in protein C (PC) and protein plasma did not prolong the aPTT of these VTE patients
S (PS) also adversely impact thrombin regulation. How- when compared with the prolongation found by APC treat-
ever, instead of limiting the activity of thrombin already ment of plasma from non-VTE controls. Several groups
formed, congenital deficiencies in PC and PS hamper subsequently demonstrated that approximately 90% of
the affected individuals ability to limit rates of thrombin patients with APC resistance have an activated form of
generation. Similar to homozygous AT deficiency, many FV that is resistant to proteolysis by APC.16 The gene
infants homozygous for PC or PS deficiency do not sur- responsible for this effect, factor VLeiden , differs from the
vive long after birth. With heterozygous deficiencies, the normal gene by a single nucleotide, producing an amino
relative surplus of Va and VIIIa that results from defective acid substitution at one of the sites where APC normally
inactivation ensures that both the tenase and prothrom- cleaves FVa, thereby rendering it refractory to inactiva-
binase complexes are able to act with enhanced kinetics, tion. Accordingly, FVaLeiden stays active in the circulation
generating an overabundance of thrombin and setting longer than normal, fostering increased thrombin gener-
the stage for VTE risk of the same order of magnitude ation.
as AT deficiency (Table 38.1). Moreover, the synthesis As the sole source of hypercoagulability, FVLeiden
of PC and PS are both vitamin K dependent, with PC is viewed as having low to intermediate procoagulant
having the shorter half-life. Accordingly, individuals who risk. Patients heterozygous for FVLeiden have a fivefold
are PC-deficient are at particular risk for thrombosis if to sevenfold increased risk of VTE (Table 38.1), whereas
warfarin therapy is initiated without the protection of the risk for homozygous carriers is increased up to 80-
initial anticoagulation with heparin. Specifically, during fold. FVLeiden may be present at a high frequency in the
the first days of warfarin treatment, before inhibition of general population. Its prevalence varies considerably in
vitamin K has decreased factors VII, IX, and XI suffi- different ethnic populations. It is present in approximately
ciently to provide the intended anticoagulation, modest 5% of people of northern European descent but rarely
suppression of PC synthesis may compound the already in patients of African or Asian descent.17 Accordingly,
subnormal PC levels, resulting in paradoxically height- depending on the ethnic makeup of the community,
ened hypercoagulability. up to 1 in 20 patients presenting for routine surgery
Considerable evidence suggests that, in addition to can be expected to have a degree of heightened risk
serving as a cofactor for APC, PS may generate additional, attributable to this gene. A very small minority of
direct anticoagulant activity. Protein S acts to inhibit the patients who demonstrate APC resistance do not carry
development of the prothrombinase complex by inhibiting the gene for FVLeiden , but their VTE risk is similar to
Xa in the absence of Va, and by inhibiting the binding that for heterozygous FVLeiden carriers, and therefore, for
of the substrate, prothrombin, to factor Va.13 Therefore, the purposes of perioperative risk, they may be treated
decreases in PS, in addition to limiting the effectiveness of identically. Currently, however, testing for APC resistance
APC, may also allow the generation of an overabundance is not indicated for routine preoperative screening in the
of the rate-limiting FXa. absence of a VTE history.
With respect to the regulatory protein, TFPI, no Another thrombophilia that operates through an
congenital or acquired deficiencies have been described increase in prothrombotic proteins is known as the
that appear to predispose to VTE at this time. One prothrombin gene mutation (G20210A). This gene was
investigation,14 however, suggests that among patients described by Poort et al. in 1996,18 noting that 18% of
with VTE of unknown etiology, the average response to VTE patients and approximately 1% of healthy controls
recombinant TFPI is reduced compared to controls, and had a mutation in the gene for prothrombin at base
individuals whose TFPI sensitivity was below the 10th 20210. This particular location is in the 3-prime region
percentile of controls had 13 times greater likelihood of a of the gene that is not translated but contains the stop
VTE. Further investigation is required to substantiate this condon. The mutation renders the end cleavage signal
observation and determine whether a reduced response to of the gene inefficient, causing additional amounts of
TFPI is attributable to alterations in FVII, FX, or another mRNA to be transcribed. Accordingly, the levels of the
pathophysiology. inactive zymogen, prothrombin, are considerably higher
in affected individuals than in the general population.
Unlike FVLeiden , which enhances levels of the active en-
zyme, the prothrombin gene mutation increases the levels
THROMBOPHILIA DUE of substrate for the enzyme complex. When this muta-
TO INCREASED tion is the only thrombophilic risk factor, the VTE risk
PROTHROMBOTIC PROTEINS is relatively low19 (Table 38.1); most carriers of this gene
will not have had an episode of VTE before age 50. The
A number of thrombophilias have been described result- importance of this thrombophilia, as for FVLeiden , resides
ing from increased levels of prothrombotic proteins. Two in the frequency of the gene, rather than its potency. Also
thrombophilias deserve particular notice because of their similar to FVLeiden , ethnicity plays a significant role in the
relatively high prevalence. Dahlback first described APC prevalence of this gene, occurring in approximately 4% of
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 559
individuals of European descent, but rarely in patients of variable, and may only become apparent with coexisting
African or Asian descent.20 dietary deficiencies of folate, vitamins B12 , and B6 . The
After excluding the thrombophilias described in the exact mechanism by which hyperhomocysteinemia pre-
preceding text, patients with VTE are more likely to have disposes to both atherosclerosis and VTE is not certain,
increased concentrations of zymogens for factors VIII, IX, but evidence points toward endothelial cell injury, perhaps
and XI compared with healthy controls.2123 Other than mediated by oxygen radicals and predisposing to plaque
theoretically providing a surplus substrate to optimize en- buildup in arterial vessels and impaired antithrombotic
zyme kinetics, it is not known exactly why elevated levels activity in their venous counterparts. Such risks for vascu-
predispose to thrombosis or what causes their plasma lar injury are certainly magnified in patients homozygous
factor levels to be higher. Genetic mutations that have for the cysteine -synthase mutation, but it is currently
yet to be elucidated may be present, possibly resulting in much less clear that mild hyperhomocysteinemia repre-
either elevated zymogen production or prolonged plasma sents another hereditary thrombophilia.
survival. The degree of risk enhancement for these in-
creased levels is low to moderate. One study23 estimated
that VTE risk increased 10% for each 100% FVIII in-
crement, and another study21 demonstrated that patients HYPERCOAGULABILITY
with FXI levels above the 90th percentile had more than a FROM THROMBOPHILIC
twofold increased VTE risk. Van Hylckama et al.22 demon-
strated a twofold to threefold increased risk for patients
COMBINATIONS
with higher FIX levels. These risks are comparable to Many VTE patients are found to have more than one
the risks for the heterozygous FVLeiden or prothrombin risk factor, whether carrying the genes for two congenital
20210A thrombophilias. thrombophilias or a combination of a congenital and ac-
quired thrombophilia. The coinheritance of FVLeiden and
prothrombin G20210A gene mutations doubles the risk
HYPERCOAGULABILITY of recurrent VTE, compared with carriers of the FVLeiden
gene mutation alone.25 The risk of symptomatic VTE
OF UNCLEAR ETIOLOGY among FVLeiden patients undergoing total hip replace-
Hyperhomocysteinemia has been identified as a risk fac- ment is increased fivefold compared to noncarriers.26
tor for both accelerated atherosclerosis and VTE, although Wahlander et al. demonstrated that the prothrombin
the magnitude of the latter is unclear at present.24 Ho- G20210A mutation had a ninefold increase in symp-
mocysteine is a thiol-containing amino acid that readily tomatic VTE after hip/knee replacement surgery despite
undergoes auto-oxidation in the plasma, forming the ox- 8 to 11 days of antithrombotic prophylaxis.27 For some
idized dimer, homocysteine, which participates in redox thrombophilias, the risk of surgery appears to synergize
reactions that produce oxygen radicals. At physiologic ho- with a baseline thrombophilic risk. Antithrombin defi-
mocysteine levels, endothelial cell-released nitrous oxide cient patients exposed to surgery have a VTE risk of
binds to homocysteine, forming S-nitrosothiol, thereby approximately 12%, compared to only 0.8% per year for
inactivating it. It is hypothesized that at supraphysiologic spontaneous VTE in AT-deficient patients and 1.2% per
homocysteine levels, nitrous oxidemediated regulation is year for VTE risk attributable to the surgery alone.11
overwhelmed, and homocysteine-derived oxygen radicals The risk of pregnancy-related VTE is also increased by
can cause local tissue injury. Hyperhomocysteinemia can the presence of a thrombophilia (see Table 38.2). Most
be found in patients with dietary deficiencies of folate and studies have found the VTE risk to be higher in the post-
B vitamins, and can also be hereditary, most commonly partum period, compared to antepartum, and the risk is
due to mutations affecting the cysteine -synthase gene higher still after cesarean section than after a vaginal de-
or the methyltetrahydrofolate reductase gene. The homo- livery, including a 10-fold higher risk of fatal pulmonary
cysteine levels associated with these mutations are highly embolism.28
TABLE 38.2 Enhanced Venous Thromboembolism Risk during Pregnancy in the Thrombophilic Parturient
Antithrombin Protein C Protein S Factor VLeiden or Factor VLeiden or
Deficiency Deficiency Deficiency Prothrombin Mutation Prothrombin Mutation
Criteria (%) (%) (%) (Homozygous) (%) (Heterozygous) (%)
No VTE history 1020 1020 <3 <3 <3
VTE in one or more 1020 1020 1020 1020 310
first-degree relative
Personal VTE history >20 1020 1020 1020 310
TABLE 38.3 Epidemiology of Venous Thromboembolism that having detectable plasma levels of IL-6, IL-8, and
(VTE) TNF-all of which are cytokines that are frequently
elevated postoperativelydoubled the risk of VTE. For
Independent Risk Factorsa Odds Ratiob IL-8 in particular, the degree of risk increased in
proportion to its levels. Cytokines IL-8 and IL-6, in
No institutionalization or recent surgery 1.00 addition to being indices of the degree of tissue damage,
Pregnancy and first 68 weeks 56 may be active modulators of coagulant activity. In vitro
postpartum exposure of monocytes to these cytokines causes an
Malignancy with chemotherapy 67 increase in monocyte TF and an associated 4.5-fold
Institutionalization without surgery 78 increase in procoagulant activity.33
Institutionalization with recent surgery 2022
Different surgical procedures in particular show
a Without
marked variability in their VTE risk, some of which may
antithrombotic prophylaxis.
b Adapted stem from factors ancillary to the surgery itself (e.g.,
from: Heit JA, Silverstein MD, Mohr DN, et al. The
epidemiology of venous thromboembolism in the community. patients age and requirements for perioperative immo-
Thromb Haemost. 2001; 86:452. bility). However, as shown in Table 38.4, multiple trauma
and orthopedic surgery carry a VTE risk at least twice
that of general surgery. These procedures are associated
with extensive muscle and bone damage. Studies have
indicated that an increased release of local TF locally may
create heightened coagulation activation and lead to a
What Conditions Lead to hypercoagulable state. Postoperative serum levels of the
thrombin-AT complex (evidence of ongoing coagulation
Acquired or Environmental activation) were elevated sixfold at 3 weeks and 2.5-fold
Hypercoagulability? at 5 weeks following hip surgery.34 This persistence of a
hypercoagulable state late after orthopedic surgery, which
correlates with a high incidence of postoperative VTE, has
prompted some type of VTE prophylaxis after discharge
SURGERY in selected patients.35
As noted in the preceding text, the specific clinical
setting is a major contributor to overall thromboembolic
risk. Table 38.3 lists the relative risks in some general PREGNANCY
clinical conditions associated with hypercoagulability,
but there is considerable variability within these broad Normal pregnancy creates a state of hypercoagulability
categories. Surgery, for all indications and all ages, stemming from increased levels of clotting factors, de-
increases the VTE risk by a factor of 7 to 8.2931 The creased PS levels, and an acquired APC resistance.36
exact etiology of this increase in postoperative deep vein The risk of DVT increases five- to six-fold during preg-
thrombosis (DVT) risk is unclear. Immobility and its nancy, and increases three- to ten-fold further in the
attendant venous stasis are unquestionably significant immediate postpartum period. Despite this increase in
factors. The inflammatory response to surgery may also DVT risk, routine thromboprophylaxis does not appear
contribute to a hypercoagulable state. As part of the to be indicated after elective cesarean section in the ab-
Leiden Thrombophilia Study, Reitsma and Rosendaal32 sence of additional risk factors.37 However, the DVT risk
demonstrated in patients presenting with their first VTE associated with pregnancy is markedly exacerbated by
From: Caprini JA, Arcelus JI, Reyna JJ. Effective risk stratification of surgical and nonsurgical patients for venous
thromboembolic disease. Semin Hematol. 2001;38(Suppl 5):12.
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 561
heritable thrombophilic disorders,38 and, indeed, illus- TABLE 38.5 Thromboembolic Risk Stratification
trates how a hitherto occult hereditary hypercoagulability for Surgery Patients
may become symptomatic when triggered by the devel-
opment of an acquired hypercoagulability (Table 38.3). Low Risk Uncomplicated surgery in patients aged
Some type of genetic hypercoagulability is identified in <40 y with minimal immobility
more than 50% of women with pregnancy-induced DVT. postoperatively and no risk factorsa
In addition to causing maternal morbidity, hereditary Moderate Any surgery in patients aged 4060 y,
thrombophilia increases the risk of recurrent pregnancy Risk major surgery in patients <40 years and
loss approximately three- to eightfold. Similarly, the no other risk factors,a minor surgery in
prevalence of thrombophilic disorders is two- to three- patients with one or more risk factorsa
fold higher in patients exhibiting complications of preg- High Risk Surgery in patients aged >60 y, major
nancy such as preeclampsia, although it is not presently surgery in patients aged 4060 y with
clear whether hypercoagulability is an etiologic factor one or more risk factorsa
in this disorder or simply aggravates the severity of Very High Major surgery in patients aged >40 y with
pregnancy-induced hypertension initiated by some other Risk previous venous thromboembolism,
mechanism.39 cancer or known hypercoagulable state,
major orthopedic surgery, elective
neurosurgery, multiple trauma, or acute
spinal cord injury
CANCER
a Riskfactors include hereditary thrombophilia, previous venous
The association between malignancy and venous throm-
bosis has been recognized for more than 100 years; thromboembolism, obesity, varices, and estrogen use.
From: Gutt CN, Oniu T, Wolkener F, et al. Prophylaxis and treatment
however, this pathophysiologyand even its true inci-
of deep vein thrombosis in general surgery. Am J Surg. 2004;189:14.
denceare still unclear.40 Autopsy data demonstrates the
presence of VTE in up to 50% of cancer patients, but
the prevalence of clinically apparent VTE in premorbid Prophylaxis strategies may take the form of phar-
cancer patients ranges from 5% to 60%, depending on the macologic or physical methods. Drugs that have proved
malignancy.41 Large retrospective studies suggest that for to be suitable for VTE prophylaxis include unfractionated
men, pancreatic and lung cancer have the greatest inci- heparin (UH), low molecular weight heparin (LMWH), the
dence of associated VTE, whereas in women, gynecologic, oral anticoagulant warfarin, direct thrombin inhibitors
pancreatic, and colorectal cancers are most commonly such as hirudin, and factor Xa inhibitors such as fonda-
associated with VTE. Certain types of chemotherapy, parinux. Large trials suggest that subcutaneous admin-
particularly when combining chemotherapeutic agents istration of UH or LMWH confers a 60% to 70% risk
with hormonal therapy or radiation treatments, increase reduction over placebo, depending on the type of surgery.
VTE risk. Cancer patients undergoing surgery have two- By contrast, aspirin provides relatively weak prophylaxis,
to threefold increased VTE risk compared to patients with a risk reduction of only 20% compared to placebo.
without malignancy who are undergoing comparable Physical methods of prophylaxis such as graded compres-
procedures.42 sion elastic stockings have a 40% to 45% risk reduction,
whereas intermittent pneumatic compression shows a risk
reduction that approaches that of UH when used as the
only prophylactic method.31 The recommendations of the
What Are the Strategies American College of Chest Physicians as adapted from
for Prevention in Patients at Turpie et al.43 are shown in Table 38.6. The management
of patients presenting for surgery who are already on oral
Risk for Thrombosis? anticoagulants will be discussed in later sections.
UH, unfractionated heparin; IU, international units; LMWH, low molecular weight heparin; INR, international normalized ratio; aPTT, activated
partial thromboplastin time.
Adapted from: Turpie AGG, Chin BSP, Lip GLH. Venous thromboembolism: Pathophysiology, clinical features, and prevention. The ABCs of
antithrombotic therapy. Br Med J. 2002;325:887.
regional anesthesia became the preferred anesthetic tech- use did not confer any lower VTE risk (p = 0.98). This
nique for this surgery and other procedures with high study did not control for pharmacologic thromboembolic
VTE risk. However, even when neuraxial anesthesia was prophylaxis and found that, in particular, the combination
combined with techniques, such as early ambulation and of pneumatic compression and warfarin prophylaxis was
intraoperative antiembolism stockings, the VTE risk was associated with a highly significant reduction in VTE risk
still unacceptably high. Postoperative prophylactic antico- (p <0.001).
agulation with drugs such as warfarin and subcutaneous Indeed, a recent meta-analysis of anesthesia for hip
heparin became the standard of care for these high-risk fracture surgery by the Cochrane Database of System-
surgeries. atic Reviews48 found that regional and general anesthesia
With the advent of routine antithrombotic prophy- appeared to produce comparable results for most out-
laxis, however, the advantages of regional over general comes studied. Seventeen trials were included, and only
anesthesia are now less clear, questioning whether neu- four included some form of pharmacologic or mechanical
raxial anesthesia still reduces the risks of VTE in patients antithrombotic prophylaxis. Accordingly, although there
receiving pharmacologic perioperative thromboprophy- was a slight reduction in VTE incidence associated with
laxis. In a randomized trial of epidural anesthesia versus regional anesthesia, it was largely a function of older
general anesthesia in 178 patients, where thromboem- studies lacking in pharmacologic prophylaxis, which did
bolic prophylaxis constituted graded elastic stockings and not translate into a significant difference in mortality.
aspirin beginning on postoperative day 1, there was a The recent U.S. Food and Drug Administration (FDA)
44% overall incidence of VTE, with no significant differ- advisory prohibiting neuraxial anesthesia in patients re-
ence between epidural and general anesthesia.46 A larger, ceiving LMWH due to increased epidural hematoma risk
retrospective study compared 297 patients who experi- may further limit the extension of regional anesthesia
enced symptomatic VTE after total hip arthroplasty to into the postoperative period. Furthermore, there is no
592 hip arthroplasty patients without VTE to identify evidence that the VTE risk reduction provided by re-
risk factors;47 regional anesthesia was not more com- gional anesthesia obviates the need for postoperative
mon among asymptomatic patients, suggesting that its pharmacologic prophylaxis. In conclusion, no particular
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 563
PLATELET ADHESION
What Are the Limitations Although the platelet plays an important role in control-
ling venous bleeding through the optimization of enzyme
of Laboratory Testing kinetics, it is the setting of arterial hemorrhage where
in Predicting Thrombotic Risk? platelet number and function become vital. The greater
velocity of arterial flow, particularly through smaller di-
ameter vessels, creates shear forces that mandate the
Laboratory tests that could predict thrombotic risk, by
binding of coagulation enzymes to the activated platelet
reserving postoperative anticoagulation with its attendant
surface. Yet, even the activated platelet itself must have a
bleeding risks only for high-risk patients, would be far
means to rapidly and reliably bind to the damaged blood
preferable to our current practice of using epidemiologic
vessel before being swept downstream by the high veloc-
data to predict thrombotic risk. Even postoperative
ity blood flow. Remarkably, the platelet uses these same,
tests capable of quantifying hypercoagulability, thereby
very high shear forces of arterial flow to foster platelet
permitting adjustments in anticoagulant dosing, would
binding to the bleeding site; essential to this process is the
be of enormous utility. Unfortunately, the multitude of
multimeric protein, vWF.51
procoagulant and anticoagulant pathways, and the fact
The base unit of vWF is synthesized by endothelial
that many of them are endothelial cell-associated, makes
cells as a polypeptide of 2813 amino acids. These base
our current whole blood and plasma testing modalities
units are then joined by disulfide bonds into dimers
less sensitive. The development of a reliable laboratory
within the endoplasmic reticulum. These dimers are
hypercoagulability measurement is still a challenge for
linked through successive N-terminal disulfide bonds into
the future.49,50
multimers, ranging in mass from 500 to 20,000 kDa.
This multimeric nature of vWF is critical to its ability to
initiate platelet adhesion.52 Some large vWF multimers
are released in a constitutive fashion into the plasma.
What Events Contribute Some are released in an abluminal direction into the vessel
to Arterial Hypercoagulability, adventitia, and the remainder are stored in endothelial cell
granules known as Weibel-Palade bodies. In plasma, vWF
and What Conditions Create It? circulates as a loosely coiled molecule with an apparent
diameter of 200 to 300 nm, showing no affinity for
Arterial occlusive disease typically takes two forms with cocirculating platelets. Disruption of the endothelial cell
distinct time courses. The first form consists of long-term layer exposes subendothelial collagen that either binds to
atheromatous plaque formation that may occur through- plasma vWF or has vWF prebound. Under the influence
out adult life, culminating in fixed, obstructive lesions of the high shear forces present along the vessel wall
that threaten oxygen delivery to tissues, particularly un- (a threshold value of 1,000 per second is required), this
der circumstances of increased oxygen demand. Risks tethered vWF uncoils to lengths as great as 1300 nm. The
for this type of arterial occlusive disease include genetic uncoiling of vWF is thought to expose a hitherto cryptic
and lifestyle factors beyond the scope of this review. The domain in the A1 region of the molecule, a domain with
second form of arterial occlusive disease is the acute for- instantaneous affinity for the platelet surface receptor,
mation of an arterial thrombus. This may occur at the site glycoprotein (GP)Ib. The coupling of this vWF domain to
of atheromatous disease precipitated by plaque rupture, GPIb is characterized by a high rate of bond formation.
or may develop in vessels free of atherosclerosis, usually in This fast-on binding briefly tethers the platelet to the
association with injury to the endothelium, vasospasm, or exposed subendothelium in the face of the high velocity
both. The potential for hypercoagulability to predispose blood flow. However, this adhesive bond is relatively
to acute arterial thrombosis is the focus of a growing body weak, and is soon overcome by shear that moves the
of evidence that, such as its venous counterpart, is criti- platelet downstream, albeit at a much slower velocity and,
cally dependent on improved understanding of the factors importantly, in proximity to the vessel wall.53 In addition
affecting both normal and pathologic arterial clot forma- to slowing the platelets velocity, the brief vWFGPIb
tion. In addition, the pharmacologic agents used to treat interaction causes a transmembrane-signaling event that
arterial hypercoagulability require a basic understand- activates the platelet.
ing of arterial hemostasis for their safe use. Accordingly, Platelet activation induced by vWFGPIb interac-
the following section will begin with an overview of the tion causes the release of intracellular granules and
events leading to arterial thrombosis, followed by some of conformational changes in a second platelet receptor,
the heritable and acquired conditions that create arterial GPIIb/IIIa. The now-activated GPIIb/IIIa is capable of
hypercoagulability. binding the more slowly moving platelet to a different
564 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
domain on vWF. Unlike the initial vWFGPIb interac- A second regulatory mechanism limiting platelet plug
tion, however, the GPIIb/IIIavWF bond has sufficient growth operates by limiting the availability of the larger
tenacity to resist local shear, and the platelet is arrested vWF multimers that are critical to platelet adhesion at
on the vWF surface.1 The larger vWF multimers are most high shear. A recently identified protein known as vWF
successful at this sequence of events, often permitting cleaving protease (vWF-CP), or ADAMTS-13, enzymatically
both platelet-binding steps to occur on a single vWF mul- reduces the larger vWF multimers into smaller, less
timer. This sequence of events is repeated several times hemostatically effective forms.57 Synthesized by the liver,
until the exposed subendothelium is covered by a mono- this protease circulates in normal blood flow without any
layer of activated platelets. Patients deficient in either vWF affinity for cocirculating vWF. However, as the platelet
(von Willebrand disease [vWD]) or the platelet receptors, plug grows and progressively narrows the arterial lumen,
GPIb, (Bernard-Soulier syndrome) and GP IIb/IIIa (Glanz- the shear rate increases. For the developing platelet
mans thrombasthenia),54 demonstrate defective platelet plug, as the shear increases, so does its dependency
adhesion to exposed subendothelium at shear rates in the on the largest vWF multimers for continued growth. If
range of 1,500 per second, rates typical for arterioles and unchecked, the sequence of platelet slowing platelet
stenotic vessels, all three associated with perioperative activation platelet adhesion that permitted the initial
bleeding. monolayer to form in the face of high shear would be
continuously repeated at the surface of the platelet plug,
finally threatening the vessel lumen. However, vWF-CP
becomes enzymatically active at higher shear and thereby
PLATELET AGGREGATION limits the availability of ultralarge vWF multimers.58 The
Once the exposed subendothelium is covered by a platelet more intermediate-sized vWF multimers, generated by the
monolayer, the next step is extension of the platelet plug activity of this enzyme, do not support platelet adhesion
by recruiting additional platelets. This is dependent on at very high shear, thereby limiting growth of the platelet
stimulation of passing platelets by agonists released from plug short of vessel occlusion.
The importance of this mechanism for limiting arterial
the activated platelets in the monolayer,55 including ADP
thrombosis is illustrated by thrombotic thrombocytopenic
and serotonin, both released from dense granules, and
purpura (TTP), a syndrome characterized by the loss of
thromboxane A2 released from the platelet cytosol. As
vWF-CP activity resulting in accumulation of ultralarge
these second-line platelets are activated, their GPIIb/IIIa
vWF multimers in plasma, consumption of platelets, and
also undergoes the conformational change that allows
arterial thrombosis. Therefore, a deficiency of this critical
it to bind a bridging ligand, in this case either vWF or
regulator of platelet plug growth allows the development
fibrinogen, already bound to GPIIb/IIIa on an adherent
of multiple thrombi to occur in vessels characterized by
platelet. Therefore, a plateletligandplatelet matrix is
high shear. In the absence of treatment, the mortality rate
formed at the monolayer surface, with either vWF or
from this disorder approaches 80%. Therapy typically in-
fibrinogen serving as the bridging ligand.
cludes transfusion of plasma to restore levels of vWF-CP.
TTP can be either hereditary or acquired. Unlike patients
with idiopathic thrombocytopenic purpura presenting for
REGULATION OF ARTERIAL splenectomy, those with TTP rarely come to the operating
room. Also, unlike idiopathic thrombocytopenic purpura,
THROMBOSIS AND ONE platelets in TTP contribute to this vasculopathy, and there-
TYPE OF ARTERIAL fore platelet transfusions must be avoided, except in cases
HYPERCOAGULABILITY of life-threatening blood loss.
Heritable Hypercoagulability
of the Arterial Vasculature Do Heritable Causes of Venous
Hypercoagulability Increase the
Genes that affect platelet adhesive proteins are cur-
rently the best candidates for heritable sources of arterial
Risk for Arterial Thrombosis?
hypercoagulability. One of these, the Kozak polymor-
phism, is a substitution of a cytosine for a thymine A logical perioperative concern for anesthesiologists car-
near the start codon for the gene encoding GPIb, part ing for a patient with venous hypercoagulability, par-
of the GPIB-IX-V complex that binds the platelet to ex- ticularly hereditary hypercoagulability, is whether that
posed subendothelial vWF. This polymorphism, present condition also increases arterial hypercoagulabilityin
in approximately 8% of whites and 17% of Japanese, particular, is the patient at increased risk of a periop-
influences messenger RNA translation and results in erative coronary or cerebral thrombosis? Cardiologists
increased expression of GPIb on the platelet surface. evaluated the prevalence of heritable hypercoagulable dis-
The Kozak polymorphism appears to increase the like- orders such as FVLeiden in patients presenting with acute
lihood of ischemic stroke, independent of other risk myocardial infarction. Meta-analysis64 suggests that, al-
factors.59 though there may be a detectable increase in risk of
Another gene affecting platelet function is the PLA coronary thrombi associated with this gene in subsets of
allele that encodes for the GPIIIa part of the GPIIb/IIIa patients such as smokers, it is relatively modest in com-
complex. Weiss et al.60 first demonstrated that the PLA2 parison to other established risk factors such as diabetes
polymorphism was more prevalent among patients with and hypercholesterolemia.
coronary thrombosis. Carter et al.61 similarly demon-
strated an increased incidence among patients with is-
chemic stroke, and our own laboratory has shown an
association between the PLA2 genotype and enhanced How Are Anticoagulated
neurocognitive decline post cardiopulmonary bypass.62
Patients Managed Before
Surgery?
The Postoperative Period: The perioperative management of patients receiving long-
A State of Acquired Arterial term anticoagulation requires special consideration of the
risks of bleeding and thrombosis.65 The risk of thrombosis
Hypercoagulability? when the preoperative patient is not effectively antico-
agulated must be weighed against the risk of bleeding
The stress of the perioperative period predisposes to during and after surgery if anticoagulation is continued
thrombosis of coronary arteries for a number of rea- perioperatively. Details of the thrombosis that warranted
sons, including catecholamine-induced increased myocar- anticoagulation, that is, the inciting thrombus, are of
dial oxygen demand, tachycardia-associated reduced oxy- primary importance. The risks associated with recurrence
gen delivery, and other local factors that predispose to of thrombosis are greatest if the inciting thrombus was ar-
plaque rupture. Whether a state of systemic arterial hy- terial, especially if associated with atrial fibrillation where
percoagulability also exists in the perioperative period recurrent embolism carries a 40% mortality. In contrast,
and contributes to such thrombotic events is arguable. recurrent VTE has a risk of associated sudden death of
If the perioperative period produced a truly systemic 6%. In addition, the time elapsed since the inciting throm-
state of arterial hypercoagulability, one would expect bus is also critical, as the risk of recurrence decreases over
other organs to be affected; proof that this particu- time for both arterial and venous thrombi.
lar clinical scenario is associated with arterial hyper- Most anticoagulated patients are managed on war-
coagulability requires very large epidemiologic studies. farin, an anticoagulation that gradually abates after
One such study of approximately 5,000 postcoronary stopping the drug.29 After discontinuing warfarin, the
artery bypass graft (CABG) surgery patients was per- international normalized ratio (INR) does not start to fall
formed by the Multicenter Study of Perioperative Is- for approximately 29 hours, and then decreases with a
chemia (McSPI) investigators.63 They found that the half-life of approximately 22 hours. If a patient is con-
approximately 3,000 patients who were restarted on as- sidered to be at high risk without anticoagulation, then
pirin within the first 48 hours postoperatively had one bridging therapy in the form of therapeutic doses of UH
third the risk of dying as those patients who did not or LMWH should be considered approximately 60 hours
receive aspirin. Indeed, for multiple systemscerebral, after the last dose of warfarin. In the case of intravenous
renal, and cardiacthe incidence of poor outcomes was UH, a window of 6 hours drug-free should be allowed
reduced by early aspirin resumption. Of course, these before surgery. For LMWH, which may be given subcuta-
results only provide indirect evidence that antiplatelet neously as an outpatient, doses should be given once or
medication is protective by reducing arterial hypercoagu- twice daily for 3 days before surgery, with the last dose
lability. no less than 18 hours preoperatively for a twice-daily dose
566 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 38.7 Rates of Thromboembolism Associated with Different Indications for Oral Anticoagulation
Thromboembolism Rate without Risk Reduction Achieved
Indication Anticoagulation by Anticoagulation (%)
VENOUS THROMBOEMBOLISM
Acute venous thromboembolism
within last month 40%/moa 80
13 mo earlier 10%/2 moa 90
Recurrent venous thromboembolismb 15%/ya 90
ARTERIAL THROMBOEMBOLISM
Nonvalvular AF 4.5%/y 75
Nonvalvular AF + previous embolism 12%/y 75
Mechanical heart valve 8%/y 75
Acute arterial embolism
Within last month 15%/mo 75
a The increased risk associated with surgery, estimated at approximately 100-fold, is not included in these rates.
b Lastepisode of venous thromboembolism >3 mo previously, but requires long-term anticoagulation because other factors suggest high risk
of recurrence.
AF, atrial fibrillation.
Reproduced with permission from: Schafer AI, Levine MN, Konkle BA, et al. Thrombotic disorders: Diagnosis and treatment.
Hematology. 2003;1:520.
(i.e., approximately 100 U per kg of LMWH) and 30 hours several days of exposure to UH, or less commonly
for a once-daily regimen (i.e., approximately 150 to 200 U to LMWH.66 The decrease in platelet count with HIT
per kg of LMWH). An additional 6-hour, drug-free inter- is generally moderate and very rarely causes clinical
val should be allowed if neuraxial anesthesia is planned. bleeding. Far more critical, however, is the associated
Table 38.7 shows recommendations for anticoagulation hypercoagulability that develops in a subset of HIT
stratified by time elapsed and location of patients initial patients in whom thrombotic complications may pro-
thrombus. duce limb amputation, pulmonary embolism, or death.
Postoperative resumption of anticoagulation requires The paradoxical potential for life- and limb-threatening
an evaluation of the risk of recurrent thrombosis and arterial and venous thrombosissimultaneous with
consideration of the degree to which surgery itself thrombocytopenia and heparin anticoagulationmakes
increases the patients hypercoagulability (e.g., minor this syndrome a unique conceptual and management
surgery versus major orthopedic surgery). This must dilemma.
be weighed against the bleeding risk associated with Between 1% to 5% of heparin-treated patients exhibit
resumption of anticoagulation. Because there is a delay a mild decrease in platelet count after 5 to 10 days of
of approximately 24 hours after warfarin administration anticoagulation, with a 50% decrease in platelet count,
before the INR increases, warfarin should generally be the generally accepted criterion for a diagnosis of HIT.
resumed as soon as possible after surgery, except in Among patients who develop HIT, as many as 30% to 50%
cases at high bleeding risk; consideration can be given will suffer venous or arterial thromboembolism; this en-
to bridging therapy with intravenous or subcutaneous tity is termed, HIT with thrombosis (HITT), and carries an
anticoagulation until the INR becomes therapeutic. associated amputation rate of 10%, a stroke rate of 10%,
and mortality of 15% to 20%.67 In a retrospective study
of more than 400 HIT patients, those with postorthopedic
surgery and trauma had the highest risk for develop-
What Are the Less Common ing VTE, and postcardiovascular surgery patients were
Causes of Hypercoagulability, highest for arterial thrombi.68 No matter how low the
platelet count, the HIT patient is generally at greatest risk
and How Are They Managed? from thrombotic complications rather than thrombocy-
topenic bleeding. The relatively low frequency and the
unpredictability of the HIT/HITT clinical course have hin-
HEPARIN-INDUCED dered prospective studies regarding its pathophysiology
THROMBOCYTOPENIA/ and management. Much of our current understanding
THROMBOSIS of perioperative HIT/HITT is gleaned from case reports
and small retrospective studies, so management requires
Heparin-induced thrombocytopenia (HIT) is an im- a basic understanding of the pathophysiology of this
mune response to heparin that can develop after disorder.
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 567
in which thrombosis is associated with antiphospholipid to enzymatic cleavage. Accordingly, the most hemostati-
antibodies.74 Antiphospholipid antibodies can be catego- cally effective larger multimers become depleted, creating
rized as anticardiolipin (aCL) antibodies or lupus anti- a risk of bleeding. Therefore, aggressive phlebotomy low-
coagulants (LAC). Studies suggest that aCL antibodies ers the risk of both thrombosis and hemorrhage in the
actually bind to 2 -GPI when it forms a complex with car- nonsurgical PV patient and, in theory, should be similarly
diolipin.75 LAC bind to phospholipids used in coagulation protective in the perioperative period.
tests, causing prolongation of the aPTT and/or the dilute
Russells viper venom time. This interference is strictly a
laboratory phenomenon and is not associated with any in-
creased risk of bleeding. Instead, both LAC and aCL may
SUMMARY
create the hypercoagulable state known as APS. APS may Hypercoagulability, a state of exaggerated coagulation ac-
play a role in 15% to 20% of all episodes of DVT, as well as tivation, plays a major role in the pathogenesis of VTE, a
in one third of new strokes in patients younger than age 50 process that affects approximately 2 million Americans
and 5% to 15% of women with recurrent miscarriages.76 annually, with an estimated mortality of 150,000 at-
No tests reliably identify the patients at greatest risk of tributable to pulmonary embolism. New heritable causes
thrombosis. Individuals with either LAC, or aCL com- of hypercoagulability are being identified, and some ge-
bined with other heritable or acquired VTE risk factors, netic predisposition to thrombosis can be identified in
appear to be at greatly increased risk for venous, but not more than 50% of DVT patients. Accordingly, anesthesi-
arterial, thrombotic events. Patients with APS presenting ologists are being asked to care for an increasing number
for surgery pose a considerable challenge for anticoagula- of patients carrying the diagnosis of hypercoagulability,
tion.77 The standard of care in APS is long-term, high-dose many of whom are chronically anticoagulated. The peri-
anticoagulation. Time spent without anticoagulation must operative period represents a high-risk time for VTE, with
be kept to a minimum so that the interval between war- selected surgeries associated with a greater than 100-fold
farin withdrawal and reestablishment should be bridged increased risk of thrombosis. Our knowledge of the opti-
with short-acting anticoagulants such as UH/LMWH. mum operative management of these patients inevitably
However, thrombosis may supervene despite optimal an- lags behind the identification of their pathophysiology,
ticoagulation, and can even progress to multiple organ making it incumbent on anesthesiologists to understand
thrombosis, the so-called catastrophic antiphospholipid the mechanisms behind hypercoagulability and, thereby,
syndrome (CAPS),78 with an exceptionally high mortality. make educated choices. Hypercoagulability plays a less
The risk of perioperative VTE in LAC- or aCL-positive clearly defined role in the pathophysiology of arterial
patients who are asymptomatic, in whom antibody detec- thrombotic events, but the high morbidity and mortality
tion was an incidental finding, has not been adequately associated with arterial occlusive events in the operative
studied to enable any definitive recommendations. patient makes staying abreast of these developments vital
to patient care.
POLYCYTHEMIA VERA
Polycythemia vera (PV) is a clonal hematopoietic disor- KEY POINTS
der that results in an increase in red cells, white cells,
and platelets, and is associated with an increased risk 1. Hypercoagulability may be considered a state of exag-
of thrombosis.79 Phlebotomy remains the mainstay of gerated activation of coagulation. Sources of hyper-
treatment for PV and, in the prephlebotomy era, throm- coagulability may be divided into two major classes:
bosis was the major cause of death, with a median life (a) congenital, often referred to as thrombophilia,
expectancy of <2 years. Whole blood viscosity increases (b) acquired, or environmental hypercoagulability.
exponentially with rising hematocrit, particularly in ves- 2. Genetic sources of hypercoagulability are lifelong
sels with relatively low shear rates. Phlebotomy (<45% conditions and only rarely are the sole contributor to
for men and <42% for women) substantially reduces, but thrombosis. In most cases of thrombosis, acquired or
does not eliminate, the risk of thrombosis in PV patients. environmental hypercoagulability serves as a trigger-
Low-dose aspirin is often recommended after phlebotomy. ing event in a patient with a thrombophilia; indeed,
Surgery presents a particularly high-risk period for PV some thrombophilia can presently be identified in
patients, as they may develop either thrombosis or bleed- more than 50% of patients with VTE.
ing complications. The increased risk of thrombosis is 3. The process leading to normal clot formation in the
the predictable combination of the PV patients baseline venous and arterial systems is not identical, and its
hypercoagulability augmented by the approximately 100- regulation is, to a degree, also distinct in the two
fold increase associated with surgery. The etiology of the types of circulation. Accordingly, risk factors that
bleeding diathesis associated with PV is often attributable predispose to thrombosis in the venous circulation
to an acquired vWD, caused by abnormally low amounts of may not necessarily predispose to arterial thrombosis,
the ultralarge vWF multimers essential to normal platelet and vice versa.
adhesion (see section on Arterial Hemostasis). The hy- 4. As a general rule, alterations affecting the speed
perviscosity associated with a high hematocrit favors the and amount of thrombin generated during clotting
conformational change in vWF that renders it vulnerable typically predispose to VTE, whereas abnormalities
C H A P T E R 3 8 / T H E H Y P E R C O A G U L A B L E P AT I E N T 569
of platelet function have a greater effect on arterial 18. Poort SR, Rosendaal FR, Reitsma PH, et al. A common
genetic variation in the 3- untranslated region of the pro-
thrombosis. The most striking exceptions to this
thrombin gene is associated with elevated plasma prothrom-
generality are two syndromes that include an element bin levels and an increase in venous thrombosis. Blood. 1996;
of immune dysfunction, specifically heparin-induced 88:3698.
thrombocytopenia and APS, where patients are at risk 19. Crowther MA, Kelton JG. Congenital thrombophilic states as-
for both arterial and venous thrombosis. sociated with venous thrombosis: A qualitative overview and
proposed classification system. Ann Intern Med. 2003;138:
REFERENCES 128.
20. Rosendaal F, Doggen CJ, Zivelin A, et al. Geographic
1. Ruggeri ZA, Dent JA, Saldivar E. Contribution of distinct distribution of the 20210 G to A prothrombin variant. Thromb
adhesive interactions to platelet aggregation in whole blood. Haemost. 1998;79:706.
Blood. 1999;94:172. 21. Meijers JC, Tekelenbur WL, Bouma BN, et al. High levels of
2. Davie EW, Ratnoff O. Waterfall sequence for intrinsic blood coagulation factor XI as a risk factor for venous thrombosis.
clotting. Science. 1964;145:1310. N Engl J Med. 2000;342:696.
3. MacFarlane R. An enzyme cascade in the blood clotting 22. van Hylckama VA, van der Linden IK, Bertina RM, et al. High
mechanism, and its function as a biological amplifier. Nature. levels of factor IX increase the risk of venous thrombosis.
1964;202:498. Blood. 2000;95:3678.
4. Hoffman M, Monroe DM III. A cell-based model of hemosta- 23. Kraaijenhagen RA, Anker PS, Koopman MM, et al. High
sis. Thromb Haemost. 2001;85:958. concentrations of factor VIII is a major risk factor for venous
5. Mann KG, Brummel K, Butenas S. What is all that thrombin thromboembolism. Thromb Haemost. 2000;83:5.
for? J Thromb Haemost. 2003;1:1504. 24. Selhub J, DAngelo A. Relationship between homocysteine
6. Bouchard BA, Butenas S, Mann KG, et al. Interaction be- and thrombotic disease. Am J Med Sci. 1998;316:129.
tween platelets and the coagulation system. In: Michelson A, 25. de Stefano V, Martinelli I, Mannucci PM, et al. The risk
ed. Platelets, Vol. 1. Amsterdam: Academic Press; 2002:229. of recurrent deep venous thrombosis among heterozygous
7. Baird TR, Walsh PN. The interaction of factor XIa with carriers of both factor VLeidein and the G20210A prothrombin
activated platelets but not endothelial cells promotes the gene mutations. N Engl J Med. 1999;341:801.
activation of factor IX in the consolidation phase of blood 26. Lindahl TL, Lindahl TH, Nilsson L, et al. APC-resistance is
coagulation. J Biol Chem. 2002;277:38462. a risk factor for postoperative thromboembolism in elective
8. Butenas S, Branda RF, vant Veer C, et al. Platelets and replacement of the hip or kneeA prospective study. Thromb
phospholipids in tissue factor-initiated thrombin generation. Haemost. 1999;81:18.
Thromb Haemost. 2001;86:660. 27. Wahlander K, Larson G, Lindahl TL, et al. Factor V Leiden
9. Solymoss S, Tucker M, Tracey P. Kinetics of inactivation of (G1691A) and prothrombin gene G20210A mutations as
membrane-bound factor Va by activated protein C. J Biol potential risk factors for venous thromboembolism after
Chem. 1988;263:14884. total hip or total knee replacement surgery. Thromb Haemost.
10. Camire RM, Kalafatis M, Simioni P, et al. Platelet-derived 2002;87:580.
factor Va/Va Leiden cofactor activities are sustained on the 28. Bonnar J. Can more be done in obstetric and gynecologic
surface of activated platelets despite the presence of activated practice to reduce morbidity and mortality associated with
protein C. Blood. 1998;91:2818. venous thromboembolism? Am J Obstet Gynecol. 1999;180:
11. Bucciarelli P, Rosendaal FR, Tripodi A, et al. Risk of venous 784.
thromboembolism and clinical manifestations in carriers of 29. Schafer A, Levine M, Konkle B, et al. Thrombotic disorders:
antithrombin, protein C, protein S, or activated protein C Diagnosis and treatment. Hematology. 2003;1:520.
resistance. Arterioscler Thromb Vasc Biol. 1999;19:1026. 30. Heit JA, Silverstein MD, Mohr DN, et al. The epidemiology
12. van Boven HH, Vandenbroucke JP, Briet E, et al. Gene- of venous thromboembolism in the community. Thromb
gene and gene-environment interactions determine the risk Haemost. 2001;86:452.
of thrombosis in families with inherited antithrombin 31. Gutt CN, Oniu T, Wolkener F, et al. Prophylaxis and
deficiency. Blood. 1999;94:2590. treatment of deep vein thrombosis in general surgery. Am J
13. Heeb MJ, Mesters RM, Tans G, et al. Binding of protein S Surg. 2004;189:14.
to factor Va is associated with inhibition of prothrombinase 32. Reitsma PH, Rosendaal FR. Activation of innate immunity in
that is independent of activated protein C. J Biol Chem. 1993; patients with venous thrombosis: The Leiden Thrombophilia
268:2872. Study. J Thromb Haemost. 2004;2:619.
14. Bombeli T, Piccapietra B, Boersma J, et al. Decreased 33. Neumann FJ, Ott I, Marx N, et al. Effect of human
anticoagulant response to tissue factor pathway inhibitor recombinant IL-6 and IL-8 on monocyte procoagulant
in patients with venous thromboembolism and otherwise no activity. Arterioscler Thromb Vasc Biol. 1997;17:3399.
evidence of hereditary or acquired thrombophilia. Thromb 34. Dahl OE, Pedersen T, Kierfulf M, et al. Sequential in-
Haemost. 2004;91:80. trapulmonary and systemic activation of coagulation and
15. Dahlback B. Resistance to activated protein C caused by the fibrinolysis during and after total hip replacement surgery.
R(506)Q mutation in the gene for factor V is a common Thromb Res. 1993;70:451.
risk factor for venous thrombosis. J Intern Med. 1997; 35. Dobesh PP. Evidence for extended prophylaxis in the setting
242(suppl):1. of orthopedic surgery. Pharmacotherapy. 2004;24(7 Pt 2):
16. Nicolaes GAF, Dahlback B. Factor V and thrombotic disease: 73S.
Description of a Janus-faced protein. Arterioscler Thromb 36. Clark P, Brennand J, Condere JA, et al. Activated protein C
Vasc Biol. 2002;22:530. sensitivity, protein C, protein S and coagulation in normal
17. Ridker PM, Miletich JP, Hennekens CH, et al. Ethnic pregnancy. Thromb Haemost. 1998;79:1166.
distribution of factor V Leiden in 4047 men and women. 37. Jacoben AF, Dorlsum A, Klow NE, et al. Deep vein throm-
Implications for venous thromboembolism screening. JAMA. bosis after elective cesarean section. Thromb Res. 2004;113:
1997;277:1305. 283.
570 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
38. Kujovich JL. Thrombophilia and pregnancy complications. 59. Baker RI, Eikelboom J, Lofthouse E, et al. Platelet glycopro-
Am J Obstet Gynecol. 2004;191:412. tein Ib Koxak polymorphism is associated with an increased
39. Kupferminc MJ. Thrombophilia and preeclampsia: The risk of ischemic stroke. Blood. 2001;98:36.
evidence so far. Clin Obstet Gynecol. 2005;48:406. 60. Weiss EJ, Bray PF, Tayback M, et al. A polymorphism of a
40. Rickles FR, Levine MN, Hull RD. Abnormalities of hemosta- platelet GP receptor as an inherited risk factor for coronary
sis in malignancy. In: Colman RW, Hirsch J, Marker VJ, et al. thrombosis. N Engl J Med. 1996;334:1090.
eds. Hemostasis and thrombosis, 2nd ed, Vol. 1. Philadelphia: 61. Carter AM, Catto AJ, Bamford JM, et al. Platelet GPIIIa
Lippincott Williams & Wilkins; 2001:1132. PlA and GPIb variable number tandem repeat polymorphisms
41. Sutherland DE, Wietz IC, Liebman HA. Thromboembolic and markers of platelet activation in acute stroke. Arterioscler
complications of cancer: Epidemiology, pathogenesis, diag- Thromb Vasc Biol. 1998;18:1124.
nosis, and treatment. Am J Hematol. 2003;72:43. 62. Mathew JP, Rinder CS, Howe JG, et al. Platelet PlA2
42. Kakkar AK, Williamson RCN. Prevention of venous throm- polymorphism enhances risk of neurocognitive decline after
boembolism in cancer patients. Semin Thromb Hemost. 1999; cardiopulmonary bypass. Ann Thorac Surg. 2001;71:663.
25:239. 63. Mangano DT. Group McSPI Research. Aspirin and mortality
43. Turpie AGG, Chin BSP, Lip GLH. Venous thromboembolism: from coronary artery bypass surgery. N Engl J Med. 2002;
Pathophysiology, clinical features, and prevention. The ABCs 347:1309.
of antithrombotic therapy. Br Med J. 2002;325:887. 64. Ye Z, Liu EH, Higgin JP, et al. Seven hemostatic gene
44. Sorensen RM, Pace NL. Anesthetic techniques during polymorphisms in coronary artery disease: Meta-analysis
surgical repair of femoral neck fractures: A meta analysis. of 66,155 cases and 91,307 controls. Lancet. 2006;367:
Anesthesiology. 1992;77:1095. 651.
45. Johnson R, Green JR, Charnley J. Pulmonary embolism and 65. Kearon C, Hirsch J. Management of anticoagulation before
its prophylaxis following the Charnley total hip replacement. and after elective surgery. N Engl J Med. 1997;336:1506.
Clin Orthop. 1977;127:123. 66. Walenga JM, Jeske WP, Prechel MM, et al. Newer insights
46. Williams-Russo P, Sharrock NE, Haas SB, et al. Randomized on the mechanism of heparin-induced thrombocytopenia.
trial of epidural versus general anesthesia: Outcomes after Semin Thromb Hemost. 2004;30(Suppl 1):57.
primary total knee replacement. Clin Orthop. 1996;331:199. 67. Kelton J. Heparin-induced thrombocytopenia: An overview.
47. White RH, Henderson MC. Risk factors for venous throm- Blood Rev. 2002;16:77.
boembolism after total hip and knee replacement surgery. 68. Greinacher A, Farner B, Kroll H, et al. Clinical features of
Curr Opin Pulm Med. 2002;8:365. heparin-induced thrombocytopenia including risk factors for
48. Parker MJ, Handoll HG, Griffiths R. Anaesthesia for hip thrombosis. Thromb Haemost. 2005;94:132.
fracture surgery in adults. Cochrane Database Syst Rev. 69. Rauova L, Poncz M, McKenzie SE, et al. Ultralarge com-
2001;18(4):CD000521. plexes of PF4 and heparin are central to the pathogenesis of
49. Crowther MA, Cook DJ, Griffith LE, et al. Neither baseline heparin-induced thrombocytopenia. Blood. 2005;105:131.
tests of molecular hypercoagulability nor D-dimer levels 70. Wallis DE, Workman DL, Lewis BE, et al. Failure of
predict deep venous thrombosis in critically ill medical- early heparin cessation as treatment for heparin-induced
surgical patients. Intensive Care Med. 2005;31:48. thrombocytopenia. Am J Med. 1999;106:629.
50. ODonnell J, Riddell A, Owens D, et al. Role of the 71. Rice L, Attisha WK, Drexler A, et al. Delayed-onset heparin-
thromboelastograph as an adjunctive test in thrombophilia induced thrombocytopenia. Ann Intern Med. 2002;136:210.
screening. Blood Coagul Fibrinolysis. 2004;15:207. 72. Warkentin TE, Greinacher A. Heparin-induced thrombo-
51. de Wit TR, van Mourik JA. Biosynthesis, processing and cytopenia and cardiac surgery. Ann Thorac Surg. 2003;76:
secretion of von Willebrand factor: Biological implications. 2121.
Best Pract Res Clin Haematol. 2001;14:241. 73. Chong BH. Heparin-induced thrombocytopenia. J Thromb
52. Savage B, Ruggeri Z. Platelet thrombus in flowing blood. Haemost. 2003;1:1471.
In: Michelson A, ed. Platelets, 1st ed, Vol. 1. Amsterdam: 74. de Groot PG, Derksen RHMW. The antiphospholipid syn-
Academic Press; 2002:215. drome: Clinical characteristics, laboratory features and
53. Frerickson BJ, Dong JF, McIntire LV, et al. Shear-dependent pathogenesis. Curr Opin Infect Dis. 2005;18:205.
rolling on von Willebrand factor of mammalian cells express- 75. Mackworth-Young CG. Antiphospholipid syndrome: Multi-
ing the platelet glycoprotein Ib-IX-V complex. Blood. 1998; ple mechanisms. Clin Exp Immunol. 2004;136:393.
92:3684. 76. Love PE, Santoro SA. Antiphospholipid antibodies: Anti-
54. Lopez J, Andrews R, Afshar-Kharghan V, et al. Bernard- cardiolipin and the lupus anticoagulant in systemic lupus
Soulier syndrome. Blood. 1998;91:4397. erythematosis (SLE) and in non-SLE disorders. Ann Intern
55. Kulkarni S, Dpheide SM, Yap CL, et al. A revised model of Med. 1990;112:682.
platelet aggregation. J Clin Invest. 2000;105:783. 77. Erkan D, Leibowitz E, Berman J, et al. Perioperative medi-
56. Marcus AJ, Broekman MJ, Drosopoulos JH, et al. The cal management of antiphospholipid syndrome: Hospital for
endothelial cell ecto-ADPase responsible for inhibition of special surgery experience, review of literature, and recom-
platelet function is CD39. J Clin Invest. 1997;99:1351. mendations. J Rheumatol. 2002;29:843.
57. Tsai H. Deficiency of ADAMTS13 causes thrombotic throm- 78. Asherson RA, Cervera R, de Groot PG, et al. Catastrophic
bocytopenic purpura. Arterioscler Thromb Vasc Biol. 2003; antiphospholipid syndrome (CAPS): International consensus
23:388. statement on classification criteria and treatment guidelines.
58. Tsai H, Sussman I, Nagel R. Shear stress enhances the Lupus. 2003;12:530.
proteolysis of von Willebrand factor in normal plasma. Blood. 79. Tefferi A. Polycythemia vera: A comprehensive review and
1994;83:2171. clinical recommendations. Mayo Clin Proc. 2003;78:174.
F. G A ST R O I N T E ST I N A L
CHAPTER POSTOPERATIVE NAUSEA
39
AND VOMITING
A
scheduled for an elective laparoscopic chole-
cystectomy. She volunteered a history of
postoperative vomiting after an elective ce-
sarean section under general anesthesia INCIDENCE
2 years ago. The anesthesiologist performed
The estimated overall incidence of postoperative nausea
induction with 2 mg per kg of intravenous (IV) propo-
and vomiting (PONV) for all surgeries and patient
fol. Muscle paralysis was obtained with rocuronium,
populations ranges between 25% and 30%, with severe,
0.6 mg per kg IV, and the patient was intubated with
intractable PONV estimated to occur in approximately
an oral endotracheal tube. Intermittent positive-pressure
0.18% of all patients.1 In high-risk groups, PONV occurs
ventilation was employed. Anesthesia was maintained
in as many as 70% of patients. PONV can delay patient
with 1% isoflurane in 1 L per minute of oxygen and
discharge from the recovery room and prolong hospital
2 L per minute of nitrous oxide. A total of 3 g per kg
stay. It may even be more distressing to patients than
of IV fentanyl was given intraoperatively for analgesia.
the postoperative pain itself.2,3 It is not uncommon for
At the end of surgery, the anesthesiologist administered
patients to develop PONV after discharge from the surgical
glycopyrrolate and neostigmine to reverse muscle relax-
center in the postdischarge period. The incidence of
ation and prophylactically gave ondansetron, 4 mg IV, to
postdischarge nausea and vomiting was found to be as
prevent postoperative vomiting. The patient developed sig-
high as 35%, with more than 70% of this group having
nificant nausea and vomited twice in the recovery room.
had no PONV in the recovery room.4
Rescue medication consisting of an additional 4 mg of
ondansetron was administered intravenously. She had
persistent vomiting despite the prophylactic and rescue
doses of ondansetron. ANATOMY AND PHYSIOLOGY
OF VOMITING
The vomiting center, situated in the lateral reticular for-
What Baseline Knowledge mation of the medulla oblongata, mediates the vomiting
reflex (see Fig. 39.1). It is closely related to the nucleus
Is Relevant? tractus solitarius and the area postrema. The chemore-
ceptor trigger zone (CTZ) is located in the area postrema.
Peripheral and central stimuli can affect both the vom-
DEFINITIONS iting center and the CTZ. Afferents from the pharynx,
gastrointestinal tract, mediastinum, renal pelvis, peri-
Nausea is defined as a subjectively unpleasant sensation toneum, and genitalia can stimulate the vomiting center.
of having the urge to vomit. Vomiting is the reflex forceful Central stimulation from the cerebral cortex, higher cor-
expulsion of gastric contents through the esophagus and tical and brainstem centers, nucleus tractus solitarius,
out of the mouth. Retching is physiologically similar to CTZ, vestibular system of the inner ear, and the visual
571
572 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Preoperative Premedication
Cortical Intraoperative Inhalational anesthetic agents
center Volatile anesthetic gases
Nitrous oxide
Intravenous anesthetic agents
Cerebellum
Vomiting Chemoreceptor Reversal agents for neuromuscular
center trigger zone
blockade
Regional anesthesia
Nucleus Postoperative Pain
Vestibular
system
tractus Movement and ambulation
solitarius
36 and 72 hours, extends its antiemetic efficacy up to to prevent chemotherapy-induced nausea and vomiting
24 hours post surgery. In adults, dexamethasone, 8 mg IV, in cancer patients.52,53 A recent study demonstrates its
has been demonstrated to be effective in preventing eme- superior antiemetic property when compared to on-
sis;38 more recent data suggest that a lower dose of 5 mg dansetron (>90% vs. 74% over 24 hours).54 Other NK-1 re-
IV may also be effective when compared to placebo.39 ceptor antagonists are undergoing clinical investigations.
in terms of recovery room costs and nursing labor costs. TABLE 39.4 Risk Factors for Postoperative Nausea and
For example, in an office-based practice, delay in patient Vomiting
discharge can directly increase personnel costs in the
PACU when overtime payment is required. Differences Patient Factors
in drug acquisition costs will also complicate the analy- Female gender
sis of the cost-effectiveness of PONV prophylaxis versus Nonsmoking status
treatment. Watcha and Smith demonstrated that the pro- History of motion sickness or postoperative nausea and
phylactic use (vs. therapeutic) of ondansetron was only vomiting
cost-effective when the incidence of PONV exceeds 33%, Anesthetic Factors
whereas the prophylactic use (vs. therapeutic) of droperi-
Use of inhalational anesthetic gases
dol was cost-effective when the incidence was as low as
Use of nitrous oxide
10%.81 In addition, there are costs that are difficult to
Use of opioids in the intraoperative and postoperative
quantify, such as nurses giving less time to other patients
period
so they can concentrate on patients with PONV. It should
also be kept in mind that prophylaxis is associated with a Surgical Factors
finite risk of adverse drug reactions and, therefore, giving Type of surgery (neurosurgery, ear-nose-throat, strabis-
antiemetic medication only when PONV occurs may be a mus, laparoscopy, breast and plastic surgery)
viable option for the lower-risk patient while also reducing Long duration of surgery
the risk of such reactions.
Patients risk of 6. Risk factors for PONV have been reliably identified
PONV and validated in various clinical studies to predict its
(based on Table 39.4) risk. Routine antiemetic prophylaxis is not required
in all patients but will undoubtedly improve patient
satisfaction in high-risk groups.
7. The strategy for PONV prophylaxis includes identify-
Low Moderate High ing patients at risk, reducing the baseline risk, and
combination therapy for prophylaxis.
REFERENCES
PONV Reduce
prophylaxis baseline risk 1. Watcha MF, White PF. Postoperative nausea and vomiting.
not indicated (Table 39.5) Its etiology, treatment, and prevention. Anesthesiology. 1992;
unless for 77:162.
surgical 2. Myles PS, Williams DL, Hendrata M, et al. Patient satisfac-
procedures tion after anaesthesia and surgery: Results of a prospective
where PONV survey of 10,811 patients. Br J Anaesth. 2000;84:6.
may be Consider 3. Macario A, Weinger M, Carney S, et al. Which clinical
detrimental nonpharmacologic anesthesia outcomes are important to avoid? The perspective
intervention if of patients. Anesth Analg. 1999;89:652.
possible 4. Carroll NV, Miederhoff P, Cox FM, et al. Postoperative
nausea and vomiting after discharge from outpatient surgery
centers. Anesth Analg. 1995;80:903.
5. Rigby M, ODonnell R, Rupniak NM. Species differences in
tachykinin receptor distribution: Further evidence that the
Moderate High
substance P (NK1) receptor predominates in human brain.
J Comp Neurol. 2005;490:335.
6. Saito R, Takano Y, Kamiya H. Roles of substance P and
NK(1) receptor in the brainstem in the development of
emesis. J Pharmacol Sci. 2003;91:87.
Combination Combination 7. Apfel CC, Laara E, Koivuranta M, et al. A simplified risk
therapy with therapy with score for predicting postoperative nausea and vomiting:
dexamethasone dexamethasone Conclusions from cross-validations between two centers.
and 5-HT3 antagonist Anesthesiology. 1999;91:693.
5-HT3 and TIVA with 8. Patel RI, Hannallah RS. Anesthetic complications following
antagonist propofol pediatric ambulatory surgery: A 3-yr study. Anesthesiology.
1988;69:1009.
9. Pusch F, Berger A, Wildling E, et al. The effects of systolic
FIGURE 39.3 Emetic risk profiling and multimodal approach arterial blood pressure variations on postoperative nausea
with combination antiemetic therapy. PONV, postoperative and vomiting. Anesth Analg. 2002;94:1652.
nausea and vomiting; 5-HT3 , 5-hydroxytryptamine type 3; TIVA, 10. Di Florio T, Goucke CR. The effect of midazolam on persis-
total intravenous anesthesia. tent postoperative nausea and vomiting. Anaesth Intensive
Care. 1999;27:38.
11. Splinter WM, MacNeill HB, Menard EA, et al. Midazolam
reduces vomiting after tonsillectomy in children. Can J
prophylaxis from at least two different classes to optimize Anaesth. 1995;42:201.
efficacy. 12. Apfel CC, Kranke P, Katz MH, et al. Volatile anaesthetics
may be the main cause of early but not delayed postoperative
vomiting: A randomized controlled trial of factorial design.
Br J Anaesth. 2002;88:659.
KEY POINTS 13. Hartung J. Twenty-four of twenty-seven studies show a
greater incidence of emesis associated with nitrous oxide
1. There are many patient-, anesthetic- and surgical- than with alternative anesthetics. Anesth Analg. 1996;83:114.
related factors associated with PONV. 14. Apfel CC, Korttila K, Abdalla M, et al. A factorial trial of six
2. It is important to understand the different classes of interventions for the prevention of postoperative nausea and
antiemetic drugs, their mechanisms of action, and the vomiting. N Engl J Med. 2004;350:2441.
various receptors located in the vomiting center and 15. Gupta A, Stierer T, Zuckerman R, et al. Comparison of
the CTZ. recovery profile after ambulatory anesthesia with propofol,
3. Both pharmacologic and nonpharmacologic interven- isoflurane, sevoflurane and desflurane: A systematic review.
Anesth Analg. 2004;98:632.
tions can be utilized to reduce the incidence of PONV.
16. Tramer M, Moore A, McQuay H. Propofol anaesthesia and
4. Combination therapy is superior to monotherapy in postoperative nausea and vomiting: Quantitative systematic
PONV prophylaxis. review of randomized controlled studies. Br J Anaesth. 1997;
5. If prophylaxis fails to prevent PONV, an antiemetic 78:247.
drug from a different class should be administered to 17. Sneyd JR, Carr A, Byrom WD, et al. A meta-analysis of nausea
treat it. and vomiting following maintenance of anaesthesia with
C H A P T E R 3 9 / P O S T O P E R AT I V E N A U S E A A N D V O M I T I N G 579
propofol or inhalational agents. Eur J Anaesthesiol. 1998; antiemetic for postoperative nausea and vomiting. Anesth
15:433. Analg. 2000;91:136.
18. Cechetto DF, Diab T, Gibson CJ, et al. The effects of propofol 38. Wang JJ, Ho ST, Liu YH, et al. Dexamethasone reduces
in the area postrema of rats. Anesth Analg. 2001;92:934. nausea and vomiting after laparoscopic cholecystectomy. Br
19. Gan TJ, El-Molem H, Ray J, et al. Patient-controlled antieme- J Anaesth. 1999;83:772.
sis: A randomized, double-blind comparison of two doses of 39. Wang JJ, Ho ST, Lee SC, et al. The use of dexamethasone for
propofol versus placebo. Anesthesiology. 1999;90:1564. preventing postoperative nausea and vomiting in females
20. Gan TJ, Glass PS, Howell ST, et al. Determination of plasma undergoing thyroidectomy: A dose-ranging study. Anesth
concentrations of propofol associated with 50% reduction in Analg. 2000;91:1404.
postoperative nausea. Anesthesiology. 1997;87:779. 40. Pearman MH. Single dose intravenous ondansetron in the
21. Gan TJ, Ginsberg B, Grant AP, et al. Double-blind, random- prevention of postoperative nausea and vomiting. Anaesthe-
ized comparison of ondansetron and intraoperative propofol sia. 1994;49(suppl):11.
to prevent postoperative nausea and vomiting. Anesthesiol- 41. Tang J, Wang B, White PF, et al. The effect of timing of on-
ogy. 1996;85:1036. dansetron administration on its efficacy, cost-effectiveness,
22. Lovstad RZ, Thagaard KS, Berner NS, et al. Neostigmine and cost-benefit as a prophylactic antiemetic in the ambula-
50 microg kg(-1) with glycopyrrolate increases postoperative tory setting. Anesth Analg. 1998;86:274.
nausea in women after laparoscopic gynaecological surgery. 42. Philip BK, McLeskey CH, Chelly JE, et al. Pooled analysis of
Acta Anaesthesiol Scand. 2001;45:495. three large clinical trials to determine the optimal dose of
23. Joshi GP, Garg SA, Hailey A, et al. The effects of antag- dolasetron mesylate needed to prevent postoperative nausea
onizing residual neuromuscular blockade by neostigmine and vomiting. The Dolasetron Prophylaxis Study Group.
and glycopyrrolate on nausea and vomiting after ambulatory J Clin Anesth. 2000;12:1.
surgery. Anesth Analg. 1999;89:628. 43. Chen X, Tang J, White PF, et al. The effect of timing of
24. Hovorka J, Korttila K, Nelskyla K, et al. Reversal of dolasetron administration on its efficacy as a prophylactic
neuromuscular blockade with neostigmine has no effect antiemetic in the ambulatory setting. Anesth Analg. 2001;93:
on the incidence or severity of postoperative nausea and 906.
vomiting. Anesth Analg. 1997;85:1359. 44. Gan TJ. Selective serotonin 5-HT3 receptor antagonists for
25. Watcha MF, Safavi FZ, McCulloch DA, et al. Effect of postoperative nausea and vomiting: Are they all the same?
antagonism of mivacurium-induced neuromuscular block on CNS Drugs. 2005;19:225.
postoperative emesis in children. Anesth Analg. 1995;80:713. 45. Rothenberg DM, Parnass SM, Litwack K, et al. Efficacy of
26. Cheng CR, Sessler DI, Apfel CC. Does neostigmine adminis- ephedrine in the prevention of postoperative nausea and
tration produce a clinically important increase in postopera- vomiting. Anesth Analg. 1991;72:58.
tive nausea and vomiting? Anesth Analg. 2005;101:1349. 46. Naguib K, Osman HA, Al-Khayat HC, et al. Prevention of
27. Borgeat A, Ekatodramis G, Schenker CA. Postoperative post-operative nausea and vomiting following laparoscopic
nausea and vomiting in regional anesthesia: A review. surgeryephedrine vs propofol. Middle East J Anesthesiol.
Anesthesiology. 2003;98:530. 1998;14:219.
28. Henzi I, Walder B, Tramer MR. Metoclopramide in the pre- 47. Oddby-Muhrbeck E, Eksborg S, Bergendahl HT, et al. Effects
vention of postoperative nausea and vomiting: A quantitative of clonidine on postoperative nausea and vomiting in breast
systematic review of randomized, placebo-controlled studies. cancer surgery. Anesthesiology. 2002;96:1109.
Br J Anaesth. 1999;83:761. 48. Handa F, Fujii Y. The efficacy of oral clonidine premedi-
29. Ferrari LR, Donlon JV. Metoclopramide reduces the inci- cation in the prevention of postoperative vomiting in chil-
dence of vomiting after tonsillectomy in children. Anesth dren following strabismus surgery. Paediatr Anaesth. 2001;
Analg. 1992;75:351. 11:71.
30. Fortney JT, Gan TJ, Graczyk S, et al. A comparison of 49. Gulhas N, Turkoz A, Durmus M, et al. Oral clonidine
the efficacy, safety, and patient satisfaction of ondansetron premedication does not reduce postoperative vomiting in
versus droperidol as antiemetics for elective outpatient children undergoing strabismus surgery. Acta Anaesthesiol
surgical procedures. S3A-409 and S3A-410 Study Groups. Scand. 2003;47:90.
Anesth Analg. 1998;86:731. 50. Gesztesi Z, Scuderi PE, White PF, et al. Substance P
31. Henzi I, Sonderegger J, Tramer MR. Efficacy, dose-response, (Neurokinin-1) antagonist prevents postoperative vomiting
and adverse effects of droperidol for prevention of postoper- after abdominal hysterectomy procedures. Anesthesiology.
ative nausea and vomiting. Can J Anaesth. 2000;47:537. 2000;93:931.
32. Domino KB, Anderson EA, Polissar NL, et al. Comparative 51. Diemunsch P, Schoeffler P, Bryssine B, et al. Antiemetic
efficacy and safety of ondansetron, droperidol, and metoclo- activity of the NK1 receptor antagonist GR205171 in the
pramide for preventing postoperative nausea and vomiting: treatment of established postoperative nausea and vomiting
A meta-analysis. Anesth Analg. 1999;88:1370. after major gynaecological surgery. Br J Anaesth. 1999;82:
33. FDA Talk Paper. FDA strengthens warnings for droperidol. 274.
Available at: http://www.fda.gov/bbs/topics/ANSWERS/2001/ 52. Navari RM. Aprepitant: A neurokinin-1 receptor antagonist
ANS01123.html. Accessed August 12, 2006. for the treatment of chemotherapy-induced nausea and
34. Gan TJ. Postoperative nausea and vomitingcan it be vomiting. Expert Rev Anticancer Ther. 2004;4:715.
eliminated? JAMA. 2002;287:1233. 53. Hesketh PJ, Grunberg SM, Gralla RJ, et al. The oral
35. Henzi I, Walder B, Tramer MR. Dexamethasone for the pre- neurokinin-1 antagonist aprepitant for the prevention of
vention of postoperative nausea and vomiting: A quantitative chemotherapy-induced nausea and vomiting: A multina-
systematic review. Anesth Analg. 2000;90:186. tional, randomized, double-blind, placebo-controlled trial in
36. Goldman AC, Govindaraj S, Rosenfeld RM. A meta-analysis patients receiving high-dose cisplatinthe Aprepitant Pro-
of dexamethasone use with tonsillectomy. Otolaryngol Head tocol 052 Study Group. J Clin Oncol. 2003;21:4112.
Neck Surg. 2000;123:682. 54. Gan T, Apfel C, Kovac A, et al. The NK1 receptor antagonist
37. Wang JJ, Ho ST, Tzeng JI, et al. The effect of timing of dex- aprepitant for prevention of postoperative nausea and
amethasone administration on its efficacy as a prophylactic vomiting [Abstract]. Anesthesiology. 2005;103:A769.
580 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
55. Yogendran S, Asokumar B, Cheng DC, et al. A prospective plus dexamethasone during the first 24 hours following
randomized double-blinded study of the effect of intravenous highly emetogenic chemotherapy. Br J Cancer. 2001;85:
fluid therapy on adverse outcomes on outpatient surgery. 1099.
Anesth Analg. 1995;80:682. 71. Lee A, Done ML. The use of nonpharmacologic techniques to
56. Maharaj CH, Kallam SR, Malik A, et al. Preoperative prevent postoperative nausea and vomiting: A meta-analysis.
intravenous fluid therapy decreases postoperative nausea Anesth Analg. 1999;88:1362.
and pain in high risk patients. Anesth Analg. 2005;100:675. 72. Gan TJ, Jiao KR, Zenn M, et al. A randomized controlled
57. Greif R, Laciny S, Rapf B, et al. Supplemental oxygen comparison of electro-acupoint stimulation or ondansetron
reduces the incidence of postoperative nausea and vomiting. versus placebo for the prevention of postoperative nausea
Anesthesiology. 1999;91:1246. and vomiting. Anesth Analg. 2004;99:1070.
58. Goll V, Akca O, Greif R, et al. Ondansetron is no more 73. Zarate E, Mingus M, White PF, et al. The use of tran-
effective than supplemental intraoperative oxygen for pre- scutaneous acupoint electrical stimulation for preventing
vention of postoperative nausea and vomiting. Anesth Analg. nausea and vomiting after laparoscopic surgery. Anesth
2001;92:112. Analg. 2001;92:629.
59. Joris JL, Poth NJ, Djamadar AM, et al. Supplemental oxygen 74. Lee A, Done ML. Stimulation of the wrist acupuncture
does not reduce postoperative nausea and vomiting after point P6 for preventing postoperative nausea and vomiting.
thyroidectomy. Br J Anaesth. 2003;91:857. Cochrane Database Syst Rev. 2004;CD003281.
60. Purhonen S, Turunen M, Ruohoaho UM, et al. Supplemental 75. Agarwal A, Bose N, Gaur A, et al. Acupressure and
oxygen does not reduce the incidence of postoperative nausea ondansetron for postoperative nausea and vomiting after
and vomiting after ambulatory gynecologic laparoscopy. laparoscopic cholecystectomy. Can J Anaesth. 2002;49:554.
Anesth Analg. 2003;96:91. 76. Wang SM, Kain ZN. P6 acupoint injections are as effective
61. Treschan TA, Zimmer C, Nass C, et al. Inspired oxygen frac- as droperidol in controlling early postoperative nausea and
tion of 0.8 does not attenuate postoperative nausea and vom- vomiting in children. Anesthesiology. 2002;97:359.
iting after strabismus surgery. Anesthesiology. 2005;103:6. 77. Somri M, Vaida SJ, Sabo E, et al. Acupuncture versus
62. Orhan-Sungur M, Sessler DI, Kranke P, et al. Supplemental ondansetron in the prevention of postoperative vomiting.
oxygen does not reduce postoperative nausea and vomit- A study of children undergoing dental surgery. Anaesthesia.
ing: A systematic review of randomized controlled trials 2001;56:927.
[Abstract]. Anesthesiology. 2005;103:A626. 78. Coloma M, White PF, Ogunnaike BO, et al. Comparison of
63. Wu O, Belo SE, Koutsoukos G. Additive anti-emetic efficacy acustimulation and ondansetron for the treatment of estab-
of prophylactic ondansetron with droperidol in out-patient lished postoperative nausea and vomiting. Anesthesiology.
gynecological laparoscopy. Can J Anaesth. 2000;47:529. 2002;97:1387.
64. Lopez-Olaondo L, Carrascosa F, Pueyo FJ, et al. Combination 79. Eberhart LH, Mayer R, Betz O, et al. Ginger does not pre-
of ondansetron and dexamethasone in the prophylaxis of vent postoperative nausea and vomiting after laparoscopic
postoperative nausea and vomiting. Br J Anaesth. 1996; surgery. Anesth Analg. 2003;96:995.
76:835. 80. Scuderi PE, James RL, Harris L, et al. Antiemetic prophy-
65. Habib AS, El-Moalem HE, Gan TJ. The efficacy of the 5- laxis does not improve outcomes after outpatient surgery
HT3 receptor antagonists combined with droperidol for when compared to symptomatic treatment. Anesthesiology.
PONV prophylaxis is similar to their combination with 1999;90:360.
dexamethasone. A meta-analysis of randomized controlled 81. Watcha MF, Smith I. Cost-effectiveness analysis of
trials. Can J Anaesth. 2004;51:311. antiemetic therapy for ambulatory surgery. J Clin Anesth.
66. Scuderi PE, James RL, Harris L, et al. Multimodal antiemetic 1994;6:370.
management prevents early postoperative vomiting after 82. Sinclair DR, Chung F, Mezei G. Can postoperative nausea
outpatient laparoscopy. Anesth Analg. 2000;91:1408. and vomiting be predicted? Anesthesiology. 1999;91:109.
67. Habib AS, White WD, Eubanks S, et al. A randomized 83. Junger A, Hartmann B, Benson M, et al. The use of an
comparison of a multimodal management strategy versus anesthesia information management system for prediction
combination antiemetics for the prevention of postoperative of antiemetic rescue treatment at the postanesthesia care
nausea and vomiting. Anesth Analg. 2004;99:77. unit. Anesth Analg. 2001;92:1203.
68. Habib AS, Gan TJ. The effectiveness of rescue antiemetics 84. Koivuranta M, Laara E, Snare L, et al. A survey of postoper-
after failure of prophylaxis with ondansetron or droperidol: ative nausea and vomiting. Anaesthesia. 1997;52:443.
A preliminary report. J Clin Anesth. 2005;17:62. 85. Eberhart LH, Hogel J, Seeling W, et al. Evaluation of three
69. Kovac AL, OConnor TA, Pearman MH, et al. Efficacy of risk scores to predict postoperative nausea and vomiting.
repeat intravenous dosing of ondansetron in controlling Acta Anaesthesiol Scand. 2000;44:480.
postoperative nausea and vomiting: A randomized, double- 86. Apfel CC, Kranke P, Eberhart LH, et al. Comparison of
blind, placebo-controlled multicenter trial. J Clin Anesth. predictive models for postoperative nausea and vomiting. Br
1999;11:453. J Anaesth. 2002;88:234.
70. de Wit R, de Boer AC, vd Linden GH, et al. Effective 87. Gan TJ, Meyer T, Apfel CC, et al. Consensus guidelines for
cross-over to granisetron after failure to ondansetron, a ran- managing postoperative nausea and vomiting. Anesth Analg.
domized double blind study in patients failing ondansetron 2003;97:62.
CHAPTER ABDOMINAL COMPARTMENT
40
SYNDROME
A
the back of a pickup truck that collided with life support, and the patient expired.
a tree. The patient arrived in the trauma This clinical scenario is an example of abdominal
resuscitation bay with a surgical airway, compartment syndrome (ACS) associated with organ
hypotension, chest wall crepitus, and fixed, dysfunction.
dilated pupils after more than 20 minutes
in transport. Bilateral chest tubes were inserted to treat
hemopneumothoraces, and a Foley catheter was inserted
and was noted to be draining frank blood. Radiologic What Is the Abdominal
studies ruled out pelvic fractures. Focused abdominal
sonography for trauma revealed the presence of intra- Compartment Syndrome?
abdominal fluid. Primary resuscitation was unsuccessful
using crystalloids and blood products, with refractory ACS occurs from increased intra-abdominal pressure.
hypotension that was reversed with continuous volume Richardson described elevated end-inspiratory pressures
infusion. The patient was transported to the operating and hypoperfusion secondary to a low cardiac output
room for emergent exploration of his abdomen. (CO) associated with ACS.1 Impaired venous return and
Perioperatively, the patient received a massive re- high peak inspiratory pressure with hypercarbia were
suscitation including crystalloids, blood products, and present, causing hypoperfusion and severe pulmonary
activated recombinant factor VII. A damage control ce- dysfunction. Early surgical decompression is mandatory,
liotomy revealed a large hemoperitoneum, which was and a better outcome is associated with early detection.
evacuated. A lacerated spleen was removed, and a large, Release of the restrictive abdominal pressure will result in
nonexpanding, left-sided retroperitoneal hematoma was the correction of organ dysfunction. Oliguria is an early
identified. The patient was coagulopathic, acidotic, and sign of ACS, but the most reliable clinical indicator is
hypothermic. The distended bowel prohibited definitive progressive failure of ventilation. A typical case of ACS
closure of the abdominal wall; the abdomen was left has a peak inspiratory pressure in the range of 85 cm
open and packed with surgical lap pads. During tempo- H2 O, with a rise in PaCo2 . A decompressive celiotomy
rary abdominal closure, using a sterile adhesive drape is indicated in the presence of abdominal distention,
and intra-abdominal drain catheters connected to con- hypercarbia, and high peak inspiratory pressures. This
tinuous suction, the patient developed sudden cardiac procedure may be performed either at bedside or in the
arrest. This event responded to the Advanced Cardiac operating room.
Life Support protocol. Immediate reexploration was per- The surgical technique performed for damage control
formed. The left retroperitoneal space was noted to be is a continuum that includes primary resuscitation,
enlarging and was now leaking frank blood. A left nephrec- damage control celiotomy, secondary resuscitation, and
tomy was performed. No further surgical bleeding was delayed reconstruction.2 Patients in extremis usually do
identified. not tolerate reconstruction. In summary, ACS is a surgical
The abdominal cavity was dressed as noted earlier. emergency that requires the damage control technique to
A postoperative computerized tomography scan of the prevent organ dysfunction. If ACS is recognized late or
head showed severe closed head injury with uncal hernia- goes unrecognized, it can lead to multiple organ failure
tion. The patient was taken to the intensive care unit, and and death.
581
582 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Saline bag
Monitor
Infusion Double-check
Monitor cable
syringe valve
Infusion
tubing
Pressure
transducer
Tape
Urinary catheter
Bladder
Valve positions
Pres
transd sure
uction
Transd Urine
u flow
occludction occlu
ed ded
Urine
flow
Measure
Drain
FIGURE 40.1 Intra-abdominal pressure monitoring device. The device is placed into the urinary
drainage circuit. By changing the valve from drain to measure, intermittent measurements may
be taken. This device is that it allows measurement of intra-abdominal pressure without breaking or
interrupting the urinary drainage circuit.
denominator in each of these conditions appears to be tumors. In the presence of one of the high-risk conditions,
the need for large volume resuscitation. For example, pa- the risk of ACS may be heightened by hypothermia, acido-
tients presenting with acute major abdominal hemorrhage sis, and/or coagulopathy due to the volume replacement
are at very high risk for ACS. Other conditions include that may be required for resuscitation. Other factors to be
nonmajor intraperitoneal hemorrhage, intra-abdominal considered include:717
catastrophes, abdominal trauma, hemorrhagic pancreati- Retroperitoneal bleeding
tis, severe ascites, hepatic transplantation, and ovarian Use of pneumatic antishock garments
584 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 40.1 Grading of the Abdominal Compartment The changes in cardiovascular parameters are due
Syndrome to increased intrathoracic pressure from intra-abdominal
hypertension.1,1821 Cardiovascular changes include an
Bladder Pressure increase in central venous and pulmonary artery wedge
Grade (mm Hg) Recommendation pressures, and systemic vascular resistance. CO decreases
progressively with increases in intra-abdominal pressure,
I 1015 Maintain normovolemia
and is dependent on the intravascular volume status.
II 1625 Hypervolemic resuscitation
Kashtan et al. showed that changes in CO were demonstra-
III 2635 Decompression
ble in the setting of ACS. Hypovolemic animals developed
IV >35 Decompression and
a 53% decline in CO, compared with a 17% decline in
reoperation
the euvolemic group and a 50% increase in the hyperv-
olemic group.19 Patients with ACS may have an increase
From: Meldrum DR, Moore FA, Moore EE, et al. Prospective
characterization and selective management of the abdominal
in CO with intravascular volume replacement, but fluid
compartment syndrome. Am J Surg. 1997;174:667. alone cannot correct the renal dysfunction and decrease
in splanchnic blood flow associated with ACS.
An early sign of the renal effects of ACS is oliguria
Abdominal wall fascial closure under tension in a high- that may progress to an anuric state.16 Intra-abdominal
risk patient hypertension may cause compression of the renal veins
Persistent surgical intra-abdominal bleeding and inferior vena cava, elevation of renal vascular
Damage control celiotomy using laparotomy pad pack- resistance, and even renal parenchymal compression, all
ing for bleeding control of which may cause severe renal dysfunction.22 In an
Patients without abdominal injury that require massive animal study, renal blood flow and glomerular filtration
fluid resuscitation for conditions like major burns or rate were 25% of normal at an intra-abdominal pressure of
severe acute pancreatitis 20 mm Hg, and only 7% of normal when intra-abdominal
ACS may be subdivided into primary, secondary, and pressure reached 40 mm Hg.23 In a swine model, elevated
tertiary variants: intra-abdominal pressure was found to decrease urine
output and upregulate the renin-angiotensin-aldosterone
PRIMARY: Primary ACS is associated with an injury or system. Abdominal decompression, in combination with
disease in the abdominopelvic region that frequently intravascular volume expansion, reversed the effects
requires early surgical or angioradiologic interven- on renal function and the renin-angiotensin-aldosterone
tion, or a condition that develops after abdominal system.24
surgery. Examples are damage control surgery, sec- In ACS, an increase in abdominal girth and intra-
ondary peritonitis, bleeding pelvic fractures, other abdominal pressure may be associated with a decrease
causes of massive retroperitoneal hematoma, liver in splanchnic blood flow and the potential development
transplant, and nonoperative management of solid of small bowel ischemia.2529 Several animal models
organ injury. demonstrated a decreased organ blood flow index with
SECONDARY: Secondary ACS is associated with con- increased intra-abdominal pressure in all major abdom-
ditions that do not originate from the abdomen, yet inal organs (except adrenal glands), and a decrease of
develop clinical findings of ACS. Examples are condi- hepatic artery, portal vein, and microcirculatory flow.
tions such as sepsis and capillary leak, major burns, and Additionally, a reduction in hepatic energy production
other conditions requiring massive fluid resuscitation. and energy level were correlated with increased intra-
TERTIARY: Tertiary (or recurrent) ACS occurs when ACS abdominal pressure.23,30,31
develops after prophylactic or therapeutic surgical or Neuropathophysiologic alterations may be seen with
medical treatment of primary or secondary ACS, when ACS. For example, one may see a functional obstruction
there is persistence of ACS after decompressive laparo- of jugular venous drainage due to increased intrapleural
tomy, or development of a new ACS episode after pressure from elevated intra-abdominal pressure sec-
definitive abdominal wall closure. The clinical man- ondary to ACS, with an associated increase in intracranial
ifestations of ACS are the result of multiple organ pressure (ICP). Bloomfield et al. demonstrated, in a
dysfunction, affecting the respiratory, renal, cardiovas- porcine model, the significant effects of lowering ICP and
cular, and neurologic systems.16 the improvement in cerebral perfusion pressure following
Respiratory manifestations are related directly to the abdominal decompression.3234 Laparotomy in patients
effect of intra-abdominal hypertension, which elevate with a combination of severe closed head injury and ab-
the hemidiaphragms, with an associated decrease in dominal injuries can result in a dramatic reduction of
intrathoracic volume and compliance. Elevated peak the ICP.35
airway pressures and pulmonary vascular resistance The eyes may also be affected in ACS, with rupture
will most often be present. In ACS, ventilatory support of retinal capillaries resulting in decreased central vision.
becomes difficult, and relatively sophisticated strategies While the retinal hemorrhage improves over time without
of mechanical ventilation are often necessary to stabilize specific treatment, an appropriate ophthalmic examina-
and improve the respiratory mechanics. Respiratory tion should, nonetheless, be performed.36 In the setting
deterioration, if it occurs, may lead to development of of an ACS, the orbital compartment also merits consider-
severe hypercarbia, worsening acidosis, and hypoxemia. ation. Sullivan et al. retrospectively reviewed 13 patients
C H A P T E R 4 0 / A B D O M I N A L C O M PA R T M E N T SY N D R O M E 585
requiring massive fluid resuscitation following thermal TABLE 40.2 Grading of Intra-Abdominal Hypertension
injury. An orbital compartment syndrome that required
lateral canthotomy and cantholysis was identified, demon- Grade Intra-Abdominal Pressure (mm Hg)
strating the importance of early diagnosis and timely
I 1215
treatment to decompress the orbit and reduce orbital
II 1620
pressure to avoid potential visual disturbances, including
III 2125
blindness.37
IV >25
Gastric decompression
Paracentesis
Rectal enemas
Gastrointestinal prokinetic agents (cisapride, metoclo-
pramide, domperidone, erythromycin)
Colonic prokinetic agents (neostigmine)
Furosemide with or without use of human albumin
20%
Continuous venovenous hemofiltration with aggressive
ultrafiltration
Sedation
Paralysis
Body positioning
Botulinum toxin into the internal anal sphincter FIGURE 40.3 Abdomen, in which closure is impossible, being
prepared for artificial closure.
Data from: Sugrue M. Abdominal compartment syndrome. Curr Opin
Crit Care. 2005;11:333.
FIGURE 40.5 Intestines being placed into the towel pack, with FIGURE 40.7 Wound just prior to placing the impermeable
drains in place. membrane over the towel-pack containing intestines.
wound infections were reported, and all 37 patients followed by intra-abdominal abscess. Respiratory failure
survived.44 and cardiovascular collapse will follow. Bowel torsion
causes ischemia and can lead to necrosis. In this situa-
tion, delayed diagnosis may allow progression to diffuse
peritonitis with the attendant large fluid resuscitation re-
What Are the Complications quirement unassociated to blood loss. Abdominal wall
if the Abdominal Compartment compliance will determine the degree of distention be-
fore development of ACS signs. Once the intra-abdominal
Syndrome Is Not Diagnosed pressure reaches 25 mm Hg, the major concerns are ex-
in a Timely Manner? tensive ischemia and necrosis of the small bowel. Short
bowel syndrome may result from radical resections of
dead bowel; this will require evaluation of nutritional sta-
Multiple organ dysfunction results from prolonged intra-
tus to prevent malnutrition. Missed colonic injuries may
abdominal hypertension. Forced abdominal wall fascial
be associated with diffuse peritonitis. In this situation,
closure should be avoided. Physiologic exhaustion can
intestinal diversion is mandatory. Temporary abdomi-
lead to multiple organ failure and death if ACS is
nal closure requires the open technique; an occlusive
allowed to progress. Prolonged bowel ischemia is as-
dressing may be used to contain the intra-abdominal con-
sociated with intestinal necrosis. Kinking of the bowel
tents, along with a suction system composed of two drain
mesentery is associated with necrosis of the bowel,
FIGURE 40.6 Impermeable membrane being prepped for FIGURE 40.8 Abdomen with intestines in a sterile pack. Note
placement over the wound. drain at superior aspect of wound.
588 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
catheters to remove the excessive fluid accumulation asso- 10. Biffl WL, Moore EE, Burch JM, et al. Secondary abdominal
ciated with ACS. Indirect measurement of pressures from compartment syndrome is a highly lethal event. Am J Surg.
the urinary bladder is important to prevent recurrent ACS. 2001;182:645.
11. Raeburn CD, Moore EE, Biffl WL, et al. The abdominal com-
partment syndrome is a morbid complication of postinjury
damage control surgery. Am J Surg. 2001;182:542.
12. Maxwell RA, Fabian TC, Croce MA, et al. Secondary abdom-
KEY POINTS inal compartment syndrome: An underappreciated mani-
1. Early identification of intra-abdominal hypertension festation of severe hemorrhagic shock. J Trauma. 1999;47:
995.
may prevent ACS.
13. Kopelman T, Harris C, Miller R, et al. Abdominal com-
2. Intra-abdominal perfusion pressure (IAPP) goal is partment syndrome in patients with isolated extraperitoneal
60 mm Hg (IAPP = Mean arterial pressure Intra- injuries. J Trauma. 2000;49:744.
abdominal pressure). 14. Mayberry JC, Goldman RK, Mullins RJ, et al. Surveyed opin-
3. Forced abdominal wall fascial closure should be ion of American trauma surgeons on the prevention of the
avoided. abdominal compartment syndrome. J Trauma. 1999;47:509.
4. Perioperative correction of hypothermia, acidosis, 15. Ivy ME, Atweh NA, Palmer J, et al. Intra-abdominal
and coagulopathy is of utmost importance during hypertension and abdominal compartment syndrome in
the acute resuscitation phase. burn patients. J Trauma. 2000;49:387.
5. Damage control surgery consists of control of hem- 16. Schein M, Wittmann DH, Aprahamian CC, et al. The
abdominal compartment syndrome: The physiological and
orrhage and contamination and identification of
clinical consequences of elevated intra-abdominal pressure.
injuries, utilizing an abbreviated laparotomy. Reop- J Am Coll Surg. 1995;180:745.
eration is indicated with refractory hypoperfusion 17. Cheatham ML. Intra-abdominal hypertension and abdomi-
associated with ongoing resuscitation. nal compartment syndrome. New Horiz. 1999;7:96.
6. Increased intra-abdominal girth combined with high 18. Ridings PC, Bloomfield GL, Blocher CR, et al. Cardiopul-
ventilatory peak airway pressures and oliguria are monary effects of raised intra-abdominal pressure before and
common manifestations of ACS. after intravascular volume expansion. J Trauma. 1995;39:
7. Oliguria is an early sign of ACS. Urinary bladder 1071.
pressure monitoring is strongly recommended. 19. Kashtan J, Green JF, Parsons EQ, et al. Hemodynamic effects
8. Temporary abdominal closure is an essential compo- of increased abdominal pressure. J Surg Res. 1981;30:249.
20. Diamant M, Benumof JL, Saidman LJ. Hemodynamics of
nent of the management of ACS. Permanent closure
increased intra-abdominal pressure: Interaction with hypov-
is considered during the postresuscitation phase,
olemia and halothane anesthesia. Anesthesiology. 1978;48:23.
and deferred until after secondary resuscitation is 21. Robotham JL, Wise RA, Bromberger-Barnea B. Effects of
completed. changes in abdominal pressure on left ventricular perfor-
mance and regional blood flow. Crit Care Med. 1985;13:803.
REFERENCES 22. Harman PK, Kron IL, Mc Laachlan HD, et al. Elevated intra-
abdominal pressure and renal function. Ann Surg. 1982;196:
1. Richardson JD, Trinkle JK. Hemodynamic and respiratory 594.
alterations with increased intraabdominal pressure. J Surg 23. Diebel LN, Wilson RF, Dulchrsky SA, et al. Effect of
Res. 1976;20:401. increased intra-abdominal pressure on hepatic arterial,
2. Rotondo MF, Schwab CW, McGonigal MD, et al. Damage portal venous, and hepatic microcirculatory blood flow.
control: An approach for improved survival in exsanguinat- J Trauma. 1992;33:279.
ing penetrating abdominal injury. J Trauma. 1993;35:375. 24. Bloomfield GL, Blocher CR, Fakhry IF, et al. Elevated intra-
3. Meldrum DR, Moore FA, Moore EE, et al. Prospective abdominal pressure increases plasma renin activity and
characterization and selective management of the abdominal aldosterone levels. J Trauma. 1997;42:997.
compartment syndrome. Am J Surg. 1997;174:667. 25. Diebel LN, Dulchavsky SA, Wilson RF. Effect of increased
4. Eddy VA, Key SP, Morris JA Jr. Abdominal compartment intra-abdominal pressure on mesenteric arterial and intesti-
syndrome: Etiology, detection, and management. J Tenn nal mucosal blood flow. J Trauma. 1992;33:45.
Med Assoc. 1994;87:55. 26. Diebel LN, Dulcharsky SA, Brown WJ. Splanchnic ischemia
5. Morris JA Jr, Eddy VA, Blinman TA, et al. The staged ce- and bacterial translocation in the abdominal compartment
liotomy for trauma: Issues in unpacking and reconstruction. syndrome. J Trauma. 1997;43:852.
Ann Surg. 1993;217:576. 27. Chang MC, Miller PR, DAgostino R, et al. Effects of
6. Barnes G, Laine GA, Giam PY, et al. Cardiovascular re- abdominal decompression on cardiopulmonary function
sponses to elevation of intra-abdominal hydrostatic pressure. and visceral perfusion in patients with intra-abdominal
Am J Physiol. 1985;248:R208. hypertension. J Trauma. 1998;44:440.
7. Saggi BH, Sugerman HJ, Ivatury RR, et al. Abdominal 28. Ivatury RR, Porter JM, Simon RJ, et al. Intra-abdominal
compartment syndrome. J Trauma. 1998;45:597. hypertension after life- threatening penetrating abdominal
8. Ertel W, Oberholzer A, Platz A, et al. Incidence and clinical trauma: Prophylaxis, incidence, and clinical relevance to
pattern of the abdominal compartment syndrome after gastric mucosal pH and abdominal compartment syndrome.
damage-control laparotomy in 311 patients with severe J Trauma. 1998;44:1016.
abdominal and/or pelvic trauma. Crit Care Med. 2000;28: 29. Friedlander MH, Simon RJ, Ivatury R, et al. Effect of
1747. hemorrhage on superior mesenteric artery flow during
9. Offner PJ, de Souza AL, Moore EE, et al. Avoidance increased intra-abdominal pressures. J Trauma. 1998;45:433.
of abdominal compartment syndrome in damage-control 30. Caldwell CB, Ricotta JJ. Changes in visceral blood flow with
laparotomy after trauma. Arch Surg. 2001;136:676. elevated intraabdominal pressure. J Surg Res. 1987;43:14.
C H A P T E R 4 0 / A B D O M I N A L C O M PA R T M E N T SY N D R O M E 589
31. Nakatani T, Sakamoto Y, Kaneko I, et al. Effects of Another untoward effect of massive resuscitation after burn
intraabdominal hypertension on hepatic energy metabolism injury. J Trauma. 2006;60:72.
in rabbits. J Trauma. 1997;43:192. 38. World Society of the Abdominal Compartment Syndrome.
32. Bloomfield GL, Ridings PC, Blocher CR, et al. Increased pleu- Available at: http://www.wsacs.org. Last accessed 23 April,
ral pressure mediates the effects of elevated intra-abdominal 2006.
pressure upon the central neurons and cardiovascular sys- 39. Sugrue M. Abdominal compartment syndrome. Curr Opin
tems. Surg Forum. 1995;46:572. Crit Care. 2005;11:333.
33. Bloomfield GL, Ridings PC, Blocher CR, et al. Effects 40. Cheatham ML, White MW, Sagraves SG, et al. Abdominal
of increased intra-abdominal pressure upon intracranial perfusion pressure: A superior parameter in the assessment
and cerebral perfusion pressure before and after volume of intra-abdominal hypertension. J Trauma. 2000;49:621.
expansion. J Trauma. 1996;40:936. 41. Simon RJ, Friedlander MH, Ivatury RR, et al. Haemorrhage
34. Bloomfield GL, Ridings PC, Blocher CR, et al. A proposed re- lowers the threshold for intra-abdominal hypertension-
lationship between increased intra-abdominal, intrathoracic, induced pulmonary dysfunction. J Trauma. 1997;42:398.
and intracranial pressure. Crit Care Med. 1997;25:496. 42. Oda J, Ueyama M, Yamashita K. Hypertonic lactated saline
35. Bloomfield GL, Dalton JM, Sugerman HJ, et al. Treatment of resuscitation reduces the risk of abdominal compartment
increasing intracranial pressure secondary to the acute ab- syndrome in severely burned patients. J Trauma. 2006;60:64.
dominal compartment syndrome in a patient with combined 43. Latenser BA, Kowal-Vern A, Kimball D. A pilot study com-
abdominal and head trauma. J Trauma. 1995;39:1168. paring percutaneous decompression with decompressive
36. Priluck IA, Blodgett DW. The effects of increased intra- laparotomy for acute abdominal compartment syndrome
abdominal pressure on the eyes. Nebr Med J. 1996;81:8. in thermal injury. J Burn Care Rehabil. 2002;23:190.
37. Sullivan SR, Ahmadi AJ, Singh CN, et al. Medical Director, 44. Scott BG, Wesh FJ, Pham HQ. Early aggressive closure of
Gainesville Fire Rescue Service. Elevated orbital pressure: the open abdomen. J Trauma. 2006;60:17.
CHAPTER ACUTE LIVER DYSFUNCTION
41
AND ANESTHESIA-INDUCED
HEPATITIS
Phillip S. Mushlin, Stuart B. Mushlin,
and Richard D. Olson
A
was admitted to the hospital for a right existing diseases are another important consideration;
hemicolectomy. She had two uneventful advanced or incipient diseases that are unrecognized pre-
surgeries under halothane anesthesia more operatively (e.g., clinically silent) may declare themselves
than 20 years earlier. Her personal and postoperatively, with a time course indistinguishable from
family history was negative for liver disease. AIH. Examples include viral hepatitis, steatohepatitis, au-
She had no allergies, had not traveled recently, and had toimmune hepatitis, and pregnancy-related diseases. This
no history of alcohol or illicit drug use. Liver chemistry chapter focuses on the diagnosis of AIH within the context
tests obtained 7 weeks earlier were normal, and imaging of confounding variables that may lead to misdiagnoses.
studies were negative for metastatic disease. We also address the treatment and prevention of AIH.
The surgery was performed under general anesthe-
sia with propofol, fentanyl, desflurane, and vecuronium.
Surgery lasted 70 minutes and was uneventful. Post-
operative pain management included Percocet (Endo
Pharmaceuticals, Chadds Ford, PA) and ibuprofen. On
Do Halogenated Anesthetics
postoperative day (POD) 5, the patient developed jaun- Cause Liver Necrosis?
dice, malaise, and anorexia, without a rash, pruritus,
or eosinophilia. Alanine aminotransferase (ALT), aspar-
tate aminotransferase (AST), and total bilirubin (2,188 IU
per L, 425 IU per L, 18 mg per dL, respectively) were CHLOROFORM
markedly elevated, but alkaline phosphatase (AP) was
The potent halogenated vapors are the only anesthetic
normal. Abdominal ultrasonography showed gallbladder
agents that have engendered concerns about postoper-
sludge, without biliary dilatation. Serum tests for iron,
ative liver dysfunction. In 1912, The American Medical
copper, ceruloplasmin, hepatitis viruses (A, B, and C),
Association issued a condemnation of chloroform anes-
cytomegalovirus, and autoantibodies (i.e., antimitochon-
thesia due to fatal cases of liver failure associated with the
drial, antismooth-muscle, and antinuclear antibodies)
use of this agent.3 The term delayed chloroform poisoning
were normal. Tests showed seropositivity for IgG for her-
(or hepatic necrosis) was generally taken to represent a
pes simplex and Epstein-Barrs viruses, consistent with
distinct pathologic entity. However, there was never scien-
prior infection.
tific proof of a causal link between chloroform anesthesia
On POD 8, AST and ALT began to decrease, but the co-
and postoperative liver injury during the seven decades
agulopathy and jaundice worsened, and encephalopathic
of its popularity. In fact, most deaths attributed to chlo-
changes occurred. While the patient was being evaluated
roform anesthesia were from cardiacrather than from
for orthotopic liver transplantation (OLT), her condition
livercomplications.4
suddenly improved. All clinical and laboratory abnormali-
ties had resolved by POD 21, and the patient was discharged
home with a diagnosis of desflurane-induced hepatitis.
HALOTHANE
An alarming number of cases of anesthesia-associated
Is This a Case of liver failure occurred a few years after halothanes entry
into the clinical arena. This prompted the Committee
Desflurane-Induced Hepatitis? on Anesthesia (of the National Academy of Sciences-
National Research Council) to investigate postoperative
What information is needed to diagnose anesthesia- hepatic necrosis. Hoping to clarify the matter quickly, the
induced hepatitis (AIH)?1,2 The diagnostic issues are committee abandoned the idea of a randomized clinical
590
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 591
trial in favor of a retrospective review known as the TABLE 41.1 Number of Reported Cases of
National Halothane Study.5 Anesthesia-Induced Hepatitis
two decades. Japan approved sevoflurane for clinical TABLE 41.2 Features of Anesthesia-Induced Hepatitis
use in 1990, and it became clinically available in the
United States in 1995. By then, sevoflurane had a solid Healthy patient
record of safety based on its use in operating theaters Brief, minor, uneventful surgery
throughout the world. More than 2 million Japanese Unremarkable recovery for a few days
patients received the anesthetic during this time, with just Syndrome presents with
four published cases of presumed sevoflurane-induced Fever 75%
liver dysfunction.1619 Anorexia, nausea 50%
Chills 30%
Myalgias 20%
DESFLURANE Rash, eosinophilia 10%
3 to 6 d later, jaundice appears
Desflurane has been in widespread clinical use since
1993. Millions and millions of patients have received this
anesthetic, yet there are but a few reports of presumed have occurred in this setting.6 The incidence of AIH is
desflurane-induced hepatitis.1,2,20 10-fold higher in patients with a history of prior halothane
anesthesia than in those who have just had their first-ever
halothane anesthetic.21 Short intervals between the two
most recent halothane anesthetics have been associated
What Is the Mechanism of with more severe liver injury. AIH afflicts women more
Anesthesia-Related Liver often than men (1.8 to 1), and obesity further increases
the risk.24,25 Age is an important but enigmatic risk
Injury? factor. AIH rarely occurs in children, and one half of
the cases have been in people over 50-years-old.21,26,27
Two notable variables that have not been found to be
PATTERNS OF LIVER INJURY risk factors for AIH are liver disease and perioperative
use of potentially hepatotoxic drugs (besides halogenated
Two distinct forms of anesthesia-induced liver injury have vapors).28
been described.21 Type 1 leads to mild, transient increases
in serum enzymes, which are detectable within hours of
surgery and resolve within 2 days. Clinical studies show
that levels of serum aminotransferases or glutathione
THE IMMUNE THEORY
S-transferase may increase in up to 20% to 50% of OF ANESTHESIA-INDUCED
patients after minor surgery under enflurane or halothane HEPATITIS
anesthesia.22,23 This type of liver injury is usually benign,
self-limiting, and clinically unimportant. Its pathogenesis Oxidative Metabolites
may involve hepatic ischemia (hypoxia) or cytotoxic
effects of anesthetic metabolites. On the other hand, type The generation of reactive intermediates (e.g., trifluo-
2, often referred to as anesthesia-induced hepatitis (AIH) roacetyl chloride) that occurs as the liver metabolizes
is a rare, idiosyncratic disorder that can lead to massive halogenated anesthetics is central to the immune theory
hepatocellular necrosis, acute liver failure, and death. of AIH. There is a strong correlation between the incidence
of AIH and the proportion of the anesthetic dose oxidized
through cytochrome P450 2E1.29,30 For halothane, enflu-
rane, isoflurane, and desflurane, the proportions oxidized
CLINICAL FEATURES are 20%, 2%, 0.2%, and 0.02%, respectively.3133 Such log-
OF ANESTHESIA-INDUCED unit differences are serendipitous and provide an elegant
HEPATITIS (AIH) probe for investigating the role of metabolites in AIH. For
example, when subjects receive equipotent doses of the
AIH is mainly an affliction of healthy patients which halogenated anesthetics, their livers produce concentra-
develops following a brief uneventful general anesthetic tions of metabolites that are dispersed over a 1,000-fold
for minor surgery. The recovery is unremarkable during range.
the first postoperative weekuntil the syndrome becomes
manifest. Among the common early clinical abnormalities Neoantigens and Autoantigens
are fever, anorexia, nausea, chills, myalgias, rash, and
eosinophilia. Jaundice usually develops 3 to 6 days later. These metabolites can covalently bind to various
This development reflects life-threatening disease, with liver macromolecules, forming trifluroacylated deriva-
a mortality rate that may approach 40%.6 Table 41.2 tives.34 Laboratory studies confirm that equivalent doses
summarizes the clinical features of AIH. (10 MAC-hours) of halogenated anesthetics lead to
Of the risk factors for AIH, the most important is widely varying amounts of tissue acylation, expressed
prior exposure to halothane; 71% to 95% of AIH cases qualitatively as halothane enflurane > isoflurane >
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 593
desflurane = oxygen.32 In other words, the extent of relates to the immune theory of AIH. For example, the
metabolite incorporation into liver proteins is high with development of hepatitis after desflurane anesthesia in
halothane, low with enflurane, even lower with isoflurane, such instances could theoretically result from traces
and often undetectable after desflurane anesthesia. The of the inciting agent (e.g., halothane) that entered
immune system may see the altered liver macromolecules the patients lungs after being aerosolized from the
as neoantigens or autoantigens. Attached portions of anes- anesthesia machine (where it previously resided). Immune
thetic molecules can seemingly act as haptens to foster crossover offers an alternative explanation of how earlier
immune recognition of the carrier proteins. Laboratory halothane (or enflurane) exposures might predispose to
and clinical evidence suggest that the immune system tar- liver injury from enflurane, isoflurane, or desflurane
gets the altered liver molecules of susceptible individuals anesthetics.4043
as nonself, and mounts an attack on the hepatocytes that
can lead to massive liver necrosis.
Misdiagnosed as 40
Anesthesia-Induced Hepatitis? 35
20
Subclinical or unrecognized liver diseases have a high
prevalence in the general population and may progress 15
to (and present as) overt postoperative liver injury or
dysfunction. Accordingly, they may be misinterpreted 10
(or summarily dismissed) as anesthesia-induced liver
5
damage.49 For example, routine screening tests were per-
formed in 7,620 healthy appearing patients (American 0
Other drugs Antibiotics Multiple NSAIDs Anesthetics
Society of Anesthesiologists [ASA] physical status classifi-
41% 31% 19% 7% 1%
cation I) scheduled for elective surgery.50 Eleven patients
(0.14%) had increases of AST, ALT, and lactate dehydro-
FIGURE 41.1 Drug-induced liver dysfunction (DILD)
genase (LDH), so their surgeries were postponed. Further
partitioned into five arbitrary categories. Bar heights show
evaluation of these patients identified subclinical liver dis-
percentages by category. Data are based on reports to the
eases (infectious mononucleosis, viral hepatitis, cirrhosis,
Swedish Adverse Drug Reactions Advisory Committee from 1985
or alcoholic hepatitis); three of the patients developed
to 2004. Acetaminophen (paracetamol) intoxications were
jaundice (i.e., 0.04%). If these individuals had undergone
excluded; nearly all cases in this study were idiosyncratic drug
surgery, their postoperative clinical course would have re-
reactions with bilirubin levels 2 times the upper limit of normal
sembled, and therefore might have been misdiagnosed as,
(ULN). The category Multiple shows the percentage of DILD
AIH. It is therefore important for the clinician to system-
cases attributed to multiple drugs from at least two different
atically work through the many possible causes of acute
categories. Other drugs shows DILDs from various categories,
liver injury before implicating anesthetic agents as the
excluding antibiotics, nonsteroidal anti-inflammatory drugs
culprit.
(NSAIDs), and anesthetics. (Common offenders in this group
included carbamazepine, ranitidine, enalapril, chlorpromazine,
sulfasalazine, omeprazole, cyclophosphamide, ticlopidine,
atorvastatin, simvastatin, and disulfiram). The overall rate of
Which Drugsbesides mortality or transplantation was 9.2%. Elevations of serum
AnestheticsCan Cause Acute bilirubin and AST were the most important predictors of death
or liver transplantation. Bilirubin levels were higher in deceased
Liver Dysfunction? or transplant recipients than in surviving patients (median
values: 18.7 vs. 5.5 ULN). The category with the highest
Many prescription medications, over-the-counter formu- mortality rate (40%) was anesthetics. All cases in this category
lations, illicit substances, and herbal preparations have involved halothane and occurred between 1985 and 1994: there
been associated with liver injury. Such adverse effects were no cases of anesthesia-induced liver dysfunction during the
can be idiosyncratic (e.g., immune mediated)51 or dose final 10 years of the study. (Data derived from 19952004,
related (e.g., acetaminophen). Susceptibility to adverse Bjornsson E, Olsson R. Outcome and prognostic markers in
drug reactions is influenced by various and complex fac- severe drug-induced liver disease. Hepatology. 2005;42:481.)
tors, including pharmacogenetics, pathological states, and
concomitant therapies. Estimated rates of overt liver in-
jury for many drugs range from 1 in 10,000 to 1 in 100,000. Acetaminophen (paracetamol) has been the single, most
Such numbers, however, tend to underestimate the prob- common cause of FHF in the United States and
lem, owing to underreporting and difficulties in detecting, Great Britain.56,57 FHF from acetaminophen often repre-
recognizing, and diagnosing adverse drug reactions.52 In- sents intentional overdoses (suicide attempts). However,
deed, the list of drugs that can cause liver dysfunction an alarming number of cases are due to uninten-
or FHF is extensive;53 prominent categories of offenders tional overdoses, which occur for two major reasons.56
include analgesics and antibiotics (see Fig. 41.1).54 First, acetaminophen is a commonand often over-
lookedingredient of many pharmaceutical preparations.
Second, some individuals are unusually susceptible to
NON-NARCOTIC ANALGESICS acetaminophen toxicity,58 owing to unrecognized risk fac-
tors, including malnutrition and alcoholism.19,59,60 In sus-
Many analgesic and anti-inflammatory drugs are hep- pected cases of acetaminophen toxicity, patients should
atotoxic and can induce injuries that range from sub- receive immediate therapy (e.g., N-acetylcysteine) to avert
clinical laboratory abnormalities to acute liver failure.55 progression of the injury to irreversible liver failure.
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 595
Liver injury has long been associated with high keep in mind that a failed attempt to find such markers
doses of aspirin.61 Such injuries are generally manifest does not rule out viral hepatitis.70 Although serologic tests
within weeks of initiating therapy, with mild-to-moderate are robust, they do, on rare occasions, yield false negative
increases of ALT and AST.62 Jaundice and clinical results. Also, history teaches us that todays viruses and
hepatitis rarely occur. Nonsteroidal anti-inflammatory tomorrows mutated variants will cause or contribute to
drugs (NSAIDs) are another common cause of hepatic liver injury in ways that are not presently understood.
injury.63 Adverse reactions to NSAIDs may be dose-related In this regard, it is notable that a high proportion of pre-
or idiosyncratic (with cross sensitivity among the various sumed halothane hepatitis cases occurred before the main
classes of NSAIDs). The associated liver injuries are rarely cause of posttransfusion hepatitis had been identified (i.e.,
severe. However, in the bigger picture, serious liver injury hepatitis C virus [HCV] was not isolated until 1989).
is much more likely to result from NSAIDs than from the
newer halogenated anesthetics.64 Risk factors for NSAID-
induced liver injury include rheumatoid arthritis (vs. STEATOHEPATITIS
degenerative arthritis)61,65 and concurrent use of other
hepatotoxic drugs. Fatty liver diseases are the most common cause of ab-
normal liver chemistry tests. The coexistence of fatty
deposits (steatosis) and inflammatory changes in hepa-
tocytes is referred to as steatohepatitis. This condition,
unlike steatosis, involves ongoing hepatocellular damage.
What Are the Early Signs of Adiposity seems to increase the susceptibility of liver cells
Life-Threatening Liver Injury? to necrosis and apoptosis. Cryptogenic cirrhosis is most
often a sequel of fatty liver disease; approximately 70% of
Drug-induced liver injuries range from asymptomatic patients with this form of cirrhosis have morbid obesity
increases of aminotransferases to FHF. The clinicians or diabetes.
ability to readily distinguish between life-threatening and Consuming large amounts of alcohol invariably leads
less serious liver injuries is an important determinant of to steatosis and steatohepatitis, but only 10% to 20%
patient outcome. Dr. Hyman Zimmerman (Hy), a pioneer of alcoholics develop cirrhosis. Alcoholism is notoriously
in drug-induced hepatotoxicity, provides useful guidance difficult to detect based on a brief patient history and
on the matter: If a drug causes enough hepatocellular dam- physical examination. Serious liver disease may elude
clinical detection because the body compensates so well
age to induce global liver dysfunction, the patients life is
for extensive liver destruction. Inflammatory fatty liver
in jeopardy.66 Jaundice in this setting reflects the inability
disease in patients without a history of alcohol use is
of the liver to adequately transport bilirubin. Hys Law
called nonalcoholic steatohepatitis (NASH).71 It occurs
teaches that a concomitant increase of serum bilirubin
mainly in middle-aged people with obesity or diabetes,
and ASTwith little or no increase of APis ominous, and
and it accounts for a high proportion of the liver test
that jaundice is a harbinger of liver failure.54,66 A hepato-
abnormalities found in healthy blood donors. Up to
cellular pattern of injury has a worse prognosis than either
20% of patients with NASH have unsuspected fibrosis or
cholestatic or mixed liver injuries. Without liver transplan-
cirrhosis. Liver disease is the third most common cause
tation, the mortality rates range from 10% to 50%.
of death, with a liver-related mortality of 11%.
may not have the telltale corneal Kayser-Fleischers rings. can lead to extensive hepatocellular necrosis and liver
Laboratory tests of copper metabolism (ceruloplasmin, failure, with a clinical picture that resembles HELLP.
serum, and urine copper levels) are not totally specific AFLP typically presents in the third trimester or
for the disease. Serial measurements of copper levels and early postpartum period. Clinical manifestations include
liver biopsy are more time consuming and invasive, but malaise, nausea, vomiting, abdominal pain (right upper
provide greater accuracy. Standard liver chemistry tests quadrant), fever, headaches, and pruritus. Management is
may be insightful, for example, an AST to ALT ratio >4 supportive (e.g., careful fluid therapy, nutritional support,
and an AP to total bilirubin ratio <2 are consistent with correction of coagulopathies) and includes prompt deliv-
Wilsons disease as the cause of FHF. ery of the fetus.85 Women usually begin to recover within
3 days of delivery. Worsening disease suggests the possi-
bility of superimposed sepsis. It is therefore important to
PREGNANCY-RELATED LIVER search for and aggressively treat sepsis. The value of OLT
is uncertain. Anecdotes have suggested its benefits, but
DYSFUNCTION women with AFLP rarely need OLT.
Liver dysfunction in pregnancy can rapidly progress
to FHF, simultaneously threatening two lives, that of Hepatitis E
the mother and fetus.77,78 Pregnancy-related disorders
associated with serious liver injury include the hemolysis, Hepatitis E virus (HEV) is mainly a disease of devel-
elevated liver enzymes, and low platelet count (HELLP) oping countries, but it may be contracted by traveling
syndrome, acute fatty liver of pregnancy (AFLP), and acute to (or having close contact with visitors from) regions
viral hepatitis.79 with HEV epidemics. Also, anti-HEV antibody titers in
the general population suggest that HEV may be more
prevalent worldwide than previously assumed. The im-
Preeclampsia and Hemolysis, Elevated
portant point is that HEV infections occur throughout the
Liver Enzymes, and Low Platelet Count world (including developed countries)86,87 and are severe
(HELLP) Syndrome in pregnancy.88,89 Vertical transmission of the virus can
lead to HEV infections of babies born to HEV-infected
Hepatic hypoperfusion develops in women with severe mothers. FHF may occur in one third of HEV-infected
preeclampsia for various reasons, such as intense vaso- parturients. Of HEV-positive mothers, 15% to 27% may
constriction, hypovolemia, or starvation. Approximately die from the HEV infection.
10% percent of preeclamptic women develop the HELLP
syndrome.80 The syndrome is associated with poor ma-
ternal and fetal outcomes. Maternal mortality may be as
high as 24%, with perinatal mortality rates up to 367 per What Is Ischemic Hepatitis?
1,000 live births. Early clinical symptoms include weak-
ness, fatigue, nausea, right upper quadrant or epigastric
pain, visual changes, and headache, followed by evidence
of overt liver failure, such as jaundice and coagulopathy.81 CLINICAL PRESENTATION
The liver histopathology includes: (i) dense fibrin deposits
Ischemic or hypoxic hepatitis should be considered in
in sinusoids; (ii) periportal and portal hemorrhages; and
the differential diagnosis of unexplained postoperative
(iii) hemorrhagic infarcts, which are usually focal, but
hepatitis. This disorder is usually manifest within hours
occasionally confluent.
of an ischemic or hypoxic insult, and is characterized by
Serial laboratory tests show rapid, large increases of
marked elevations (greater than 20-fold above normal)
ALT and AST, owing to ongoing hepatocellular necrosis.82
in serum aminotransferases and LDH. Patients develop
Hemolysis and liver dysfunction lead to hyperbilirubine-
symptoms of acute liver disease, sometimes with elevated
mia in 40% of cases. In severe cases, the hepatic vascu-
serum glucose levels, mental confusion, and reversible
lature becomes friable and the liver is highly susceptible
renal failure.90,91 Liver abnormalities begin to abate
to rupture.83 Hepatic rupture complicates 2% of HELLP
shortly after resolution of the inciting event (unless the
cases, with a mortality rate of 40% to 60%. (The HELLP
severity of the patients underlying disease precludes this).
syndrome is discussed in greater detail in Chapter 49).
The usual histologic finding is centrilobular necrosis
with little or no inflammatory response.92 In the most
Acute Fatty Liver of Pregnancy extreme cases, however, massive hepatic necrosis occurs,
with massive elevations of serum transaminases (e.g.,
AFLP is a rare disorder (1 of 10,000 to 15,000 pregnancies)
exceeding 30,000 IU per L).93
and should be recognized as a cause of FHF.84,85 AFLP is
a microvesicular steatosis (similar to Reye Syndrome, or
valproic acid or tetracycline toxicity); the defect is in the
fatty acid -oxidation spiral of mitochondria.84 With mild INCIDENCE OF ISCHEMIC
disease, there are foamy hepatocytes with intact hepatic HEPATITIS
architecture and minimal inflammation or necrosis. AFLP
is therefore less likely than HELLP to cause large increases The incidence of ischemic hepatitis varies greatly with
of serum transaminases and LDH. However, severe AFLP the patient population studied. For example, in one
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 597
study, approximately 0.5% of patients admitted to medical respiratory disease.97 Many of the patients in the study ex-
intensive care units developed ischemic hepatitis, which perienced severe, refractory hypoxemia during flare-ups of
usually followed episodes of hypotension or acute heart their respiratory disease, and ischemic (hypoxic) hepatitis
failure.94 However, in another study, the overall incidence developed in 7% of them. Three independent risk factors
was 2.6% in patients in a coronary care unit, and in for ischemic hepatitis were identified in these patients:
the subgroup of patients with low cardiac output, the (i) Cor pulmonale, (ii) PaO2 <46 mm Hg, (iii) serum blood
occurrence was almost 22%.95 urea nitrogen (BUN) >36 mg per dL. The subgroup of pa-
tients that developed hypoxic hepatitis had worse clinical
outcomes: They needed mechanical ventilation twice as
often and had a twofold higher mortality rate than those
CAUSES OF ISCHEMIC without hypoxic hepatitis.
HEPATITIS
The liver is susceptible to hypoxic injury, although such
RISK FACTORS FOR ISCHEMIC HEPATITIS
injuries are often subclinical. Perioperative events that Important risk factors for ischemic or hypoxic hepati-
can lead to overt hypoxic liver injury include: (i) hepatic tis include severe hypotension, cardiac disease, sepsis,
hypoperfusion, from systemic arterial hypotension or hep- primary hepatic circulatory disturbances, and intrahep-
atic venous obstruction; (ii) hypoxemia, from respiratory atic or posthepatic venous obstruction (e.g., Budd-Chiari
failure, anemia, low FIO2 ; and (iii) sepsis. One large study, syndrome).
conducted over a 10-year period, identified 142 episodes of
ischemic hepatitis.96 Patients were categorized into four
groups based on their clinical presentations: 13% had res-
piratory failure; 13% had septic shock; 14% had acute What Are the Anesthetic Issues
heart failure; and 56% had decompensated congestive
heart failure. Associated with Liver
Dysfunction?
Left versus Right Heart Failure
Patients with severe liver disease are at increased risk for
Intraoperative hypotension is common, but by itself, perioperative hepatic complications. For example, clinical
seldom causes ischemic hepatitis. One explanation for studies have shown high rates of morbidity and mortality
this observation comes from a cohort study in trauma in patients with acute alcoholic hepatitis. As the severity
patients.91 All patients exhibited episodes of hypotension of liver disease increases, so does the likelihood of post-
(systemic BP <75 mm Hg) for >15 minutes. However, operative hepatic complications, especially for patients
ischemic hepatitis developed only in the cohort with undergoing abdominal or emergency surgery.98 Studies
underlying cardiac disease (94% with right heart failure). show that the incidence of postoperative complications
This and other studies suggest that right heart failure, (e.g., liver failure, bleeding, infection, sepsis, renal failure,
with resultant hepatic venous congestion, predisposes to pulmonary failure, ascites) relates directly to the sever-
hepatic injury induced by hypotensive events.91 ity of hepatic dysfunction (e.g., Child-Pugh classification:
A, 10%; B, 31%; C, 76%)99,100 (see Table 41.3).
Hypoxemia Liver disease renders patients susceptible to hypoxic
or ischemic liver injury, because it impairs compensatory
An observational study looked at the importance of hy- physiologic mechanisms that help preserve liver blood
poxemia in patients with right heart failure and endstage flow under adverse circumstances. Close monitoring
of the circulation during major surgery is therefore fulminant liver failure occurs, the mortality rate can be
essential to enable rapid detection and careful treatment 50%, and when encephalopathy occurs shortly after the
of hypovolemia and hemodynamic instability. development of jaundice, the mortality rate may be as
Which anesthetic technique best preserves hepatic high as 80%. The importance of early recognition of
function? There is no clearly identifiable anesthetic irreversible liver disease should not be underestimated,
technique that is best for the liver. The goals of the because complications of advanced hepatic failure can
anesthetic should include optimizing cardiopulmonary render patients ineligible for liver transplantation. Of
function and preserving hepatic perfusion, particularly patients that recover from AIH, liver function usually
during surgeries that decrease splanchnic blood flow, such returns to normal. Nonetheless, periodic monitoring of
as laparotomy. Anesthetic agents such as isoflurane and liver chemistries may help detect the rare cases of cirrhosis
fentanyl (or remifentanil) would be reasonable choices that follow idiosyncratic drug reactions.101
because they tend to preserve cardiac output and favorably
affect the hepatic oxygen supply and demand relationship.
The ideal anesthetic would be an inert gas that causes no
cardiovascular depression, preserves hepatic blood flow, What Measures
and is not metabolized (and would therefore be free
of hepatotoxicity). Clinical studies with such an agent Should Be Taken to Prevent
(xenon) are currently in progress. Anesthesia-Induced Hepatitis?
Careful preoperative evaluations can help identify indi-
viduals at risk for AIH. When there is a history of AIH,
What Is the Approach to it seems prudent to avoid using any of the volatile agents
Unexplained Postoperative because of the possibility of immune crossover and the
remote chance that trace amounts from an anesthesia
Liver Dysfunction? machine could trigger AIH. This decision to avoid inhaled
anesthetics is predicated on the assumption that there
is a suitably safe alternative (e.g., total intravenous or
GET THE DIAGNOSIS RIGHT regional anesthesia) for the surgery. Avoiding the use of
halothane is the single, most effective way to decrease the
AIH lacks pathognomonic features. It is therefore a di- incidence of AIH. In countries that still use halothane (for
agnosis of exclusion that can be neither proved nor economic reasons), it would seem reasonable to chose
disapproved. The more severe the liver injury, the more an alternative anesthetic for patients who have received
important it is to assemble a multidisciplinary intensive halothane during the prior 6 weeks, because recent ex-
care team (including a hepatologist and liver transplant posure to halothane is the most important risk factor for
surgeon) to efficiently sort through the diagnostic possi- AIH (even in children).
bilities and determine the prognosis. The starting point
is a thorough history and physical examination, liver
chemistry tests, serologic panels, imaging studies, and,
occasionally, endoscopic procedures and liver biopsy (per- SUMMARY
cutaneous or transjugular). It is important to conduct a
This chapter begins with a clinical case that is based on a
thorough search for (and to treat) reversible causes of the
recent report of desflurane-induced hepatitis.1,2 It contin-
liver injury (e.g., sepsis, extrahepatic biliary obstruction).
ues by identifying important clinical issues that must be
Medications that could possibly cause or contribute to
considered to properly evaluate purported cases of AIH.
liver injury should be discontinued.
AIH is a rare, idiosyncratic, life-threatening condition.
Its pathogenesis seemingly involves oxidative metabo-
lites of anesthetics (through cytochrome P450 2E1) that
MANAGING THE covalently bind to and modify various hepatic macro-
COMPLICATIONS OF POLD molecules, leading to autoantigen and neoantigen for-
mation. Despite the strong association between AIH and
AIH is a small subset of unexplained POLD, for which autoantibodies against hepatocellular proteins, the signif-
there is no specific therapy. Treatment is mainly sup- icance of these antibodies is unclear; many individuals
portive. Patients depend on skillful, timely interventions develop such autoantibodies without developing AIH.
of a multidisciplinary health care team to get the best Most of the reported cases of AIH involve halothane.
possible outcomes. This includes maintaining fluid and Desflurane and sevoflurane have been used extensively
electrolyte balance, correcting hypoglycemia, supporting for more than a decade. Yet, there are but a few
hemodynamics and pulmonary function, providing proper published reports of unexplained hepatic failure following
nutritional support, and judiciously managing coagula- the use of these anesthetics. Indeed, the rarity of AIH
tion abnormalities. with the newer volatile agents makes AIH a statistically
Patients with rapidly deteriorating liver function improbable diagnosis. Postoperative liver injury is much
should be evaluated as soon as possible for OLT. When more likely to result from preexisting liver disease than
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 599
from AIH. It is therefore important to conduct a thorough 6. Despite the rarity of AIH with the newer halogenated
search for all possible causes of liver injury in patients vapors, the possibility of immune crossover may be
with unexplained postoperative liver dysfunction. The reason enough to avoid future use of any volatile
diagnostic possibilities include common and uncommon anesthetic in a patient with a history of AIH.
liver diseases, such as viral hepatitis, steatohepatitis,
autoimmune hepatitis, and Wilsons disease. If extensive
evaluations fail to turn up a cause, the diagnosis of AIH REFERENCES
should be considered. 1. Tung D, Yoshida EM, Wang CS, et al. Severe desflurane
For patients with a history of AIH (as with malignant hepatotoxicity after colon surgery in an elderly patient. Can
hyperthermia-susceptible patients), it is prudent to avoid J Anaesth. 2005;52:133.
using volatile anesthetics if an equally acceptable and safe 2. Martin JL. Volatile anesthetics and liver injury: A clinical
alternative is available. This recommendation is based update or what every anesthesiologist should know. Can J
on the possibility that anesthesia machines contaminated Anaesth. 2005;52:125.
3. Thorpe CM, Spence AA. Clinical evidence for delayed
with residual amounts of a volatile agent can, along with
chloroform poisoning. Br J Anaesth. 1997;79:402.
immune crossover, trigger AIH. 4. Dykes MHM. The early years: 18461912. Int Anesthesiol
The newer volatile agents are less likely than their Clin. 1970;8:175.
predecessors to cause liver injury. In the continued 5. JAMA. Summary of the National Halothane Study. Possible
quest for safer anesthetics, some clinical investigators association between halothane anesthesia and postoper-
have recently focused on the anesthetic, xenon. It is ative hepatic necrosis. Report by Subcommittee on the
inert, with minimal cardiovascular effects. It is not National Halothane Study of the Committee on Anesthesia,
metabolized by the liver, and therefore cannotaccording National Academy of Science. 1966;197:775.
to the immune theorycause AIH. Unfortunately, inert 6. Touloukian J, Kaplowitz N. Halothane-induced hepatic
gases are expensive and not cost-effective for routine disease. Semin Liver Dis. 1981;1:134.
7. Bunker JP. Final report of the National Halothane Study.
clinical use. Nevertheless, inert anesthetics will be useful
Anesthesiology. 1968;29:231.
tools to clarify the interrelationships between anesthetic 8. Neuberger J, Williams R. Halothane hepatitis. Dig Dis.
metabolism, immunopathology, and AIH. 1988;6:52.
9. Inman WH, Mushin WW. Jaundice after repeated exposure
to halothane: An analysis of Reports to the Committee on
Safety of Medicines. Br Med J. 1974;1:5.
KEY POINTS 10. Kenna JG, Jones RM. The organ toxicity of inhaled
anesthetics. Anesth Analg. 1995;81:S51.
1. AIH is a rare, idiosyncratic disorder that should not 11. Lewis JH, Zimmerman HJ, Ishak KG, et al. Enflurane
be diagnosed until the clinical team has excluded all hepatotoxicity. A clinicopathologic study of 24 cases. Ann
known or more likely causes of unexplained post- Intern Med. 1983;98:984.
operative fever and jaundice. AIH lacks any unique 12. Eger EI, Smuckler EA, Ferrell LD, et al. Is enflurane
or pathognomonic features to prove or disprove its hepatotoxic? Anesth Analg. 1986;65:21.
13. Carrigan TW, Straughen WJ. A report of hepatic necrosis
existence.
and death following isoflurane anesthesia. Anesthesiology.
2. The differential diagnosis of POLD encompasses the
1987;67:581.
many preexisting diseases (e.g., viral or autoim- 14. Sinha A, Clatch RJ, Stuck G, et al. Isoflurane hepatotox-
mune hepatitis) and perioperative disorders (e.g., icity: A case report and review of the literature. Am J
adverse drug reactions, mesenteric ischemia) that Gastroenterol. 1996;91:2406.
can progress rapidly from a subclinical entity to liver 15. Stoelting RK, Blitt CD, Cohen PJ, et al. Hepatic dysfunction
failure. after isoflurane anesthesia. Anesth Analg. 1987;66:147.
3. In patients with drug-induced hepatocellular injury, 16. Ogawa M, Doi K, Mitsufuji T, et al. Drug induced hepatitis
the development of jaundice should be considered life following sevoflurane anesthesia in a child. Masui. Jap J
threatening (Hys Law). Anesth. 1991;40:1542.
17. Shichinohe Y, Masuda Y, Takahashi H, et al. A case of
4. AIH is most likely an immune-mediated disease
postoperative hepatic injury after sevoflurane anesthesia.
that stems from the metabolism of anesthetics by
Masui. Jap J Anesth. 1992;41:1802.
cytochrome P450 2E1. Reactive metabolites of these 18. Watanabe K, Hatakenaka S, Ikemune K, et al. A case
anesthetics covalently bind to and transform liver of suspected liver dysfunction induced by sevoflurane
proteins into antigenic molecules. For reasons that anesthesia. Masui. Jap J Anesth. 1993;42:902.
remain unclear, the immune systemin a tiny 19. Kayashima K, Matsumoto T, Sata T, et al. A case of
fraction of the populationtargets these altered severe liver dysfunction after sevoflurane anaesthesia [in
macromolecules for destruction, leading to massive Japanese]. J Clin Anesth Jpn. 1995;19:53.
hepatic necrosis. 20. Martin JL, Plevak DJ, Flannery KD, et al. Hepatotoxicity
5. The most effective way to decrease the incidence of after desflurane anesthesia. Anesthesiology. 1995;83:1125.
AIH is to avoid the use of halothane. The incidence of 21. Neuberger J, Williams R. Halothane anaesthesia and liver
damage. Br Med J Clin Res Ed. 1984;289:1136.
halothane hepatitis in the subset of patients who have
22. Allan LG, Hussey AJ, Howie J, et al. Hepatic glutathione
had multiple or recent halothane anesthetics may be S-transferase release after halothane anaesthesia: Open
as high as 1 in 3,500. By comparison, AIH with the randomised comparison with isoflurane. Lancet. 1987;1:
newer halogenated vapors is extremely rare. 771.
600 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
23. Hussey AJ, Aldridge LM, Paul D, et al. Plasma glutathione 44. Klatskin G, Kimberg DV. Recurrent hepatitis attributable
S-transferase concentration as a measure of hepatocellu- to halothane sensitization in an anesthetist. N Engl J Med.
lar integrity following a single general anaesthetic with 1969;280:515.
halothane, enflurane or isoflurane. Br J Anaesth. 1988; 45. Sutherland DE, Smith WA. Chemical hepatitis associated
60:130. with occupational exposure to halothane in a research
24. Mushin W, Rosen M, Jones E. Post-halothane jaundice in laboratory. Vet Hum Toxicol. 1992;34:423.
relation to previous administration of halothane. Br Med J. 46. Neuberger J, Vergani D, Mieli-Vergani G, et al. Hepatic
1971;3:18. damage after exposure to halothane in medical personnel.
25. Voigt MD, Workman B, Lombard C, et al. Halothane Brit J Anaesth. 1981;53:1173.
hepatitis in a South African populationfrequency and the 47. Njoku DB, Greenberg RS, Bourdi M, et al. Autoantibodies
influence of gender and ethnicity. S Afr Med J. 1997;87:882. associated with volatile anesthetic hepatitis found in the
26. Kenna JG, Neuberger J, Mieli-Vergani G, et al. Halothane sera of a large cohort of pediatric anesthesiologists. Anesth
hepatitis in children. Br Med J Clin Res Ed. 1987;294:1209. Analg. 2002;94:243.
27. Warner LO, Beach TP, Garvin JP, et al. Halothane and 48. Eger EI II. Good news, bad news. Anesth Analg. 2002;94:239.
children: The first quarter century. Anesth Analg. 1984;63: 49. Wataneeyawech M, Kelly K. Hepatic diseasesunsuspected
838. before surgery. N Y State J Med. 1975;75:1278.
28. Brown BJ. General anesthetics and hepatic toxicity. Ariz 50. Schemel WH. Unexpected hepatic dysfunction found by
Med. 1977;34:5. multiple laboratory screening. Anesth Analg. 1976;55:810.
29. Kharasch ED, Thummel KE, Mautz D, et al. Clinical 51. Liu ZX, Kaplowitz N. Immune-mediated drug-induced liver
enflurane metabolism by cytochrome P450 2E1. Clin disease. Clin Liver Dis. 2002;6:755.
Pharmacol Ther. 1994;55:434. 52. Lee WM, Senior JR. Recognizing drug-induced liver in-
30. Garton KJ, Yuen P, Meinwald J, et al. Stereoselective jury: Current problems, possible solutions. Toxicol Pathol.
metabolism of enflurane by human liver cytochrome P450 2005;33:155.
2E1. Drug Metab Dispos. 1995;23:1426. 53. Bjornsson E, Jerlstad P, Bergqvist A, et al. Fulminant
31. Carpenter RL, Eger EI, Johnson BH, et al. The ex- drug-induced hepatic failure leading to death or liver trans-
tent of metabolism of inhaled anesthetics in humans. plantation in Sweden. Scand J Gastroenterol. 2005;40:1095.
Anesthesiology. 1986;65:201. 54. Bjornsson E, Olsson R. Outcome and prognostic markers in
32. Njoku D, Laster MJ, Gong DH, et al. Biotransformation of severe drug-induced liver disease. Hepatology. 2005;42:481.
halothane, enflurane, isoflurane, and desflurane to triflu- 55. Manov I, Motanis H, Frumin I, et al. Hepatotoxicity of anti-
oroacetylated liver proteins: Association between protein inflammatory and analgesic drugs: Ultrastructural aspects.
acylation and hepatic injury. Anesth Analg. 1997;84:173. Acta Pharmacol Sin. 2006;27:259.
33. Kharasch ED. Metabolism and toxicity of the new anes- 56. Larson AM, Polson J, Fontana RJ, et al. Acetaminophen-
thetic agents. Acta Anaesthesiol Belg. 1996;47:7. induced acute liver failure: Results of a United States
34. Clarke JB, Thomas C, Chen M, et al. Halogenated anes- multicenter, prospective study. Hepatology. 2005;42:1364.
thetics form liver adducts and antigens that cross-react 57. OGrady JG. Acute liver failure. Postgrad Med J. 2005;81:148.
with halothane-induced antibodies. Int Arch Allergy Imm. 58. Kwan D, Bartle WR, Walker SE. Abnormal serum transam-
1995;108:24. inases following therapeutic doses of acetaminophen in the
35. Shiraishi Y, Ikeda K. Uptake and biotransformation of absence of known risk factors. Dig Dis Sci. 1995;40:1951.
sevoflurane in humans: A comparative study of sevoflurane 59. Whitcomb DC, Block GD. Association of acetaminophen
with halothane, enflurane, and isoflurane. J Clin Anesth. hepatotoxicity with fasting and ethanol use. JAMA. 1994;
1990;2:381. 272:1845.
36. Kharasch ED. Biotransformation of sevoflurane. Anesth 60. Zimmerman HJ, Maddrey WC. Acetaminophen (paraceta-
Analg. 1995;81(Suppl 6):27. mol) hepatotoxicity with regular intake of alcohol: Analy-
37. Kharasch ED, Armstrong AS, Gunn K, et al. Clinical sevoflu- sis of instances of therapeutic misadventure. Hepatology.
rane metabolism and disposition. II. The role of cytochrome 1995;22:767.
P450 2E1 in fluoride and hexafluoroisopropanol formation. 61. Fries JF, Ramey DR, Singh G, et al. A reevaluation of
Anesthesiology. 1995;82:1379. aspirin therapy in rheumatoid arthritis. Arch Intern Med.
38. Kharasch ED, Karol MD, Lanni C, et al. Clinical sevoflurane 1993;153:2465.
metabolism and disposition. I. Sevoflurane and metabolite 62. Prescott LF. Liver damage with non-narcotic analgesics.
pharmacokinetics. Anesthesiology. 1995;82:1369. Med Toxicol. 1986;1(Suppl 1):44.
39. Mushlin PS, Gelman S. Liver dysfunction after anesthesia. 63. Singh G, Ramey DR, Morfeld D, et al. Comparative toxicity
In: Benumof JL, Saidman LJ, eds. Anesthesia and perioper- of non-steroidal anti-inflammatory agents. Pharmacol Ther.
ative complications, 2nd ed. St. Louis: Mosby; 1999:441. 1994;62:175.
40. Sigurdsson J, Hreidarsson AB, Thjodleifsson B. Enflurane 64. Garcia Rodriguez LA, Perez Gutthann S, Walker AM, et al.
hepatitis. A report of a case with a previous history of The role of non-steroidal anti-inflammatory drugs in acute
halothane hepatitis. Acta Anaesthesiol Scand. 1985;29:495. liver injury. Br Med J. 1992;305:865.
41. Christ DD, Satoh H, Kenna JG, et al. Potential metabolic 65. Garcia Rodriguez LA, Williams R, Derby LE, et al. Acute
basis for enflurane hepatitis and the apparent cross- liver injury associated with nonsteroidal anti-inflammatory
sensitization between enflurane and halothane. Drug Metab drugs and the role of risk factors. Arch Intern Med. 1994;
Dispos. 1988;16:135. 154:311.
42. Christ DD, Kenna JG, Kammerer W, et al. Enflurane 66. Lewis JH. Hys law, the Rezulin Rule, and other predictors
metabolism produces covalently bound liver adducts recog- of severe drug-induced hepatotoxicity: Putting risk-benefit
nized by antibodies from patients with halothane hepatitis. into perspective. Pharmacoepidemiol Drug Saf . 2006;15:221.
Anesthesiology. 1988;69:833. 67. Douglas HJ, Eger EI, Biava CG, et al. Hepatic necrosis
43. Hasan F. Isoflurane hepatotoxicity in a patient with a associated with viral infection after enflurane anesthesia.
previous history of halothane-induced hepatitis. Hepato- N Engl J Med. 1977;296:553.
gastroenterology. 1998;45:518. 68. Morley TS. Halothane hepatitis. JAMA. 1973;225:1659.
C H A P T E R 41/ A C U T E L I V E R D Y S F U N C T I O N A N D A N E S T H E S I A - I N D U C E D H E P AT I T I S 601
69. Vento S, Garofano T, Renzini C, et al. Fulminant hepatitis 85. Ko H, Yoshida EM. Acute fatty liver of pregnancy. Can J
associated with hepatitis A virus superinfection in patients Gastroenterol. 2006;20:25.
with chronic hepatitis C. N Engl J Med. 1998;338:286. 86. Kwo PY, Schlauder GG, Carpenter HA, et al. Acute hepatitis
70. Brechot C, Degos F, Lugassy C, et al. Hepatitis B virus DNA E by a new isolate acquired in the United States. Mayo Clin
in patients with chronic liver disease and negative tests for Proc. 1997;72:1133.
hepatitis B surface antigen. N Engl J Med. 1985;312:270. 87. Waar K, Herremans MM, Vennema H, et al. Hepatitis E is
71. Clark JM. The epidemiology of nonalcoholic fatty liver a cause of unexplained hepatitis in The Netherlands. J Clin
disease in adults. J Clin Gastroenterol. 2006;40:S5. Virol. 2005;33:145.
72. Takahashi A, Ohira H, Onizawa M, et al. A recovery 88. Khuroo MS, Kamili S. Aetiology, clinical course and
case of acute-onset autoimmune hepatitis presenting as outcome of sporadic acute viral hepatitis in pregnancy.
fulminant hepatic failure, who received living donor-liver J Viral Hepat. 2003;10:61.
transplantation. Intern Med. 2006;45:1217. 89. Singh S, Mohanty A, Joshi YK, et al. Mother-to-child
73. Kessler WR, Cummings OW, Eckert G, et al. Fulminant hep- transmission of hepatitis E virus infection. Indian J Pediatr.
atic failure as the initial presentation of acute autoimmune 2003;70:37.
hepatitis. Clin Gastroenterol Hepatol. 2004;2:625. 90. Gitlin N, Serio KM. Ischemic hepatitis: Widening horizons.
74. Hay JE, Czaja AJ, Rakela J, et al. The nature of unexplained Am J Gastroenterol. 1992;87:831.
chronic aminotransferase elevations of a mild to moderate 91. Seeto RK, Fenn B, Rockey DC. Ischemic hepatitis: Clinical
degree in asymptomatic patients. Hepatology. 1989;9:193. presentation and pathogenesis. Am J Med. 2000;109:109.
75. Teubner A, Tillmann HL, Schuppan D, et al. Prevalence of 92. Gibson PR, Dudley FJ. Ischemic hepatitis: Clinical features,
circulating autoantibodies in healthy individuals. Med Klin. diagnosis and prognosis. Aust N Z J Med. 1984;14:822.
2002;97:645. 93. Bynum TE, Boitnott JK, Maddrey WC. Ischemic hepatitis.
76. Schiodt FV, Atillasoy E, Shakil AO, et al. Etiology and Dig Dis Sci. 1979;24:129.
outcome for 295 patients with acute liver failure in the 94. Fuchs S, Bogomolski-Yahalom V, Paltiel O, et al. Ischemic
United States. Liver Transpl Surg. 1999;5:29. hepatitis: Clinical and laboratory observations of 34 pa-
77. Riely CA. Hepatic disease in pregnancy. Am J Med. 1994; tients. J Clin Gastroenterol. 1998;26:183.
96:18S. 95. Henrion J, Descamps O, Luwaert R, et al. Hypoxic hepatitis
78. Pereira SP, ODonohue J, Wendon J, et al. Maternal and in patients with cardiac failure: Incidence in a coronary
perinatal outcome in severe pregnancy-related liver disease. care unit and measurement of hepatic blood flow [see
Hepatology. 1997;26:1258. comments]. J Hepatol. 1994;21:696.
79. Jaiswal SP, Jain AK, Naik G, et al. Viral hepatitis during 96. Henrion J, Schapira M, Luwaert R, et al. Hypoxic hepatitis:
pregnancy. Int J Gynaecol Obstet. 2001;72:103. Clinical and hemodynamic study in 142 consecutive cases.
80. Weinstein L. Syndrome of hemolysis, elevated liver en- Medicine (Baltimore). 2003;82:392.
zymes, and low platelet count: A severe consequence of 97. Henrion J, Minette P, Colin L, et al. Hypoxic hepatitis
hypertension in pregnancy. Am J Obstet Gynecol. 1982;142: caused by acute exacerbation of chronic respiratory failure:
159. A case-controlled, hemodynamic study of 17 consecutive
81. Duffy BL. HELLP syndrome and the anaesthetist. Anaes- cases. Hepatology. 1999;29:427.
thesia. 1988;43:223. 98. Friedman L, Maddrey W. Surgery in the patient with liver
82. Portis R, Jacobs MA, Skerman JH, et al. HELLP syndrome disease. Med Clin North Am. 1987;71:454.
(hemolysis, elevated liver enzymes, and low platelets) patho- 99. Bloch RS, Allaben RD, Walt AJ. Cholecystectomy in patients
physiology and anesthetic considerations. AANA J. 1997; with cirrhosis. A surgical challenge. Arch Surg. 1985;120:
65:37. 669.
83. Chan AD, Gerscovich EO. Imaging of subcapsular hep- 100. Brown MW, Burk RF. Development of intractable ascites
atic and renal hematomas in pregnancy complicated by following upper abdominal surgery in patients with cirrho-
preeclampsia and the HELLP syndrome. J Clin Ultrasound. sis. Am J Med. 1986;80:879.
1999;27:35. 101. Andrade RJ, Lucena MI, Kaplowitz N, et al. Outcome of
84. Ibdah JA. Acute fatty liver of pregnancy: An update on acute idiosyncratic drug-induced liver injury: Long-term
pathogenesis and clinical implications. World J Gastroen- follow-up in a hepatotoxicity registry. Hepatology. 2006;44:
terol. 2006;12:7397. 1581.
G. ENDOCRINE
CHAPTER ACUTE HYPERGLYCEMIA
42
AND HYPOGLYCEMIA
CASE SUMMARY
What Is Diabetes Mellitus?
52-year-old man with a past medical his-
A
tory of diabetes mellitus presents with chest
Diabetes mellitus develops from the failure of the body
pain, shortness of breath, nausea, and a
to produce an adequate amount of insulin to avoid an
cold leg which has been symptomatic for
increase in serum glucose. Type 1 is an autoimmune
3 hours. Review of systems indicates that he
disease of the -pancreatic cells resulting in essentially
has polyuria. On physical examination, he is
little, if any, endogenous insulin production. However,
diaphoretic and tachypneic. He is slightly confused and
95% of diabetics have type 2 disease. In this disease, the
has fruity breath. His pulse is 120, and blood pressure is
patient has some inadequacy of insulin production, often
90/50. His breathing is labored and deep, with a respiratory
accompanied by resistance to the action of insulin.1
rate in the mid-30s. The cardiopulmonary examination is
The prevalence of diabetes has increased over the last
otherwise unremarkable. His skin is cool to the touch, but
decade. The 40% increase is almost entirely attributed
his right leg is mottled. No pulses are obtainable in that ex-
to type 2 diabetes. The trend has been associated with
tremity. His electrocardiogram shows some minor degree
an increase in obesity.1 In fact, there is some suggestion
of ST depression in leads II, III, and aVF that do not quite
that obesity, hypertension, and hyperlipidemia may be
reach 1 mm. His chest radiograph is unremarkable. Labo-
clustered together with insulin resistance as part of a
ratory abnormalities are: Glucose 655 mg per dL, sodium
metabolic syndrome. It is unclear whether patients with
132 mmol per L, potassium 5.5 mmol per L, magnesium
this syndrome carry an increase in complications such as
1.2 mmol per L, phosphate 3.0 mmol per L, creatinine
cardiovascular disease.2
3.2 mg per dL, troponin 0.7 ng per mL, and a blood gas pH
of 6.9 with a base excess of 21. In addition, urine is posi-
tive for ketones, and his serum betahydroxybutyrate is 2.9.
The patient receives 8 units of regular insulin as a bolus,
and will start on a continuous insulin infusion at a rate of What Is Diabetic Ketoacidosis?
4 units per hour. He receives judicious fluid resuscitation
with normal saline to maintain an adequate blood pres- Diabetic ketoacidosis (DKA) is the most common hyper-
sure and a urine output of at least 0.5 mL/kg/hour. Central glycemic emergency. The presence of a high serum blood
venous and arterial access are obtained. After a moder- glucose, low serum pH, and increase in ketone production
ate amount of fluid resuscitation, cautious introduction are the cornerstone of pathophysiology. The annual inci-
of intravenous metoprolol or esmolol will be attempted. dence of DKA per the Centers for Disease Control is 3 to
A heparin infusion will be started. Electrolytes will be cor- 8 per 1000 persons. The mortality from DKA is 2% to 5%
rected. While these therapies are ongoing and the patient in developed countries and 6% to 24% in underdeveloped
has some degree of stability, surgery to relieve the poor countries. These statistics have worsened over the last
perfusion to his right leg will be undertaken. Eventually, 20 years.3
the patient will reside in the intensive care unit (ICU). Although previous conventional wisdom reported that
He will have ongoing surveillance for cardiac ischemia, DKA only developed in type 1 diabetics, it does present
intravascular volume status, insulin administration, and in type 2 patients but with a much lower incidence. Only
electrolyte correction. approximately 5% to 10% of diabetic have type 1. The
602
C H A P T E R 4 2 / A C U T E H Y P E R G LY C E M I A A N D H Y P O G LY C E M I A 603
increased incidence of DKA is related to the increased in a hyperosmolar hyperglycemic state, hyperglycemia
prevalence of type 2 diabetes. DKA can occur in any may drop with fluid resuscitation alone. Insulin should
setting where a patient has absolute or relative insulin be considered in much smaller doses only after initial
deficiency.4 Type 1 diabetic patients have a near-total resuscitation has occurred in these patients. Once blood
absence of endogenous insulin. During times of stress, glucose falls below 250 mg per dL, fluid resuscitation
such as an infection, patients may fail to administer should continue, with dextrose in normal saline at a
the appropriate amount of exogenous insulin. Type 2 slower rate of 150 to 250 mL per hour. The endpoint
diabetic patients have peripheral insulin resistance, in for DKA should be clearing ketones from the urine, which
addition to varying degrees of inadequate endogenous in- is slower than serum. The endpoint for hyperosmolar
sulin production.5 In both situations, circulating carbohy- hyperglycemic state is when the plasma osmolarity is
drates are not utilized because of insufficient functioning <310 mOsm per kg.8
insulin. Patients frequently have hyperkalemia when they
In addition, increased levels of counterregulatory
present in DKA due to the extracellular potassium shifts
hormones are also present and result in insulin resistance.
secondary to acidosis. However, total body potassium is
Glucagon, catecholamines, cortisol, and growth hormone
depleted on average of 3 to 5 mmol per kg body weight.3
stimulate lipolysis, glycogenolysis, gluconeogenesis, and
Therefore, potassium supplementation should begin as
proteolysis. An increased stimulation of these ketogenic
soon as potassium level falls below 5.4 mEq per L.8 The
pathways develops. At some point, ketones cannot be
combination of clearing the acidosis and administering
utilized peripherally and begin to accumulate. As a
result, there is hyperglycemia and an accumulation of insulin will drive potassium into the cells, making it safe
ketoacids. -hydroxybutyric acid is the ketoacid typically to start replacing the deficit. It is important to assure that
measured in plasma.6 Proinflammatory cytokines have urine output is maintained, but potassium clearance is
also been shown to be elevated in DKA. Interleukin not affected unless the creatinine clearance is <15 mL
(IL)-10, IL-8, IL-6, IL-1, and tumor necrosis factor per minute. This is due to extensive tubular secretion
(TNF-) are increased and can further exacerbate of potassium, which exceeds its glomerular filtration
insulin resistance and aggravate hyperglycemia. After rate.9
treatment with insulin and resolution of acidosis, most The patient with DKA often also requires magnesium
levels of inflammatory mediators decrease. In addition, replacement as part of their treatment. Losses of this
markers of cardiovascular risk, such as homocysteine, are divalent cation are frequently underappreciated. While
also elevated and respond to insulin, which suggest the hypomagnesemia is certainly associated with deficiency, a
potential for increased risk of cardiovascular problems normal serum magnesium does not rule it out.10 The only
secondary to DKA.7 way to accurately assess total body magnesium is through
In contradistinction to DKA, the hyperosmolar hy- tissue levels that are not readily available. Diuresis re-
perglycemic state typically associated with type 2 diabetes sults in increased renal magnesium wasting, and most
usually results in more severe electrolyte disturbances patients with DKA have had an osmotic diuresis from
and glucouric-induced volume depletion. Owing to the hyperglycemia. In addition, up to 40% of hospitalized pa-
relatively minimal degree of acidosis, the patient experi- tients are likely to be magnesium-deficient, and 25% to
ences less severe symptoms and thus presents later in the 39% of diabetics in the general population are also likely
course. Therefore, blood sugar may be significantly higher to be magnesium-deficient secondary to chronic mag-
as is serum osmolarity, but the serum pH is relatively nor- nesiumuria.10 Therefore, a patient with DKA will need
mal. The initial therapy for DKA and hyperosmolar states magnesium replacement. Since magnesium is difficult to
is similar. Aggressive fluid resuscitation and electrolyte accurately measure, and a significant amount is lost in
replacement are paramount. Insulin therapy may be more
the urine despite ongoing deficiency, high doses (up to
important in DKA, but still has a role in hyperosmolar
8 g in 24 hours) are required.11 Rapidly infused magne-
states as well.8
sium is renally excreted at a high rate, and therefore slow
The physical examination in a patient with DKA
continuous infusion provides more complete replacement
frequently demonstrates mental status changes and car-
of deficits. Significant renal dysfunction, which would
diovascular effects that are consistent with dehydration.
limit potassium replacement, should also limit magne-
These can be extreme, depending on the degree of de-
hydration, but more so depending on the severity of the sium replacement. Recent investigation into the effects of
inciting cause of the episode. The deep, rapid breathing magnesium deficiency suggests that it may play a role in
secondary to acidosis is called Kussmaul respirations. Di- insulin resistance and producing diabetic complications
abetic keotacidosis can also be associated with delayed such as cardiovascular disease.10
gastric emptying or intestinal ileus.8 DKA is commonly associated with phosphate deple-
Initial fluid resuscitation should include a 1-L bolus tion that can be as high as 1.5 mmol per kg body weight.3
of normal saline over an hour or less, followed by 250 to Patients can present initially with hyperphosphatemia,
500 mL per hour continuous infusion. Blood glucose which is also related to cellular shifts that will reverse
should be checked hourly during resuscitation. For a with the initiation of DKA therapy. Phosphate repletion
patient in DKA, insulin should also be started with a bolus does not need to be prophylactic, but should be initiated
of 0.15 units per kg, followed by an infusion. Table 42.1 is once the level falls to <1.5 mg per dL to prevent complica-
one example of an insulin infusion protocol. For patients tions such as white cell, red cell, and platelet dysfunction,
604 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
shock if the blood glucose is 150 to 180 mg per dL or DCCT trial were all diabetic patients managing their own
higher.21 Trauma patients are also a group that has shown insulin at home. Over time, diabetic patients will have
some survival benefit from better glycemic control. These a blunted response of their counterregulatory system to
patients may be more susceptible to the deleterious effects hypoglycemia and will display only minimal elevation in
of hyperglycemia than other patients in a surgical ICU.22,23 glucagon and epinephrine.28 As a result, patients will not
In 2001, Professor Van den Berghe et al. released a only have a suboptimal increase in serum glucose, but also
study performed in surgical ICU patients where blood will not experience warning symptoms of hypoglycemia.
glucose was maintained at or below 110 mg per dL.24 Using intensive insulin therapy in hospitalized pa-
Before this, maintaining the serum glucose below 200 mg tients usually requires a continuous insulin infusion.
per dL was generally considered satisfactory. Essentially, Therefore, the patient is likely to require hourly blood glu-
the goal was to avoid problems such as osmotic diuresis, cose measurement so that a trend toward hypoglycemia
electrolyte abnormalities, and acidbase imbalances as- can be identified earlier, and either treated or avoided.
sociated with progressively uncontrolled hyperglycemia, Certainly, close attention is mandated in a longer term
with the added benefit that neutrophil and macrophage diabetic patient. A subsequent retrospective comparative
function was likely improved at a lower level as well.25 study by Krinsley in 2004 used a more gentle goal of
Aggressive insulin management represents a paradigm 140 mg per dL. There was no difference between groups
shift. in the rate of hypoglycemia to the level of 40 mg per dL or
In this prospective, randomized study that was less, and only 0.5% increase to the level of 40 to 59 mg per
dominated by postoperative cardiac surgical patients dL. Some of the beneficial effects, including a decrease
(approximately two thirds of the patients), all of whom in mortality, renal failure, and blood transfusion, were
initially required mechanical ventilation, one group of seen despite the less aggressive goal.29 Given the data in
patients were maintained between 180 mg per dL and the preceding text, a mixed goal may be an appropriate
200 mg per dL (conventional) compared to the other compromise. A patient in the operating room or short stay
group that was maintained no higher than 110 mg in the ICU could be maintained at a level of 140 mg per
per dL. Mortality in the ICU was decreased by 32%. dL or less. Once the ICU duration becomes >3 to 5 days,
Overall in-hospital mortality was also decreased by 34%. a more intensive insulin regimen may be warranted given
Bloodstream infections, renal failure requiring dialysis or that these are the patients that showed the most benefit.30
hemofiltration, red cell transfusion, and critical illness Most of the data available assesses patients who re-
polyneuropathy were all decreased by 40% to 50%. quire an ICU admission. There are no data in perioperative
There was also a tendency to require the ventilator patients who have not had a stroke, myocardial infarction,
for fewer days. All the patients were in the surgical or required an ICU stay. A few studies have documented
ICU, and therefore were likely undergoing some of the the decrease in complications from similar endpoints to
above alterations in their hormonal milieu. Because those described in the preceding text when glucose control
most of the patients were not diabetic, they were likely is maintained perioperatively in addition to the ICU ad-
experiencing postoperative insulin resistance.24 This is mission. In one study, the benefit was concluded to be re-
somewhat similar to what was found in the DCCT trial lated to the intraoperative management alone because the
in that any degree of normoglycemia maintained at any level of control was similar in the ICU within 12 hours.31
time point in patients who were either newly diagnosed It is not clear whether insulin or normoglycemia
or long-term, poorly controlled diabetics produced some is producing whatever benefit is seen. A regression
benefit.19 This is the first time that prospective evidence analysis of Van den Berghes study showed that it was
showed benefit in a nondiabetic patient. the reduction of hyperglycemia, and not the amount
Despite some methodologic questions regarding of insulin, that was associated with an improvement
mode of feeding and overall morbidity and mortality in mortality.32 Finney et al. not only showed that
ratios, the study by Van den Berghe et al., as well as a com- hyperglycemia was associated with adverse outcomes
panion follow-up in medical ICU patients, has driven the in surgical ICU patients, but that increasing insulin
evolving recommendations for improved glycemic control doses potentially increased mortality.33 Although it is
in perioperative patients.26 An ongoing Australian and understood that insulin decreases hyperglycemia, other
New Zealand Intensive Care Society Clinical Trials Group effects of insulin are less well understood. Insulin may
and Canadian Critical Care Trials Group study is prospec- have beneficial effects on inflammation based on cytokine
tively randomizing 5,000 ICU patients to further guide profiles as reviewed earlier in this chapter, but it may
practitioners as to the level of beneficial glycemic control aggravate others such as nuclear factor B upregulation.34
in critically ill patients.27
The major drawback to keeping blood glucose at
near normal is an increase in hypoglycemia. In Van den
Berghes study, hypoglycemia was defined as a level 40 mg What Are the Signs and
per dL or less; 39 of 765 patients had hypoglycemia in Symptoms of Hypoglycemia
the intensive insulin therapy group compared to 6 in
the conventionally treated group.24 The worst symptoms and How Should It Be Treated?
reported from this hypoglycemia were sweating and
agitation. Increased episodes of hypoglycemia were also Hypoglycemia results in two classes of symptoms. Neu-
found in the DCCT trial. However, the patients in the rogenic symptoms are related to the response of the
C H A P T E R 4 2 / A C U T E H Y P E R G LY C E M I A A N D H Y P O G LY C E M I A 607
autonomic nervous system to low blood sugar. These in- cerebral ischemia. In addition, abnormal serum glucose
clude tremulousness, palpitations, anxiety, hunger, sweat- has a direct effect on axonal loss due to an increased
ing, and paresthesias. Neuroglycopenic symptoms are axonal intracellular fluid and a decrease in microfil-
related to low blood sugar in the brain. These include aments.40 Poor glycemic control also exacerbates the
confusion, weakness, a sensation of warmth, cognitive intracellular acidosis that already exists during brain is-
failure, seizure, and coma. Part of the blunted response to chemia through an increase in anaerobic metabolism.43
hypoglycemia in a diabetic patient may also be due to an These effects may explain the increase in critical illness
alteration in the brain of the hypoglycemic threshold.35 polyneuropathy in hyperglycemic ICU patients and wors-
Patients who progress to coma are then likely to suffer ened outcome in patients with hyperglycemic stroke.
from other factors related to loss of consciousness such Hyperglycemia causes dilation of the afferent arteri-
as lack of airway protection and respiratory depression, ole leading to glomerular hyperfiltration in the kidney, and
which can result in death. Six percent of all deaths in results in an elevated hydrostatic pressure. Over time, the
young diabetic patients who have no other health prob- extracellular matrix in the mesangial area expands, caus-
lems may be related to nocturnal hypoglycemia.36 ing a decline in the glomerular filtration rate. This helps
Oral treatment of hypoglycemia includes a number to explain why angiotensin-converting enzyme blockers
of options. A few teaspoons or packets of sugar in a have slowed the progression of diabetic nephropathy.40
glass of water provide a 20 g dose of carbohydrate in the Hyperglycemia can lead to glycation of hemoglobin and
form of D-glucose which can increase blood glucose by alteration in membrane phospholipids, which results in
approximately 50 mg per dL in 15 to 20 minutes. Other less deformability, slower transit time, and a shorter
supplemental carbohydrate sources include lifesavers, life span of erythrocytes.44 Normalizing serum glucose
honey, juice, and milk, but the onset is longer than from the 200 to 250 mg per dL, ranging <120 mg per
15 to 20 minutes, and the increase in glycemic level is dL, can dramatically improve polymorphonuclear (PMN)
less.37 If a patient is not alert enough to take oral glucose, leukocyte phagocytic capability. There is an association
1 amp of D50 provides approximately 25 g of carbohydrate between elevated serum glucose and elevated PMN in-
intravenously and averages an increase of approximately tracellular calcium causing this dysfunction.45 Clearly, a
160 mg per dL, with a range of 50 to 300 mg per dL.38 D50 difference exists in the potential for infection with im-
can be administered through a peripheral intravenous line paired PMN phagocytosis. An increased tendency toward
in an adult. Finally, 1 mg of glucagon may be administered ventilator-associated pneumonia may be partly because
subcutaneously, with an increase of approximately 150 mg of the delayed extubation in patients with poor glycemia
per dL in blood glucose within 60 minutes.39 control. However, this may be an incomplete picture of
pulmonary complications associated with hyperglycemia.
For instance, patients with idiopathic interstitial pneumo-
nia (a form of pulmonary fibrosis) have a tendency to be
What Are the Effects diabetic.46
of Hyperglycemia on Different
Organ Systems?
How Are the Perioperative
The vascular system plays a major role in the development Risks Different for Diabetic
of retinal, renal, and cardiac diseases, and is often
the most commonly cited target when considering the Patients?
effects of hyperglycemia. Much of the problem has been
associated with the increase in free radical oxygen species. Patients with diabetes have a number of disease processes
It is likely much more of a multifactorial phenomenon for which they are at increased risk compared to the gen-
that includes thickening of the endothelial basement eral population. These include retinopathy, nephropathy,
membrane, decreasing the degree of endothelium-derived neuropathy, delayed wound healing, stroke, myocardial
nitric oxideinduced vasodilatation and changing the infarction (MI), aspiration pneumonia, and sudden death.
actions of smooth muscle cells which leads to atheroma Although some of these processes may have variable im-
evolution.40,41 pact, changes in vision are the least likely around the
There are a number of less, well known effects time of surgery. Prospective study of perioperative vision
that become interesting when considering the impact changes has not documented diabetes as a significant risk
of short term glycemic control. For instance, diabetic factor.47 Wound healing is a concern in diabetic patients
patients are 75% more likely than other age-matched because it is five times more likely to be complicated by a
controls to develop cardiomyopathy that is independent bacterial or fungal infection than in nondiabetic patients.
of atherosclerotic disease. In fact, hyperglycemia alone This is not a problem in superficial wounds requiring
can lead to the altered function of the ryanodine receptor, epidermal repair if normoglycemia is maintained. How-
as well as endoplasmic reticulum calcium ATPase that ever, all impaired healing processes cannot be corrected
can be responsible for the decease in cardiac function by maintaining a normal serum glucose if the wound is
that develops over time.42 deeper and involves collagen formation.48
Certainly, vascular atherosclerotic disease will exac- Nephropathy evolves over time in patients with
erbate the damage of diabetes on distal neuropathy and diabetes. The earliest form is microalbuminuria that
608 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
An oral agent should only be used in hospitalized pa- TABLE 42.4 Medications with Significant Potential
tients when their oral intake has been established to avoid to Alter Glucose Tolerance
hypoglycemia.
An insulin infusion is best for patients whose glycemic Tendency Tendency
control will be relatively constant, in addition to providing toward toward
an amount of basal insulin for those patients who may Medication Hyperglycemia Hypoglycemia
require it to prevent ketoacidosis. For instance, patients
ACE inhibitor +
on parenteral nutrition or continuous tube feedings
agonists +
and/or are receiving steroids throughout a 24-hour period
-blocker +
are going to experience ongoing sources of glucose
agonists ++ +
generation. An example of an insulin infusion algorithm
-blockers + ++
is shown in Table 42.1. Steroids tend to increase insulin
Calcium channel ++
resistance and hepatic glucose production. As has already
blockers
been reviewed earlier in this chapter, infection, sepsis, and
Corticosteroids +++
stress also tend to increase hyperglycemia. An elevated
Cyclosporine ++
serum glucose itself can also increase insulin resistance.
Diazoxide +++
Therefore, patients may not only need continuous insulin
Disopyramide ++
but also require a significant amount, often reaching
Fibric acid + +
double digits every hour. It can be much easier to capture
derivatives
the patient if an infusion is utilized. If patients are
Lidocainea +
not receiving any feedings, they will need a source of
Lithium + +
glucose, such as some fluid with D5, once the blood
Minoxidil ++
sugar decreases below 250 mg per dL to help prevent
Monoamine oxidase ++
the development of ketosis. Chromium deficiency may
inhibitor
participate in the development of insulin resistance, but
Niacin ++
no conclusive evidence exists to substantiate its use in
Octreotide + +
type 1, type 2, steroid-induced, or gestational diabetes.69
Oral contraceptives +
Pentamidine +++ ++
Phenothiazines +
Phenytoin +
Who Can Become Quinineb +++
Hypoglycemic? Salicylatesa ++
Streptozotocin +++
Hypoglycemia is most often considered in relation to the Sulfamethoxyzole +
administration of insulin or an oral hypoglycemic agent. Thiazide diuretics +++
There are a number of other situations in which a patient Thyroid +
may experience hypoglycemia. The most common cause a Documented only in overdose.
in a healthy appearing patient is related to medications. b Associated with malaria.
Please see Table 42.4 for a list of medications that alter ACE, angiotensin-converting enzyme.
glucose tolerance. In addition, an insulinoma or factitious Data from: Pandit MK, Burke J, Gustafson AB, et al. Drug-induced
hypoglycemia should be considered. disorders of glucose tolerance. Ann Intern Med 1993;118:529.
The two situations in which a hypoglycemic disorder
can be ruled out are as follows. One is a normal plasma level. Blunted responses to glucagon are related to liver
glucose level obtained during symptoms or in the absence dysfunction due to the inability to store glycogen, such as
of signs or symptoms during a witnessed 72-hour fast.70 in liver failure and critical illness.70 A few rare conditions
Many patients have symptoms without having one of the to consider in an adult include glycogen storage dis-
rare disorders causing hypoglycemia. Insulin, C-peptide, ease, hypopituitarism, galactosemia, carnitine deficiency,
and proinsulin levels are increased from an insulinoma, sepsis, renal failure, congestive heart failure, starvation,
as well as secondary to sulfonylureas. Obviously, the two and insulin-antibody hypoglycemia. Disorders to consider
can be distinguished by checking serum sulfonylurea in children include small for gestational age, Bechwith-
levels. Factitious hypoglycemia from exogenous insulin Wiedemann syndrome, erythroblastosis fetalis, and Reye
is identified by significant suppression of C-peptide levels. syndrome.70
This is most often seen in female health care workers.70
Patients with coexisting disease carry a much more
broad potential of hypoglycemic disorders. Hypoglycemia
related to ethanol is associated with suppressed in- KEY POINTS
sulin and C-peptide levels. Non-cell tumors causing
low plasma glucose also have an elevated insulin-like, 1. The incidence of type 2 diabetes continues to
growth factor. Adrenal insufficiency can produce de- grow and appears to be reaching near-epidemic
creased glucose in the serum associated with a low cortisol proportions.
C H A P T E R 4 2 / A C U T E H Y P E R G LY C E M I A A N D H Y P O G LY C E M I A 611
2. A significant number of type 2 diabetic patients are 4. Newton CA, Raskin P. Diabetic ketoacidosis in Type 1 and
Type 2 diabetes mellitus. Arch Intern Med. 2004;164:1925.
unaware of their diagnosis and may be first identified
5. Chan JL, Abrahamson MJ. Pharmacological management
during the perioperative period. of Type 2 diabetes mellitus: Rationale for rational use of
3. Ninety-five percent of diabetes mellitus is re- insulin. Mayo Clin Proc. 2003;78:459.
lated to insulin resistance and associated with 6. Foster DW, McGarry JD. The metabolic derangements and
obesity. treatment of diabetic ketoacidosis. N Engl J Med. 1983;309:
4. Although type 1 diabetes is far more commonly 159.
associated with DKA, it can present in type 2 diabetic 7. Stentz FB, Umpierrez GE, Cuerro R, et al. Proinflammatory
patients. cytokines, markers of cardiovascular risk, oxidative stress,
5. DKA continues to be associated with significant and lipid peroxidation in patients with hyperglycemic crisis.
morbidity and mortality. Diabetes. 2004;53:2079.
8. Goldberg PA, Insucchi SE. Critical issues in endocrinology.
6. Glucagon, cathecholamines, and growth hormone
Clin Chest Med. 2003;24:583.
can increase insulin resistance. 9. Leaf A, Camara AA, Albertson B. Renal tubular secretion of
7. DKA and a hyperosmolar hyperglycemic state re- potassium in man. J Clin Invest. 1949;28:1526.
quire the aggressive replacement of intravascular 10. Hans CP, Sialy R, Bansal DD. Magnesium deficiency and
volume, and the electrolyte replacement of potas- diabetes mellitus. Curr Sci. 2002;83:1456.
sium, magnesium, and phosphate, as well as variable 11. Hamill-Ruth RJ, McGory R. Magnesium repletion and its
amounts of insulin. effect on potassium homeostasis in critically ill adults. Crit
8. Infection is the most common cause of DKA; other Care Med. 1996;24:38.
causes include surgery, trauma, burns, myocardial 12. Viallon A, Zeni F, Lafond P, et al. Does bicarbonate therapy
ischemia, stroke, pancreatitis, thyroid storm, and improve the management of severe diabetic ketoacidosis?
Crit Care Med. 1999;27:2690.
noncompliance with diabetic medications.
13. Glaser N, Barnett P, McCaslin I, et al. Risk factors for cerebral
9. Current recommendations for the optimal level of edema in children with diabetic ketoacidosis. N Engl J Med.
perioperative serum glucose control range from 80 2001;344:264.
to 140 mg per dL, with some advocating the higher 14. Joshi N, Caputo GM, Weitekamp MR, et al. Infections in
level to avoid hypoglycemic complications while patients with diabetes mellitus. N Engl J Med. 1999;341:1906.
providing the best risk:benefit ratio. 15. Ramsey SD, Newton K, Blough D, et al. Incidence, outcomes,
10. Bloodstream infections, renal failure, red cell trans- and cost of foot ulcers in patients with diabetes. Diabetes
fusion requirements, critical illness polyneuropathy Care. 1999;22:382.
incidence, number of ventilator-free days, and mor- 16. Desborough JP. The stress response to trauma and surgery.
tality were all improved with better glucose control Br J Anaesth. 2000;85:109.
17. Schricker T, Gougeon R, Eberhart L, et al. Type 2 di-
in selected perioperative patients.
abetes mellitus and the catabolic response to surgery.
11. The rule of 1,500 describes a patients insulin sen-
Anesthesiology. 2005;102:320.
sitivity. 1,500 Dividing by a patients total daily 18. Takala J, Ruokonen E, Webster NR, et al. Increased mortality
insulin requirement yields the mg per dL glu- associated with growth hormone treatment in critically ill
cose decrease expected from each unit of insulin adults. N Engl J Med. 1999;341:785.
administered. 19. The Diabetes Control and Complications Trial Research
12. Twenty grams of D-glucose (2 packets of sugar) can Group. The effect of intensive treatment of diabetes on the
increase serum blood glucose PO 50 mg per dL in development and progression of long term complications in
20 minutes; alternatively, 1 amp of D50 can increase insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:
serum blood glucose approximately 160 mg per dL. 977.
20. Weir CJ, Murray GD, Dyker AG, et al. Is hyperglycemia an
13. Diabetic patients are at higher risk for neuropa-
independent predictor of poor outcome after acute stroke?
thy, nephropathy, delayed wound healing, stroke,
Br Med J. 1997;314:1303.
myocardial infarction, aspiration pneumonia, and 21. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia
death during the perioperative period than their and increased risk of death after myocardial infarction in
nondiabetic counterparts. patients with and without diabetes. Lancet. 2000;355:773.
14. Epidural anesthesia can suppress counterregula- 22. Laird AM, Miller PR, Kilgo P, et al. Relationship of early hy-
tory hormones from surgery and decrease hyper- perglycemia to mortality in trauma patients. J Trauma. 2004;
glycemia. 56:1058.
15. Insulin is ideally used as an infusion during the 23. Vogelzang M, Nijboer JMM, van der Horst ICC, et al.
perioperative period. Hyperglycemia has a stronger relation with outcome in
trauma patients than in other critically ill patients. J Trauma.
16. Medications are the most common cause of hypo-
2006;60:873.
glycemia.
24. Van Den Berghe G, Wouters P, Weekers F, et al. Intensive
insulin therapy in critically ill patients. N Engl J Med.
REFERENCES 2001;345:1359.
25. Rayfield EJ, Ault MJ, Keusch GT, et al. Infection and diabetes:
1. Permutt MA, Wasson J, Cox N. Genetic epidemiology of The case for glucose control. Am J Med. 1982;72:439.
diabetes. J Clin Invest. 2005;115:1431. 26. American Association of Clinical Endocrinologists. Posi-
2. Kahn R, Buse J, Ferrannini E, et al. The metabolic syndrome: tion statement on inpatient diabetes and metabolic control.
Time for a critical appraisal. Diabetes Care. 2005;28:2289. Available at: http://www.aace.com/pub/positionstatements/
3. Lebovitz HE. Diabetic ketoacidosis. Lancet. 1995;345:767. Accessed July 6, 2006.
612 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
27. The George Institute. Normoglycemia in intensive care 50. Juul AB, Wetterslev J, Kofoed-Enevoldsen A. Long term
evaluation and survival using glucose algorithm regu- postoperative mortality in diabetic patients undergoing
lation (NICE-SUGAR STUDY). Available at: http://www major non-cardiac surgery. Eur J Anaesthesiol. 2004;21:
.clinicaltrials.gov/ct/gui/show/NCT00220987 Accessed July 6, 523.
2006. 51. Watkins PJ, Thomas PK. Diabetes mellitus and the nervous
28. Cryer PE, Gerich JE. Glucose counterregulation, hypo- system. J Neurol Neurosurg Psychiatry. 1998;65:620.
glycemia, and intensive insulin therapy in diabetes mellitus. 52. Marik PE. Aspiration pneumonitis and aspiration pneumo-
N Engl J Med. 1985;313:232. nia. N Engl J Med. 2001;344:665.
29. Krinsley JS. Effect of an intensive glucose management 53. Haffner SM, Lehto S, Ronnemaa T, et al. Mortality from
protocol on the mortality of critically ill adult patients. Mayo coronary heart disease in subjects with type 2 diabetes and
Clin Proc. 2004;79:992. in nondiabetic subjects with and without prior myocardial
30. Malhotra A. Intensive insulin in intensive care. N Engl J Med. infarction. N Engl J Med. 1998;339:229.
2006;354:516. 54. Aguilar D, Solomon SD, Kober L, et al. Newly diagnosed and
31. Gandhi GY, Nuttall GA, Abel MD, et al. Intraoperative hy- previously known diabetes mellitus and one year outcomes
perglycemia and perioperative outcomes in cardiac surgery of acute myocardial infarction. Circulation. 2004;110:1572.
patients. Mayo Clin Proc. 2005;80:862. 55. Juul AB, Wetterslev J, Gluud C, et al. Effect of perioper-
32. Van Den Berghe G, Wouters PJ, Bouillon R, et al. Outcome ative beta blockade in patients with diabetes undergoing
benefit of intensive insulin therapy in the critically ill: Insulin major non-cardiac surgery: Randomized placebo controlled,
dose versus glycemia control. Crit Care Med. 2003;31:359. blinded multicentre trial. Br Med J. 2006;332:1482.
33. Finney SJ, Zekveld C, Elia A, et al. Glucose control and 56. McAnulty GR, Robertshaw HJ, Hall GM. Anaesthetic man-
mortality in critically ill patients. JAMA. 2003;290:2041. agement of patients with diabetes mellitus. Br J Anaesth.
34. Montagnani M, Golovchenko I, Kim I, et al. Inhibition of 2000;85:80.
phosphatidylinositol 3-kinase enhances mitogenic actions 57. Enquist A, Brandt MR, Fernandes A, et al. The blocking
of insulin in endothelial cells. J Biol Chem. 2002;277: effect of epidural analgesia on the adrenocortical and
1794. hyperglycemic responses to surgery. Acta Anesthesiol Scand.
35. Cryer PE. Symptoms of hypoglycemia, thresholds for their 1977;21:330.
occurrence, and hypoglycemia unawareness. Endocrinol 58. Bromage PR, Shibata HR, Willoughby HW. Influence of
Metab Clin North Am. 1999;28:495. prolonged epidural blockade on blood sugar and cortisol
36. Sovik O, Thordarson H. Dead-in-Bed syndrome in young responses to operations upon the upper part of the abdomen
diabetic patients. Diabetes Care. 1999;22:1240. and thorax. Surg Gynaecol Obstet. 1971;21:330.
37. Brodows RG, Williams C, Amatruda JM. Treatment of insulin 59. Traynor C, Hall GM. Endocrine and metabolic changes
reactions in diabetics. JAMA. 1984;252:3378. during surgery: Anaesthetic implications. Br J Anaesth. 1981;
38. Adler PM. Serum glucose changes after administration of 53:153.
50% dextrose solution: Pre-and in-hospital calculations. Am 60. Halter JB, Pflug AE. Effect of sympathetic blockade by
J Emerg Med. 1986;4:504. spinal anesthesia on pancreatic islet cell function in man.
39. Cryer PE, Fisher JN, Shamoom H. Hypoglycemia. Diabetes Am J Physiol. 1980;239:E150.
Care. 1994;17:734. 61. Peyton PJ, Myles PS, Silbert BS, et al. Perioperative epidural
40. Sheetz MJ, King GL. Molecular understanding of hyper- analgesia and outcome after major abdominal surgery in
glycemias adverse effects for diabetic complications. JAMA. high-risk patients. Anesth Analg. 2003;96:548.
2002;288:2579. 62. Schricker T. The catabolic response to surgery: How can it be
41. Creager MA, Luscher TF, Consentino F, et al. Diabetes and modified by the anesthesiologist? Can J Anaesth. 2001;48:R1.
vascular disease: Pathophysiology, clinical consequences, 63. Metz SA, Halter JB, Robertson RP. Induction of defective
and medical therapy. Circulation. 2003;108:1527. insulin secretion and impaired glucose tolerance by cloni-
42. Poornima IG, Parikh P, Shannon RP. Diabetic cardiomyopa- dine. Selective stimulation of metabolic alpha adrenergic
thy: The search for a unifying hypothesis. Circ Res. 2006; pathways. Diabetes. 1978;27:554.
98:596. 64. Kitamura A, Hoshino T, Kon T, et al. Patients with diabetic
43. Wass CT, Lanier WL. Glucose modulation of ischemic brain neuropathy are at risk of a greater intraoperative reduction
injury: Review and clinical recommendations. Mayo Clin in core temperature. Anesthesiology. 2000;92:1311.
Proc. 1996;71:801. 65. Umpierrez GE, Isaacx SD, Bazargan N, et al. Hyperglycemia:
44. Babu N, Singh M. Influence of hyperglycemia on aggrega- An independent marker of in-hospital mortality in pa-
tion, deformability and shape parameters of erythrocytes. tients with undiagnosed diabetes. J Clin Endocrinol Metab.
Clin Hematol Microcirc. 2004;31:273. 2002;87:978.
45. Alexiewicz JM, Kumar D, Smogorzewski M. Polymorphonu- 66. Queale WS, Siedler AJ, Brancati FL. Glycemic control and
clear leukocytes in non-insulin dependent diabetes mellitus: sliding scale insulin use in medical inpatients with diabetes
Abnormalities in metabolism and function. Ann Intern Med. mellitus. Arch Intern Med. 1997;157:545.
1995;123:919. 67. Ratner RE, Hirsch IB, Neifing JL, et al. Less hypoglycemia
46. Usuki J, Enomoto T, Azuma A, et al. Influence of hyper- with insulin glargine in intensive insulin therapy for type 1
glycemia to the severity of pulmonary fibrosis. Chest. 2001; diabetes. Diabetes Care. 2000;23:639.
120:71S. 68. Glister BC, Vigersky RA. Perioperative management of type 1
47. Warner ME, Fronapfel PJ, Hebl JR, et al. Perioperative vision diabetes mellitus. Endocrinol Metab Clin North Am. 2003;
changes. Anesthesiology. 2002;96:855. 32:411.
48. King L. Impaired wound healing in patients with diabetes. 69. Cefalu WT, Hu FB. Role of chromium in human health and
Nurs Stand. 2001;15:39. diabetes. Diabetes Care. 2004;27:2741.
49. Nathan DM. Long-term complications of diabetes mellitus. 70. Service FJ. Hypoglycemic disorders. N Engl J Med. 1995;
N Engl J Med. 1993;328:1676. 332:1144.
CHAPTER DISORDERS OF THE ADRENAL
43
GLAND
A
emergency department with a history of re- The right vein is only a few millimeters long and enters
current headaches, sweating, palpitations, the vena cava; the left vein is longer and enters the left
tightness of the chest, and severe hyper- renal vein. One potential complication of removing the
tension. She has been treated with benzo- left adrenal gland is the risk of inadvertent ligation of the
diazepines by her primary care physician. apical renal arterial branch to the upper pole, which lies in
Her past medical history is otherwise negative. During close contact to the inferior border of an adrenal tumor.
physical examination, she was found to have an excep- A complication of removing the right adrenal gland is
tionally high blood pressure of 230/120 mm Hg, with a injury to the vena cava.
heart rate of 100 beats per minute; temperature of 37 C;
oxygen saturation 98%; lungs were clear without wheezes
or rales; cardiac rhythm was regular without obvious
murmur; abdomen was soft and nontender; neurology BIOCHEMISTRY
consult-nonfocal. Is this another anxiety attack? AND PHYSIOLOGY
The adrenal cortex constitutes approximately 70% of the
adrenal gland and is made up of three layers, which are
distinct in regard to histology, cellular organization, and
What Baseline Knowledge their relationship to blood supply. The three layers, from
about the Adrenal Gland the outside in, are:2
Is Relevant? ZONA GLOMERULOSA: This layer comprises approxi-
mately 15% of the gland and is the main site of
mineralocorticoid production. The major mineralocor-
ticoid is aldosterone (95%), with the balance being
ANATOMY made up by corticosterone and deoxycorticosterone.
ZONA FASICULATA: This layer is the major part of the
The adrenal glands lie alongside the upper part of each
gland and is the primary site for glucocorticoid produc-
kidney. The adrenal gland is alternatively referred to as
tion (mainly cortisol and, to a much lesser extent, cor-
the suprarenal gland.1 The right adrenal gland lies between
tisone).
the inferior vena cava and the right crus of the diaphragm,
ZONA RETICULARIS: This is the site of both anabolic and
its right border projecting to the right of the vena cava
sex hormone synthesis. The adrenal steroids contain
and its upper part coming into contact with the bare area
either 19 or 21 carbon atoms. C19 steroids are mainly an-
of the liver. Only its lower half has a peritoneal covering.
drogenic in activity, whereas the C21 steroids have either
The left adrenal gland is crescent shaped and drapes the
glucocorticoid or mineralocorticoid properties. The
medial border of the left kidney above the hilum. Its lower
hormones are transported bound to plasma proteins.
pole is covered in front by the body of the pancreas and
the splenic artery. The rest of the gland is covered by Cortisol binds to -globulin (transcortin or cortisol-
peritoneum and forms part of the stomach bed. It lies on binding globulin),3 a high affinity, low-capacity system.
the left crus of the diaphragm. The medial border is to Normally <5% of circulating cortisol is unbound. Small
the left of the celiac ganglion and overlapped by the left amounts are carried bound to albumin, which is a low-
gastric vessels. affinity, high-capacity protein.4 Only the unbound cortisol
613
614 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
and its biologically inactive metabolites can be filtered at release from the anterior pituitary. ACTH is derived from
the glomerulus. The average production rate of cortisol the prohormone, propiomelanocortin, which undergoes
is 12 to 15 mg/m2 /day. Cortisol is rapidly metabolized, complex processing within the pituitary to produce ACTH
mainly in the liver. One of the main enzymes that and a number of other peptides.7 Many of these pep-
regulates cortisol metabolism is 11 -hydroxysteroid tides, including ACTH, contain melanocyte-stimulating
dehydrogenase. The major metabolites are formed by hormone-like sequences which cause pigmentation when
the reduction of the double bond and ketone groups to levels of ACTH are markedly elevated. Circulating ACTH
produce tetrahydrocortisol, tetrahydrocortisone, cortol, stimulates cortisol production in the adrenal gland. The
and cortolone.4 These metabolites are excreted into the cortisol secreted (or any other synthetic corticosteroid ad-
urine in conjugation with glucoronic acid, and account for ministered to the patient) causes negative feedback to the
approximately 60% to 70% of the total cortisol produced. hypothalamus and pituitary to inhibit further CRH and
A small amount of cortisol (up to 50 g per day) is excreted ACTH release. The set point of this system clearly varies
unchanged in the urine and represents the unbound throughout the day according to the circadian rhythm
cortisol in plasma that is filtered through the kidney. (with maximal activity taking place soon after awaken-
Aldosterone has less protein binding than cortisol. ing and the lowest concentrations between 10:00 PM and
An ultrafiltrate of plasma contains as much as 50% of 2:00 AM), which is usually overridden by severe stress
circulating aldosterone, which explains the shorter half- (e.g., trauma, surgery, intense exercise). ACTH is also re-
life (20 minutes for aldosterone compared to 2 hours for leased during stress, independent of the circulating serum
cortisol) and high metabolic clearance rate. The average cortisol level. CRH, vasopressin, and norepinephrine act
secretion rate of aldosterone is 100 to 200 g per day. synergistically to increase ACTH release during stress.
Both the kidney and the liver are sites of metabolic Endorphinergic pathways also play a role in ACTH regu-
clearance, with only a small percentage of aldosterone lation. The acute administration of morphine stimulates
excreted remaining unchanged in the urine. The ma- ACTH release, whereas chronic administration blocks it.
jor androgens produced by the adrenal cortex are de-
hydroepiandrosterone (DHEA) sulfate, androstenedione,
DHEA, 11-hydroxyandrostenedione, and small amounts
of testosterone.
PHYSIOLOGY OF ADRENAL
Normal adrenal function is important for modulat- MEDULLARY
ing: (i) intermediary metabolism and immune responses
through glucocorticoid action; (ii) blood pressure, vas-
CATECHOLAMINES
cular volume, and electrolytes through mineralocorticoid Biosynthesis begins with tyrosine, which can be ob-
action; and (iii) secondary sexual characteristics (in fe- tained from the diet or synthesized from phenylalanine.
males) through androgenic action.5 The adrenal axis plays Tyrosine is actively transported from the bloodstream
an important role in the stress response by rapidly in- into the adrenal gland, and converted into 3, 4-di-
creasing cortisol levels. Cortisol and aldosterone are the hydroxyphenylalanine (DOPA) by the rate-limiting, mi-
two essential hormones, whereas the adrenal androgens tochondrial enzyme, tyrosine hydroxylase. Feedback
are of relatively less physiologic significance in adults. inhibition is exerted by norepinephrine. Decarboxylation
Adrenal disorders include hyperfunction (Cushings syn- of DOPA to dopamine is catalyzed by the cytosolic en-
drome) and hypofunction (adrenal insufficiency), as well zyme, aromatic L-amino acid decarboxylase. Dopamine,
as a variety of genetic abnormalities of steroidogenesis. in turn, must be actively transported into granulated vesi-
cles that contain dopamine -hydroxylase in order to be
converted to norepinephrine. For most chromaffin tis-
STEROID BIOSYNTHESIS sue and neurons, the synthesis ends with norepinephrine
binding to the granule. In the nonadrenergic neuron,
The precursor of all steroids is cholesterol. Part of the this granule containing norepinephrine is secreted into
cholesterol is synthesized from acetate, but most of it the synapse during depolarization. As with other chro-
is taken up from low-density lipoprotein in the circula- maffin tissue, the adrenal medulla stores and releases
tion. A common and important rate-limiting step for the norepinephrine in granules. The adrenal medulla is the
synthesis of all steroid hormones is cleavage of the side bodys primary source of both epinephrine and the cy-
chain from cholesterol (C27) to yield pregnenolone (C21), tosolic enzyme, phenyethanolamine N-methyltransferase,
the common branch point for the synthesis of progestins, which produces epinephrine from norepinephrine. Be-
corticoids, androgens, and, hence, estrogens.6 Expression cause phenyethanolamine N-methyltransferase activity is
of the side chain cleavage enzyme, cytochrome P450 scc, so dependent on glucocorticoids (cortisol in humans), it is
which converts cholesterol to pregnenolone, is one of the also dependent on ACTH and the rest of the hypothalamic-
unique features of steroidogenic cells that participate in pituitary-adrenal (HPA) axis. The ACTH also has a tropic
de novo steroid synthesis. Glucocorticoid production by effect on tyrosine hydroxylase activity of the adrenal
the adrenal gland is under hypothalamic-pituitary con- medulla. For the enzymatic reaction to take place, nore-
trol. Corticotropin-releasing hormone (CRH) is secreted pinephrine must be released from its storage granule in the
in the hypothalamus in response to circadian rhythm, cytoplasm to produce epinephrine, and then transported
stress, and other stimuli and travels down the portal sys- back into another granule. The storage granules in the
tem to stimulate adrenal corticotropic hormone (ACTH) adrenal medulla contain epinephrine or norepinephrine,
CHAPTER 43/DISORDERS OF THE ADRENAL GLAND 615
and the release can be selective. Under normal physiologic metabolite are more likely to be significantly elevated in
conditions, approximately 80% of the catecholamine out- children with a pheochromocytoma. The etiology for the
put from the adrenal medulla is epinephrine. increased production and secretion of catecholamines is
not clear. It is conceivable that the negative feedback
mechanism of norepinephrine on tyrosine hydroxylase is
altered so that sensitivity to feedback is decreased, or per-
What Are the Facts haps metabolism or release is so rapid that feedback does
not transpire in the usual manner. Small tumors tend
on Pheochromocytoma? to secrete high levels of circulating catecholamines. With
intracellular metabolism being more prevalent in large
tumors, high levels of metabolites tend to be released, and
INCIDENCE free catecholamine secretion is reduced.
Approximately 50% of patients with a pheochromocy-
Pheochromocytomas are catecholamine-secreting tumors toma have sustained hypertension, 45% are normotensive
of chromaffin tissue and are a rare cause of hypertension.8 with paroxysms of hypertension, and 5% are normotensive
Less than 0.1% of the hypertensive population has a or even hypotensive.12,14 These differences, in part, relate
pheochromocytoma.9,10 The hypertension caused by these to the patterns of catecholamine secretion; bursts produce
tumors is usually curable. Surgery on a patient with an hypertensive episodes. Patients with sustained hyperten-
unrecognized pheochromocytoma can be fatal; similarly sion and some normotensive patients can have high or
the administration of -adrenergic-blocking drugs can normal levels of norepinephrine. How sustained levels
have untoward side effects. of high norepinephrine concentrations result in persis-
These tumors can be associated with other poten- tent hypertension is not understood. However, if elevated
tially fatal but curable diseases. The high incidence of catecholamine secretion persists, - and -receptors may
pheochromocytoma in families as a primary disease, in become desensitized, or even downregulated. Hemody-
association with multiple endocrine neoplasia or other namic mechanisms will no longer respond to elevated
familial diseases such as Von Hippel Lindau syndrome levels of norepinephrine, and blood pressure will nor-
and neurofibromatosis I, indicates the need for genetic malize. Catecholamine levels and blood pressure do not
counseling.11 Approximately 16% of pheochromocytomas usually correlate well, but a significant change in cat-
will be associated with other endocrine disorders, such echolamine concentration will elicit a blood pressure
as multiple endocrine syndrome 2, which is comprised of response. Patients with the rare, exclusively epinephrine-
medullary thyroid carcinoma, pheochromocytoma, and producing tumors can present with normotension, or even
parathyroid hyperplasia. hypotension, secondary to the vasodilating properties of
epinephrine. Orthostatic hypotension is another result of
the ganglionic-blocking activity of excessive amounts
of catecholamines.
PATHOPHYSIOLOGY
The incidence of pheochromocytoma as a benign tumor in
one of the adrenal glands is 80%. Twenty percent are extra- CLINICAL PRESENTATION
adrenal, with half located below the diaphragm in areas
such as along the aorta, near the urinary bladder, and in The classic triad of symptoms in patients with a pheochro-
the organ of Zuckerkandl; the other half is located above mocytoma consists of episodic headache, sweating, and
the diaphragm in areas along the aorta, in the lungs or tachycardia, all of which are largely due to the phar-
heart, or in the neck or carotid bodies. Ten percent occur macologic effects of the catecholamines secreted from
in children. In nonfamilial disease, the classical teaching these tumors.12,15 Other signs and symptoms include pal-
is that 10% of patients have bilateral adrenal tumors and lor, orthostatic hypotension, visual blurring, papilledema,
10% have multiple extra-adrenal tumors; however, in fa- weight loss, polyuria, polydipsia, increased erythrocyte
milial disease, more than 80% are bilateral or in multiple sedimentation rate, hyperglycemia, psychiatric disorders,
sites. The incidence of a malignant pheochromocytoma is and, rarely, secondary erythropoesis consistent with over-
10%. The occurrence of a pheochromocytoma is evenly production of erythropoietin. The abnormalities in car-
distributed between the sexes and can occur at any age, bohydrate metabolism are directly related to increases
although the peak incidence is between the fourth and in catecholamine production; these changes resolve after
sixth decades. Catecholamine secretion is responsible for adrenalectomy.
the signs and symptoms of a pheochromocytoma.12 It is There are two rare presentations of a pheochromocy-
unusual that a tumor will grow large enough or be so toma: Episodic hypotension in patients with tumors that
invasive as to interfere with the function of the surround- secrete only epinephrine and rapid cyclic fluctuations of
ing organs. The manifestations of a pheochromocytoma hypertension and hypotension.16,17 The latter group of pa-
are primarily the result of the excessive secretion of nore- tients can be treated by fluid repletion and -adrenergic
pinephrine, epinephrine, and dopamine.13 The most com- antagonists. These patients can exhibit significant base-
mon combination is predominantly norepinehrine and line electrocardiographic (ECG) changes due to the toxic
epinephrine. Some tumors secrete only norepinehrine, effects exerted on the myocardium by the excessively high
but <10% secrete only epinephrine. Dopamine and its levels of catecholamines. Patients may also present with
616 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
chest pain and ECG changes suspicious for ischemia. De- from as high as 45% to as low as 0% to 3% after the
spite striking repolarization changes, many patients who pheochromocytoma is excised, has encouraged the ad-
proceed to coronary angiography preoperatively are found ministration of -antagonists for preoperative therapy
to have no obstruction of their coronary arteries. Anec- which is usually initiated once the diagnosis is estab-
dotal reports suggest ECG changes to be a manifestation lished. Phenoxybenzamine, a long-acting (24 to 48 hours),
of toxic myocarditis. In addition to the ECG changes men- noncompetitive, presynaptic, -adrenergic antagonist, is
tioned, many patients with a pheochromocytoma are noted typically initiated at least 7 days (usually for 2 to 4 weeks)
to have a long QTc interval which may predispose to ven- before surgery and added incrementally until the blood
tricular arrhythmias.18 After removal of the pheochromo- pressure is controlled and paroxysms disappear. The ini-
cytoma, the QTc intervals tend to normalize. An elevated tial dose is usually 10 mg every 12 hours. Most patients
temperature more commonly reflects a catecholamine- require between 80 and 200 mg per day. The combina-
mediated increase in the metabolic rate and diminished tion of -adrenergic receptor blockade and a liberal salt
heat dissipation secondary to vasoconstriction. Polyuria intake will restore the contracted plasma volume towards
is an occasional finding, and rhabdomyolysis with resul- normal.19 Selective -blockers, such as doxazosin, pra-
tant myoglobinuric renal failure may result from extreme zosin and terazosin, have also been used effectively, but
vasoconstriction and ensuing muscle ischemia. their role in preoperative management may be limited
to the treatment of individual paroxysms. Nitroprus-
side, calcium channel blockers (where it was found that
treatment extended postoperatively with calcium channel
DIAGNOSIS blockers is also effective in treating the clinical manifesta-
The diagnosis of pheochromocytoma is usually suggested tions of catecholamine-induced myocarditis and coronary
by the history in a symptomatic patient or the family vasospam),20 and angiotensin-converting enzyme in-
history in a patient with familial disease, and can hibitors all can be used to reduce blood pressure in
usually be confirmed by measurements of urinary and these patients. Intraoperatively, nitroprusside is benefi-
plasma catecholamines and metabolites and radiologic cial in the treatment of hypertensive crises. -adrenergic
tests. Specific tests for diagnosing pheochromocytoma blockade is initiated more than 3 days before surgery in
are outlined in Table 43.1. patients with persistent tachycardia or reflex tachycardia
related to the initiation of -blockade, or in those having
arrhythmias. It is important to initiate -blockade before
-blockade to avoid the situation of unopposed -agonism
ANESTHETIC whereby the patient suffers from intense vasoconstriction
CONSIDERATIONS from the -adrenergic excess and is at risk for extreme
hypertension as well as a dramatic increase in myocardial
Preparing the patient for surgical removal of a pheochro- workload. Low doses often suffice; a reasonable starting
mocytoma entails the institution of - and -adrenergic dose is 10 mg propranolol, three to four times per day, and
blockade. The reduction in perioperative mortality rates, is increased as needed to control the pulse rate. Labetolol,
Data from: a Sawka AM, Jaeschke R, Singh RJ, et al. A comparison of biochemical tests for pheochromocytoma: Measurement of
fractionated plasma metanephrines compared with the combination of 24h urinary metanephrines and catecholamines. J Clin Endocrinol
Metab. 2003;88:553.
b Lenders JW, Pacak K, Walthar MM, et al. Biochemical diagnosis of pheochromocytoma: Which test is best? JAMA. 2002;287:1427.
c Mukherjee JJ, Peppercorn PD, Reznick RH, et al. CT imaging of pheochromocytoma: Effects of non-ionic contrast medium on circulating
a -adrenergic antagonist with some -blocking activity, volatile agent. If a potent volatile anesthetic is used solely to
is effective as a second-line medication, but can increase maintain anesthesia, the drop in the blood pressure can be
the blood pressure if used alone, presumably because of dangerous after tumor resection. After tumor ligation and
its -blocking effect being much more pronounced than resection, the drop in blood pressure can be dangerously
its -blockade. The pharmacologic adrenergic blockade abrupt; however, this can be anticipated through close
helps to blunt the intense surges in blood pressure that communication with the surgical team. Many patients re-
occur during surgery and tumor manipulation. quire a vasoconstrictor (e.g., phenylephrine infusion) to
Intravascular volume is also contracted in patients support the blood pressure for a period of hours after the
with a pheochromocytoma. This is manifested by hemo- tumor is removed. If postoperative hypertension ensues,
concentration and orthostatic changes in blood pressure. it may indicate the presence of occult tumor or volume
-adrenergic-mediated vasoconstriction, and possibly al- overload. After successful surgery, catecholamine excre-
tered capillary permeability, is thought to be responsible tion returns to normal in about two weeks.
for these findings. Preoperatively, -adrenergic blockade
enables the patient to replete intravascular volume. The
presence or absence of orthostasis and changes in hema-
tocrit should be assessed at the time of the preoperative How Can Adrenal Insufficiency
visit. Despite the fact that hypotension commonly oc-
curs after vascular isolation of the tumor, most clinicians
Lead to Critical Illness?
continue -blockade up until the morning of the surgery.
Preoperative treatment with -methyl-para-tyrosine
results in depletion of tumor catecholamine stores caused
GLUCOCORTICOIDS
by the competitive inhibition of tyrosine hydroxylase, and Critical illness, whether from sepsis, trauma, surgery, or
decreases both blood pressure lability and intraoperative any condition associated with hemodynamic compromise,
blood loss.10 This medication is currently reserved for stimulates the HPA axis, with a resultant increase in ACTH
patients with metastatic disease or for situations in which and cortisol secretion.25 Exogenous glucocorticoids can
surgery is contraindicated and long-term medical therapy suppress the HPA axis, and the patient on chronic gluco-
is needed. corticoids may not produce sufficient levels of ACTH and
Potentially life-threatening fluctuations in blood pres- cortisol in the perioperative period to meet physiologic
sure indicate the need for direct arterial pressure mon- demands. The incidence of perioperative adrenal insuffi-
itoring, and large intraoperative fluid shifts underscore ciency in patients treated with glucocorticoids is difficult
the importance of excellent intravenous access and uri- to determine but must be low; earlier reports were incon-
nary output monitoring. Young patients with healthy clusive and lacked biochemical confirmation of adrenal
hearts may need only central venous pressure monitoring, insufficiency.26,27
whereas those with a history of cardiac disease, includ-
ing catecholamine-induced cardiomyopathy, may benefit Major Physiologic Actions
from transesophageal echocardiography or a pulmonary
artery catheter. Endotracheal intubation should be at- The HPA axis regulates the adrenal output of glucocor-
tempted only after a deep level of anesthesia is attained. ticoids. Hypothalamic release of the CRH stimulates the
Intraoperative hypertension can be effectively treated with pituitary to produce ACTH, which, in turn, stimulates the
phentolamine (a short-acting -adrenergic antagonist that adrenal cortex to produce cortisol, thereby completing
may be given as an intravenous bolus of 2 to 5 mg the cycle by providing negative feedback for both hor-
or by continuous infusion), nitroprusside, or nicardipine mones. The usual cortisol output of the adrenal gland in
titrated to effect. Other agents that can be used include normal, nonstressed situations is between 15 and 30 mg
nitroglycerin, fenoldopam,21 diltiazem, prostaglandin E1 , per day, which can be amplified to 60 to 100 mg/m2 /day
and magnesium sulfate (as an infusion).22 during times of stress.
Anesthetic drugs and techniques that stimulate the Glucocorticoids increase blood glucose levels, thereby
sympathetic nervous system, such as ephedrine, ketamine, facilitating the delivery of glucose to the cells during stress
and hypercarbia; potentiate the arrhythmogenic effects by increasing the rate of hepatic gluconeogenesis and in-
of catecholamines (e.g., halothane); inhibit the parasym- hibiting glucose uptake by adipose tissue. It is, therefore,
pathetic nervous system (e.g., pancuronium); or release not surprising that patients with adrenal insufficiency
histamine (e.g., morphine sulfate, atracurium) may pre- present with hypoglycemia.28 Glucocorticoids also stim-
cipitate hypertension and are best avoided. During laparo- ulate proteolysis and lipolysis, again to provide the cell
scopic surgery, creation of the pneumoperitoneum may with the energy and substrate required for the response
trigger the release of catecholamines and huge fluctuation to stress and repair from injury. Small amounts of glu-
in hemodynamics that can be controlled with a vasodilator. cocorticoids must be present for a number of metabolic
Interestingly, the hemodynamic changes are typically very reactions to occur; this is called permissive action and in-
similar when pheochromocytomas are resected laparo- cludes catecholamines to produce pressor responses and
scopically compared to when they are removed through bronchodilatation.
laparotomy.23,24 Blood loss and length of stay are less with Glucocorticoids are required for a normal car-
the laparoscopic than the open procedure.23 Anesthesia diovascular response to angiotensin II; epinephrine,
is often maintained with an opioid analgesic and a potent and norepinephrine, which support the maintenance of
618 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
cardiac contractility, vascular tone and integrity, and Effects of Disease of the Supra-Renal Capsules, which was
blood pressure.29 Glucocorticoids decrease the production subsequently named after him, that is, Addisons Disease.31
of nitric oxide.30 They also have an inhibitory effect on the Worldwide, tuberculosis remains the most common cause
endothelial production of prostacyclin. Relative glucocor- of adrenal insufficiency, often presenting many years after
ticoid deficiency therefore allows enhanced prostacyclin the initial diagnosis. Normal gland function is usually not
production and results in a vasodilated state. recovered, even after effective antituberculous therapy. In
Glucocorticoids possess anti-inflammatory and im- the developed world, autoimmune adrenal destruction is
munosuppressive effects by influencing most cells that play the culprit in more than 80% of cases.
a role in inflammatory reactions and enhancing the release Autoimmune polyendocrinopathy syndrome type I
of anti-inflammatory factors such as interleukin-1 recep- (also referred to as the autoimmune polyendocrinopathy-
tor antagonist, soluble tumor necrosis factor receptor, and candidiasis-ectodermal dystrophy syndrome) is a rare, auto-
interleukin-10. Glucocorticoid receptors have been found somal recessive disorder and usually presents in childhood
in almost every nucleated cell in the body, which suggests
(age 10 to 15 years).32,33 It is more common in adulthood,
their widespread influence throughout the body.
with a prevalence in women, and has several modes of
inheritance, with an age of onset usually between 20 and
40 years. It is characterized by autoimmune thyroid dis-
What Is the Etiology of Primary ease, insulin-dependent diabetes mellitus, adrenal failure,
pernicious anemia, and primary gonadal failure. The type
Adrenal Insufficiency? II syndrome with primary adrenal insufficiency and au-
toimmune thyroid disease was formerly referred to as
In 1855, Thomas Addison first described adrenal insuffi- Schmidts syndrome. Other causes of primary adrenal in-
ciency in his classic paper, On the Constitutional and Local sufficiency are detailed in Table 43.2.
Data from: a Dluhy RG. The growing spectrum of HIV-related endocrine abnormalities (editorial). J Clin Endocrinol Metab. 1990;70:563.
b Streeten DHP. Adrenal hemorrhage. Endocrinologist. 1996;6:277.
c Rao RH, Vagnucci AH, Amico JA. Bilateral massive adrenal hemorrhage: Early recognition and treatment. Ann Intern Med. 1989;110:227.
d Xarli VP, Steele AA, Davis PJ, et al. Adrenal Hemorrhage in the adult. Medicine. 1978;57:277.
e Seidenwurm DJ, Elmer EB, Kaplan LM, et al. Metastases to the adrenal glands and the development of Addisons disease. Cancer. 1984;54:552.
f Wagner RL, White PF, Kan PB, et al. Inhibition of adrenal steroidogenesis by the anesthetic etomidate. N Engl J Med. 1984;310:1415.
g Sonino N. The use of Ketoconazole as the inhibitor of steroid production. N Engl J Med. 1987;317:812.
h Keilholz U, Guthrie GP, Jr. Adverse effect of phenytoin on mineralocorticoid replacement with fludrocortisone in adrenal insufficiency. Am J
Neurol. 1984;16:628.
j Stuhrmann M, Heilbronner H, Reis A, et al. Characterization of a Xp21 microdeletion syndrome in a 2-year-old boy with muscular dystrophy,
glycerol kinase deficiency and adrenal hypoplasia congenital. Hum Genet. 1991;86:414.
ACTH, adrenal corticotropic hormone; CRH, corticotropin-releasing hormone.
CHAPTER 43/DISORDERS OF THE ADRENAL GLAND 619
feedback control on the HPA axis by suppressing CRH se- other parts of the body. The risk factors for develop-
cretion and, in turn, ACTH secretion, which usually leads ment of adrenal suppression from topical steroids for
to adrenal atrophy and loss of cortisol secretory ability. dermatologic indications include:
Normal adults secrete approximately 20 mg of cortisol
Application to a large surface area of the skin
and 100 g of aldosterone per day. The daily amount of
Application for a prolonged period of time
cortisol secreted by the adrenal gland is approximately
Use of occlusive dressings
50 mg during a minor procedure or surgery, whereas
Use of highly potent glucocorticoid agents
75 to 100 mg per day are produced with major surgery.
The amount of cortisol secretion can reach 200 to even Similarly, the development of adrenal suppression
500 mg per day under severe stress. Perioperative stress from inhaled steroids is related to dose, duration of ther-
is related to the degree of trauma and the depth of apy, and use of a potent agent (specifically fluticasone).
anesthesia. Deep general or regional anesthesia causes the Because the risk is low, it is advisable to consider sup-
usual intraoperative glucocorticoid surge to be postponed plementation for any patient who has received steroids
to the postoperative period.44 within a year, including patients on topical and inhaled
Immediate therapy for an adrenal crisis is manda- glucocorticoids. The question is how much steroid to give?
tory and consists of fluid and electrolyte resuscitation and It is commonly recommended to intravenously adminis-
steroid replacement. General perioperative management ter the maximum amount of glucocorticoid that the body
should include the avoidance of etomidate as an induc- needs in response to maximal stress (i.e., approximately
tion agent, as adrenal suppression, even after a single 200 mg per day hydrocortisone phosphate per 70 kg body
doseand particularly in septic patientsis seen with its weight), although 100 mg per day per 70 kg body weight is
use.4548 Etomidate has the salutatory effect of being a usually sufficient for minor surgical procedures. This dose
mild -agonist,49 but inhibits the enzymatic formation of is usually decreased by approximately 25% per day until
adrenal steroids. A single induction dose of etomidate has enteral feeding resumes, at which point the usual mainte-
been shown to increase the risk of adrenal insufficiency nance dose of glucocorticoids can be administered. A low
12-fold in intensive care unit (ICU) patients.50 It has been dose cortisol replacement regimen using 25 mg of cortisol
suggested that any critically ill patient who receives eto- equivalents before the induction of anesthesia, followed
midate should also receive stress steroid coverage. In the by a continuous infusion of cortisol equivalents in the
setting of adrenal insufficiency, the following measures next 24 hours, has been advocated, especially for pa-
are advised: Infusion of sodium-containing fluids; mini- tients undergoing minor (e.g., hernia repair) to moderate
mal doses of any anesthetic to avoid increased sensitivity (e.g., hip replacement) surgical stress. Considering that
to drug-induced myocardial depression; invasive monitor- the physiologic stress of local anesthesia is low, patients
ing of hemodynamics, glucose, and electrolytes; decreased need take only their usual daily glucocorticoid dose before
initial doses of muscle relaxants; and the use of a periph- procedures done under local anesthesia.
eral nerve stimulator. The risks of a single dose of steroids
are minimal.5153
In any case, it would be reasonable not to supplement
perioperatively with a dose lower than what the patient KEY POINTS
has already been receiving; however, the following patient PHEOCHROMOCYTOMA
population may not have had suppression of their HPA
axis: 1. A rare cause of hypertension.
2. -adrenergic blockade should never be instituted
Patients who received any dose of glucocorticoid for before -adrenergic blockade.
<3 weeks, although HPA suppression may ensue after 3. -adrenergic blockade is usually not required and
only five daily doses of 20 mg of prednisone should not be given unless a patient has persistent
Patients treated with alternate-day glucocorticoid tachycardia and supraventricular arrhythmias.
therapy 4. Twenty five percent of those who die in hospitals
Patients who have received morning doses of 5 mg per of pheochromocytoma crisis do so during the in-
day of prednisone or its equivalent duction of anesthesia, during stressful perioperative
In addition, patients may develop adrenal insuffi- periods, or during labor and delivery.
ciency from topical glucocorticoids, inhaled glucocorti- 5. Adrenergic blockade is adequate when: (a) There
coids, and regional administration of glucocorticoids to are no blood pressure readings >165/90 mm Hg for
CHAPTER 43/DISORDERS OF THE ADRENAL GLAND 621
48 hours; (b) orthostatic hypotension is present, 5. Parker LN. Adrenal androgens in clinical medicine. San Diego:
Academic Press; 1989.
but blood pressure on standing up should not
6. Miller WM. Molecular biology of steroid hormone synthesis.
be <80/45 mm Hg; (c) ECG is free of ischemic Endocr Rev. 1988;9:295.
changes. 7. Dallman MF, Akana SF, Cascio CS, et al. Regulation of ACTH
6. Anesthetic agents that block catecholamine reuptake secretion. Recent Prog Horm Res. 1987;43:113.
or cause catecholamine release are to be avoided. 8. Bravo EL, Gifford RW. Pheochromocytoma. Endocrinol
7. During surgery, tumor manipulation and isolation of Metab Clin North Am. 1993;22:329.
the vessels draining the tumor can result in a 1,000- 9. Ross NS, Aron DC. Hormonal evaluation of the patient
fold increase in plasma catecholamine concentration with an incidentally discovered adrenal mass. N Engl J Med.
within minutes. 1990;323:1401.
8. When bilateral adrenalectomy is being performed, 10. Pacak K, Linehan WM, Eisenhofer G, et al. Recent
advances in genetics, diagnosis, localization, and treat-
adrenal cortical insufficiency should be treated
ment of pheochromocytoma. Ann Intern Med. 2001;134:
with stress doses of hydrocortisone intraoperatively 315.
and postoperatively until stable. Mineralocorticoid 11. Chew SL, Reznek RH, Sheaves R, et al. The diagnosis of
should be replaced in the postoperative period. bilateral pheochromocytomas in Von Hippel-Lindau disease
9. Hypotension is the most common complication in by adrenal vein samples. Q J Med. 1994;87:49.
the immediate postoperative period; the treatment 12. Bravo EL. Evolving concepts in the pathophysiology, diag-
is aggressive volume expansion and phenylephrine nosis, and treatment of pheochromocytoma. Endocrinal Rev.
infusion. 1994;15:356.
10. Blood glucose levels should be monitored regularly 13. Bravo EL. Pheochromocytoma: New concepts and future
for several hours after the procedure. trends. Kidney Int. 1991;40:544.
14. Baxter MA, Hunter P, Thompson GR, et al. Pheochromo-
cytomas as a cause of hypotension. Clin Endocrinol (Oxf).
ADRENAL INSUFFICIENCY 1992;37:304.
1. The most common cause of primary adrenal failure, 15. Stein PP, Black HR. A simplified diagnostic approach
to pheochromocytoma. A review of the literature and
Addisons disease, is autoimmune in nature.
report of our institutions experience. Medicine (Baltimore).
2. The most common cause of secondary adrenal
1991;70:46.
insufficiency is suppression of corticotropin (ACTH) 16. Gangaly A, Grim CE, Weinberger MH, et al. Rapid cyclic
release by prior glucocorticoid therapy. fluctuations of blood pressure associated with an adrenal
3. A characteristic pattern of refractory hypotension due pheochromocytoma. Hypertension. 1984;6:281.
to decreased systemic vascular resistance and, to a 17. Oishi S, Sasaki M, Ohno M, et al. Periodic fluctuations
lesser degree, decreased cardiac output; hypovolemia of blood pressure and its management in patients with
and electrolyte disturbances are seen in patients with pheochromocytoma. Case report and review of the literature.
adrenal insufficiency. Jpn Heart J. 1988;29:389.
4. The adrenal cortex normally produces 25 to 30 mg of 18. Liao WB, Liu CF, Chiang CW, et al. Cardiovascular
manifestations of pheochromocytoma. Am J Emerg Med.
cortisol per day. The amount increases in response to
2000;18:622.
minor stress (50 mg per day) and major stress (75 to
19. Bogolioubev A, Keefe D, Groeger J. Circulatory shock. Crit
100 mg per day). The production rate of cortisol Care Clin. 2001;17:697.
seldom exceeds 200 to 300 mg per day. 20. Bravo EL, Tagle R. Pheochromocytoma: State-of-the-art and
5. Current dosing recommendations for supplemental future prospects. Endocr Rev. 2003;24:539.
corticosteroids in patients with adrenal insufficiency 21. Cooper Z, Mihm F. Blood pressure control with fenoldopam
are: Minor stress, 25 mg of hydrocortisone; moderate during excision of a pheochromocytoma. Anesthesiology.
stress, 50 to 75 mg per day or 25 mg intraoperatively, 1999;41:558.
followed by an intravenous infusion of 100 mg over 22. Pivalizza E. Magnesium sulfate and epidural anesthesia
24 hours; major stress, 100 to 150 mg per day or 200 in pheochromocytoma and severe coronary artery disease.
Anesth Analg. 1995;81:414.
to 300 mg per 70 kg body weight in divided doses per
23. Sprung J, OHara JF, Gill IS, et al. Anesthetic aspects of la-
day.
paroscopic and open adrenalectomy for pheochromocytoma.
6. Even a single dose of etomidate can cause secondary Urology. 2000;55:339343.
adrenal insufficiency. If etomidate is used, subsequent 24. Gill IS. The case for laparoscopic adrenalectomy. J Urol.
care providers should be made aware of its use. 2001;166:429.
25. Irvine NJ, Taft AD, Feek CM. Addisons disease. In:
James VHT, ed. Comprehensive endocrinology. The adrenal
REFERENCES
gland. New York: Raven Press; 1979:131.
1. McMinn RM. Abdomen. In: Last anatomy, 8th ed. New York: 26. Mohler JL, Michael KA, Freedman AM, et al. The evaluation
Churchill Livingstone; 1990:373. of postoperative function of the adrenal gland. Surg Gynecol
2. Gannong WF. The adrenal medulla and adrenal cortex. In: Obstet. 1985;161:551.
Review of medical physiology, 22nd ed. McGraw-Hill; 2005: 27. Alford WC Jr, Meador CK, Mihalevich J, et al. Acute adrenal
356. insufficiency following cardiac surgical procedures J Thorac
3. James VHT. Comprehensive endocrinology: the adrenal gland. Cardiovasc Surg. 1979;78:489.
New York: Raven Press; 1992. 28. Pilkis SJ, Granner DK. Molecular physiology of the regu-
4. Makin HJL. Biochemistry of steroid hormones, 2nd ed. lation of hepatic gluconeogenis and glycolysis. Annu Rev
Oxford: Blackwell Scientific; 1984. Physiol. 1992;54:885.
622 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
29. Sakane M, Hoffman BB. Glucocorticoids induce transcrip- 41. Rasmusan S, Olsson T, Hagg E. A low dose ACTH test to
tion and expression of the alpha 1B adrenergic receptor assess the function of the hypothalamic-pituitary-adrenal
gene in DTT1 MF-2 smooth muscles. J Clin Invest. 1991;88: axis. Clin Endocrinol. 1996;44:151.
385. 42. Tordjman K, Jaffe A, Grazas N, et al. The role of the low dose
30. Walker G, Pfeilschifter J, Kunz D. Mechanisms of sup- (1 microgram) adrenocorticotropin test in the evaluation of
pression of inducible nitric-oxide synthase (iNOS) ex- patients with pituitary diseases. J Clin Endocrinol Metab.
pression in interferon (IFN)-gamma-stimulated Raw264.7 1995;80:1301.
cells by dexamethasone: Evidence of glucocorticoid-induced 43. Abdu TA, Elhadd TA, Neary R, et al. Comparison of the
degradation of iNOS protein by caplain as a key step low dose short test (1 microg), the conventional dose short
in post-transcriptional regulation. J Biol Chem. 1997;272: Synacthen test (250 microgr), and the insulin tolerance test
16679. for assessment of the hypothalamo-pituitary-adrenal axis
31. Addison T. On the constitutional and local effects of disease of in patients with pituitary disease. J Clin Endocrinol Metab.
the suprarenol capsules. London: Highley; 1855. 1999;84:838.
32. Ahonen P, Myllarniemi S, Sipila I, et al. Clinical variation 44. Ayama T, Taniguchi K, Jin T, et al. Effects of anaesthesia
of autoimmune polyendocrinopathy-candidiasis-ectodermal and surgery on plasma aldosterone concentration and renin
dystrophy (APECED) in a series of 68 patients. N Engl J Med. activity in man. Br J Anaesth. 1979;51:747.
1990;322:1829. 45. Wagner Rl, White PF, Khan PB, et al. Inhibition of adrenal
33. Neufeld M, Maclaren NK, Blizzard RM. Two types of au- steroidogenesis by the anesthetic etomidate. N Engl J Med.
toimmune Addisons disease associated with different poly- 1984;310:1415.
glandular autoimmune (PGH) syndromes. Medicine. 46. Absalom A, Pledger D, Kong A, et al. Adrenocortidal function
981;160:355. in critically ill patients 24h after a single dose of etomidate.
34. Barnett AH, Espiner EA, Donald RA. Patients presenting Anaesthesia. 1999;54:861.
with Addisons disease need not be pigmented. Postgrad 47. Jackson WL Jr. Should we use etomidate as an induction
Med J. 1982;58:690. agent for endotracheal intubation in patients with septic
35. Annane D, Sebille V, Charpentier C, et al. Effect of treatment shock? Chest. 2005;127:1031.
with low doses of hydrocortisone and fludrocortisone on 48. Annane D. ICU physicians should abandon the use of
mortality in patients with septic shock. JAMA. 2002;288: etomidate! Intensive Care Med. 2005;31:325.
862. 49. Pascoe PJ, Ilkiw JE, Haskin SC, et al. Cardiopulmonary
36. Oelkers W. Adrenal Insufficiency. N Engl J Med. 1996; effects of etomidate in hypovolemic dogs. Am J Vet Res. 1992;
335:1206. 53:2178.
37. Marik PE, Zaloga GP. Adrenal insufficiency during septic 50. Malerba G, Romano-Girard F, Cravoisy A, et al. Risk
shock. Crit Care Med. 2003;31:141. factors of relative adrenocortical deficiency in ICU pa-
38. Richards ML, Caplan RH, Wickus GG, et al. The rapid tients needing mechanical ventilation. Intensive Care Med.
low-dose (1 microgram) Cosyntropin test in the immediate http://dx.doi.org/10.1007/s004-2550-8, 2006.
postoperative period: Results in elderly subjects after major 51. Lampe GH, Roizen MF. Anesthesia for patients with
abdominal surgery. Surgery. 1999;125:431. abnormal function of the adrenal cortex. Anesthesiol Clin
39. Marik PE, Kiminyo K, Zaloga GP. Adrenal insufficiency North America. 1987;5:245.
in critically ill HIV-infected patients. Crit Care Med. 2002; 52. Symreng T, Karlberg BE, Kagedal B, et al. Physiological
30:1267. cortisol substitution of long term steroid treated patients
40. Mayenknecht J, Diedrich S, Bahr V, et al. Comparison of undergoing major surgery. Br J Anaesth. 1981;53:949.
low and high dose corticotropin stimulation tests in patients 53. Udelsman R, Ramp J, Gallucci WT, et al. Adaptation during
with pituitary disease. J Clin Endocrinol Metab. 1998;83: surgical stress: A reevaluation of the role of glucocortico-
1558. steroid. J Clin Invest. 1986;77:1377.
CHAPTER MORBID OBESITY
44 Patrick J. Neligan
A
of his electrocardiogram (ECG) reveals left bundle branch
Roux-en-Y gastric bypass (RYGB). He has
block. His fasting blood glucose is 170 mg per dL. Other
a body mass index (BMI) of 65. He has
laboratory values are within normal limits. On physical
a history of hypertension, congestive heart
examination he has central obesity, blood pressure of
failure (CHF) (he has an ejection fraction of
150/90 mm Hg, heart rate 76 bpm. There is no jugular
28%) and obstructive sleep apnea (OSA). His
venous distension. His neck circumference is 20.5 in.
apneahypopnea index is 108, and he receives continuous
(52 cm). He has good mouth opening, with Mallampati
positive airway pressure (CPAP) 15 cm H2 O at night.
class 2 airway, and moderate neck extension. He reports
Pulmonary function tests reveal a mixed restrictive
dyspnea on mild exertion and occasional orthopnea.
obstructive picture. Blood gas, on room air reveals a
P A R T I EPIDEMIOLOGY
623
624 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
distribution, most of which are women, and those with 2. Increased expression of angiotensinogen by adipose
central (apple-shaped) fat distribution, most of which are tissue leading to activation of the renin-angiotensin-
men. This is conventionally recorded in whites as waist- aldosterone axis
to-hip ratio. A ratio of >1 in men, and >0.85 in women 3. Increased renal sodium retention secondary to hyper-
suggest central obesity.9 insulinemia
Data from the Framingham study has revealed a
significant increase in the incidence of heart failure in
MO patients.11 For each increment of 1 in BMI, there
What Medical Problems Does was an increase in the risk of heart failure of 5% for
This Patient Have? men and 7% for women. As compared with subjects with
a normal BMI, obese individuals had double the risk of
heart failure.11
This patient has several problems characteristic of MO.
These include cardiovascular disease (characterized by
hypertension, coronary heart disease [CAD], and CHF),
pulmonary disease (characterized by OSA and obesity METABOLIC SYNDROME
hypoventilation syndrome [OHS]), and metabolic disease
MO is associated with a significant increase in the risk
(the metabolic syndrome characterized by central obe-
of CAD. The relative risk is 2.80 for men and 2.71 for
sity, dyslipidemia, insulin resistance-hyperglycemia, and
women.12 These patients are also at risk for dyslipi-
hypertension). In addition, he is likely to have undiag-
demia and type 2 diabetes mellitus, with all the asso-
nosed liver disease.
ciated complications. The combination of central obesity,
MO predisposes patients to cardiovascular disease.
insulin resistance, hypertension, dyslipidemia, and im-
This includes hypertension, CAD, CHF, cerebrovascular
paired glucose tolerance has been termed the metabolic
disease, varicose veins, and deep venous thrombosis.
syndrome. First described by Reaven in 1988,13 and
Hypertension is seen in 50% to 60% of obese patients.
known by several monikers (including Syndrome X),
There is a 3 to 4 mm Hg increase in systolic blood
this disorder has recently been defined by the National
pressure and a 2 mm Hg increase in diastolic pressure per
Cholesterol Education Program (NCEP)14 and the World
10 kg weight gained.10 Obesity contributes to hypertension
Health Organization (WHO)15 for research and practi-
through the following:
cal purposes. The definitions are slightly different (see
1. Increased vascular tone, secondary to increased sympa- Table 44.1).16
thoadrenal activity and reduced bioavailability of nitric Depending on which definition is used, between
oxide, due to increased oxidative stress 25.1%17 and 27% (NCEP definition)18 of the population
a Blood pressure criteria generally treated operationally by researchers as (systolic blood pressure) or, (diastolic blood pressure) or,
although not included in original definitions, antihypertensive treatment.
NCEP ATPIII, National Cholesterol Education Program Adult Treatment Panel III; WHO, World Health Organization; HDL, high density
lipoprotein.
Data from: (1) Third report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of
high blood pressure in adults (Adult Treatment Panel III): Final report. Circulation. 2002;106:3143 and (2) Alberti KG, Zimmet PZ. Definition,
diagnosis and classification of diabetes mellitus and its complications. Part 1: Diagnosis and classification of diabetes mellitus provisional
report of a WHO consultation. Diabet Med. 1998;15:539.
CHAPTER 44/MORBID OBESITY 625
(WHO definition) have metabolic syndrome. Again, levels of fructose in the diet,28 predisposes patients
African American women and Mexican Americans of both to nonalcoholic (fatty) liver disease. As body weight
genders are at particular risk. Obesity is not an essen- increases, there is a progressive increase in the risk
tial component of metabolic syndrome; however, there of development of nonalcoholic steatohepatitis. This is
is a strong correlation between visceral fat deposits and an inflammatory disease that is reversible in its early
metabolic syndrome. Hence definitions of metabolic syn- stages with weight loss. However, sustained liver injury
drome emphasize waist circumference rather than BMI. It leads to fibrosis and cirrhosis in 10% to 25% of affected
is possible to be metabolically obese and of normal weight, individuals.16
or obese without metabolic syndrome (metabolically
healthy obese).19 This distinction is important because
metabolic syndrome, not BMI, predicts future cardiovas-
cular disease in women.20 Amongst Finnish men, the
OBSTRUCTIVE SLEEP APNEA
prevalence of metabolic syndrome ranged from 8.8% OSA-hypopnea syndrome occurs in up to 70% of mor-
(WHO definition) to 14.3% (NCEP definition). Patients bidly obese patients undergoing bariatric surgery.29 This
with metabolic syndrome were 2.9 to 3.3 times more likely is characterized by five or more episodes of apnea or hy-
to die of coronary arterial disease. In a posthoc analysis popnea per hour with daytime somnolence, or 15 episodes
of two cardiovascular trials, patients with metabolic syn- without. Hypopnea is a 30% reduction in airflow for
drome were 1.5 times more likely to have major coronary 10 seconds or longer, together with at least a 4% reduc-
events versus those without it.21 The presence of diabetes tion in oxygen saturation. There is no direct relationship
worsens the risk of metabolic syndrome. In a large cohort between OSA and BMI,30 although there is a correla-
of patients, the prevalence of coronary arterial disease was tion with central obesity. OSA is caused by narrowing
19.2% in patients with metabolic syndrome and type 2 di- of the upper airway due to fat in the pharyngeal wall
abetes, 13.9% with metabolic syndrome alone, and 7.5% (at the level of the soft palate and submental area), with
with diabetes alone. The presence of coronary arterial loss of pharyngeal dilator activity during sleep. In addi-
disease increases perioperative risk. Hence the metabolic tion, there is an abnormality of the central control of
syndrome should be (until prospective epidemiologic data breathing.
is available) considered an independent perioperative risk OSA is quantified by performing sleep studies
factor. (polysomnography). This generates either an apnea-
Metabolic syndrome is an inflammatory disorder. hypopnea index (AHI) or respiratory disturbance index
Adipose tissue, and in particular visceral fat, is an en- (RDI). An AHI or RDI >30 signifies severe OSA. The
docrine, paracrine, and immunologic organ. Obesity is patient described in the case summary has an AHI >100;
a state of chronic inflammation.22 Insulin is an anti- this should be considered very high risk.
inflammatory hormone. Increased circulating free fatty The treatment for OSA is CPAP, with or without in-
acids, derived from highly metabolic visceral fat, can spiratory pressure support. CPAP is probably beneficial to
reduce insulin activity and promote hepatic steatosis. postoperative patients with a history of OSA, particularly
Tissue macrophages invade adipose tissue and release at the time of rapid eye movement (REM) sleep on day 3
tumor necrosis factor alpha (TNF-). This, in turn, or 4.31 Evidence that this intervention improves outcomes
causes the release of interleukin (IL)-1, IL-6, and other is lacking. The incidence and severity of OSA significantly
cytokines. There is an alteration in the relative concen- diminishes following gastric bypass surgery.32
trations of adipose-derived hormones, collectively known In addition to OSA, this patient also has the OHS,
as adiopkines. Leptin, the first adiopkine described, is which is also referred to as sleep hypoventilation syndrome.
involved in the control of satiety and is markedly proin- This is characterized by chronic respiratory insufficiency,
flammatory. Leptin levels are raised in patients with with both an obstructive and restrictive pattern on pul-
the metabolic syndrome. Conversely, adiponectin, which monary function tests, and hypercarbia in medically
is thought to be anti-inflammatory and enhances in- complicated obesity. At its extreme, the patient develops
sulin sensitivity, is reduced in these patients. Resistin, pulmonary hypertension and right ventricular dysfunc-
an adipokine that antagonizes insulin, is elevated in tion (cor pulmonale). This is colloquially referred to as
the metabolic syndrome, and hence the metabolic syn- Pickwickians syndrome, after the rotund Dickens char-
drome produces an inflammatory picture analogous to acter. Not all patients with OHS have OSA, and not all
low grade sepsis. Interestingly, there are preliminary patients with OSA are obese.33
data that this adipokine picture is associated with an in- There is little doubt that OHS results in worse
crease in the risk of myocardial ischemia.23 Recent studies intermediate term outcomes in MO.10 Although it is
have highlighted the contribution of inflammation to my- universally accepted that the presence of OSA increases
ocardial ischemia and infarction.24,25 Long-term therapy perioperative risk, particularly in terms of postoperative
for metabolic syndrome includes lifestyle modification, airway problems (narcotic-induced obstruction of the
weight loss, tight control of hypertension and diabetes, airway), there is little published data to support this
-blockade, statin and perhaps fibrate administration, and contention.34 The American Society of Anesthesiologists
nicotinic acid and thiazolidinedione (insulin sensitizer) has recently approved guidelines for the perioperative
therapy.26,27 management of these patients.35 (see Table 44.2)
The presence of high levels of free fatty acids in Postoperative atelectasis, with associated hypoxemia
the liver, consequence of insulin resistance and high and increased pulmonary workload is a significant
626 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
problem for morbidly obese patients. During general than Mallampati score, for difficulty in the process of
anesthesia, there is a significant reduction in total res- intubation.
piratory system compliance.36 This leads to significantly Brodsky et al.43 investigated difficult intubation in
lower lung volumes, higher intra-abdominal pressure, and 100 patients with a BMI >40. Patients were preoperatively
ventilationperfusion mismatch.37 In addition, morbidly measured for height, weight, neck circumference, width
obese patients have a significantly higher airways resis- of mouth opening, sternomental distance, thyromental
tance than normal.36 distance, and Mallampati score. All patients underwent
Morbidly obese patients also have significantly more rapid sequence induction and direct laryngoscopy. There
atelectasis than nonobese patients before induction (2.1% was one failed intubation and 12 problematic intubations.
of total lung area vs. 1%, p < 0.01), after tracheal ex- There was no association between difficult intubation and
tubation (7.6% vs. 2.8%, p < 0.05), and 24 hours (9.7% increased BMI. Indeed, the incidence of problematic or
vs. 1.9%, p < 0.01) following laparoscopic surgery.38 This difficult intubation was not greater than that previously
was noted despite the application of 6 cm H2 O pos- reported in the general population.44 In the logistic regres-
itive end-expiratory pressure (PEEP) to both groups, sion analysis, neck circumference was the only patient
intraoperatively. This leads to a significantly increased characteristic that had a significant effect on the proba-
perioperative risk in terms of primary and secondary bility of problematic intubation.43 This was significantly
respiratory failure. Vital capacity decreases following associated with male gender, higher Mallampati score,
extubation. This relation varies linearly with BMI.39 At- and grade 3 views during laryngoscopy.
electasis increases the workload of breathing; hence, in There appears to be a relation between the presence of
the recovery room, the combination of partial neuromus- OSA, and difficult tracheal intubation.45 In a case-matched
cular blockade, opioids, and segmental lung collapse may study of 15 patients who had proven difficult intubation,
lead to acute respiratory distress requiring reintubation. there was a significant relationship with the apnea-
Of more concern is the progressive increase in atelectasis hypopnea index (AHI).46 Using ultrasonography of the soft
that occurs over the first 24 hours, at which stage the tissue of the neck, Ezri et al. showed that obese patients
patients are often less supervised on the ward. Atelec- who are difficult to intubate have more paratracheal soft
tasis and hypoventilation, secondary to opioids leading tissue.45 This factor may be of more importance than BMI.
to hypercapnia induced somnolence, may lead to airway There is little evidence to support the routine use
obstruction and respiratory arrest. of fiberoptic intubation on the basis of BMI. Although
In summary, this patient has several factors that have the large studies on this issue identify problematic and
the potential to complicate his/her perioperative course: difficult intubation, there is little to indicate that the
incidence of failed intubation is higher than in the general
Risk of myocardial ischemia due to CAD, hypertension,
population. Indeed, in the Brodsky study,43 all patients
and metabolic syndrome
were intubated by trainees.
Risk of acute pulmonary edema due to his history of
CHF
Risk of perioperative hypoxemia due to low respiratory
system compliance
Risk of airway obstruction due to severe OSA How Should I Induce
Risk of abnormal drug metabolism due to undiagnosed Anesthesia?
liver disease
How then to induce anesthesia in a morbidly obese
patient? This is really a matter of choice, experience,
and confidence in ones technical abilities. Mask ven-
What Perioperative Challenges tilation of morbidly obese patients is notoriously dif-
Does the Anesthesiologist ficult.47 Many anesthesiologists prefer rapid sequence
induction and airway security in view of the perceived
Anticipate? difficulty with mask ventilation. It is important to note
that succinylcholine should be dosed on the basis of
The widely accepted dogma that obesity is an independent actual, rather than lean, body weight. Crucial to the
risk factor for difficult intubation40 has been challenged.41 success of intubation and prevention of hypoxemia is
A study of 764 consecutive patients undergoing general patient positioning during induction.48 The morbidly
anesthesia without airway pathology failed to show a obese patient should be placed with the head, upper
correlation between BMI and difficult laryngoscopy.41 body, and shoulders significantly elevated above the
Using the intubation difficulty scale,42 Juvin et al. claimed chest. An imaginary horizontal line should connect the
a difficult intubation rate of 15.5% in patients with patients sternal notch with the external auditory mea-
BMI >35 compared to 2.2% of controls (BMI <30); tus.48 This can be achieved by using a stack of linen
an absolute risk increase of 13%. However, the rate of positioned behind the patients shoulders (a ramp) or
difficult laryngoscopy was similar for both groups (10.4% with a commercially available, foam elevation pillow.49
vs. 10.1%). No patients proved impossible to intubate. This positioning guarantees a higher rate of successful
The study was underpowered to detect reasons, other intubation.43
628 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
PREVENTING ATELECTASIS
P A R T II
AND POSTOPERATIVE HYPOXEMIA
presumably prolonged direct compression of dorsal and TABLE 44.3 Indications for Bariatric Surgery
gluteal muscles against the operating bed.
Highly lipophilic drugs have an increased volume of BMI >40 or BMI >35 with associated medical comor-
distribution in bariatric patients: These include barbitu- bidity worsened by obesity
rates, benzodiazepines, fentanyl, and sufentanil. However, Failed dietary therapy
pharmacokinetics in this circumstance are difficult to Psychological stability
predict, given the interpatient variability. Less lipophilic Knowledgeable about operation and its sequelae
drugs have normal volumes of distribution. Drugs such as Motivated
nondepolarizing neuromuscular blockers can be dosed on Medical problems do not preclude likely survival from
ideal body weight or lean body mass. Steroid based neu- surgery
romuscular blockers have a prolonged duration of action.
Benzylisoquinoliniums (atracurium and cis-atracurium) BMI, body mass index.
have predictable pharmacokinetic and pharmacodynamic
effects.75 Succinylcholine should be dosed according to
and reduces long-term morbidity, mortality, and health
actual body weight.76 In general, drugs that are rapidly
care use.12,8587
metabolized and those with low lipophilic properties are
The Swedish Obese Subjects (SOS) study compared
preferable in this patient population, and hence, morphine
different types of obesity surgery versus conservative
and hydromorphone are probably preferable to fentanyl
treatment in a matched pair design.88,89 This was a non-
for perioperative analgesia. Lipid insoluble volatile agents
randomized observational trial. Patients with a BMI >34
are more suitable in this regard. Sevoflurane has supe-
were studied over 2 years. There was a significantly greater
rior emergence properties to isoflurane,77 but recovery
weight loss after surgical than nonsurgical treatment,
from anesthesia is significantly faster with desflurane
which resulted in significant improvements of comor-
than with sevoflurane.78 Desflurane is associated with
bidities, such as diabetes (from a prevalence at baseline of
more rapid emergence from anesthesia and better oxy-
19% to 10% after 2 years), hypertension (from 53% to 31%
genation in the recovery room.79 There is no association
after 2 years), sleep apnea (from 23% to 8% after 2 years),
between increased BMI and postoperative nausea and
dyspnea when climbing stairs (from 87% to 19% after
vomiting.80
2 years), and chest pain when climbing stairs (from 28%
to 4% after 2 years). There was a significant difference in
the perceived quality of life amongst patients, particularly
women, dependent on absolute weight loss. The more
Are Bariatric Patients at weight the patients lost, the greater their quality of life.
Increased Risk for Aspiration For example, there was a significant difference in quality
of life between women with a 30 to 40 kg weight loss and
Pneumonitis? those with >40 kg weight loss. The authors suggest that
bariatric surgery should target the greatest possible weight
The risk of pulmonary aspiration due to high gastric loss.88,89 Weight loss significantly reduces cardiovascular
residual volume in morbidly obese patients appears to risk.12 (see Table 44.4)
be exaggerated.8183 Remarkably, there is little evidence Historically two distinct surgical approaches to bari-
that obese patients have a higher incidence of gastroe- atric surgery evolved: restrictive and malabsorptive proce-
sophageal reflux or aspiration pneumonitis. In fact, a dures (see Table 44.5). Restrictive procedures involve the
study of 256 fasted patients revealed significantly lower creation of a small gastric pouch, which fills rapidly, lead-
gastric residual volumes in obese patients (BMI >30) ing to early satiety. These include a variety of gastroplas-
compared with controls.81 Patients with a BMI >30, who ties and adjustable gastric banding. Malabsorptive proce-
drank 300 mL of clear fluid 2 hours preoperatively, had no dures involve bypassing a large section of the small bowel,
increase in gastric residual volume or decrease in gastric
pH compared to fasting controls,84 and therefore rapid se- TABLE 44.4 Cardiovascular Risk Reduction Benefits
quence induction for airway protection is unnecessary of Weight Loss
for most of the obese patients.
For every kilogram of weight loss the following favorable
changes occur:
Fasting serum cholesterol 1%
What Is Bariatric Surgery, and Low density lipoprotein cholesterol 0.7%
Is It Safe? Triglycerides 1.9%
High density lipoprotein cholesterol +0.2%
Systolic blood pressure 0.5%
The surgical treatment of MO is known as bariatric surgery. Diastolic blood pressure 0.4%
Morbidly obese patients rarely respond to medical and Blood glucose 0.2 mM
dietary therapy. Obesity surgery should be considered in
adult patients with a documented BMI 35 and related Data from: Anderson JW, Konz EC. Obesity and disease manage-
comorbidity, or a BMI of at least 40 (see Table 44.3). ment: Effects of weight loss on comorbid conditions. Obes Res.
Bariatric surgery reduces obesity related complications 2001;9(Suppl 4):326S.
630 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 44.5 Bariatric Operations: Mechanism of Action Postoperative surgical complications following RYGB
may be early or late. The most significant complications of
Restrictive Vertical banded RYGB are anastomotic leakage and bleeding.93 The pres-
gastroplasty ence of fever, tachycardia, and tachypnea following this
Adjustable gastric banding operation should alert the clinician to the possibility of
anastomotic leak. Most patients can be treated by drainage
Largely restrictive/mildly Roux-en-Y gastric bypass with or without oversewing. In some cases, surgery is re-
malabsorptive (RYGB) quired to correct the leak. This can be done through the
open or laparoscopic approach. Bleeding usually occurs
Largely Biliopancreatic diversion
at the staple line and can be treated conservatively. In-
malabsorptive/mildly Duodenal switch
deed, reoperation rarely identifies the source of bleeding.
restrictive Stomal ulceration occurs in up to one in six patients on the
thereby reducing the surface area for absorption. The jejunal side of the anastomosis due to acid leakage from
introduction of high osmolar material into the jejunum the gastric remnant.93 Stoma stenosis due to anastomotic
leads to a dumping syndrome and avoidance of food. The strictures usually occurs during the first postoperative
months.94 Most cases are amenable to endoscopic dilata-
more simple the procedure, the greater the likelihood of
tion, but some require surgical correction for persistence
weight regain, and the more complex the procedure, the
of stenosis or perforation caused by dilatation.
more likely complications will ensue.
The RYGB operation for severe obesity is now the Fernandez et al.8 looked at all-cause perioperative
most commonly performed bariatric operation in the outcomes following bariatric surgery. The mortality rate
United States (70% of bariatric operations).90 There are was 1.5%; the incidence of anastomotic leaks 3%, small
three major variations: Open, laparoscopic, and hand bowel obstruction 3%, and pulmonary embolism 1.14%.
assisted. The operation is similara 15 to 30 mL gastric There was a significant difference in mortality in patients
pouch is created, isolated from the distal stomach by a who had laporascopic-assisted and short-limbed gastric
21-mm stapled, circular anastomosis (internal diameter bypasses versus open and long-limbed bypasses. The latter
12 to 14 mm). A 75 to 150 cm, antecolic, antegastric Roux group was represented by patients with a significantly
limb is created, and a stapled side-side jejunojejunostomy higher BMI (57.8% vs. 46%, p < 0.0001). Preoperative
is fashioned. The purpose of the Roux limb is to divert the weight (BMI >50), hypertension, postoperative leak, and
pancreatic juice and bile. The longer the Roux limb, the pulmonary embolism significantly increased risk.
greater the degree of malabsorption. Flum et al.95 investigated a total of 16,155 patients
RYGB should be considered the gold standard weight who underwent bariatric procedures (mean age, 47.7 years
loss surgery with excellent short, intermediate, and long- [standard deviation (SD), 11.3 years]; 75.8% women).
term weight loss results. Sugerman et al. studied more The rates of 30-day, 90-day, and 1-year mortality were
than 1,025 patients who had undergone gastric bypass 2%, 2.8%, and 4.6%, respectively. There was a significant
surgery over 20 years.91 There was a strong correlation increase in risk for men and elderly patients. For example,
between weight loss and resolution of type 2 diabetes there was a fivefold increase in the risk of death for
and hypertension. Within 1 year postoperatively, patients patients older than 75 years.
had lost 66% of excess body weight, and hypertension Dindo et al. studied a prospective cohort of
had resolved in 69% and diabetes in 83%. There was a 6,336 patients undergoing elective general (nonbariatric)
strong relation between the magnitude of weight loss and surgery,71 of which 25% were classified as obese or mor-
improvement or resolution of comorbidities. This relation bidly obese, and no evidence of a relation between obesity
continued with follow-up beyond 5 years. Reddy et al. and mortality emerged. The only area in which obesity
reported a 33% mean excess weight loss following open increased risk was with wound infections (4% vs. 3%,
RYGB in 103 patients, with a mean follow-up time of p = 0.03).
5 months.92
POSTOPERATIVE MANAGEMENT
P A R T III
OF THE MORBIDLY OBESE PATIENT
traditional open gastric bypass surgery, there is good Liberation from mechanical ventilation is particularly
evidence that epidural anesthesia improves cardiovascular challenging. Tracheostomy bypasses redundant upper
function,97 and perhaps postoperative pulmonary func- airway tissue that predisposes to postextubation airway
tion. Conversely, Charghi et al.96 compared intravenous obstruction. Liberation to negative pressure ventilation
patient-controlled morphine to epidural analgesia, using should be achieved from higher levels of PEEP/CPAP
fentanyl and bupivicaine in patients undergoing gastric than would usually be expected.104 This ensures optimal
bypass surgery. The type of analgesia did not affect the alveolar recruitment at the time of extubation. In this
quality of pain control at rest, the frequency of nau- light, extubation to CPAP/BiPAP may be beneficial.105
sea and pruritus, the time to ambulation and return of Morbidly obese patients, whether in the intensive care
gastrointestinal function, or the length of hospital stay. unit or operating room, should always be extubated in
Patients receiving epidural analgesia had a greater risk the sitting or reverse Trendelenburg position to optimize
of wound infection than subjects with patient-controlled pulmonary mechanics immediately following extubation,
anesthesia. which is the highest risk period.39
Epidural analgesia is technically more difficult in MO
patients. It is useful to sit the patient upright to visualize
the midline structures. Longer needles are often required.
It is important to note that local anesthesia requirements What Other Postoperative
are usually 20% to 25% lower in this patient population
as compared to normal.
Complications Should I Be
For RYGB, a balanced multimodal approach is Aware of?
recommended.98 Intraoperative fentanyl was associated
with greater postoperative sedation than the use of a Morbidly obese patients are at increased risk of venous
medley of non-narcotic agents.99 Preoperative wound thromboembolism due to immobility and the procoagu-
infiltration significantly reduces postoperative pain.100 lant response to surgery. The pharmacokinetic effects of
various antithrombotics are unaltered in these patients.
Nevertheless, the predictable response to low molecular
weight heparinoids suggests superior utility. The opti-
How Do I Prevent and Treat mal dose of enoxaparin is 40 mg b.i.d.106 Although the
Postoperative Respiratory prophylactic use of inferior vena caval filters has been ad-
vocated in the super-obese, the efficacy of this approach
Failure? is unproven.
Meticulous attention must be applied to nutrition
The risk of postoperative respiratory failure and airway and pressure points. Bariatric patients are particularly
obstruction in patients with OSA (in particular those prone to protein malnutrition. Although the clinical per-
with AHI >30) cannot be overemphasized. All patients ception is of a patient who can afford a period of
should be recovered in the semirecumbent or reverse fasting, this is not the case. Protein, required for mus-
Trendelenburg position for the duration of hospitaliza- cular function, is metabolized preferentially, and lipid
tion. Early mobilization leads to lung recruitment and stores remain unchanged. Obese patients cannot toler-
should be encouraged. All patients with a diagnosis of ate significant further reduction in physiologic reserve.
OSA should receive CPAP or bilevel positive airway pres- The administration of glucose containing fluids predis-
sure (BiPAP) in the recovery room (this is titrated to poses the patient to hyperglycemia without providing
response) and at night while they sleep, unless verified as nutrition.
unnecessary by pulse oximeter. If patient-controlled anes- Tissue perfusion is significantly reduced in critically
thesia is used for analgesia, continuous (basal) infusions ill bariatric patients, particularly with extensive edema.
should be avoided. In patients with severe sleep apnea Hence, these patients are unusually vulnerable to pres-
and OHS, prophylactic tracheostomy is an option to con- sure ulceration, that is, ulcers that heal poorly. Patients
sider. cannot reposition themselves. Regular turning along with
The morbidly obese patient presents a particularly the use of specially designed bariatric beds may alle-
difficult challenge to critical care professionals. Extreme viate the problem. Early and aggressive mobilization
obesity (BMI >40) is an independent risk factor for death is imperative, even if this involves walking the patient
in prolonged stay, critically ill patients.101 In patients with a battery-driven mechanical ventilator. This en-
with acute lung injury, there is a dramatic increase in sures the maintenance of muscular integrity and lung
lung and chest wall elastance. This requires a modi- expansion.
fied approach to the mechanical ventilation strategy. The Rigorous screening for infections should be under-
lung injury is dominated by extensive atelectasis. Higher taken, with rapid source control. Diagnostic tests can
levels of PEEP are required to keep the lung open, be particularly challenging. Chest radiographs are fre-
and frequently higher transalveolar pressures than would quently uninterpretable because of the density of chest
normally be considered safe102 are required. Overdis- wall tissues. Computerized tomography is often impos-
tension injuries are unlikely due to severe stiffening of the sible because of weight limitations of 180 kg on most
chest wall.103 Tidal volumes should generally not exceed machines. Ultrasonographic examination of the abdomen
6 mL per kg. and lower limbs is often extremely limited.
632 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
CONCLUSION REFERENCES
1. Kuczmarski RJ, Flegal KM, Campbell SM, et al. Increasing
The patient described in the clinical scenario has a number prevalence of overweight among US adults. The National
of important medical risk factors. He is at elevated risk Health and Nutrition Examination Surveys, 19601991.
for perioperative myocardial ischemia, hypoxic respira- JAMA. 1991;272:205.
tory failure, airway obstruction, deep venous thrombosis, 2. Flegal KM, Carroll MD, Ogden CL, et al. Prevalence and
pressure ulceration, and wound infection. The combina- trends in obesity among US adults, 19992000. JAMA. 2002;
tion of a very high BMI, central obesity, and an AHI 288:1723.
>100 is of particular concern with regard to airway man- 3. Hedley AA, Ogden CL, Johnson CL, et al. Prevalence of
overweight and obesity among us children, adolescents,
agement. Although intubation can usually be performed
and adults, 19992002. JAMA. 2004;291:2847.
following the induction of anesthesia, one should, as
4. Freedman DS, Khan LK, Serdula MK, et al. Trends and
part of planning, consider performing a preemptive tra- correlates of class 3 obesity in the United States from 1990
cheostomy to prevent postoperative airway obstruction. through 2000. JAMA. 2002;288:1758.
In addition, postoperative mobilization may be difficult, 5. Calle EE, Thun MJ, Petrelli JM, et al. Body-mass index and
and prophylactic placement of an inferior vena cava (IVC) mortality in a prospective cohort of U.S. adults. N Engl J
filter should be considered as well. Following extuba- Med. 1999;341:1097.
tion, in the sitting position, this patient should receive 6. Mokdad AH, Bowman BA, Ford ES, et al. The continuing
CPAP immediately and in the recovery room. Owing to epidemics of obesity and diabetes in the United States.
JAMA. 2001;286:11951200.
his unique risk profile, admission to a high dependency
7. Steinbrook R. Surgery for severe obesity. N Engl J Med.
or intensive care unit for postoperative monitoring is 2004;350:1075.
recommended. 8. Fernandez AZ Jr, Demaria EJ, Tichansky DS, et al. Multi-
variate analysis of risk factors for death following gastric
bypass for treatment of morbid obesity. Ann Surg. 2004;
239:698702.
9. Bjorntorp P. Aging and body composition. Nutrition.
KEY POINTS 1997;13:572.
1. Obesity is a multisystem metabolic disease. 10. Nowbar S, Burkart KM, Gonzales R, et al. Obesity-
2. Morbidly obese patients are at increased risk for associated hypoventilation in hospitalized patients: Preva-
lence, effects, and outcome. Am J Med. 2004;116:1.
hypertension, coronary artery disease, and CHF.
11. Kenchaiah S, Evans JC, Levy D, et al. Obesity and the risk
3. The metabolic syndrome, which combines central of heart failure. N Engl J Med. 2002;347:305.
obesity, impaired glucose tolerance, dyslipidemia 12. Anderson JW, Konz EC. Obesity and disease management:
and hypertension, significantly increases medium- Effects of weight loss on comorbid conditions. Obes Res.
term and presumably perioperative risk. 2001;9(Suppl 4):326S.
4. The OSA syndrome occurs in 70% of morbidly 13. Reaven GM. Banting lecture 1988. Role of insulin resistance
obese patients and increases the risk of postoper- in human disease. Diabetes. 1988;37:1595.
ative airway obstruction. Postoperative noninvasive 14. National Cholesterol Education Program. Third report of
ventilation is recommended. the National Cholesterol Education Program (NCEP) expert
panel on detection, evaluation, and treatment of high blood
5. Morbidly obese patients develop significant periop-
pressure in adults (Adult Treatment Panel III): Final report.
erative atelectasis; this can be reduced by preinduc-
Circulation. 2002;106:3143.
tion CPAP and intraoperative PEEP. 15. Alberti KG, Zimmet PZ. Definition, diagnosis and classifi-
6. Mask ventilation and laryngoscopy can be difficult cation of diabetes mellitus and its complications. Part 1:
in morbidly obese patients; careful positioning with Diagnosis and classification of diabetes mellitus provi-
the patients head above the chest and abdomen sional report of a WHO consultation. Diabet Med. 1998;15:
ameliorates this problem. 539.
7. Many drug pharmacokinetics are affected by dra- 16. Moller DE, Kaufman KD. Metabolic syndrome: A clinical
matic increases in adipose tissue, but particularly and molecular perspective. Annu Rev Med. 2005;56:45.
17. Ford E, Giles W. A comparison of the prevalence of
succinylcholine should be dosed on the basis of
the metabolic syndrome using two proposed definitions.
actual, not ideal, body weight.
Diabetes Care. 2003;26:575.
8. Bariatric surgery is the most effective method 18. Ford ES, Giles WH, Mokdad AH. Increasing prevalence of
of inducing weight loss. Procedures are either the metabolic syndrome among U.S. adults. Diabetes Care.
restrictive or malabsorptive. RYGB combines both 2004;27:2444.
features. 19. Karelis AD, St Pierre DH, Conus F, et al. Metabolic and
9. Medical complications of bariatric surgery include body composition factors in subgroups of obesity: What do
respiratory failure, pulmonary embolism, pressure we know? J Clin Endocrinol Metab. 2004;89:2569.
sores, and deep venous thrombosis. Surgical com- 20. Kip KE, Marroquin OC, Kelley DE, et al. Clinical im-
portance of obesity versus the metabolic syndrome in
plications include anastomotic dehiscence, stomal
cardiovascular risk in women: A report from the Womens
ulceration, and stomal herniation.
Ischemia Syndrome Evaluation (WISE) Study. Circulation.
10. Morbidly obese, critically ill patients are at in- 2004;109:706.
creased risk. Extreme care should be taken with 21. Girman C, Rhodes T, Mercuri M. The metabolic syn-
respiratory care, pressure points, nutrition, and mo- drome and risk of major coronary events in the Scan-
bilization. dinavian Simvastatin Survival Study (4S) and the Air
CHAPTER 44/MORBID OBESITY 633
Force/Texas Coronary Atherosclerosis Prevention Study 42. Adnet F, Borron SW, Racine SX, et al. The intubation
(AFCAPS/TexCAPS). Am J Cardiol. 2004;93:136. difficulty scale (IDS): Proposal and evaluation of a new score
22. Wisse BE. The inflammatory syndrome: The role of adipose characterizing the complexity of endotracheal intubation.
tissue cytokines in metabolic disorders linked to obesity. Anesthesiology. 1997;87:1290.
J Am Soc Nephrol. 2004;15:2792. 43. Brodsky JB, Lemmens HJ, Brock-Utne JG, et al. Morbid
23. Pischon T, Girman CJ, Hotamisligil GS, et al. Pasma obesity and tracheal intubation. Anesth Analg. 2002;94:732.
adiponectin levels and risk of myocardial infarction in men. 44. Wilson ME, Spiegelhalter D, Robertson JA, et al. Predicting
JAMA. 2004;291:1730. difficult intubation. Br J Anaesth. 1988;61:211.
24. Pai JK, Pischon T, Ma J, et al. Inflammatory markers and 45. Ezri T, Gewurtz G, Sessler DI, et al. Prediction of difficult
the risk of coronary heart disease in men and women. laryngoscopy in obese patients by ultrasound quantification
N Engl J Med. 2004;351:2599. of anterior neck soft tissue. Anaesthesia. 2003;58:1111.
25. Ridker PM, Rifai N, Pfeffer M, et al. Elevation of tumor 46. Hiremath AS, Hillman DR, James AL, et al. Relationship
necrosis factor-{alpha} and increased risk of recurrent coro- between difficult tracheal intubation and obstructive sleep
nary events after myocardial infarction. Circulation. 2000; apnoea. Br J Anaesth. 1998;80:606.
101:2149. 47. Hillman DR, Platt PR, Eastwood PR. The upper airway
26. Koh KK, Han SH, Quon MJ. Inflammatory markers during anaesthesia. Br J Anaesth. 2003;91:31.
and the metabolic syndrome: Insights from therapeutic 48. Brodsky JB, Lemmens HJ, Brock-Utne JG, et al. Anesthetic
interventions. J Am Coll Cardiol. 2005;46:1978. considerations for bariatric surgery: Proper positioning is
27. Moller DE, Kaufman KD. Metabolic syndrome: A clinical important for laryngoscopy. Anesth Analg. 2003;96:1841.
and molecular perspective. Annu Rev Med. 2005;56:45. 49. Rich JM. Use of an elevation pillow to produce the head-
28. Wylie-Rosett J, Segal-Isaacson CJ, Segal-Isaacson A. Car- elevated laryngoscopy position for airway management in
bohydrates and increases in obesity: Does the type of morbidly obese and large-framed patients. Anesth Analg.
carbohydrate make a difference? Obes Res. 2004;12:124S. 2004;98:264.
29. Frey WC, Pilcher J. Obstructive sleep-related breathing 50. Berthoud MC, Peacock JE, Reilly CS. Effectiveness of
disorders in patients evaluated for bariatric surgery. Obes preoxygenation in morbidly obese patients. Br J Anaesth.
Surg. 2003;13:676. 1991;67:464.
30. OKeeffe T, Patterson EJ. Evidence supporting routine 51. Edmark L, Kostova-Aherdan K, Enlund M, et al. Optimal
polysomnography before bariatric surgery. Obes Surg. 2004; oxygen concentration during induction of general anesthe-
14:23. sia. Anesthesiology. 2003;98:28.
31. Knill RL, Moote CA, Skinner MI, et al. Anaesthesia with 52. Rusca M, Proietti S, Schnyder P, et al. Prevention of at-
abdominal surgery leads to intense REM sleep during the electasis formation during induction of general anesthesia.
first postoperative week. Anesthesiology. 1990;73:52. Anesth Analg. 2003;97:1835.
32. Guardiano SA, Scott JA, Ware JC, et al. The long-term 53. Herriger A, Frascarolo P, Spahn DR, et al. The effect
results of gastric bypass on indexes of sleep apnea. Chest. of positive airway pressure during pre-oxygenation and
2003;124:1615. induction of anaesthesia upon duration of non-hypoxic
33. Kessler R, Chaouat A, Schinkewitch P, et al. The obesity- apnoea. Anaesthesia. 2004;59:243.
hypoventilation syndrome revisited: A prospective study of 54. Coussa M, Proietti S, Schnyder P, et al. Prevention of
34 consecutive cases. Chest. 2004;120:369. atelectasis formation during the induction of general anes-
34. Meoli AL, Rosen CL, Kristo D, et al. Upper airway thesia in morbidly obese patients. Anesth Analg. 2004;98:
management of the adult patient with obstructive sleep 1491.
apnea in the perioperative periodavoiding complications. 55. Hedenstierna G, Rothen HU. Atelectasis formation during
Sleep. 2003;26:1060. anesthesia: Causes and measures to prevent it. J Clin Monit
35. American Society of Anesthesiologists Task Force. Practice Comput. 2000;16:329.
guidelines for the perioperative management of patients 56. Gander S, Frascarolo P, Suter M, et al. Positive end-
with obstructive sleep apnea: A report by the american expiratory pressure during induction of general anesthesia
society of anesthesiologists task force on perioperative increases duration of nonhypoxic apnea in morbidly obese
management of patients with obstructive sleep apnea. patients. Anesth Analg. 2005;100:580.
Anesthesiology. 2006;104:1081. 57. Pelosi P, Ravagnan I, Giurati G, et al. Positive end-
36. Auler JO Jr, Miyoshi E, Fernandes CR, et al. The effects expiratory pressure improves respiratory function in obese
of abdominal opening on respiratory mechanics during but not in normal subjects during anesthesia and paralysis.
general anesthesia in normal and morbidly obese patients: Anesthesiology. 1999;91:1221.
A comparative study. Anesth Analg. 2002;94:741. 58. Yoshino J, Akata T, Takahashi S. Intraoperative changes in
37. Pelosi P, Croci M, Ravagnan I, et al. The effects of body mass arterial oxygenation during volume-controlled mechanical
on lung volumes, respiratory mechanics, and gas exchange ventilation in modestly obese patients undergoing laparo-
during general anesthesia. Anesth Analg. 1998;87:654. tomies with general anesthesia. Acta Anaesthesiol Scand.
38. Eichenberger A, Proietti S, Wicky S, et al. Morbid obesity 2003;47:742.
and postoperative pulmonary atelectasis: An underesti- 59. Tusman G, Bohm SH, Melkun F, et al. Effects of the alveolar
mated problem. Anesth Analg. 2002;95:1788. recruitment manoeuver and PEEP on arterial oxygenation
39. Ungern-Sternberg BS, Regli A, Schneider MC, et al. Effect in anesthetized obese patients. Rev Esp Anestesiol Reanim.
of obesity and site of surgery on perioperative lung volumes. 2002;49:177.
Br J Anaesth. 2004;92:202. 60. Perilli V, Sollazzi L, Modesti C, et al. Comparison of pos-
40. Rose DK, Cohen MM. The airway: Problems and predictions itive end-expiratory pressure with reverse Trendelenburg
in 18,500 patients. Can J Anaesth. 1994;41:372. position in morbidly obese patients undergoing bariatric
41. Ezri T, Warters RD, Szmuk P, et al. The incidence of class surgery: Effects on hemodynamics and pulmonary gas ex-
zero airway and the impact of Mallampati score, age, sex, change. Obes Surg. 2003;13:605.
and body mass index on prediction of laryngoscopy grade. 61. Sprung J, Whalley DG, Falcone T, et al. The effects of tidal
Anesth Analg. 2001;93:10731075. volume and respiratory rate on oxygenation and respiratory
634 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
mechanics during laparoscopy in morbidly obese patients. 81. Harter RL, Kelly WB, Kramer MG, et al. A comparison of
Anesth Analg. 2003;97:268. the volume and pH of gastric contents of obese and lean
62. Boyce JR, Ness T, Castroman P, et al. A preliminary study of surgical patients. Anesth Analg. 1998;86:147.
the optimal anesthesia positioning for the morbidly obese 82. Juvin P, Fevre G, Merouche M, et al. Gastric residue is not
patient. Obes Surg. 2003;13:4. more copious in obese patients. Anesth Analg. 2001;93:1621.
63. Perilli V, Sollazzi L, Bozza P, et al. The effects of the reverse 83. Vaughan RW, Bauer S, Wise L. Volume and pH of gastric
trendelenburg position on respiratory mechanics and blood juice in obese patients. Anesthesiology. 1975;43:686.
gases in morbidly obese patients during bariatric surgery. 84. Maltby JR, Pytka S, Watson NC, et al. Drinking 300 mL
Anesth Analg. 2000;91:1520. of clear fluid two hours before surgery has no effect on
64. Casati A, Comotti L, Tommasino C, et al. Effects of gastric fluid volume and pH in fasting and non-fasting
pneumoperitoneum and reverse Trendelenburg position obese patients. Can J Anaesth. 2004;51:111.
on cardiopulmonary function in morbidly obese patients 85. Gould JC, Garren MJ, Starling JR. Laparoscopic gastric
receiving laparoscopic gastric banding. Eur J Anaesthesiol. bypass results in decreased prescription medication costs
2000;17:300. within 6 months. J Gastrointest Surg. 2004;8:983.
65. Baraka AS, Hanna MT, Jabbour SI, et al. Preoxygenation 86. Christou NV, Sampalis JS, Liberman M, et al. Surgery
of pregnant and nonpregnant women in the head-up versus decreases long-term mortality, morbidity, and health care
supine position. Anesth Analg. 1992;75:757. use in morbidly obese patients. Ann Surg. 2004;240:416.
66. Dixon BJ, Dixon JB, Carden JR, et al. Preoxygenation is 87. Ferchak CV, Meneghini LF. Obesity, bariatric surgery and
more effective in the 25 degrees head-up position than in type 2 diabetesa systematic review. Diabetes Metab Res
the supine position in severely obese patients: A randomized Rev. 2004;20:438.
controlled study. Anesthesiology. 2005;102:1110. 88. Karlsson J, Sjostrom L, Sullivan M. Swedish obese subjects
67. Altermatt FR, Munoz HR, Delfino AE, et al. Pre-oxygenation (SOS)an intervention study of obesity. Two-year follow-up
in the obese patient: Effects of position on tolerance to of health-related quality of life (HRQL) and eating behavior
apnoea. Br J Anaesth. 2005;95:706709. after gastric surgery for severe obesity. Int J Obes Relat
68. Pelosi P, Croci M, Calappi E, et al. Prone positioning Metab Disord. 1998;22:113.
improves pulmonary function in obese patients during 89. Karason K, Lindroos AK, Stenlof K, et al. Relief of cardiores-
general anesthesia. Anesth Analg. 1996;83:578. piratory symptoms and increased physical activity after
69. Juvin P, Blarel A, Bruno F, et al. Is peripheral line placement surgically induced weight loss: Results from the Swedish
more difficult in obese than in lean patients? Anesth Analg. Obese Subjects study. Arch Intern Med. 2000;160:1797.
2003;96:1218. 90. Schauer PR, Ikramuddin S. Laparoscopic surgery for
70. Gilbert TB, Seneff MG, Becker RB. Facilitation of internal morbid obesity. Surg Clin North Am. 2001;81:1145.
jugular venous cannulation using an audio-guided Doppler 91. Sugerman HJ, Wolfe LG, Sica DA, et al. Diabetes and
ultrasound vascular access device: Results from a prospec- hypertension in severe obesity and effects of gastric bypass-
tive, dual-center, randomized, crossover clinical study. Crit induced weight loss. Ann Surg. 2003;237:751.
Care Med. 1995;23:60. 92. Reddy RM, Riker A, Marra D, et al. Open Roux-en-Y gastric
71. Dindo D, Muller MK, Weber M, et al. Obesity in general bypass for the morbidly obese in the era of laparoscopy.
elective surgery. Lancet. 2003;361:2032. Am J Surg. 2002;184:611.
72. Kabon B, Nagele A, Reddy D, et al. Obesity decreases 93. Byrne TK. Complications of surgery for obesity. Surg Clin
perioperative tissue oxygenation. Anesthesiology. 2004;100: North Am. 2001;81:1181.
274. 94. Sanyal AJ, Sugerman HJ, Kellum JM, et al. Stomal
73. Schulz-Stubner S. Bilateral occipital neuropathy as a complications of gastric bypass: Incidence and outcome
rare complication of positioning for thyroid surgery in a of therapy. Am J Gastroenterol. 1992;87:1165.
morbidly obese patient. Acta Anaesthesiol Scand. 2004;48: 95. Flum DR, Salem L, Broeckel Elrod JA, et al. Early mortality
126. among medicare beneficiaries undergoing bariatric surgical
74. Khurana RN, Baudendistel TE, Morgan EF, et al. Postoper- procedures. JAMA. 2005;294:1903.
ative rhabdomyolysis following laparoscopic gastric bypass 96. Charghi R, Backman S, Christou N, et al. Patient controlled
in the morbidly obese. Arch Surg. 2004;139:73. i.v. analgesia is an acceptable pain management strategy in
75. Ogunnaike BO, Jones SB, Jones DB, et al. Anesthetic morbidly obese patients undergoing gastric bypass surgery.
considerations for bariatric surgery. Anesth Analg. 2002;95: A retrospective comparison with epidural analgesia. Can J
1793. Anaesth. 2003;50:672.
76. Rose JB, Theroux MC, Katz MS. The potency of succinyl- 97. Gelman S, Laws HL, Potzick J, et al. Thoracic epidural vs
choline in obese adolescents. Anesth Analg. 2000;90:576. balanced anesthesia in morbid obesity: An intraoperative
77. Torri G, Casati A, Albertin A, et al. Randomized comparison and postoperative hemodynamic study. Anesth Analg. 1980;
of isoflurane and sevoflurane for laparoscopic gastric 59:902.
banding in morbidly obese patients. J Clin Anesth. 2001;13: 98. Schumann R, Jones SB, Ortiz VE, et al. Best practice
565. recommendations for anesthetic perioperative care and
78. De Baerdemaeker LE, Struys MM, Jacobs S, et al. Opti- pain management in weight loss surgery. Obes Res. 2005;13:
mization of desflurane administration in morbidly obese 254.
patients: A comparison with sevoflurane using an inhala- 99. Feld JM, Laurito CE, Beckerman M, et al. Non-opioid
tion bolus technique. Br J Anaesth. 2003;91:638650. analgesia improves pain relief and decreases sedation after
79. Juvin P, Vadam C, Malek L, et al. Postoperative recovery gastric bypass surgery. Can J Anaesth. 2003;50:336.
after desflurane, propofol, or isoflurane anesthesia among 100. Schumann R, Shikora S, Weiss JM, et al. A comparison of
morbidly obese patients: A prospective, randomized study. multimodal perioperative analgesia to epidural pain man-
Anesth Analg. 2000;91:714. agement after gastric bypass surgery. Anesth Analg. 2003;
80. Bellon F, Lapresle C. Chemical structure of two fragments 96:469.
of human serum albumin and their location in the albumin 101. Nasraway SA Jr, Albert M, Donnelly AM, et al. Morbid
molecule. Biochem J. 1975;147:585. obesity is an independent determinant of death among
CHAPTER 44/MORBID OBESITY 635
surgical critically ill patients. Crit Care Med. 2006;34: College of Chest Physicians; The American Association for
964. Respiratory Care; and the American College of Critical Care
102. Moloney ED, Griffiths MJ. Protective ventilation of patients Medicine. Chest. 2001;120:375S.
with acute respiratory distress syndrome. Br J Anaesth. 105. Joris JL, Sottiaux TM, Chiche JD, et al. Effect of bi-
2004;92:261. level positive airway pressure (BiPAP) nasal ventilation on
103. Dreyfuss D, Saumon G. Ventilator-induced lung injury: the postoperative pulmonary restrictive syndrome in obese
Lessons from experimental studies. Am J Respir Crit Care patients undergoing gastroplasty. Chest. 1997;111:665.
Med. 1998;157:294. 106. Scholten DJ, Hoedema RM, Scholten SE. A comparison of
104. MacIntyre NR, Cook DJ, Ely EW Jr, et al. Evidence- two different prophylactic dose regimens of low molecular
based guidelines for weaning and discontinuing ventilatory weight heparin in bariatric surgery. Obes Surg. 2002;
support: A collective task force facilitated by the American 12:19.
H . D I SOR D E R S OF
T E M P E R AT U R E R EG U L AT I O N
CHAPTER UNINTENTIONAL PERIOPERATIVE
45
HYPOTHERMIA
Anthony G. Doufas
CASE SUMMARY
What Baseline Knowledge Is
65-year-old, 67-kg woman with a history of
Relevant?
A
hypertension undergoes a vaginal hysterec-
tomy for a uterine myoma. Her preoperative
hemoglobin and blood pressure are 11 g Tc is among the most tightly controlled physiologic
per dL and 160/87 mm Hg, respectively. An parameters in humans. Although normal circadian and
epidural catheter is placed for postoperative other physiology-driven variations in Tc exist, at any
pain management. After induction of general anesthe- given time robust thermoregulatory control does not allow
sia, the patient is placed in a dorsal lithotomy position. more than a few tenths of a degree deviation from the
Her core body temperature (Tc) (esophageal) is 36.5 C expected set point. In awake humans, any type of thermal
before surgical incision. Thermal management includes insult is readily counteracted by specific physiologic
intravenous fluid warming and application of traditional responses, the intensities of which are proportional to the
blankets on the upper body. One hour after incision, Tc need for adequate temperature control. During anesthesia
is 35.5 C and, despite adequate analgesia and hypnosis, and surgery, inhibition of the normal thermoregulatory
heart rate and blood pressure gradually increase. The sur- defenses is the major cause for a typical decrease by 1 C
geon recognizes technical difficulties in completing the to 3 C in the Tc of the unwarmed patient. Therefore,
operation vaginally and proceeds with the abdominal hys- albeit a widely accepted definition of perioperative mild
terectomy. Upper body forced-air warming is started, and hypothermia is lacking, an unvoiced consensus defines it
8 mL of 0.25% bupivacaine is administered through the as a Tc between 34 C and 36 C.
epidural catheter. Thirty minutes later, blood pressure is
100/66 mm Hg and Tc is 34.9 C. After a 3 1/2-hour proce-
dure, during which the patient received 4 L crystalloids
and 4 units of blood, she arrives in the recovery area with What Is the Underlying
her trachea intubated, with a Tc of 35.3 C and hemoglobin
of 9 g per dL. Blood pressure is 170/90 mm Hg, and heart Physiology of Intraoperative
rate is 110 beats per minute with 6 to 10 multiform, Hypothermia?
ventricular extrasystoles per minute. Full body forced-air
warming is applied, and 10 mg of labetalol, IV bolus is
administered while an epidural analgesic infusion is ini-
tiated for pain relief. The patients trachea is extubated TEMPERATURE MONITORING
1 hour later when her oral temperature is 36 C, and she
is discharged to the ward after an overall recovery time of The core thermal compartment is comprised of highly
2 1/2 hours. perfused tissues, with a temperature that is uniform and
636
C H A P T E R 4 5 / U N I N T E N T I O N A L P E R I O P E R AT I V E H Y P O T H E R M I A 637
34
C
EVOLUTION
OF HYPOTHERMIA Anesthetic 1 2 3 4 5
induction
The administration of anesthesia is almost invariably as- Time (h)
sociated with a decrease in Tc of 1 C to 3 C, depending
on the type and dose of the anesthetic, amount of sur- FIGURE 45.1 Schematic presentation of the three phases of
gical exposure, and ambient temperature. Intraoperative intraoperative hypothermia, based on data from Matsukawa
hypothermia follows a characteristic pattern that consists et al.4 and Kurz et al.3 The human body, as shown in the inset, is
of three distinct phases: Redistribution, linear decrease, denoted by a 5-limb star with a dense black core representing
and Tc plateau (see Fig. 45.1).3,4 the high-heat central thermal compartment and a low-heat
surrounding gray area indicating the peripheral tissues. A: The
Redistribution typical preoperative patient is vasoconstricted with a large
core-to-peripheral tissue temperature gradient and a clear
In awake humans, the core thermal compartment con- distinction between the two thermal compartments. B: After
sists of well perfused tissues of the trunk and head that anesthesia-induced peripheral vasodilation, heat driven by the
are maintained at a 2 C to 4 C higher temperature than temperature gradient flows toward the periphery of the body.
the rest of the body. This normal core-to-peripheral tissue Therefore, core temperature decreases, and the two body
temperature gradient is maintained by tonic thermoreg- compartments become more homogeneous (redistribution
ulatory vasoconstriction of arteriovenous shunts in the hypothermia, AB). C: Radiation- and convection-mediated
fingers and toes.5 losses have the higher impact on systemic heat balance during
General anesthesia inhibits tonic thermoregulatory the linear phase in the core temperature decrease (BC). At this
vasoconstriction by both a central and a peripheral vasodi- point, hypothermia activates thermoregulatory vasoconstriction
lating effect. Vasodilation promotes the redistribution of in an effort to constrain heat in the body core and restore the
heat from the core compartment to the peripheral tissues normal core-to-periphery temperature gradient. D: During
of the body, resulting in a relatively hypothermic core thermal plateau (CD), core temperature becomes stable or
(Fig. 45.1, AB). Up to 80% of the typical decrease of even increases slightly. However, the total heat content of the
1.5 C in Tc during the first hour of anesthesia is attributed body continues to decline, largely at the expense of the
to heat redistribution.4 peripheral thermal compartment. (Data from: Matsukawa T,
Sessler DI, Sessler AM, et al. Heat flow and distribution during
Linear Phase induction of general anesthesia. Anesthesiology. 1995;82:662 and
Kurz A, Sessler DI, Christensen R, et al. Heat balance and
The redistribution phase of hypothermia is followed by distribution during the core-temperature plateau in anesthetized
a slow, linear decrease in Tc that represents a net heat humans. Anesthesiology. 1995;83:491.)
loss to the environment (Fig. 45.1, BC) through the skin
or the operating field.3 Total cutaneous heat loss can be
considered as a linear function of the skin-to-ambient Radiation is the transfer of heat between two sur-
temperature difference and is mediated through four faces through photons and depends on the temperature
different mechanisms: Radiation, conduction, convection, difference between the two bodies, as well as their ability
and evaporation. Radiation is the most important of those to absorb and emit heat (emissivity). Human skin has a
mechanisms and, when combined with convection, is high emissivity, meaning that it absorbs and emits heat
responsible for 70% to 90% of the total heat loss in the very efficiently. Conduction, on the other hand, is respon-
intraoperative patient.2,6 sible for the direct transfer of heat between two adjacent
638 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
a Only prospective, randomized human trials are included; subjective responses were evaluated by observers blinded to treatment group and core temperature.
b Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events. A randomized clinical
trial. JAMA. 1997;14:1127.
c Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. Study of Wound
Infection and Temperature Group. N Engl J Med. 1996;19:1209.
d Schmied H, Kurz A, Sessler DI, et al. Mild hypothermia increases blood loss and transfusion requirements during total hip arthroplasty. Lancet.
1996;347(8997):289.
e Winkler M, Akca O, Birkenberg B, et al. Aggressive warming reduces blood loss during hip arthroplasty. Anesth Analg. 2000;4:978.
f Widman J, Hammarqvist F, Sellden E. Amino acid infusion induces thermogenesis and reduces blood loss during hip arthroplasty under spinal anesthesia. Anesth
Analg. 2002;6:1757.
g Johansson T, Lisander B, Ivarsson I. Mild hypothermia does not increase blood loss during total hip arthroplasty. Acta Anaesthesiol Scand. 1999:10:1005.
h Lenhardt R, Marker E, Goll V, et al. Mild intraoperative hypothermia prolongs postanesthetic recovery. Anesthesiology. 1997;6:1318.
i Kurz A, Sessler DI, Narzt E, et al. Postoperative hemodynamic and thermoregulatory consequences of intraoperative core hypothermia. J Clin Anesth. 1995;5:359.
N, total number of subjects; P, percent of subjects who developed consequence; Trmcore , difference in core temperature between the treatment groups.
Different outcomes of the same study are presented in separate rows; NS, nonsignificant; VAS is a 100-mm long visual analog scale (0 mm = intense cold,
100 mm = intense heat).
enhances the ability of platelets to respond to activat- cell-dependent antibody production.39 Approximately a
ing stimuli in vitro28 and heightens aggregation.29 These 1 C decrease in intraoperative Tc was associated with a
findings show that, in the range of temperatures com- suppressed lymphocyte activation and reduced immunity-
monly encountered intraoperatively, the inhibition of promoting cytokines at 24 and 48 hours after surgery.39 In
intrinsic platelet function is not the cause of coagulopathy. addition, the in vitro phagocytic capacity of neutrophils,
Furthermore, these observations lend support to the hy- as well as the intraoperative production of reactive
pothesis that hypothermia may promote coagulopathy by oxygen species in surgical patients, declines during
reducing the availability of platelet activators.28 mild hypothermia in a temperature-dependent manner.40
When assayed at hypothermic temperatures, plasma Because oxidative bacterial killing is partly dependent
behaves as if it is clotting factor-deficient. At a Tc of on oxygen availability,41 any reduction of the latter
35 C, clotting is prolonged equal to that caused by an may indirectly impair neutrophil function. Consequently,
18% to 35% reduction in the various clotting factors.30 mild intraoperative hypothermia during the first few
As a result, coagulation function tests are greatly pro- decisive hours following bacterial contamination42 may
longed during mild intraoperative hypothermia, although weaken the local response to infection by triggering
the clinical importance of this prolongation remains de- subcutaneous vasoconstriction, thereby producing tissue
batable. Conversely, the balance between clot formation hypoxia.
and lysis is not affected by mild-to-moderate hypothermia.
Thromboelastographic evidence from patients undergo- Clinical Infection and Impaired Healing
ing cardiopulmonary bypass suggests that hypothermia
does not affect clot strength and delays clot formation of the Surgical Wound
rather than facilitates clot degeneration.31
Consistent with this evidence, patients whose intraopera-
tive Tc was only 1.9 C lower than the normothermic group
Surgical Blood Loss had three times the incidence of surgical wound infection
following colon surgery.43 These infections were clinically
In 1996, a double-blind, controlled, randomized trial significant as indicated by the fact that infected patients,
demonstrated that patients undergoing elective hip on the average, were hospitalized 1 week longer than
arthroplasty with a Tc only 1.6 C lower than normother- uninfected patients. Interestingly, preoperative warming
mic patients had a 500 mL (30%) increased blood loss and decreased the incidence of wound infection after clean
a significantly augmented allogeneic transfusion require- surgery, although the standard treatment group was not
ment.32 The same investigators confirmed the hemostatic hypothermic.44 These findings indirectly support a role for
benefits of maintaining intraoperative normothermia in a decreased peripheral tissue perfusion in the pathogenesis
subsequent retrospective analysis.33 In contrast, another of wound infection.
study of blood loss during hip arthroplasty failed to iden- Hypothermia increased the duration of hospitaliza-
tify a beneficial effect for intraoperative normothermia.34 tion by 20%, even when infected patients were excluded
Possible explanations for the different findings between from the analysis. This finding probably indicates an im-
these two, well-designed studies may include differences paired wound healing process, as was demonstrated by
in the methods employed to evaluate blood loss or even decreased collagen deposition around the surgical inci-
in the applied surgical technique. Recently, two more sion.43 It is consistent with a previous study showing
randomized controlled trials confirmed that only approx- that mild hypothermia aggravates postoperative protein
imately 0.5 C-core hypothermia increases blood loss by wasting.45 Furthermore, dysfunctional hemostasis may
200 to 300 mL in patients undergoing hip arthroplasty also lead to a slow healing process, as platelet activation,
and spinal anesthesia.35,36 There seems to be little doubt which is particularly affected by hypothermia, plays a
that hypothermia causes a clinically important coagulo- primary role in initiating wound healing.46
pathy because all but one randomized trial have identified
increased blood loss.
ALTERED ANESTHETIC
PHARMACOLOGY
SURGICAL SITE INFECTION
Surgical site infection is the most common, preventable,
Muscle Relaxants
adverse outcome after a major operation. Patients who Adductor pollicis temperature is primarily determined by
develop a surgical site infection have a twofold increase in the Tc of the blood perfusing the muscle. In the absence
the length of hospital stay and the risk of death, although of muscle relaxants, adductor pollicis twitch response
the associated health care cost is greatly augmented.37 decreases by approximately 10% per degree centigrade
reduction in body temperature during mild hypothermia.
Hypothermia-Impaired Wound Immunity This effect is potentiated by muscle relaxants (20% per
degree centigrade in the presence of vecuronium). How-
Experimental and human evidence indicate that mild core ever, the direct effect of hypothermia on muscle strength
hypothermia directly impairs various components of the is probably of limited clinical importance compared to
immune system, such as natural killer cell activity38 and the marked effects on drug kinetics.
C H A P T E R 4 5 / U N I N T E N T I O N A L P E R I O P E R AT I V E H Y P O T H E R M I A 641
Leg up position
Volatile Anesthetics
Animal studies support a decrease in the minimum alveo-
lar concentration of halothane and isoflurane by roughly 37C 34C
5% per degree centigrade reduction in Tc.48 This effect
has also been demonstrated in children during isoflurane
anesthesia.49 The increased tissue solubility of volatile
agents in lower-than-normal temperatures may be re-
sponsible for both increasing the potency and delaying Pre induction warming
the washout of inhalational anesthetics. Pre induction peripheral vasodilation
from reduced intercompartmental clearances and is not PEEP application in leg up position
associated with decreased hepatic blood flow.50 The same
degree of hypothermia, however, did not change the FIGURE 45.2 Important factors that have been found to
propofol requirement for loss of responsiveness during promote () or prevent the development of intraoperative
anesthesia for a craniotomy.51 hypothermia (see text for details). PEEP, positive end-expiratory
pressure.
DELAYED POSTANESTHETIC
RECOVERY How Can Perioperative
A prospective, randomized trial demonstrated that mild Hypothermia Be Prevented or
hypothermia delayed the discharge of adult patients Treated?
from the postanesthesia care unit by approximately
40 minutes. When normothermia (Tc >36 C) was also
Hypothermia augments the risk for adverse outcomes in
included in the discharge criteria, the difference between
many surgical patients or, at the very least, provokes shiv-
the two groups increased to approximately 2 hours.52
ering and thermal discomfort postoperatively. Currently,
This effect of hypothermia has not been demonstrated
there is no reliable model that can be used to predict
in infants and children, although these patients were
the magnitude or consequences of hypothermia based on
not randomly assigned to specific intraoperative thermal
preoperative patient parameters.
management.53
The most identified hypothermia-related complica-
tions are established intraoperatively; it is, therefore,
important to keep surgical patients warm rather than to
THERMAL COMFORT allow them to cool and then warm them postoperatively.
An effective plan for the intraoperative maintenance of
Mild hypothermia produces substantial postoperative normothermia must focus on minimizing heat redistri-
thermal discomfort.54 Major thermoregulatory re- bution and preventing heat loss during anesthesia and
sponses55 and hypothermia-related complications are surgery (see Fig. 45.2).
primarily determined by Tc. However, core body and
mean skin temperatures contribute almost equally toward
thermal comfort sensation.56 As a consequence, although MINIMIZE REDISTRIBUTION
cutaneous warming is not immediately effective in restor-
ing Tc, it greatly improves thermal comfort in mildly Redistribution is the primary cause of intraoperative hy-
hypothermic patients recovering from anesthesia.57 pothermia 2 to 3 hours after the induction of general
642 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
anesthesia.4 Two major factors influence the extent of A high body surface area-to-weight ratio (e.g., infants
redistribution hypothermia: (i) the degree of anesthesia- and small children), large surgical exposure, and a cool
induced, central inhibition of thermoregulatory vasocon- environment are all associated with greater heat loss and
striction and (ii) the magnitude of the core-to-peripheral hypothermia. Also, central baroreceptor loading in pa-
tissue temperature gradient before anesthetic induction. tients undergoing surgery in the leg-up position aggravates
The latter is inversely proportional to the patients total intraoperative hypothermia through a centrally medi-
initial heat content. Increasing the systemic heat con- ated inhibition of thermoregulatory vasoconstriction.70
tent by warming the peripheral body tissues decreases In long operations, this effect may delay the activation
the core-to-periphery temperature gradient and, there- of thermoregulatory defenses to hypothermia (thermal
fore, inhibits the natural drive for heat redistribution. plateau) and result in further heat loss. The effect of
In addition, preinduction warming activates a heat dis- body position on thermoregulation is readily reversed by
sipation process, causing peripheral vasodilation. The baroreceptor unloading through the application of posi-
subsequent induction of anesthesia has only a minimal tive end-expiratory pressure (PEEP).70
vasomotor effect because the centrally mediated vasocon- Effective body insulation and active cutaneous warm-
striction is already impeded and results in diminished ing can diminish systemic heat loss during anesthesia and
redistribution.58 surgery. Because Tc is the single most important body
Any chronic59 or acute60,61 pharmacologic inter- temperature, effective warming depends directly on the
vention that promotes central59,61 and/or peripheral60,61 amount of heat transferred to the core. Heat flow within
inhibition of thermoregulatory vasoconstriction before the body is a function of vasomotor tone, which influ-
anesthesia similarly decreases the core-to-periphery tem- ences both the amount of blood flow to extremities and
perature gradient, thereby preventing or minimizing the extent to which countercurrent heat exchange between
postinduction redistribution hypothermia. Of course, an the arterial and venous side of circulation reduces heat
important physiologic requirement for redistribution hy- transfer to the core.71 Therefore, the reduction of total
pothermia is an adequate intravascular volume, which heat loss and restoration of core normothermia are best
serves as the heat carrier in the body. A recent study achieved intraoperatively when anesthesia-induced pe-
demonstrated that conservative perioperative fluid man- ripheral vasodilation facilitates intercompartmental heat
agement (1 vs. 8 mL/kg/hour) in children undergoing transfer. This relation typically occurs during the redis-
minor surgery not only ameliorated redistribution hy- tribution and linear phases of Tc decrease, before the
pothermia, but also increased intraoperative Tc to higher- emergence of active thermoregulatory vasoconstriction
than-preinduction values.62 (thermal plateau).
The perioperative infusion of amino acids has been An alternative, internal warming method is the pe-
shown to prevent hypothermia and several hypothermia- rioperative infusion of amino acids that considerably
related complications36,63,64 in surgical patients receiving increases metabolic heat production in anesthetized pa-
general63,65,66 or neuraxial67 anesthesia. Amino acids tients.65 In addition to the beneficial effects on ther-
augment peripheral thermogenesis65 and increase the set mogenesis, amino acids also increase the threshold for
point for all autonomous thermoregulatory defenses.68 As thermoregulatory vasoconstriction72 and hasten the emer-
a result, total body heat content in the preinduction stage gence of centrally mediated defenses to core hypother-
rises, and anesthesia-induced redistribution does not lead mia. The latter mechanism facilitates the conservation of
to core hypothermia. metabolically produced heat.
Body morphology is another important factor that
influences redistribution hypothermia after anesthetic in-
duction. To facilitate the dissipation of metabolic heat PATIENT WARMING SYSTEMS
from their well-insulated bodies, obese patients spend
much of their time vasodilated, which differs from the An effective warming system must modulate cutaneous
constricted state most patients maintain preoperatively. heat loss because roughly 90% of metabolic heat is
As a result, obesity is associated with a decreased core- lost through the skin surface.73 The noninvasive sys-
to-periphery temperature gradient and reduced redistri- tems presently available can be categorized as passive
bution. Conversely, very thin patients redistribute more insulation or active cutaneous heating. However, internal
than average-weight patients.69 This relation is impor- chemical and invasive warming methods also exist.
tant. Patients with a preoperative thermoregulatory status Single- and multilayer passive insulators reduce heat
particularly prone to redistribution hypothermia should loss by 30% to 50%.74 This amount is clinically important
receive special preanesthetic care. and sometimes is sufficient to restore thermal steady state.
The efficacy of passive insulators depends on the size of
the skin area covered, rather than the material of the in-
REDUCE HEAT LOSS sulator per se,74 because passive insulators operate mainly
by eliminating convection-mediated heat loss. The still air
During redistribution, heat loss is not the major cause that is trapped between the cover and skin surface retains
of hypothermia. Rather, it is during the linear, and to most of the heat. In the theoretical scenariothat is, when
some extent the thermal, plateau phases of hypothermia the perfect insulation reduces heat loss to zerobody tem-
(Fig. 45.1) that factors influencing total heat loss from the perature would increase by only 1 C per hour. In practice,
body can have an apparent effect. even the best insulation rarely reduces heat loss by even
C H A P T E R 4 5 / U N I N T E N T I O N A L P E R I O P E R AT I V E H Y P O T H E R M I A 643
FIGURE 45.3 Passive insulation, active cutaneous heating with Negative Pressure Warming
forced-air or circulating-water garments, and internal warming
Negative pressure warming is another noninvasive warm-
are the most effective warming methods for the perioperative
ing method that uses a slight vacuum applied to the hand
patient.
and forearm to facilitate peripheral-to-core heat transfer.
The high rates of body core rewarming (10 C per hour)
that were originally demonstrated by the inventor of the
50%. As a result, active cutaneous warming is commonly technology80 were not confirmed by subsequent studies in
required to compensate for the extensive heat losses asso- patients81 or volunteers.82
ciated with major surgery in a cool environment. Active
systems maintain normothermia better than passive in-
sulators,75 and their effectiveness is proportional to the
Internal Warming Methods
treated skin surface area. Forced air warming, resistive Heat loss caused by cold intravenous fluids becomes
heating, and circulating-water garments are currently the significant when large amounts of crystalloid solution
most effective noninvasive options (see Fig. 45.3). How- or blood are administered to the patient. However, fluid
ever, other noninvasive methods such as negative-pressure warming does not warm the patients to any important
and internal warming have also been used. extent, because fluids cannot be heated much above the
normal body temperature. Therefore, fluid warmers alone
Forced-Air and Resistive Heating will not keep patients normothermic and should not
be used as substitutes for passive insulation or active
Forced-air warming reduces radiation heat loss by replac- warming methods. Similarly, the heat amount that is
ing cool operating room surfaces with a warm cover. Most transferred directly to the core of the body through
importantly, it transfers heat to the body through convec- heating and humidifying inspiratory gases is not sufficient
tion, or facilitated conduction. Heat conduction from the to maintain intraoperative normothermia.73
still air to the body surface increases by an order of mag- In contrast, invasive catheter systems that use
nitude if the air moves rapidly over the skin. Therefore, countercurrent heat exchange mechanisms have been
convection by forced-air warming increases heat gain successfully applied for the thermal management of the in-
by 30 to 50 W.76 In contrast, passive insulation reduces traoperative patient.83 Also, amino acid infusion, has been
normal cutaneous heat loss from approximately 100 to effectively used to ameliorate intraoperative hypothermia
70 W.74 Clinical studies suggest that resistive heating (elec- as a result of an increase in metabolic heat production
tric) blankets are equally effective as forced-air warming and centrally mediated activation of thermoregulatory
systems in maintaining intraoperative normothermia.77 vasoconstriction.72
644 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
8. Sessler DI, Rubinstein EH, Eger EI II. Core temperature 28. Faraday N, Rosenfeld BA. In vitro hypothermia enhances
changes during N2 O fentanyl and halothane/O2 anesthesia. platelet GPIIb-IIIa activation and P-selectin expression.
Anesthesiology. 1987;67:137. Anesthesiology. 1998;88:1579.
9. Sessler DI, McGuire J, Hynson J, et al. Thermoregulatory 29. Scharbert G, Kalb M, Marschalek C, et al. The effects of test
vasoconstriction during isoflurane anesthesia minimally temperature and storage temperature on platelet aggrega-
decreases cutaneous heat loss. Anesthesiology. 1992;76:670. tion: A whole blood in vitrostudy. Anesth Analg. 2006;102:
10. Belani K, Sessler DI, Sessler AM, et al. Leg heat content 1280.
continues to decrease during the core temperature plateau in 30. Johnston TD, Chen Y, Reed RL II. Functional equivalence
humans anesthetized with isoflurane. Anesthesiology. 1993; of hypothermia to specific clotting factor deficiencies.
78:856. J Trauma. 1994;37:413.
11. Bissonnette B, Sessler DI. Thermoregulatory thresholds 31. Kettner SC, Kozek SA, Groetzner JP, et al. Effects of
for vasoconstriction in pediatric patients anesthetized with hypothermia on thrombelastography in patients undergoing
halothane or halothane and caudal bupivacaine. Anesthesi- cardiopulmonary bypass. Br J Anaesth. 1998;80:313.
ology. 1992;76:387. 32. Schmied H, Kurz A, Sessler DI, et al. Mild hypothermia
12. Simbruner G, Weninger M, Popow C, et al. Regional increases blood loss and transfusion requirements during
heat loss in newborn infants. Part I. Heat loss in healthy total hip arthroplasty. Lancet. 1996;347(8997):289.
newborns at various environmental temperatures. S Afr 33. Schmied H, Schiferer A, Sessler DI, et al. The effects of
Med J. 1985;68:940. red-cell scavenging, hemodilution, and active warming on
13. Matsukawa T, Sessler DI, Christensen R, et al. Heat flow allogenic blood requirements in patients undergoing hip or
and distribution during epidural anesthesia. Anesthesiology. knee arthroplasty. Anesth Analg. 1998;86:387.
1995;83:961. 34. Johansson T, Lisander B, Ivarsson I. Mild hypothermia does
14. Frank SM, Nguyen JM, Garcia CM, et al. Temperature not increase blood loss during total hip arthroplasty. Acta
monitoring practices during regional anesthesia. Anesth Anaesthesiol Scand. 1999;43:1005.
Analg. 1999;88:373. 35. Winkler M, Akca O, Birkenberg B, et al. Aggressive warming
15. Sessler DI, Ponte J. Shivering during epidural anesthesia. reduces blood loss during hip arthroplasty. Anesth Analg.
Anesthesiology. 1990;72:816. 2000;91:978.
16. Joris J, Ozaki M, Sessler DI, et al. Epidural anesthesia impairs 36. Widman J, Hammarqvist F, Sellden E. Amino acid infusion
both central and peripheral thermoregulatory control during induces thermogenesis and reduces blood loss during hip
general anesthesia. Anesthesiology. 1994;80:268. arthroplasty under spinal anesthesia. Anesth Analg. 2002;95:
17. Sessler DI. Complications and treatment of mild hypother- 1757.
mia. Anesthesiology. 2001;95:531. 37. Kirkland KB, Briggs JP, Trivette SL, et al. The impact of
18. Doufas AG. Consequences of inadvertent perioperative surgical-site infections in the 1990s: Attributable mortality,
hypothermia. Best Pract Res Clin Anaesthesiol. 2003;17:535. excess length of hospitalization, and extra costs. Infect
19. Frank SM, Fleisher LA, Olson KF, et al. Multivariate Control Hosp Epidemiol. 1999;20:725.
determinants of early postoperative oxygen consumption 38. Ben-Eliyahu S, Shakhar G, Rosenne E, et al. Hypothermia
in elderly patients. Effects of shivering, body temperature, in barbiturate-anesthetized rats suppresses natural killer
and gender. Anesthesiology. 1995;83:241. cell activity and compromises resistance to tumor metas-
20. Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative tasis: A role for adrenergic mechanisms. Anesthesiology.
maintenance of normothermia reduces the incidence of mor- 1999;91:732.
bid cardiac events. A randomized clinical trial. JAMA. 1997; 39. Beilin B, Shavit Y, Razumovsky J, et al. Effects of mild
277:1127. perioperative hypothermia on cellular immune responses.
21. Frank SM, Higgins MS, Breslow MJ, et al. The cate- Anesthesiology. 1998;89:1133.
cholamine, cortisol, and hemodynamic responses to mild 40. Wenisch C, Narzt E, Sessler DI, et al. Mild intraoperative
perioperative hypothermia. A randomized clinical trial. Anes- hypothermia reduces production of reactive oxygen inter-
thesiology. 1995;82:83. mediates by polymorphonuclear leukocytes. Anesth Analg.
22. Frank SM, Satitpunwaycha P, Bruce SR, et al. Increased my- 1996;82:810.
ocardial perfusion and sympathoadrenal activation during 41. Borregaard N, Kragballe K. Role of oxygen in antibody-
mild core hypothermia in awake humans. Clin Sci (Lond). dependent cytotoxicity mediated by monocytes and neu-
2003;104:503. trophils. J Clin Invest. 1980;66:676.
23. Backlund M, Lepantalo M, Toivonen L, et al. Factors 42. Burke JF. The effective period of preventive antibiotic
associated with post-operative myocardial ischaemia in action in experimental incisions and dermal lesions. Surgery.
elderly patients undergoing major non-cardiac surgery. Eur 1961;50:161.
J Anaesthesiol. 1999;16:826. 43. Kurz A, Sessler DI, Lenhardt R. Perioperative normothermia
24. Frank SM, Cattaneo CG, Wieneke-Brady MB, et al. Threshold to reduce the incidence of surgical-wound infection and
for adrenomedullary activation and increased cardiac work shorten hospitalization. Study of Wound Infection and
during mild core hypothermia. Clin Sci (Lond). 2002;102:119. Temperature Group. N Engl J Med. 1996;334:1209.
25. Frank SM, el-Gamal N, Raja SN, et al. Alpha-adrenoceptor 44. Melling AC, Ali B, Scott EM, et al. Effects of preopera-
mechanisms of thermoregulation during cold challenge in tive warming on the incidence of wound infection after
humans. Clin Sci (Lond). 1996;91:627. clean surgery; A randomised controlled trial. Lancet. 2001;
26. Tanaka M, Nagasaki G, Nishikawa T. Moderate hypothermia 358(9285):876.
depresses arterial baroreflex control of heart rate during, 45. Carli F, Emery PW, Freemantle CA. Effect of peroperative
and delays its recovery after, general anesthesia in humans. normothermia on postoperative protein metabolism in
Anesthesiology. 2001;95:51. elderly patients undergoing hip arthroplasty. Br J Anaesth.
27. Kettner SC, Sitzwohl C, Zimpfer M, et al. The effect of graded 1989;63:276.
hypothermia (36 degrees C-32 degrees C) on hemostasis in 46. Deuel TF, Kawahara RS, Mustoe TA, et al. Growth factors
anesthetized patients without surgical trauma. Anesth Analg. and wound healing: Platelet-derived growth factor as a model
2003;96:1772. cytokine. Annu Rev Med. 1991;42:567.
646 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
47. Heier T, Caldwell JE. Impact of hypothermia on the 67. Kasai T, Nakajima Y, Matsukawa T, et al. Effect of
response to neuromuscular blocking drugs. Anesthesiology. preoperative amino acid infusion on thermoregulatory
2006;104:1070. response during spinal anaesthesia. Br J Anaesth. 2003;90:58.
48. Eger EI II, Johnson BH. MAC of I-653 in rats, including a test 68. Nakajima Y, Takamata A, Matsukawa T, et al. Effect of
of the effect of body temperature and anesthetic duration. amino acid infusion on central thermoregulatory control in
Anesth Analg. 1987;66:974. humans. Anesthesiology. 2004;100:634.
49. Liu M, Hu X, Liu J. The effect of hypothermia on isoflurane 69. Kurz A, Sessler DI, Narzt E, et al. Morphometric influences
MAC in children. Anesthesiology. 2001;94:429. on intraoperative core temperature changes. Anesth Analg.
50. Leslie K, Sessler DI, Bjorksten AR, et al. Mild hypothermia 1995;80:562.
alters propofol pharmacokinetics and increases the duration 70. Nakajima Y, Mizobe T, Takamata A, et al. Baroreflex mod-
of action of atracurium. Anesth Analg. 1995;80:1007. ulation of peripheral vasoconstriction during progressive
51. Leslie K, Bjorksten AR, Ugoni A, et al. Mild core hypothermia hypothermia in anesthetized humans. Am J Physiol Regul
and anesthetic requirement for loss of responsiveness Integr Comp Physiol. 2000;279:R1430.
during propofol anesthesia for craniotomy. Anesth Analg. 71. Sessler DI, Sessler AM. Experimental determination of heat
2002;94:1298. flow parameters during induction of general anesthesia.
52. Lenhardt R, Marker E, Goll V, et al. Mild intraoperative hy- Anesthesiology. 1998;89:657.
pothermia prolongs postanesthetic recovery. Anesthesiology. 72. Taguchi A, Ratnaraj J, Kabon B, et al. Effects of a
1997;87:1318. circulating-water garment and forced-air warming on body
53. Bissonnette B, Sessler DI. Mild hypothermia does not impair heat content and core temperature. Anesthesiology. 2004;100:
postanesthetic recovery in infants and children. Anesth Analg. 1058.
1993;76:168. 73. Hynson JM, Sessler DI. Intraoperative warming therapies:
54. Kurz A, Sessler DI, Narzt E, et al. Postoperative hemody- A comparison of three devices. J Clin Anesth. 1992;4:
namic and thermoregulatory consequences of intraoperative 194.
core hypothermia. J Clin Anesth. 1995;7:359. 74. Sessler DI, McGuire J, Sessler AM. Perioperative thermal
55. Cheng C, Matsukawa T, Sessler DI, et al. Increasing mean insulation. Anesthesiology. 1991;74:875.
skin temperature linearly reduces the core-temperature 75. Krenzischek DA, Frank SM, Kelly S. Forced-air warming ver-
thresholds for vasoconstriction and shivering in humans. sus routine thermal care and core temperature measurement
Anesthesiology. 1995;82:1160. sites. J Post Anesth Nurs. 1995;10:69.
56. Frank SM, Raja SN, Bulcao CF, et al. Relative contribution of 76. Sessler DI, Moayeri A. Skin-surface warming: Heat flux and
core and cutaneous temperatures to thermal comfort and au- central temperature. Anesthesiology. 1990;73:218.
tonomic responses in humans. J Appl Physiol. 1999;86:1588. 77. Negishi C, Hasegawa K, Mukai S, et al. Resistive-heating and
57. Alfonsi P, Nourredine KE, Adam F, et al. Effect of postoper- forced-air warming are comparably effective. Anesth Analg.
ative skin-surface warming on oxygen consumption and the 2003;96:1683.
shivering threshold. Anaesthesia. 2003;58:1228. 78. Gali B, Findlay JY, Plevak DJ. Skin injury with the use of a
58. Hynson JM, Sessler DI, Moayeri A, et al. The effects of water warming device. Anesthesiology. 2003;98:1509.
preinduction warming on temperature and blood pressure 79. Janicki PK, Stoica C, Chapman WC, et al. Water warming
during propofol/nitrous oxide anesthesia. Anesthesiology. garment versus forced air warming system in prevention
1993;79:219. of intraoperative hypothermia during liver transplantation:
59. Kudoh A, Takase H, Takazawa T. Chronic treatment with A randomized controlled trial [ISRCTN32154832]. BMC
antidepressants decreases intraoperative core hypothermia. Anesthesiol. 2002;2:7.
Anesth Analg. 2003;97:275. 80. Grahn D, Brock-Utne JG, Watenpaugh DE, et al. Recovery
60. Vassilieff N, Rosencher N, Sessler DI, et al. Nifedipine from mild hypothermia can be accelerated by mechanically
and intraoperative core body temperature in humans. distending blood vessels in the hand. J Appl Physiol. 1998;
Anesthesiology. 1994;80:123. 85:1643.
61. Toyota K, Sakura S, Saito Y, et al. The effect of pre-operative 81. Smith CE, Parand A, Pinchak AC, et al. The failure of negative
administration of midazolam on the development of intra- pressure rewarming (Thermostat) to accelerate recovery
operative hypothermia. Anaesthesia. 2004;59:116. from mild hypothermia in postoperative surgical patients.
62. Ezri T, Szmuk P, Weisenberg M, et al. The effects of Anesth Analg. 1999;89:1541.
hydration on core temperature in pediatric surgical patients. 82. Taguchi A, Arkilic CF, Ahluwalia A, et al. Negative pressure
Anesthesiology. 2003;98:838. rewarming vs. forced air warming in hypothermic postanes-
63. Sellden E, Lindahl SG. Amino acid-induced thermogenesis thetic volunteers. Anesth Analg. 2001;92:261.
reduces hypothermia during anesthesia and shortens hospi- 83. Doufas AG, Akca O, Barry A, et al. Initial experience with
tal stay. Anesth Analg. 1999;89:1551. a novel heat-exchanging catheter in neurosurgical patients.
64. Saitoh Y, Kaneda K, Tokunaga Y, et al. Infusion of amino Anesth Analg. 2002;95:1752.
acid enriched solution hastens recovery from neuromuscular 84. Xiong J, Kurz A, Sessler DI, et al. Isoflurane produces
block caused by vecuronium. Br J Anaesth. 2001;86:814. marked and nonlinear decreases in the vasoconstriction and
65. Sellden E, Brundin T, Wahren J. Augmented thermic effect of shivering thresholds. Anesthesiology. 1996;85:240.
amino acids under general anaesthesia: A mechanism useful 85. Sessler DI, Rubinstein EH, Moayeri A. Physiologic re-
for prevention of anaesthesia-induced hypothermia. Clin Sci sponses to mild perianesthetic hypothermia in humans.
(Lond). 1994;86:611. Anesthesiology. 1991;75:594.
66. Sellden E, Branstrom R, Brundin T. Preoperative infusion 86. Szmuk P, Ezri T, Sessler DI, et al. Spinal anesthe-
of amino acids prevents postoperative hypothermia. Br J sia speeds active postoperative rewarming. Anesthesiology.
Anaesth. 1996;76:227. 1997;87:1050.
CHAPTER HYPERTHERMIA
46 Barbara W. Brandom
CASE SUMMARIES
What Is the Relationship
CASE #1: A child presented for computerized
tomographic-guided biopsy of nodules on the diaphragm, between Hyperthermia and
placement of a Broviac catheter, and lumbar puncture Thermoregulation?
with general anesthesia. In the last 2 weeks, abdominal
girth had been increasing. Ascites and pleural effusions
Hyperthermia is an elevation of core body temperature
were noted. Oral temperature was 39.5 C. Is this hyper- above 38 C for any reason. Normally, core temperature
thermia? How should it be treated? is maintained in a narrow range, 0.2 C, by thermo-
CASE #2: A healthy adolescent underwent elective regulatory reflexes,1 although an interperson variability of
orthopedic surgery during inhalation anesthesia. A lar- 2 C in normal core temperature2 and diurnal variability
yngeal mask airway was used. Axillary temperature was of approximately 1 C can occur.3,4
<36 C, so the lower body was warmed with 43 C forced
air. When laryngospasm was perceived, an endotracheal
tube and an esophageal temperature probe were placed. HEAT PRODUCTION
Esophageal temperature was >41 C. Is this hyperthermia?
How should it be treated? Heat production is divided into two processes: Obligatory
and facultative.5 Obligatory thermogenesis is a result of
CASE #3: A middle-aged woman with a body mass index the metabolic processes necessary to sustain life. Facul-
>40 kg per m2 underwent laparoscopic Roux-en-Y gastric tative thermogenesis is a rapidly inducible process driven
bypass. Preoperative laboratory evaluation was normal. by the hypothalamus and sympathetic nervous system
Her course in the operating room and postanesthesia in response to cold or excess food intake.6 Facultative
care unit (PACU) was unremarkable. Seven hours after thermogenesis can increase the thermoregulatory set-
the operation, the patients heart rate was 120 bpm, and point by 0.3 C.7
her axillary temperature was 38.2 C. Blood pressure was
140/60 mm Hg. A Foley catheter relieved urinary retention.
She stated that patient-controlled analgesia was providing
good pain relief. Hematologic laboratory tests and serum
HEAT LOSS
electrolytes were normal. Tachycardia and fever persisted Evaporation
overnight despite the administration of 2 L of isotonic
crystalloid intravenously. The maintenance of thermal steady state requires the dissi-
The anesthesiologist may encounter a patient pation of heat produced by the metabolism to the relatively
with elevated core temperature preoperatively, intra- cool environment. Almost all of this heat moves through
operatively, or after surgery and anesthesia have been the skin, the rest being lost through the respiratory tract.
completed. This chapter presents the basic elements of When core temperature rises above a reproducible thre-
thermoregulation that are relevant to maintaining body shold value, sweating and vasodilation occur. Skin blood
temperature greater than normal. Several situations that flow can increase by 8 L per minute.8 In a dry environment
produce more heat than expected or prevent dissipation of with moving air, evaporation of sweat can dissipate heat
heat will be presented. The life-threatening consequences at 10 times the basal metabolic rate.9 Significant amounts
of critical temperature will be reviewed, and interventions of salt and up to 2 L per hour of water can be lost in
that can be used to decrease body temperature will be sweat.10,11 At rest, without sweating, evaporative heat loss
described. is only approximately 5% of the basal metabolic rate.
647
648 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Conduction 300
Sweat (g/m2/h)
as a linear function of the difference between skin and 200
ambient temperatures. Conduction is the direct transfer
0% 0.8% 1.2%
of heat from one surface to an adjacent surface. Insulation
between two surfaces will reduce conduction. 100
Convection
0
Convection is loss of heat to moving air, in proportion 36 37 38 39 40
to the square root of the velocity of the air. With a Esophageal temperature (C)
15 to 20 mph wind velocity, heat loss by convection is
at a maximum. FIGURE 46.1 Increasing isoflurane exposure results in a greater
increase in esophageal temperature to elicit the same rate of
Radiation sweat production. (From Washington DE, Sessler DI, Moayeri A,
et al. Thermoregulatory responses to hyperthermia during
Radiation is the transfer of heat from one body to isoflurane anesthesia in humans. J Appl Physiol. 1993;74:82.)
another through photons. Therefore, radiation does not
depend on the temperature of the surrounding air.12 When
environmental temperature is less than core temperature, TEMPERATURE GRADIENTS
vasodilation allows core temperature to decrease, because
heat is lost to the environment through conduction and Under normal physiologic conditions in the absence of
convection. When the patient is not insulated from the anesthesia, a gradient in body temperature of 2 C to 4 C
cooler environment by materials such as heavy clothes, is present from the body core to the extremities. Skin
heat loss by conduction, convection, evaporation, and temperature is 5 C to 9 C lower than core body tem-
radiation is greater. perature measured at the forehead,15 pulmonary artery,
esophagus, rectum, or bladder. This heat gradient nor-
mally is maintained by vasoconstriction that is controlled
by sympathetic reflexes. Heat will be constrained to the
Why Does Hyperthermia core compartment with increased vasoconstriction. This
situation may result from low blood volume, low cardiac
Occur? output,16 or some other source of elevated plasma cate-
cholamines. A 10% decrease in plasma volume is sufficient
Hyperthermia can be passive or active. It is the result to compromise convective thermoregulation.17 Retention
of one or more of the following factors: Decreased heat of heat by the core compartment contributes to the ele-
loss to the environment, constraint of heat to the core vated temperature that is one of the signs of untreated
thermal compartment, increased metabolic production pheochromocytoma and to fever after cardiopulmonary
of heat,13 or excessive delivery of exogenous heat. With bypass in the presence of low cardiac output. Mechanical
current forced-air warming devices, it is possible to de- obstruction to regional blood flow that prevents heat dis-
liver excessive amounts of heat to a patient during the sipation also is seen. Leg tourniquets have increased core
administration of anesthesia. Decreased environmental heat sufficient to raise temperature more than 1 C in an
heat loss occurs when the patients insulation is effective anesthetized child18 (see Fig. 46.2).
and prevents heat loss by conduction, convection, radia-
tion, or evaporation. Evaporative environmental heat loss
is reduced when humidity is high.
During anesthesia, the threshold temperature at which How Is Fever Produced?
sweating begins is increased. For example, exposure to
1.2% isoflurane increases the sweating threshold from an Metabolism is usually reduced approximately 15% by
average of 36.6 C to 38.1 C in men and from 37.1 C to general anesthesia. However, many factors can cause
38.3 C in women14 (see Fig. 46.1). However, the maxi- increased metabolism and increased heat production. The
mum sweating intensity and the gain of this response (the most frequent of these is fever, and hyperthermia is most
increase in sweat production for each unit increase in tem- often caused by fever. However, it is important to differ-
perature) are not altered by an increased anesthetic dose. If entiate between fever and hyperthermia. Fever is defined
the patient cannot produce sweat, heat loss by evaporation by a core temperature over 38.5 C because of pyrogens.
depends on the application of exogenous liquid (water or
alcohol) to the skin. Heat loss to the environment will not
occur by conduction or convection when environmental THE ROLE OF PYROGENS
temperatures are greater than body temperature. There-
fore, in a hot and humid environment, otherwise normal Many tissue injuries produce proteins and protein frag-
anesthetized patients can easily become hyperthermic. ments. These and bacterial lipopolysaccharide toxins may
CHAPTER 46/HYPERTHERMIA 649
POSTOPERATIVE FEVER
be called exogenous pyrogens.19 These exogenous pyrogens
stimulate the release of endogenous pyrogens, including Even when no infection is present, postoperative fever is
interleukin 1 (IL-1), IL-6, and tumor necrosis factor alpha very common. After major noncardiac surgery in adults,
(TNF-) from macrophages and monocytes. Endogenous the median temperature was 38 C, and 25% of patients
pyrogens, in turn, promote the release of prostaglandin had temperature >38.5 C 11 hours after surgery.24 This
(PG) E2 in the preoptic area of the hypothalamus where temperature elevation was associated with increased IL-6.
temperature is regulated.20 Following cardiopulmonary bypass, IL-6 concentration
Normal thermoregulatory responses and an elevated increases, and fever of 38 C to 39 C is common.25 Major
set-point are seen with fever. For example, the normal intracranial neurosurgical procedures may be followed
range of core temperaturein which there is no auto- by fever over 38.5 C in more than 80% of children,
nomic response that could produce gain or loss of heatis although meningitis is present in only approximately
0.2 C to 0.3 C. If core temperature is below the lower 10%.26 During the rehabilitation phase following brain
limit of this interthreshold range, vasoconstriction, non- injury, fever is much less common. In the absence of
shivering thermogenesis, and shivering occur. When the infection, temperature was less than, 38.2 C, and fever
interthreshold range is exceeded, sweating and then va- occurred only in those patients with traumatic brain injury
sodilation follow. Endogenous pyrogens such as ILs 1, 2, or aneurysmal subarachnoid hemorrhage.27
and 6, TNF, and interferon- increase the core tempera- Fever also is common after routine tonsillectomy,28
ture set-point in the middle of the interthreshold range. orthopedic surgery,29 and gynecologic surgery.30 Non-
As a result, body temperature is maintained above 37.2 C infectious processes are associated with 50% to more
and even above 38 C. Such temperature elevation may than 90% of fevers after gynecologic surgery.31 Years ago,
stimulate immune function, thereby aiding recovery from it was recognized that a temperature elevation >38 C
infection.21 The common finding of preoperative fever in occurs in 25% to 30% of children during the first 3
an infected patient does not mean that normothermia postoperative days. Clinical evaluation was more useful
should be the goal of an anesthetic administered to fa- than laboratory tests in determining the cause. Fever was
cilitate surgical treatment. Shivering in a patient with associated with surgery longer than 2-hours duration,
elevated core temperature is a sign that pyrogens have intraoperative blood transfusion, preexisting infection,
raised the temperature set-point, and normal thermo- and the administration of preoperative antibiotics. In <2%
regulation is acting to maintain the elevated temperature. of these children, sepsis was the reason for fever.32 These
Fever may be decreased following the administration of fevers are not complications of anesthesia as such; they are
drugs that inhibit PGs. Fever will not respond to external the result of pyrogens released by the surgical procedure.
cooling measures without the inhibition of PGs, and they Rarely, postoperative fever is caused by an en-
rarely produce temperatures above 41.1 C. docrine abnormality, such as hyperthyroidism,33 which
650 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
was not active preoperatively. Case reports of malignant and impedes heat loss, but also stimulates both 1 - and
hyperthermia (MH) susceptibility, with or without 3 -adrenoreceptors. The activation of 3 -adrenoreceptors
rhabdomyolysis, or Duchenne muscular dystrophy with stimulates cyclic adenosine monophosphate (cAMP)-
exacerbated rhabdomyolysis, presenting with postopera- dependent liberation of intracellular free fatty acids. There
tive fever have been published. However, a series of cases are proteins, known as uncoupling proteins, which func-
selected for evaluation of postoperative fever rarely finds tion as catalysts for the movement of anionic portions
a muscular cause for the temperature elevation. of fatty acids across mitochondrial membranes. In the
Paradoxical hyperthermia on exposure to cold has presence of uncoupling proteins, increased intracellular
been observed experimentally in cold-adapted rats34 ; these free fatty acid concentrations result in the decreased gen-
animals have an increased metabolic rate. A central eration of adenosine triphosphate (ATP) and increased
injection of PGE1 also increased core temperature by thermogenesis.45 Therefore, uncoupling proteins regulate
1.9 C in these animals in contrast to a 0.9 C increase in the balance between ATP production and thermogenesis
animals made hypermetabolic to a similar degree with in muscle and brown adipose tissue46,47 (see Fig. 46.3).
thyroxine. Whether similar mechanisms are active in Activation of 1 -adrenoreceptors strongly potentiates the
humans is unknown. thermogenic effect of 3 -adrenoreceptor activation.48
Perhaps MDMA will very rarely be found in patients
presenting for anesthesia, but other phenethylamine sym-
DRUGS THAT PRODUCE pathomimetics may well be seen. Amphetamine, metham-
HYPERTHERMIA phetamine, and structurally different stimulants such as
cocaine can have similar uncoupling of thermogenesis and
Fevers caused by adverse reactions to drugs that are ATP production in muscle and other cells, as can disease
administered during anesthesia are anesthetic compli- states in which endogenous catecholamines are elevated.
cations. Antibiotics, antihistamines, and barbiturates are Recognition of the pharmacologic mechanisms that
among those that have been associated with such drug- can produce hyperthermia is important. Epinephrine and
induced fever. In some cases, these are allergic reactions. dobutamine do not activate the 3 - adrenoreceptor and do
Diphenhydramine, a commonly administered antihis- not have thermogenic effects. Conversely, thyroid function
tamine, has significant anticholinergic effects which con- has a significant effect on sympathomimetic-mediated
tribute to hyperthermia primarily by decreasing sweat thermogenesis. Expression of uncoupling proteins in
formation.35 Fever can be present in cases of salicylate different tissues is regulated through transcription by
toxicity.36 Postoperative hyperthermia may follow the ad- thyroid hormone and dietary fatty acids.49
ministration of ketamine.37
A different form of drug toxicity can produce Serotonin and Neuroleptic Malignant
hyperthermia. In some situations, the factors that produce Syndromes
elevation of the thermoregulatory set-point are complex
and not initiated by pyrogens. For example, diatrizoate, an Other hyperthermic syndromes in which nonadrenergic
ionic contrast dye used for myelography, affects dopamin- stimulation of the central nervous system may play a part
ergic neurotransmission and produces increased motor include the serotonin syndrome and the neuroleptic malig-
activity and thermoregulatory set-point elevation.38 An el- nant syndrome (NMS). The serotonin syndrome includes
evated temperature after the spread of this dye into the cognitive, autonomic, and neuromuscular dysfunction, as
cerebrospinal fluid is a result of increased metabolic heat well as hyperthermia, clonus, and muscle rigidity in a
from increased muscle activity and set-point elevation patient taking one or more of the serotoninergic drugs.50
due to dopaminergic stimulation. However, dopaminer- Monoamine oxidase inhibitors, tricyclic antidepressants,
gic drugs can also produce hypothermia.39 The ascending serotonin reuptake inhibitors, and lithium have been
tonic-clonic syndrome includes hyperthermia and has associated with the serotonin syndrome, as have the anal-
been noted after cerebrospinal fluid introduction of di- gesics meperidine, dextromethorphan, and tramadol.51
atrizoate and two other similar compounds, metrizamide Recently, a few cases of serotonin syndrome have been
and metrizoate. It is frequently fatal.40 reported after the addition of oxycodone to the treat-
Similarly, 3, 4-methylenedioxymethamphetamine ment of patients taking other drugs that can alter the
(MDMA, aka ecstasy) alters serotonin,41 dopamine,42 serotoninergic and dopaminergic systems.52 Perhaps in-
and norepinephrine43 in the hypothalamus, all of which creased dopamine and glutamate concentrations in the
have an effect on thermoregulation. An early effect hypothalamus contribute to the pathophysiology of the
of MDMA is to release neuronal serotonin which up- serotonin syndrome.53
regulates dopamine biosynthesis and release by activation Encephalopathy, skeletal muscle rigidity, autonomic
of serotonin 2A postsynaptic receptors. The subsequent dysfunction, and hyperthermia are signs of the NMS
activation of D1 receptors plays an essential role in which occur in patients taking neuroleptic, antipsychotic
the hyperthermic response to MDMA.44 MDMA can re- drugs. NMS occur in apparently normal people receiving
set thermoregulation in the hypothalamus, as well as these drugs as antiemetics or sedatives and in those
produce effects in the rest of the body that increase being treated for a range of neuropsychiatric disorders.54
production and reduce elimination of heat. The adminis- Haloperidol is most frequently associated with NMS, but
tration of MDMA is followed by a very marked elevation of all classes of drugs that block D2 dopamine receptors
plasma norepinephrine, which produces vasoconstriction have been implicated.54 Cases of NMS and deaths after
CHAPTER 46/HYPERTHERMIA 651
NAD+ ADP
H+
Mitochondria
NADH ATP
Cell
FIGURE 46.3 Links are proposed between the redox state of coenzymes in the mitochondria and
adenosine triphosphate (ATP) production. The electrochemical gradient is modified by exchange of
anions through A, anions and protons through S, and protons through uncoupling proteins (UCP).
RC is the respiratory chain. F0 F1 ATPases are regenerating oxidized coenzymes. One of the
mechanisms that increases the flexibility of mitochondrial production of NADH and ATP is the
dissipation of the proton gradient across the inner mitochondrial membrane through UCP. When
respiration is altered in this manner, increased oxidation generates heat as a byproduct. NAD,
nicotinamide adenine dinucleotide; ADP, adenosine diphosphate; NADH, reduced nicotinamide
adenine dinucleotide. (From Ricquier D, Bouillard, F. The uncoupling protein homologues: UCP1,
UCP2, UCP3, StUCP and AtUCP. Biochemistry. 2000;J354:161.)
the administration of prochlorperazine, metoclopramide, syndrome can be more difficult to treat, especially with
droperidol, and promethazine have been reported. For oral medications.59
these and other reasons, the hypoactivity of central
nervous system dopamine is thought to be the principle
cause of NMS. Both plasma and urinary catecholamine
concentrations may be elevated in patients with NMS, as What Patient Factors
they are in the presence of pheochromocytoma. Halting
the administration of neuroleptic drugs usually is followed
Predispose to Hyperthermia?
by resolution of encephalopathy and hyperthermia.
Increased metabolic rate, with or without elevation of
Drug Withdrawal the thermoregulatory set-point, can contribute to hyper-
thermia. Such changes occur in several diseases including
Withdrawal from several drugs, including alcohol, hyperthyroidism, MH, and mitochondrial diseases. Fever
sedative-hypnotics and opiates, can produce elevated tem- and tachycardia are two of the constant features of hy-
perature. Other less commonly administered drugs also perthyroidism.60 Because human chorionic gonadotropin
are potential causes for concern when they are withdrawn (hCG) is a weak thyrotropin agonist, symptoms of hy-
in the perioperative period. For example, disruption of perthyroidism, including hyperthermia, may occur in
intrathecal baclofen administration, used to treat severe obstetric conditions with very high hCG.61
spasticity, produces a withdrawal syndrome consisting
of itching, nausea, headache, malaise, increased skeletal
muscle reflexes, increased heart rate, hypertension, in- DISEASE STATES
creased temperature, and sometimes delirium55 and hal- AND SYNDROMES
lucinations.56 Intrathecal baclofen withdrawal syndrome
may be a form of serotonergic syndrome resulting from Malignant Hyperthermia
the loss of -aminobutyric acid B, receptor-mediated,
presynaptic inhibition of serotonin.57 Persistent fever Abnormal function of the ryanodine receptor due to
with autonomic instability also has been observed after mutations is the most common known cause of MH. When
withdrawal from oral baclofen.58 These symptoms may exposed to inhalation anesthetics, the abnormal ryanodine
resolve within 24 hours of administration of the usual receptor is more sensitive to agonists, more calcium moves
dose of oral baclofen. However, when intrathecal baclofen through the ryanodine receptor from the sarcoplasmic
administration is stopped suddenly, as may occur if the reticulum into the muscle cell, and muscle metabolism
catheter is displaced during spinal surgery, the withdrawal increases as a result.62 Metabolic rate can rapidly increase
652 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
10-fold in MH and, therefore, increased carbon dioxide maintain a normal temperature. When such patients
production and tachycardia are the earliest signs of are exposed to pyrogens, they may experience extreme
MH. Temperature elevation may be a late sign in the temperature elevation. In other neurologic diseases, Hunt-
MH syndrome. Other metabolic diseases,63,64 including ington disease in particular, patients may also be prone
congenital abnormalities of mitochondrial function,65 to the development of NMS or serotonin syndrome.75
may produce similar symptoms without exposure to In Angelman syndrome, extreme temperature elevation
anesthetic drugs. Elevation of temperature to a critical has been associated with severe generalized myoclonus.76
level can occur in any of these diseases. High temperature has also been observed in myotonic
syndromes such as the Schwartz-Jampel syndrome or the
Pheochromocytoma and Other Adrenal Stuve-Wiedemann syndrome.
Tumors Tumors
In some diseases, such as pheochromocytoma, at least two
Hyperthermia may complicate the operative management
mechanisms may contribute to hyperthermia, a frequent
of oncology patients. For example, a toddler with a
finding in this disease.66 Indeed, elevated temperature
3-month history of increasing irritability, tachycardia,
is observed in up to 10% of patients with adrenal
tachypnea, and increasing sweating was anesthetized
tumors, without specifying histologic type.67 The elevated
for resection of a suprarenal mass with intrathoracic
endogenous levels of norepinephrine in patients with
extension. Preoperatively, the 12-month-old, emaciated
pheochromocytomas may induce uncoupling proteins
child had a heart rate of 144 bpm, blood pressure of
which increase heat production. Furthermore, some
110/70 mm Hg, and axillary temperature of 35.8 C.77 After
pheochromocytomas are associated with increased IL-6
induction of anesthesia with thiopental and neuromuscu-
levels. Increased IL-6 will increase the set-point of
lar block by pancuronium, the esophageal temperature
temperature regulation. Pharmacologic treatment with
was 38 C. Isoflurane was added to maintain anesthesia.
- and -blockers or by surgical removal of the tumor
The esophageal temperature rose to 40 C over the next
can be followed by reduction of temperature elevation
40 minutes, and heart rate increased from 120 to 144 bpm.
and decreased concentrations of serum IL-6.68 Of note,
Arterial blood gas analysis showed a PaCo2 of 38 mm Hg,
IL-6 can induce the production of nitric oxide. The
HCO3 of 21 mmol per L, and pH 7.35. The temperature
resulting vasodilation may counter the hypertensive effect
decreased to 38 C after 90 minutes of surgery. Rectal tem-
of excessive norepinephrine, which usually is observed in perature on admission to the PACU was 40 C. After the
patients presenting with this type of tumor. administration of acetaminophen suppositories and ex-
posure to cool mist and an iced damp cloth, temperature
Osteogenesis Imperfecta dropped to 38 C. Follow-up cultures of blood, urine, spu-
tum, and cerebrospinal fluid were negative. Biopsy results
Patients with osteogenesis imperfecta (OI) frequently
confirmed the diagnosis of neuroblastoma. The tempera-
develop fever during or after surgery.69 OI is due to one
ture elevation in this patient could have been due to vol-
of several mutations in the pro-1 or the pro-2 chains
ume contraction, IL-6, and the effects of norepinephrine.
of the type I collagen heterotrimer.70 Type I collagen
Because acetaminophen was effective, IL-6 or other py-
also is found in teeth, ligaments, skin, and sclera, all
rogens must have been present. Significant perioperative
of which are abnormal in OI. A bleeding disorder in
hyperthermia also has been reported in very ill patients
patients with OI has been attributed to abnormal platelet
with lymphoma78,79 and myelogenous leukemia.80
function. Temperature may reach 40 C in OI patients with
no evidence of a muscle disorder, specifically MH, which Miscellaneous Disease States
would be expected to increase metabolism.71 In contrast
to patients with MH susceptibility, patients with OI have Acute intermittent porphyria can present with fever, acute
increased basal body temperatures. Increased thyroxine is muscle pain, and weakness.81 In the presence of varied
present in OI patients with increased oxygen consumption genetic conditions, fever can be severe with or with-
and decreased body weight.72 In general, patients with OI out exposure to anesthesia. This has been observed in
can receive inhalation anesthetics without producing a the Prader-Willi syndrome,82 glycogen storage disease,83
critical temperature elevation when their intravascular hypohidrotic ectodermal dysplasia, and autoinflamma-
volume status is adequate and they are allowed to tory diseases.84 In Prader-Willi Syndrome, life-threatening
loose heat by conduction and convection. An increased fever may be a common element of illness in children
temperature postoperatively is to be expected in patients younger than 2 years.
with OI in the absence of infection, especially when
surgery lasts longer than 1 hour.73
more heat is produced or absorbed by the body than TABLE 46.1 Symptomatic Treatment of Noncritical
can be dissipated, and therefore, core temperature will Hyperthermia
rise. Cytokines, inflammatory and anti-inflammatory, are
produced in response to endogenous or exogenous heat. Remove clothing and blankets
Reduction of body temperature to normal does not result Monitor breathing and heart rate
in the suppression of these cytokines.85,86 Give 10 to 20 mL/kg of isonatremic IV fluid
Apply cold compresses to neck and groin
Bathe in tepid water
ACUTE RESPONSE Increase air movement
The acute phase response to heat stress includes Do not give anything by mouth
fever, leucocytosis, muscle catabolism, stimulation of
IV, intravenous.
the hypothalamic-pituitary-adrenal system, and increased
synthesis of acute phase proteins.87 This sequence of
events is similar to that produced by sepsis.88 Increased and removing the source of pyrogens93 or other causes of
core temperature can increase free radical production, hyperthermia and supporting the processes that remove
in particular F2 isoprostanes, and deplete cellular glu- heat during normal thermoregulation. If hyperthermia
tathione.89 IL-6, produced during heat stress, stimulates has been induced by a drug, such as a sympathetic
the hepatic production of anti-inflammatory, acute phase nervous system stimulant94 or antipsychotic medication,
proteins, which inhibit the production of reactive oxy- or the patient has underlying endocrine or muscle
gen. Core temperatures over 40.5 C are life-threatening. pathology, specific pharmacologic treatment95,96 may be
At 41.5 C, brain injury can occur. Increased IL-1 and necessary. Antipyretics, which can block IL effects on
TNF- induced by heat are associated with elevated intra- thermoregulatory reflexes and reset core temperature to a
cranial pressure, decreased cerebral blood flow, and severe lower level, should be given. Any deficits in plasma volume
neuronal injury. should be repaired generously as tolerated. Cardiac output
should be increased with vasodilators if possible. The
Multiorgan System Failure patient can remain uncovered, but it is not always
advisable to institute physical measures, such as washing
The persistence of core temperature over 41.5 C for more with cold water, in an attempt to increase convective or
than 45 minutes can produce multiorgan system failure.87 conductive heat loss93 (see Figs. 46.4A and B).
Injury to the liver, heart, and skeletal muscle occurs,
and endotoxin escapes from the gut. Cardiovascular dys-
function, coagulopathy, and renal failure are common. IMMERSION IN COLD WATER
Encephalopathy, acute respiratory disease, intestinal is-
chemia, pancreatic injury, and disseminated intravascular Because the conductivity of water is approximately
coagulation with thrombocytopenia may also result from 25 times that of air, heat can be lost most efficiently
heat injury. Reduction of core temperature to normal in- during immersion of the trunk in cold water. This
hibits fibrinolysis, but not activation of coagulation.90 At method has been tested in runners and in anesthetized
core temperature of 45 C, brain death is nearly certain. experimental subjects. Water at 14 C was as effective as
In response to severe heat stress, cardiac output ice water at 5 C in decreasing core temperature after
increases by up to 20 L per minute.8 A patient whose 12 minutes of treatment.97 Indeed, cooling by immersion
cardiovascular function is limited by intravascular volume may produce temperature several degrees cooler than
depletion, intrinsic cardiac disease, or medications that desired. If immersion is impractical, placing bags of ice
decrease cardiac output will be more susceptible to heat on the groin and in the axillae and around the neck where
injury. Proteins produced by many cells in response to large blood vessels are relatively close to the skin can help
heat (heat-shock proteins) induce a transient tolerance to to decrease core temperature. When the trachea has been
a second, potentially lethal insult. A low level of expression
of heat-shock protein, as in the elderly when there has
TABLE 46.2 Treatment of Critical Hyperthermia Greater
been no acclimatization to heat and in some genetic
than 41.5 C
polymorphisms, makes heat injury more likely.91,92
Initiate symptomatic treatment of hyperthermia
Immerse in cold water
Do not give anything by mouth
How Should Hyperthermia Aggressively pursue potential root causes
Be Managed? Consider drug treatment
Dantrolene
Carvedilol
GENERAL MEASURES Antibiotics
Diazepam
Therapy is guided by the severity of the fever (see Bromocriptine
Tables 46.1 and 46.2). Treatment consists of identifying
654 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
1.8
Heat production
(kcal/kg/h)
1.4
Core temperature (C)
39
38 1.0
37
3
Skin temperature
35
1
Control
(C)
31
0
27
Cooling 2 Self-adjust
Systemic heat balance
Control
70 0
(kcal/kg)
VAS (mm)
50
2 Cooling
30
10
4
0 1 2 3 4 5 6 7 8 3 4 5 6 7 8
A Elapsed time (h) B Elapsed time (h)
FIGURE 46.4 A: Core temperature increases after administration of interleukin 2. Cooling the skin
does not reduce core temperature significantly. Cooling the skin is uncomfortable for the patient.
VAS, visual analog scale. B: Cooling the skin of a febrile patient may result in a negative systemic
heat balance, but cooling the skin also produces the greatest increase in heat production. (From
Lenhardt R, Negishi C, Sessler DI, et al. The effects of physical treatment on induced fever in
humans. Am J Med. 1999;106:550.)
intubated, iced saline can be used to lavage the stomach. instituted. Although controlled studies of efficacy are lack-
However, gastric lavage can be complicated by abdominal ing, physical methods have been applied in attempts to
cramping and diarrhea.98 Bladder lavage and lavage of any diminish cutaneous vasoconstriction and shivering in re-
open body cavities with cool saline can also decrease core sponse to the application of cold to the skin. The patient
temperature, but forced-air or circulating-water cooling may be vigorously massaged, sprayed with water at 40 C,
methods are at least twice as effective as bladder lavage. or exposed to hot moving air at 45 C, either at the same
Such active cooling measures should be reserved time that cooling methods are applied or in alternating
for situations in which elevated temperature is life- manner.99,101
In heat stroke, decreased consciousness necessitates
threatening. When fever <40.5 C was induced with IL-2,
tracheal intubation and controlled mechanical ventila-
active cooling did not significantly reduce peak or inte-
tion. Neuromuscular blockade with vecuronium in the
grated over time core temperatures.99 Furthermore, active
presence of low dose isoflurane anesthesia, 0.6 minimum
cooling with forced air at 15 C increased oxygen con-
alveolar concentration, in subjects who had received IL-2
sumption by 35% to 40% in subjects with a fever of decreased peripheral tissue heat content rather than al-
38.5 C. Active cooling increased arterial blood pressure lowing it to increase. Less increase in core heat content
and plasma norepinephrine and epinephrine concentra- was seen in those who had received vecuronium with
tions. Subjects exposed to cool air shivered and reported isoflurane compared to isoflurane alone.102 Nondepolariz-
that they were uncomfortable. When shivering begins, ing neuromuscular blockers block shivering and therefore
there is a sustained increase in oxygen consumption.100 decrease the heat generated in response to pyrogens. How-
Pharmacologic intervention to prevent these side effects ever, there was no evidence that neuromuscular blockers
should be considered when active cooling measures are could alter the set-point for thermoregulation. Similarly,
CHAPTER 46/HYPERTHERMIA 655
in healthy volunteers who had not received pyrogens or No complications were noted during intubation,
anesthetics, dantrolene decreased the shivering threshold other than a transient decrease in oxygen saturation that
temperature from an average of 36.7 C to 36.3 C. More was relieved with positive-pressure ventilation following
importantly, dantrolene greatly reduced the rate of in- endotracheal tube placement. Tumor lysis syndrome
crease in oxygen consumption for each degree decrease in occurred in the following 24 hours, but the child recovered
core temperature.100 without dialysis. No other complications occurred the
If core temperature has been >40 C, and sympa- following week. Future anesthetic choices need not be
thetic nervous system stimulants produced hyperther- restricted for this patient.
mia, the administration of 3 -adrenoreceptor antagonists,
such as carvedilol, may block facultative thermogene- Case 2
sis. If NMS is the cause, bromocriptine may ameliorate
thermogenesis. If hyperthyroidism is the cause of hy- In the second case, the critical fever presumably was
perthermia, -blockers may decrease metabolism until produced by abnormally increased muscle metabolism
antithyroid medications become effective. Similarly, if following exposure to inhalation anesthetics. This sce-
pheochromocytoma or another catecholamine-secreting nario represented acute MH. An arterial bloodgas anal-
tumor contributed to the problem, - and -blockade may ysis in the hour before tracheal intubation showed a
be necessary to blunt thermogenesis, while at the same mild mixed acidosis. A second arterial bloodgas analysis
time, restoration of adequate plasma volume is carried after tracheal intubation demonstrated increasing acido-
out. Elevated metabolism due to increased intracellular sis. Treatment for MH was initiated with dantrolene,
muscle calcium resulting from abnormal function of the ice packs, and cold intravenous fluids. The anesthetic
ryanodine receptor necessitates the rapid administration was terminated. More than 8 mg per kg of dantrolene
of dantrolene. Anesthesia providers must not focus only were given in the first hour of treatment. Excessive
on physical abnormalities that increase heat production. bleeding was noted when additional vascular catheters
Increased production without concurrent heat dissipation were placed. Although acidbase status was corrected,
can lead to life-threatening hyperthermia. and the core temperature was reduced to 35 C, rhab-
domyolysis was extreme. The patient died more than
24 hours later with disseminated intravascular coagula-
CASE DISCUSSIONS tion and hyperkalemia, despite continued treatment with
dantrolene.
Hyperthermia was present in all three cases that opened No evidence of a myopathy due to a defect in the
this chapter, because core temperature was >38 C. Initial structure of muscle proteins was found at autopsy, and no
treatment was symptomatic, but definitive treatment ryanodine receptor mutation was identified in the three
depends on the etiology of the temperature elevation. hot spots of this gene in blood from this patient. The
ryanodine receptor gene (RYR1) is the primary genetic
Case 1 locus associated with MH. There are 106 exons, sections
of DNA that are transcribed into messenger RNA, that
Preoperative fever in the first instance most likely was contribute to the ryanodine receptor protein. Most of the
due to cytokines produced by the tumor. Temperature el- mutations in RYR1 that have been associated with MH
evation may have been exacerbated by volume depletion, are in one of three areas: The N-terminus, central, or
because oral intake had been minimal for days before ad- C-terminus hot spots. This patients course is consistent
mission to the hospital the evening before the anesthetic with classic MH. However, a significant percentage of
was administered. The tentative diagnosis was aggressive patients with MH susceptibility will not have a mutation
non-Hodgkins lymphoma. Rapid growth of this malig- in the 12 to 15 exons in the hot spots of RYR1, which are
nancy may also have increased the basal metabolic rate. examined in the current clinical diagnostic genetic test of
Covering the child with warm blankets during isoflurane MH susceptibility.103,104 All first-degree relatives of this
anesthesia led to a further increase in esophageal and patient are considered to be MH-susceptible until they
nasal temperatures. Blankets were removed to increase have exhibited a normal caffeinehalothane contracture
convective and conductive heat loss. A fluid challenge of test. See www.mhaus.org for the addresses of active
20 mL per kg of lactated Ringers solution was given, in MH Diagnostic Centers. With the tentative diagnosis of
addition to continuing the alkaline fluid that had been MH-susceptible, these relatives should not receive potent
ordered preoperatively by the oncologist. inhalation anesthetics or succinylcholine.
The temperature was 37 C on admission to the Ongoing work may improve the sensitivity of the
PACU and increased to 38 C after discharge from the ryanodine receptor gene test in the diagnosis of MH
PACU. Any anesthetic drugs could have been given to susceptibility.105 More mutations causative for MH are
this child. Cisatracurium was administered to facilitate identified every year. Potential patients and physicians
tracheal intubation due to concern about the decreased are welcome to discuss their questions regarding this
clearance of other drugs, owing to reduced hepatic and test with the Center for Medical Genetics, University
renal blood flow (mass effect of the intra-abdominal of Pittsburgh (800-454-8155). The North American MH
tumor) and because succinylcholine can increase mas- Registry welcomes submission of deidentified reports
seter muscle tension in the presence of temperature of cases similar to this one. See www.mhreg.org for
elevation. instructions and report forms.
656 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
26. Kossoff EH, Vining EP, Pyzik PL, et al. The postoperative 48. Zhao J, Cannon B, Nedergaard J. Alpha 1-Adrenergic
course and management of 106 hemicortications. Pediatr stimulation potentiates the thermogenic action of beta3-
Neurosurg. 2002;37:298. adrenoreceptor-generated cAMP in brown fat cells. J Biol
27. Clinchot DM, Otis S, Colachis SC. Incidence of fever in the Chem. 1997;272:32847.
rehabilitation phase following brain injury. Am J Phys Med 49. Sprague JE, Mallett NM, Rusyniak DE, et al. UCP3
Rehabil. 1997;76:323. and thyroid hormone involvement in methamphetamine-
28. Panarese A, Clarke RW, Yardley MP. Early post-operative induced hyperthermia. Eur J Pharmacol. 2004;68:1339.
morbidity following tonsillectomy in children: Implications 50. Keck PE: Serotonin syndrome. In: Mann SC, Caroff SN,
for day surgery. J Laryngol Otol. 1999;113:1089. Keck PE, et al. eds. Neuroleptic malignant syndrome and
29. Merjanain RB, Kiriakos CR, Dorey FJ, et al. Normal related conditions, 2nd ed. Washington, DC: American
postoperative febrile response in the pediatric orthopaedic Psychiatric Publishing; 2003:86.
population. J Pediatr Orthop. 1998;18:497. 51. Houlihan DJ. Serotonin syndrome resulting from the co-
30. Schey D, Salom EM, Papadia A, et al. Extensive fever administration of tramadol, venlafaxine, and mirtazapine.
workup produces low yield in determining infectious Ann Pharmacother. 2004;38:411.
etiology. Am J Obstet Gynecol. 2005;192:1729. 52. Karunatilake H, Buckley NA. Serotonin syndrome in-
31. Schwandt A, Andrews SJ, Fanning J. Prospective analysis duced by fluvoxamine and oxycodone. Am Pharmacother.
of a fever evaluation algorithm after major gynecologic 2006;40:155.
surgery. Am J Obstet Gynecol. 2001;184:1066. 53. Shioda K, Nisijima K, Yoshino T, et al. Extracellular
32. Yeung RS, Buck JR, Filler RM. The significance of fever serotonin, dopamine and glutamate levels are elevated
following operations in children. J Pediatr Surg. 1982; in the hypothalamus in a serotonin syndrome animal
17:347. model induced by tranylcypromine and fluoxetine. Prog
33. Lynch BA, Dolan JP, Mann M. Thyrotoxicosis after gastric Neuropsychopharmacol Biol Psychiatry. 2004;28:633.
bypass surgery prompting operative re-exploration. Obes 54. Caroff SN. Neuroleptic malignant syndrome. In: Mann SC,
Surg. 2005;15:883. Caroff SN, Keck PE, et al. eds. Neuroleptic malignant
34. Petervari E, Konesecsko-Gaspar M, Balasko M, et al. syndrome and related conditions, 2nd ed. Washington, DC:
Increased thermoregulatory responsiveness in cold adapted American Psychiatric Publishing; 2003:6.
but not in hyperthyroid hypermetabolic rats. Acta Physiol 55. Leo RJ, Baer D. Delirium associated with baclofen with-
Hung. 2003;90:1. drawal: A review of common presentations and manage-
35. Reilly JF, Weisse ME. Topically induced diphenhydramine ment strategies. Psychosomatics. 2005;46:503.
toxicity. J Emerg Med. 1990;8:59. 56. Kao LW, Amin Y, Kirk MA, et al. Intrathecal baclofen
36. Candy JM, Morrison C, Paton RD, et al. Salicylate toxicity withdrawal mimicking sepsis. J Emerg Med. 2003;24:423.
masquerading as malignant hyperthermia. Pediatr Anesth. 57. Meythaler JM, Roper JF, Brunner RC. Cyproheptadine for
1998;8:421. intrathecal baclofen withdrawal. Arch Phys Med Rehabil.
37. Lees DE, Macnamara T. Ketamine-induced hyperthermia- 2003;84:638.
postictal or malignant? Anesthesiology. 1977;47:390. 58. Cunningham JA, Jelic S. Baclofen withdrawal: A cause of
38. Lasser EC, Lamkin GE. Effects of intrathecal injection of prolonged fever in the intensive care unit. Anaesth Intensive
diatrizoate on dopamine receptors. Acad Radiol. 2002;9:826. Care. 2005;33:534.
39. Mann SC. Thermoregulatory mechanisms and antipsy- 59. Douglas AF, Weiner HL, Schwartz DR. Prolonged intrathe-
chotic drug-related heatstroke. In: Mann SC, Caroff SN, cal baclofen withdrawal syndrome. Case report and discus-
Keck PE, et al. eds. Neuroleptic malignant syndrome and sion of current therapeutic management. J Neurosurgery.
related conditions, 2nd ed. Washington, DC: American Psy- 2005;102:1133.
chiatric Publishing; 2003:57. 60. Migneco A, Ojetti V, Testa A, et al. Management of
40. Sandow BA, Donnal JF. Myelography complications and thyrotoxic crisis. Eur Rev Med Pharmacol Sci. 2005;9:69.
current practice patterns. Am J Rad. 2005;185:768. 61. Hershman JM. Human chorionic gonadotroin and the thy-
41. Rothwell NJ. CNS regulation of thermogenesis. Crit Rev roid: Hyperemesis gravidarum and trophoblastic tumors.
Neurobiol. 1994;8:1. Thyroid. 1999;9:653.
42. Cox B, Lee TF. Further evidence for a physiological role for 62. Nelson T. Malignant hyperthermia: A pharmacogenetic
hypothalamic dopamine in the thermoregulation in the rat. disease of Ca++ regulating proteins. Curr Mol Med. 2002;2:
J Physiol (London). 1980;300:1. 347.
43. Mallick BN, Jha SK, Islam F. Presence of alpha-1 63. Hollander AS, Olney RC, Blackett PR, et al. Fatal malignant
adrenoreceptors on thermosensitive neurons in the medial hyperthermia-like syndrome with rhabdomyolysis compli-
preoptico-anterior hypothalamic area in rats. Neurophar- cating the presentation of diabetes mellitus in adolescent
macology. 2002;42:697. males. Pediatrics. 2003;111:1447.
44. Mechan AO, Esteban B, OShea E, et al. The pharmacology 64. Blanc P, Portas M, Paupe A, et al. Deficiency in medium
of the acute hyperthermic response that follows admin- chain acyl coA dehydrogenase manifested as febrile coma.
istration if 3,4-methylenedioxymethamphetamine (MDMA, Ann Pediatr (Paris). 1993;40:313.
ecstasy) to rats. Br J Pharmacol. 2002;135:170. 65. Melegh B, Bock I, Gati I, et al. Multiple mitochondrial
45. Jezek P, Freisleben HJ. Fatty acid binding site of the DNA deletions and persistent hyperthermia in a patient
mitochondrial uncoupling protein. Demonstration of its with Brachmann-de Lange phenotype. Am J Med Genet.
existence by EPR spectroscopy of 5-DOXYL-stearic acid. 1996;65:82.
FEBS Lett. 1994;343:22. 66. Gordon DL, Atamian SD, Brooks MH, et al. Fever in
46. Jezek P, Engstova H, Zackova M, et al. Fatty acid pheochromocytoma. Arch Int Med. 1992;152:1269.
cycling mechanism and mitochondrial uncoupling proteins. 67. Klausner JM, Nakash R, Inbar M, et al. Prolonged fever
Biochim Phiophys Acta. 1998;1365:319. as a presenting symptom in adrenal tumors. Oncology.
47. Ricquier D, Bouillaud F. The uncoupling protein homo- 1988;45:15.
logues: UCP1, UCP2, UCP3, StUCP and AtUCP. Biochem J. 68. Kang JM, Lee WJ, Kim WB, et al. Systemic inflamma-
2000;354:161. tory syndrome and hepatic inflammatory cell infiltration
658 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
caused by an interleukin-6 producing pheochromocytoma. 87. Bouchama A, Knochel JP. Heat stroke. N Engl J Med. 2002;
Endocr J. 2005;52:193. 346:1978.
69. Ghert M, Allen B, Davids J, et al. Increased postoperative 88. Kurahashi K, Kajikawa O, Sawa T, et al. Pathogenesis of
febrile response in children with osteogenesis imperfecta. septic shock in Pseudomonas aeruginosa pneumonia. J Clin
J Pediatr Orthop. 2003;23:261. Invest. 1999;104:743.
70. Deak SB, Scholz PM, Amenta PS, et al. The substitution 89. McAnulty SR, McAnulty L, Pascoe DD, et al. Hyperther-
of arginine for glycine 85 of the alpha 1(I) procollagen mia increases exercise-induced oxidative stress. Physiol
chain results in mild osteogenesis imperfecta. The mutation Biochem. 2005;26:188.
provides direct evidence for three discrete domains of 90. Bouchama A, Bridey F, Hammami MM, et al. Activation
cooperative melting of intact type I collagen. J Biol Chem. of coagulation and fibrinolysis in heatstroke. Thromb
1991;266:21827. Haemost. 1996;76:909.
71. Porsborg P, Astrup G, Bendixen D, et al. Osteogenesis 91. Moseley PL. Heat shock proteins and heat adaptation of the
imperfecta and malignant hyperthermia. Is there a rela- whole organism. J Appl Physiol. 1997;83:1413.
tionship? Anaesthesia. 1996;51:863. 92. Milner CM, Campbell RD. Polymorphic analysis of the
72. Cropp G, Myers DN. Physiologic evidence of hyperme- three MHC-linked HSP70 genes. Immunogenetics. 1992;36:
tabolism in osteogenesis imperfecta. Pediatrics. 1972;49: 357.
375. 93. Lenhardt R, Negishi C, Sessler DI, et al. The effects of
73. Kenan S, Liebergall M, Simchen E, et al. Fever following physical treatment on induced fever in humans. Am J Med.
orthopedic operations in children. J Pediatr Orthop. 1998; 1999;106:550.
18:497. 94. Sprague JE, Moze P, Caden D, et al. Carvediol reverses
74. Williams MS, Rooney BL, Williams J, et al. Investigation of hyperthermia and attenuates rhabdomyolysis induced by
thermoregulatory characteristics in patients with Prader- 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy) in
Willi syndrome. Am J Med Genet. 1994;49:302. an animal model. Crit Care Med. 2005;33:1311.
75. Gaasbeek D, Naarding P, Stor T, et al. Drug-induced hyper- 95. Rusyniak DE, Sprague JE. Toxin-induced hyperthermic
thermia in Huntingtons disease. J Neurol. 2004;251:454. syndromes. Med Clin North Am. 2006;90:261.
76. Stecker MM, Myers SM. Reserpine responsive myoclonus 96. Coffey RJ, Edgar TS, Francisco GE, et al. Abrupt withdrawal
and hyperpyrexia in a patient with Angelman syndrome. from intrathecal baclofen: Recognition and management
Clin Neurol Neurosurg. 2003;105:183. of a potentially life-threatening syndrome. Arch Phys Med
77. Mayhew JF. Intraoperative hyperthermia in a child with Rehabil. 2002;83:735.
neuroblastoma. Pediatr Anesth. 2006;16:890. 97. Clements JM, Casa DJ, Knight J, et al. Ice-water immersion
78. Tseuda K, Dubick MN, Wright BD, et al. Intra-operative and cold-water immersion provide similar cooling rates in
hyperthermia crisis in two children with undifferentiated runners with exercise-induced hyperthermia. J Athl Train.
lymphoma. Anesth Analg. 1978;57:511. 2002;37:146.
79. Lees D, Gadde PL, MacNamara TE. Malignant hyperther- 98. Plattner O, Kruz A, Sessler DI, et al. Efficacy of in-
mia in association with Burketts lymphoma: Report of a traoperative cooling methods. Anesthesiology. 1997;87:
third case. Anesth Analg. 1980;59:514. 1089.
80. Simons PS, Smithson WA, Gronert GA, et al. Acute myel- 99. Graham BS, Lichtenstein MJ, Hinson JM, et al. Nonexer-
ogenous leukemia and malignant hyperthermia in a patient tional heat stroke: Physiologic management and cooling in
with type 1b glycogen-storage disease. Pediatrics. 1984; 14 patients. Arch Intern Med. 1986;146:87.
105:428. 100. Lin CM, Neeru S, Doufas AG, et al. Dantrolene re-
81. Cohen PL, Hadler NM, Starkenburg R. Acute intermittent duces the threshold and gain for shivering. Anesth Analg.
porphyria presenting as acute muscle pain, fever, and 2004;98:1318.
weakness. Arthritis Rheum. 1997;40:586. 101. Al-Aska AK, Abu-Aisha H, Yaqub B, et al. Simplified cooling
82. Ince E, Ciftci E, Tekin M, et al. Characteristics of bed for heatstroke. Lancet. 1987;1:381.
hyperthermia and its complications in patients with Prader 102. Lenhardt R, Negishi C, Sessler DI, et al. Paralysis only
Willi syndrome. Pediatr Int. 2005;47:550. slightly reduces the febrile response to interleukin-2 during
83. Edelstein G, Hirshman CA. Hyperthermia and keto-acidosis isoflurane anesthesia. Anesthesiology. 1998;89:648.
during anesthesia in a child with glycogen-storage disease. 103. Sei Y, Sambuughin NN, Davis EJ, et al. Malignant
Anesthesiology. 1980;52:90. hyperthermia in North America: Genetic screening of the
84. Fietta P. Autoinflammatory diseases: The hereditary peri- three hot spots in the type I ryanodine receptor gene.
odic fever syndromes. Acta Biomed Ateneo Parmense. 2006; Anesthesiology. 2004;101:824.
75:92. 104. Brandom BW. The genetics of malignant hyperthermia.
85. Chang DM. The role of cytokines in heat stroke. Immunol Anesthesiol Clin North America. 2005;23:615.
Invest. 1993;22:553. 105. Sambuughin N, Holley H, Muldoon S, et al. Screening of the
86. Hammami MM, Bouchama A, Al-Sedairy S, et al. Concen- entire ryanodine receptor type 1 coding region for sequence
trations of soluble tumor necrosis factor and interleukin-6 variants associated with malignant hyperthermia suscepti-
receptors in heatstroke and heatstress. Crit Care Med. bility in the North American population. Anesthesiology.
1997;25:1314. 2005;102:515.
I . O B ST ET R I C
CHAPTER LABOR AND DELIVERY
CASE SUMMARY form of analgesia are generally safe for mother and fetus,
there are several associated complications with which con-
27-year-old woman presents to the emer- sulting anesthesiologists should familiarize themselves.
A
gency room with a persistent headache of The parturient should be well informed of the associated
24 hours duration. Five days ago, she un- risks including failure of analgesia, noninfectious mater-
derwent combined spinal epidural (CSE) nal fever, neurologic injury, hypotension, alteration of
labor analgesia for an uncomplicated vaginal progress of labor, and PDPH.
delivery. Following evaluation, emergency
room physicians make a diagnosis of postdural puncture
headache (PDPH). Five hundred milligrams of caffeine FAILURE OF ANALGESIA
are administered intravenously without improvement of
symptoms. The obstetric anesthesia service is called to One of the most important and common complications of
evaluate the patient for placement of an epidural blood neuraxial techniques is failure of or inadequate analgesia.
patch (EBP). Evaluation reveals a nonpositional headache The overall failure rate of epidural catheters used during
without diplopia, tinnitus, or relieving factors. The patient labor has been reported to be as high as 12%.2 The
is also complaining of epigastric pain. Before caffeine ad- term, failure, itself is quite broad and includes elements
ministration, the patients blood pressure was 160/110. such as failure to produce an adequate block, need
The obstetric anesthesiologist suggests obtaining a urine for catheter replacement, catheter migration/dislodgment,
sample for evaluation of urine protein. In addition, an and inability to insert a catheter despite cannulation of
obstetric consult is recommended to evaluate the patient the epidural space. By making an attempt to identify the
for possible postpartum preeclampsia. The urine dipstick etiology of failure, it may be possible to improve the
reveals 3+ proteinuria, and the patient is admitted for effectiveness of a neuraxial blockade.
treatment of postpartum preeclampsia. On her way to the
ward, she has an eclamptic seizure. Mechanisms of Failed Analgesia
Midline Structures
Several factors may contribute to epidural block failure.
What Are the Complications The presence of epidural midline structures may be re-
Associated with Analgesia for sponsible for some failed blocks. Autopsy, imaging, and
endoscopy have confirmed the existence of the plica me-
Labor? diana dorsalis, a midline band in the epidural space. This
band may cause a deflection of the Touhy needle to one
Many parturients choose neuraxial analgesia to alleviate side of the epidural space, uneven spread of local anes-
pain associated with labor and delivery. More than 50% thetic, or unilateral block. The variability in spinal nerve
of women in the United States receive either an epidural root diameter also may play a role in some block failures.
or CSE analgesia during labor.1 The popularity of these Larger spinal nerve roots such as sacral roots have longer
neuraxial techniques has increased because they are the diffusion distances for local anesthetics, which could result
most effective forms of providing analgesia for labor. Al- in incomplete blockade of those nerves. Failure of sacral
though the techniques to perform and administer this nerve root blockade can occur up to 17% of the time. A body
659
660 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
mass index >30 and extremes of stature have been asso- analgesia. In addition, CSE is faster in onset5 and has a
ciated with higher failure rates and greater probability of lower failure and wet tap rate.6,7 In a series of more
pain, respectively; however, it is unclear how these varia- than 19,000 patients, the failure rate for CSE was found
tions in body habitus contribute to epidural failure.3 to be 10% compared with 14% when epidural catheters
are used alone.2 Despite these results, there are several
Catheter Position and Defects theoretic disadvantages to CSE, including the following:
Several factors related to methodology and equipment 1. Unproven epidural catheter effectiveness
should be considered as well. Initial catheter misplace- 2. The risk of threading the epidural catheter intrathecally
ment can occur. In a study identifying failed catheter 3. Excessively high blocks
placement using epidurograms, three types of malpo- 4. Increased risk of PDPH
sitions were revealed, including transforaminal escape, 5. Increased risk of fetal bradycardia from spinal opioids
passage to the anterior epidural space, and paravertebral 6. Maternal respiratory depression
location. Catheter migration or dislodgment results in the 7. The risk of introduction of metal into the subarachnoid
local anesthetic not being delivered to the epidural space. space
Catheter defects such as reduced patency of the ports 8. Increased equipment costs
and aberrant placement of the distal hole in the catheter Nonetheless, these potential disadvantages and/or
also have been reported in failed epidural blocks. The complications do not occur at a greater rate than with
number of catheter ports (uniport vs. multiport) has been epidural catheters alone.5 In addition, one study found
found to affect block failure rates, with single-orifice and that CSE did not result in lower catheter failure rates;8
multi-orifice catheters having a 14.3% and 9.3% failure however, these investigators did not administer a sub-
rate, respectively. Other evidence has, however, suggested arachnoid drug after dural puncture. Although the poten-
no difference in the block failure rate between the two tial etiologies for failure of neuraxial analgesia (including
catheter types.3 choice of technique) are many, it is important to effec-
tively manage a failed block to maximize patient safety
Catheter Placement Techniques and satisfaction.
The preferences of those placing the epidural catheters can
also affect the incidence of catheter failure. A randomized
study in parturients found that use of air for loss of
resistance resulted in a higher incidence of inadequate What Is the Approach for the
analgesia compared with using saline.4 In addition, the Patient with Failed Labor
volume of local anesthetic injected into the epidural space
may be important for block success. Using the same
Analgesia?
milligram amount of the drug, a higher volume has been
found to provide better analgesia compared to a smaller
volume. HISTORY AND PHYSICAL
The distance of catheter insertion also influences the EXAMINATION
incidence of failed block. The optimal distance of catheter
insertion into the epidural space is probably in the range Failures of both CSE and epidural catheters are evaluated
of 2 to 6 cm, although 5 cm is the distance that many similarly. Soliciting a detailed history of the nature
clinicians use to reduce the risk of dislodgment, provide of the pain is the first step. Questions that focus
adequate analgesia, and reduce the risk of intravascular on the location (e.g., unilateral, perineal, abdominal),
cannulation. temporal association, and intensity may help further
Lastly, the technical skill and experience of the anes- elucidate the etiology. The physical examination should
thesiologist, which integrates some of the aforementioned focus on the bilateral levels of sensory and motor
factors, may greatly contribute to the success of the block. blockade in the lower extremities. It is important to
Indeed, failure of blockade varies inversely with experi- rule out symptoms of femoral, obturator, and sciatic
ence.3 neuropathies as potential etiologies of motor blockade.
The history and physical examination will usually guide
the anesthesiologist toward the proper treatment, even
if the etiology remains unclear. Ultimately, the specific
etiology of the pain may never be discovered, but more
Is a Combined Spinal Epidural importantly, adequate analgesia should be achieved.
Better Than Epidural
Analgesia? TREATMENT
One factor that may be overlooked when considering the Volume of Anesthetic
etiology of neuraxial failure is the choice of either epidural
or CSE for labor analgesia. There is a growing body One of the most common maneuvers employed in the
of evidence suggesting that CSE provides more effective treatment of inadequate labor analgesia is to bolus
C H A P T E R 47 / L A B O R A N D D E L I V E R Y 661
Effective?
Yes No Patient-Controlled Analgesia
Yes No Another option is the use of patient-controlled epidural
Consider replacing analgesia (PCEA), if available. Major advantages include
catheter vs. IV-PCA
Replace catheter decreased local anesthetic use, and increased patient
satisfaction. In addition, because the patient can self-
Increase rate of infusion, administer a bolus of the drug to the point of comfort, the
demand dose, or concentration workload for the anesthesiologist is reduced.5
the mother and neonate. Opioids are most effective when cesarean rates between parturients who received epidu-
administered by patient-controlled analgesia (PCA). Fen- ral analgesia and those who did not.11 These studies arose
tanyl is a good choice for PCA administration because of from a policy change in the Department of Defense, which
its rapid onset and short duration. The use of PCA also ruled that all military hospitals providing obstetric care
leads to greater patient satisfaction and lower overall opi- must make epidural analgesia available to all patients in
oid consumption when compared to intermittent bolus labor who request it. In military hospitals, before 1993,
dosing by nurses.10 Ketamine also can be added to the very few women received epidural analgesia for labor.
PCA to reduce fentanyl use or as an excellent alternative Shortly thereafter, most parturients elected to have epidu-
for short-term analgesia (usually during late stage 2 of ral analgesia. This type of study design should ensure
labor). When administered by intermittent bolus, doses that at least the characteristics of the patient population
of 10 mg up to 1 mg per kg produce intense analgesia are similar. Because these studies were not controlled for
with minimal respiratory depression. Doses >1 mg per changes in the practice style (e.g., increased instrumen-
kg may result in oversedation or unconsciousness with tal deliveries by obstetricians because patients had good
loss of airway reflexes. This is an undesirable situation, analgesia) that may have incited increased epidural ad-
especially in the obstetric population. ministration, these studies were flawed, thereby limiting
Inadequate analgesia is, unfortunately, a common their capacity in drawing definitive conclusions.
scenario for labor and delivery, and there are numerous Fortunately, included in the aforementioned studies
potential etiologies. Regardless of the cause, troubleshoot- was a series of randomized, controlled trials conducted
ing should be promptly performed to ensure patient to analyze the question of whether the administration
satisfaction and safety. Catheter replacement should be of epidural analgesia for labor increases the rate of
considered early because all parturients are at risk for ce- cesarean delivery.11 Randomized, controlled trials (prefer-
sarean delivery. Fentanyl PCA and IV ketamine should be ably double-blinded) serve as the gold standard for
considered options of last resort when effective neuraxial answering these types of clinical research questions. How-
analgesia cannot be achieved. ever, blinding in these studies is impossible, considering
that epidural analgesia clearly is superior when compared
with systemic opioids or no analgesia. Of more than 12
clinical trials, only 1 has shown an increased rate of ce-
What Are the Effects of Labor sarean delivery with epidural administration.12 That study
has been criticized for being underpowered to make its
Analgesia on the Progress and primary conclusion from the data. The study was termi-
Outcome of Labor? nated early for an unacceptably high cesarean rate in the
epidural group, and one more cesarean delivery in the con-
The influence of epidural analgesia on the progress and trol group would have eliminated statistical significance.
outcome of labor and delivery has been one of the most The other trials and meta-analyses of these trials have
controversial areas of obstetric anesthesia. Many studies shown that epidural analgesia does not increase cesarean
have evaluated the effects of epidural analgesia on the delivery rates.10,13
progress of labor and rates of cesarean and instrumental A recent study by Wong et al. further substantiated the
deliveries, as well as fetal and neonatal outcome. Lim- conclusion that epidural analgesia does not increase rates
itations in the design of many studiesincluding small of cesarean delivery.14 In this randomized controlled trial,
sample size, lack of randomization and retrospective two groups were compared. In the first group, parturients
analysishave led to erroneous interpretations and con- received intrathecal fentanyl at their first request for
clusions with respect to the effects of epidural analgesia analgesia if cervical dilation was <4 cm. At the second
on labor and outcome of delivery.11 request for analgesia, the women were administered
A series of retrospective studies from 1989 to 1996 epidural analgesia with bupivacaine. In the second group,
first suggested that epidural analgesia was associated IV hydromorphone was administered at first request if
with increased rates of cesarean delivery secondary to cervical dilation was <4 cm. Epidural analgesia was then
dystocia.11 These studies demonstrated up to a sixfold administered when the cervical dilation was >4 cm or at
increase in rates of cesarean delivery in parturients re- the third request for analgesia. No significant differences
ceiving epidural analgesia compared with those receiving were found in rates of cesarean delivery between the two
systemic opioids or no analgesia. This group of studies groups. Collectively, these randomized controlled trials
suffered from several design flaws that make it difficult strongly suggest that epidural analgesia does not increase
to draw any definitive conclusions. First, there are in- the cesarean delivery rate compared to systemic opioid
herent biases attributed to retrospective study designs. analgesia.
Most importantly, the two groups being compared may
not have shared equivalent risks, which was the case in
several studies. In most retrospective studies, women re- LENGTH OF LABOR
ceiving epidural analgesia more often had induced labor,
were frequently nulliparous, and had a smaller pelvis with The study by Wong et al. also demonstrates important
larger babies. secondary outcome results (e.g., duration of labor).14 The
One group of population-based studies that were median time from initiation of analgesia to complete di-
inherently less biased demonstrated no difference in lation and to vaginal delivery was significantly reduced in
C H A P T E R 47 / L A B O R A N D D E L I V E R Y 663
the early epidural group compared with the late epidural neonates born of mothers who had epidural analgesia had
group (90 and 80 minutes, respectively). It should be increased fetal base excess compared to those of mothers
noted that the women in the systemic analgesia group who received systemic opioids.19 These results are not
had lesser degrees of cervical dilation at first request for surprising, given that systemic opioids readily cross the
analgesia, thereby explaining some of the differences in placenta to a greater degree than epidural medications.
time to complete dilation and vaginal delivery. Before this Another finding determined that fetal base excess at birth
study, most agreed that epidural analgesia prolonged both was improved when epidural analgesia was administered
stage 1 and 2 of labor by approximately 20 to 50 minutes during labor.19
and 15 to 30 minutes, respectively.11 This was observed
in both randomized trials and meta-analyses. However, Fetus
these studies were significantly affected by high patient
crossover rates from systemic analgesia to epidural anal- An additional area of concern has been the effect of
gesia. Intention-to-treat analysis still has demonstrated epidural analgesia on fetal heart tones (FHT). There is
a significant difference. Overall, the data suggests that a theoretic risk of fetal bradycardia in the setting of
epidural analgesia prolongs stage 1 and 2 of labor, but epidural analgesia, secondary to hypotension induced by
further study may be necessary to determine whether this sympathetic blockade. One study comparing the effects of
increase in length of labor is clinically relevant. Since epidural analgesia on FHT to IV meperidine demonstrated
there is no evidence that links early epidural placement no evidence of adverse effects.20 However, CSE techniques
to adverse outcomes, the American Society of Anesthe- have produced different results with regard to FHTs.
siologists and the American College of Obstetricians and The use of CSE has been shown to increase the rate
Gynecologists issued a joint statement declaring that a of emergency cesarean delivery for fetal bradycardia in
womans request for pain relief at any time during labor several studies, especially when intrathecal opioids are
is sufficient indication to provide such relief.15 administered;21 on the other hand, other studies have
found no such relation.22,23 Collectively, the evidence
to date suggests that epidural analgesia and CSE have
minimal effects on fetal and neonatal outcome.
OTHER CONCERNS Most of the controversy surrounding the effects of
In addition to rates of cesarean delivery and length of epidural analgesia on the progress of labor has abated.
labor, there are other concerns regarding the effects of Current evidence suggests that epidural analgesia for
epidural analgesia for labor. Given that parturients with labor does not result in higher rates of cesarean delivery.
epidurals tend to have longer labors, there has been an The length of labor is likely prolonged by epidural
increase in oxytocin use and rate of operative vaginal analgesia, but the clinical significance of this increase
delivery.16 However, it is unclear whether obstetricians is unclear. Operative vaginal delivery, perineal injury, and
are more comfortable with inducing labor or performing oxytocin use occur at increased rates when parturients
operative vaginal deliveries in the setting of excellent receive epidural analgesia. Despite these concerns, there
pain relief. The increase in operative vaginal delivery, is no evidence in the literature to suggest that adverse
particularly forceps delivery, has resulted in a higher rate fetal or neonatal effects result from epidural analgesia
of perineal injury, leading to fecal incontinence in those administered to the laboring mother.
women who receive epidural analgesia for labor.16 There
has also been speculation that epidural analgesia results
in more fetal head malposition.17,18 Current evidence
regarding fetal malposition has been controversial, with What Are the Neurologic
some studies showing an increased rate of occiput Complications Associated with
posterior presentation, whereas others have not.
Neuraxial Labor Analgesia?
neuraxial blocks were administered over a 10-year period, Once this diagnosis is suspected, it is imperative to
determined that the incidence of injury in parturients quickly confirm the diagnosis to prevent long-term adverse
was approximately 1 per 25,000 for epidural blockade.25 outcomes. A neurologic surgery consult and emergent
More neurologic complications occurred with epidural computed tomographic (CT) scan should be obtained. It
than with spinal anesthesia. Additionally, more complica- may also be helpful to notify operating room personnel of
tions were found in surgical blocks than obstetric blocks. a possible laminectomy to evacuate the hematoma, so that
These injuries included spinal hematoma, epidural ab- a room may be prepared and on stand by should the case
scess, spinal cord lesions, subdural hematoma, permanent proceed. To avoid permanent neurologic injury (e.g., para-
abducens paralysis, and Horners syndrome with facial plegia), a spinal hematoma should be evacuated as soon
pain. as possible, preferably before neurologic signs develop.24
In a recent retrospective study, the incidence was found needle trauma will usually be accompanied by significant
to be 1 per 200,000 labor epidural blocks, the same in- pain. When pain is encountered during needle placement,
cidence as that for spinal hematoma.25 The pattern of insertion should cease, and the needle should be redi-
signs and symptoms shares some similarities with spinal rected. Injection of local anesthetic into the spinal cord or
hematomas, with a few notable exceptions. The onset of a spinal nerve will likely result in permanent neurologic
symptoms occurs approximately one week after placement injury. Pain on injection is an important warning that
of the epidural catheter. Fever and leukocytosis typically the injection should be stopped. However, needle trauma
accompany lower extremity motor and sensory deficits. without injection will most likely result in a nerve deficit
Back pain may also be present. The diagnosis and treat- that resolves over a period of days or months.
ment of epidural abscesses are essentially identical to those
for spinal hematoma, but in some cases, epidural abscess
has been treated with antibiotics in lieu of a laminectomy.
However, this form of treatment should only be considered DRUG TOXICITY
in cases with no evidence of neurologic compromise.27 The last mechanism of direct neural injury is drug
Simple precautions may aid in the prevention of toxicity. This usually results from unintentional injection
epidural abscesses. Wearing of masks, hats, and sterile of the wrong substance. Many different drugs including
gloves may help reduce infection. Sterile preparation and ephedrine, phenol, thiopental, and potassium chloride
draping of the area to be punctured should also be strictly have unintentionally been injected into the epidural
practiced. Despite these precautions, epidural abscess will space.27,30 Fortunately, injury by this mechanism is
still occur because most result from hematogenous spread rare because the epidural space seems to be very
from distant sites other than the epidural space.27 forgiving. Vigilance on the part of anesthesia providers can
dramatically reduce the incidence of these complications.
No Yes
INCIDENCE Conservative:
Consider further Bed rest
PDPH remains one of the most frequent complications of workup Caffeine
epidural analgesia provided during labor. The incidence Ergot alkaloids
of PDPH resulting from accidental dural puncture ranges Abdominal binder
from 0.5% to 6%.30 Much of this variation is dependent
on the experience of the person placing the block, with Epidural blood patch Failure Success
anesthesiology residency training programs having a
higher rate of PDPH than in other settings.32 In addition to
Failure Success
practitioner experience, there are many technical aspects
of epidural placement that can impact the incidence of
PDPH. Proper diagnosis and treatment of a PDPH can Repeat epidural blood patch
greatly affect the mothers ability to effectively care for
her neonate during the first days of life. Failure Success
Consider alternative
SIGNS AND SYMPTOMS diagnosis
In most patients, the diagnosis of PDPH is usually straight- FIGURE 47.2 Suggested algorithm for approach to postpartum
forward because the constellation of signs and symptoms headache.
are relatively distinct.33 The classic presentation is de-
scribed as a postural headache, with pain being at its
worst in the upright position and relieved by the supine decreases with age, with most cases occurring younger
position. Pain typically localizes to the frontal-occipital than 40 years. Gender also plays a role, with women
region, with occasional radiation to the neck. It is usually having a twofold increase in incidence compared with
described as dull, throbbing, and can range from mild to men. Patients with a history of PDPH are at increased
incapacitating. Patients often will complain of neck stiff- risk for recurrent headaches with subsequent neuraxial
ness. Other associated symptoms may include nausea and blocks. Although pregnancy also has been suggested as
vomiting, ocular (e.g., diplopia) and auditory (e.g., hear- a risk factor, it is not well established. In cases where
ing loss, tinnitus) disturbances resulting from traction on multiple attempts at epidural catheter placement have
cranial nerves and, rarely, seizures. The onset of PDPH is been made or difficulty encountered, the risk of PDPH
1 to 2 days after dural puncture, and the duration (when increases. There are several mitigating factors for PDPH
untreated) is usually <1 week, although a PDPH lasting in the setting of neuraxial blockade, including morbid
from months to years has been reported. The differential obesity, continuous spinal analgesia, and CSE analgesia.
diagnosis for PDPH is broad. It is imperative to con-
sider alternative diagnoses when the patients headache
symptoms are not consistent with PDPH (see Fig. 47.2).
Differential diagnoses include preeclampsia, migraine, THERAPEUTIC STRATEGIES
hypertension, brain tumor, subdural hematoma, sub- Given the uncertainty of the effectiveness of the various
arachnoid hemorrhage, cortical vein thrombosis, cerebral treatments for PDPH, one must employ all necessary
ischemia or infarction, pseudotumor cerebri, sinusitis, measures to prevent these headaches. Regardless of risk
meningitis, caffeine withdrawal, lactation headache, and factors, once an individual has been diagnosed with
pneumocephalus. Some of these entities can have devas- PDPH, effective treatment becomes paramount. Once
tating consequences for the patient and, therefore, one dural puncture has occurred, injection of intrathecal
must be thorough in pursuing a diagnosis when PDPH saline prophylactically may prevent headache, but its
seems unlikely. efficacy is unclear.31 There are a host of therapies that
relieve symptoms but do not cure the headache;31 these
may be useful in patients who refuse an EBP. Bed
RISK FACTORS rest is frequently employed for PDPH and is effective
because, as already mentioned, the headache is postural
While the pathophysiology of PDPH remains unclear, risk and diminishes in the supine position. Hydration has not
factors have been well described.33 The incidence of PDPH been shown to be effective, and the use of an abdominal
C H A P T E R 47 / L A B O R A N D D E L I V E R Y 667
binder has also been mentioned in the literature, although patches are less rigorously tested alternatives that may
its effectiveness is unclear. Certain medications including prove useful in the future. If other measures fail, surgery
caffeine, theophylline, and sumatriptan have limited, to close a dural tear may be the last resort to cure a PDPH.
short-term effectiveness in reducing symptoms of PDPH. Despite the risks, a prophylactic EBP is known to reduce
the duration of headache.34
Epidural Blood Patch
Choice of Needle
Efficacy In addition to performing prophylactic procedures once
Despite all these options, the EBP is considered the most a wet tap has occurred, there are several modifications
definitive treatment for PDPH, though there is very little of a technique that may minimize the risk of PDPH
evidence for its efficacy.27,31 Although the mechanism for before block placement.33 The choice of needle for
relief of PDPH by EBP has not been fully elucidated, it dural puncture (if spinal or CSE is performed) can
is known that the epidural blood clot initially compresses greatly influence the risk of headache. Smaller needles
the dural sac, thereby raising CSF pressure. However, reduce the risk of PDPH (a 27-gauge size has a low
the clot has been shown to dissolve within 24 hours of incidence of PDPH). The use of pencil-point needles (e.g.,
blood injection. Animal studies have shown fibroelastic Sprotte, Whitacre, Pencan) results in fewer headaches
activity in the area of clot formation 7 days after an than cutting needles (e.g., Quincke). If a Quincke needle
EBP. Most patients (88% to 96%), including obstetric and is used, insertion of the bevel parallel to the dural fibers
nonobstetric, obtain immediate relief, but only 60% to (longitudinal) has been reported to reduce the incidence of
75% have permanent cure of their headache after the PDPH.36 Finally, using air for epidural loss of resistance
first blood patch.27 However, in parturients, only 33% technique can theoretically worsen PDPH by causing a
to 64% experience permanent pain relief after a single pneumocephalus.
blood patch. A second EBP can be placed when the first
fails, with a much higher rate of permanent resolution
of symptoms; however, if the second blood patch is Dealing with Dural Puncture
ineffective, the diagnosis of PDPH should be questioned. When faced with an unintended dural puncture or wet tap,
In general, an EBP is safe and complications are rare.33 there are several options. First, one can simply remove the
epidural needle and attempt placement at another level,
preferably higher to avoid threading the catheter through
Timing and Volume the dural puncture site. Another option that has gained
The optimal timing of EBP placement is >24 hours after some support in the literature for reducing the incidence
dural puncture.33 When injecting the blood, best results of PDPH is to insert the epidural catheter intrathecally for
have been obtained with volumes in the 15 to 20 mL range. labor analgesia.37 After delivery, the catheter should be
Twenty milliliters most frequently results in success, but left in place for as long as possible, up to 24 hours. Some
injection volume may be limited by patient tolerance (e.g., have advocated the injection of saline into the intrathecal
back pain).31 Injecting more than 20 mL of blood will space before removal of the catheter, but evidence for
probably not increase the effectiveness of the procedure.33 its effectiveness is sparse.38 Because the efficacy of
intrathecal catheter placement has been refuted by several
Placement studies,39 one should consider the difficulty of the initial
placement when deciding whether to thread the epidural
Before prophylactic EBP placement, it is important to catheter intrathecally or replace it at another interspace.
allow the local anesthetic effects to wear off for several PDPH is one of the most frequent complications of
reasons.33 First, pain is used as a signal to stop the neuraxial labor analgesia. The risk can be minimized by
injection, so if pain sensation is reduced by the local several choices of technique; however, when PDPH occurs,
anesthesia still present in the system, this important the only definitive treatment is an EBP.
feedback will be lost. Second, the local anesthetic may
inhibit coagulation of the injected blood. Finally, a
prophylactic EBP could potentially result in a high block
if an epidural or spinal block is already present.
What Physiologic Changes
Complications Post Epidural Blood Patch
Occur with Maternal
A prophylactic blood patch would seem to be a reasonable Hypotension?
option; nevertheless, this approach has shown mixed re-
sults.31,34 Serious, rare complications including paralysis Hypotension, defined as a 20% to 30% decrease in
and cauda equina syndrome are reported in the litera- blood pressure from baseline, frequently complicates both
ture.35 Meningitis can occur following EBP placement, epidural and CSE analgesic techniques for labor. Studies
with high fever and sepsis as contraindications for a sec- suggest that maternal hypotension accompanies epidu-
ond EBP. Back, neck, or radicular pain can present either ral analgesia 12% to 29% of the time.40 Hypotension
during or after the procedure. Unfortunately, effective al- can cause symptoms that include dizziness and nausea
ternatives to the EBP are lacking. Dextran and fibrin glue and vomiting; however, the effect of more concern is
668 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
reduced uterine blood flow. Since uterine blood flow is has generally been defined as 500 to 1,000 mL of
proportionate to maternal blood pressure, reductions in crystalloid given immediately before the injection of a
blood pressure can adversely affect the fetus, especially local anesthetic.43 In recent years, several randomized,
when uteroplacental perfusion is already compromised controlled trials have assessed the effectiveness of IV fluid
(e.g., preeclampsia). Reductions in maternal blood pres- preloading and questioned the benefit of fluid loading
sure are known to produce fetal asphyxia, resulting in before neuraxial analgesia. One trial using a higher dose
abnormal FHTs and, in some cases, fetal bradycardia. of epidural analgesia for labor (0.25% bupivacaine) found
Several mechanisms have been proposed for the no difference in the incidence of hypotension between
changes in maternal and fetal physiology that occur as parturients who received an IV preload and those who
a result of maternal hypotension secondary to epidural did not.44 In studies where low-dose epidural analgesia
analgesia. Conduction anesthesia causes direct inhibition or CSE was used, fluid preloading did not reduce the
of sympathetic nerve activity in a segmental manner.41 incidence of hypotension in parturients.4547 In addition,
Lumbar epidural analgesia for labor inhibits several levels IV preloading has not been found to affect the frequency
of sympathetic innervation, resulting in vasodilation, of fetal bradycardia that results from epidural analgesia.
especially in the blood vessels of the lower extremities No study has shown a difference in rates of instrumental
and abdominal cavity. This sympathectomy leads to vaginal or cesarean delivery as a result of IV preloading;43
relative hypovolemia due to venous pooling, which results however, IV preloading has been shown to decrease the
in decreased venous return to the heart and reduced rate of uterine contractions when administered before
cardiac output. Ultimately, the reduction in cardiac output the initiation of labor with epidural analgesia.44 This
causes a decrease in blood pressure. In addition, pain effect could result in increased oxytocin administration
relief from epidural analgesia, combined with sympathetic or misinterpretation of the progress of labor, which leads
denervation of the adrenal gland, results in reduced levels to increased intervention. Overall, IV preloading with
of circulating catecholamines.40 The reduced levels of modest amounts of fluid before epidural analgesia for
catecholamines, in particular epinephrine, also contribute labor does not prevent maternal hypotension or fetal
to a reduction in blood pressure. bradycardia, but it may reduce the severity of hypotension
Fetal bradycardia can accompany a decreased ma- and improve the response to vasopressors.
ternal blood pressure, but the degree and duration of
hypotension necessary to produce fetal bradycardia is un- Ephedrine
certain. The primary cause of fetal bradycardia is thought
to be due to the decreased uteroplacental perfusion that Although fluid preloading is not effective in preventing
accompanies maternal hypotension;40 however, other pos- maternal hypotension and fetal bradycardia, ephedrine
sible mechanisms for fetal bradycardia exist. During labor, administration is known to both prevent and treat these
catecholamines are increased, resulting in 2 -receptor conditions. In a prospective, double blind, randomized
stimulation and tocolysis. Since epidural analgesia causes trial in which the treatment group received 25 mg of in-
a decrease in blood epinephrine levels, 2 agonism also tramuscular (IM) ephedrine before CSE, the incidence
decreases, with resultant uterine hyperactivity. In turn, of maternal hypotension was reduced.40 A significant
uterine hyperactivity impairs oxygen delivery to the pla- reduction in the incidence and frequency of fetal late
centa, which leads to fetal hypoxia and bradycardia. Other decelerations was also noted. However, there was an in-
mechanisms that may contribute to fetal bradycardia in- creased incidence of fetal tachycardia in the treatment
clude occult aortocaval compression, absorption of local group, which was related to the placental transfer of
anesthetic from the epidural space resulting in uterine ephedrine. In another randomized trial, parturients in
artery constriction, and fetal local anesthetic systemic the treatment group were given 10-mg ephedrine boluses,
toxicity.42 followed by continuous ephedrine infusions.42 Ephedrine
administration resulted in fewer episodes of fetal brady-
cardia while preventing maternal hypotension. Ephedrine
is clearly beneficial in the prevention and treatment of
PREVENTION maternal hypotension due to epidural analgesia for labor.
AND TREATMENT
Considering the risks of maternal hypotension for both
Phenylephrine
the mother and fetus, it is imperative to either prevent While ephedrine administration has been effective in pre-
or promptly treat this complication. Left uterine dis- venting and treating maternal hypotension, a growing
placement is a simple maneuver that may increase both body of evidence in the literature concerning cesarean de-
maternal blood pressure and uteroplacental perfusion. livery under spinal anesthesia suggests phenylephrine may
be the better choice. Although phenylephrine may main-
Intravenous Preloading tain uteroplacental perfusion pressure, there have been
concerns that it produces uteroplacental vasoconstric-
Besides left uterine displacement, two commonly used tion, thereby lowering villous perfusion and producing
methodsfluid administration and vasopressor therapy placental and fetal stress. However, in studies evaluat-
are used to prevent and/or treat maternal hypotension. ing the fetal effects of vasopressor therapy, ephedrine has
IV fluid preloading has been practiced widely, and been consistently associated with lower umbilical cord pH
C H A P T E R 47 / L A B O R A N D D E L I V E R Y 669
when compared with phenylephrine. In a case series of of epidural analgesia versus IM meperidine, the authors
337 parturients undergoing cesarean delivery under spinal demonstrated that epidural analgesia was associated with
anesthesia, multiple linear regression analysis found that a 1 C increase in body temperature every 7 hours com-
ephedrine administration for treatment of hypotension pared to constant temperature measurements for the IM
was associated with increased uterine artery base deficit meperidine group. However, the ambient temperature
and decreased pH, while direct -agonist use was not of the labor rooms, 24 C to 26 C, confounded the re-
associated with these laboratory findings.48 In addition, sults of this study. A follow-up study in which labor
a randomized, controlled trial determined that the ad- room temperature was cooler (20 C to 21 C) also revealed
ministration of phenylephrine infusions after initiating an association between epidural analgesia and maternal
spinal anesthesia for cesarean delivery prevents the fe- fever.53 However, in this study, the difference in tempera-
tal metabolic acidosis associated with hypotension under ture between the two groups was not clinically significant
spinal anesthesia.49 Furthermore, meta-analytic compar- (e.g., <1 C).
isons of ephedrine versus phenylephrine administration Later studies found clinically and statistically sig-
for treatment of hypotension during spinal anesthesia for nificant differences between parturients who received
cesarean delivery demonstrated that ephedrine was inde- epidural analgesia and those who did not. In a two-
pendently predictive of lower umbilical cord pH, whereas part study with retrospective review and nonrandomized
phenylephrine had no effect on fetal acidbase status.50 prospective analysis, a 0.07 C per hour increase in body
Although the reason for these differences are unclear, temperature was observed in the epidural group com-
there has been speculation that ephedrine causes fetal pared to those who did not receive epidural analgesia for
acidemia by increasing metabolic demand in the fetus labor.54 A greater number of women with epidural anal-
because it readily crosses the placenta. gesia were nulliparous with longer labors. These factors,
Although phenylephrine has been shown to be a nulliparity and long labor, were independent variables for
better choice than ephedrine for treatment of hypotension maternal fever and maternal infection.53,55 This evidence
during cesarean delivery under spinal anesthesia, there led to the hypothesis that the etiology for maternal fever
is no evidence that this is true for hypotension following associated with epidural analgesia is actually maternal
epidural analgesia for labor. However, one could speculate infection.
that the physiologic principles that govern the maternal Another study solidified the relationship between
hypotension resulting in fetal metabolic acidosis under maternal infection and epidural analgesia.56 In this study,
both conditions are similar. Therefore, it may be possible a series of placentas were harvested from women who
to extrapolate that phenylephrine may be a better choice delivered 6 hours or more after rupture of membranes.
than ephedrine for the treatment of maternal hypotension Again, women administered epidural analgesia were
following epidural analgesia. more likely to be febrile. What this study demonstrated,
Maternal hypotension resulting from epidural anal- however, was that epidural analgesia was only associated
gesia must be treated aggressively to prevent abnormal with fever in the presence of placental inflammation.
FHTs. If left untreated, serious complications can result, These results suggested that maternal infection was
including fetal bradycardia. Fluid preloading is not par- responsible for fever associated with epidural analgesia.
ticularly useful in preventing hypotension but may reduce The request for epidural analgesia probably served as a
the severity of hypotension. Ephedrine and phenylephrine marker for pain associated with chorioamnionitis. While
are both effective in the prevention and treatment of this study seemed to explain the etiology of maternal fever
maternal hypotension, but phenylephrine may be the associated with epidural analgesia, it could not explain
superior agent for preventing fetal metabolic acidosis. the lack of fever in those who acquired infection while
Future studies of phenylephrine administration for the receiving IV opioids.
treatment of hypotension that occurs during epidural Two studies have evaluated the lack of association
analgesia for labor are needed. between IV opioid use for labor analgesia and fever in
the setting of chorioamnionitis. In one study, a group of
volunteers were injected with interleukin-2, a pyrogen,
followed by IV fentanyl, epidural analgesia with local
Is There a Causal Relationship anesthetic, or epidural analgesia with local anesthetic
between Labor Analgesia and and fentanyl.57 Fentanyl had an antipyretic effect in
those receiving it intravenously, but the other two groups
Noninfectious Maternal Fever? that received epidural analgesia had the predicted febrile
response to interleukin-2. These results may explain
Maternal fever associated with epidural analgesia for why maternal temperature increases in the setting of
labor commonly occurs. There has been considerable chorioamnionitis with epidural analgesia, but not with IV
controversy regarding the validity and significance of this fentanyl administration.
finding. In the general operating arena, the use of lumbar Another cohort study looked to confirm the an-
epidural analgesia causes a decrease in patient temper- tipyretic effect of opioids.58 This study had four groups:
ature, presumably due to redistribution of cooler blood (i) no labor analgesia; (ii) IV and IM nalbuphine for
from the lower extremities to the central core.51 Increased analgesia; (iii) epidural analgesia with bupivacaine and
temperature associated with epidural analgesia for labor fentanyl; and, (iv) epidural analgesia with IVIM nal-
was first reported in 1989.52 In this nonrandomized trial buphine. No significant difference in the incidence of fever
670 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
was found between the no-analgesia group and the IVIM However, febrile fetuses did not have lower Apgar scores
nalbuphine group (1% vs. 0.3%) and between the two or umbilical cord blood pH at birth. Another randomized
epidural groups (17% vs. 17%). However, a significant dif- trial demonstrated that neonates born to mothers who
ference was noted between the no-analgesia group and the received epidural analgesia for labor had increased rates
epidural-only group (1% vs. 17%). The authors argued that of sepsis workup and antibiotic therapy.61 Interestingly,
if opioid administration could have an antipyretic effect the rate of sepsis workup was increased, even in women
in those who might otherwise be febrile (i.e., women with who did not have fever after epidural analgesia. Although
chorioamnionitis), there should be no difference in the neonatal sepsis workups were increased in the epidural
incidence of fever between the no-analgesia and epidural- group, confirmed cases of sepsis were rare and not
only groups. Neither group would have demonstrated the related to labor analgesia, suggesting that it may be
antipyretic effect of systemic opioids. Therefore, they con- suboptimal to unnecessarily subject neonates to antibiotic
cluded that systemic opioid use does not account for the administration and performance of diagnostic tests.
difference in the incidence of fever for those receiving The relation between epidural analgesia for labor
epidural analgesia. Despite these results, it is difficult to and maternal fever is complex. A combination of several
directly compare the two studies, given that fentanyl was factors likely contributes to fever in this setting. It is prob-
used in the first while nalbuphine was used in the most able that women with chorioamnionitis are more likely
recent. The difference in findings could be explained by to request epidural analgesia for pain associated with
several factors. First, the study designs were different: the infection. Those who do not receive epidural analge-
nonrandomized trial versus cohort. Additionally, fentanyl sia probably continue to have pain and receive systemic
is a potent opioid agonist, while nalbuphine is a combined opioids, which may mask the fever associated with the
agonist-antagonist. Finally, the first study used healthy infection. Epidural analgesia may also alter thermoreg-
male volunteers, whereas the cohort study used parturi- ulation, decrease sweating due to sympathectomy, and
ents in labor. A prospective, randomized trial may help increase shivering, thus, leading to heat production. The
resolve this controversy in the future. magnitude of these effects in the context of other ex-
Although some studies have focused on infection and perimental findings remains unclear. Finally, one must
the antipyretic effect of opioids as a possible etiology consider that there have been no adverse maternal or
for fever associated with epidural analgesia, at least one neonatal outcomes associated with epidural-related ma-
study suggests that fever may be due to alteration of ternal fever.
the maternal thermoregulatory physiology by epidural
analgesia.59 This large retrospective, cohort study found
a significant difference in maternal fever associated with
the use of epidural analgesia.59 Two study time periods
KEY POINTS
were compared. In the first period, 1992 to 1993, epidural
analgesia was only available to women with preeclampsia 1. Aggressive troubleshooting should be performed for
and severe cardiovascular disease (1% of parturients). In inadequate epidural analgesia for labor with early
the second study period, 1995 to 1996, epidural analgesia consideration of catheter replacement because all
was available on demand to many more women (83% parturients are at risk for cesarean delivery.
of parturients). Fever, defined as >38 C, was present in 2. Current evidence suggests that epidural analgesia for
3/498 patients in the first group and 63 of 572 patients in labor does not result in higher rates of cesarean
the second group. In this study, the investigators found delivery, but does result in increased length of labor,
no statistically significant difference in the duration of rates of operative vaginal delivery, perineal injury,
membrane rupture or number of vaginal examinations and oxytocin use. The clinical significance of these
between groups. No evidence was found to support effects is unclear
infection as the etiology of fever; however, this was not 3. Neurologic complications from epidural analgesia for
the primary focus of the study. labor are rare. Evaluation of injury requires an orga-
nized approach to rule out catastrophic etiologies.
4. PDPH is one of the most frequent complications of
neuraxial labor analgesia, but risk can be minimized
IMPLICATIONS FOR by different choices of technique. EBP is the only
NEONATAL OUTCOME definitive treatment. Other etiologies for postpartum
headache must be considered when patient history is
The controversy has been considerable regarding the not consistent with PDPH.
effects of fever associated with epidural analgesia for 5. Ephedrine and phenylephrine are effective in the
labor and neonatal outcome. One study in the early prevention and treatment of maternal hypotension.
1990s used an intrauterine probe to measure fetal skin Phenylephrine may be the superior agent to prevent
temperature.60 In the study, results from 57 women with fetal metabolic acidosis. Fluid preloading is not par-
ruptured membranes in active labor were reported. Fetal ticularly useful in preventing hypotension but may
temperature >38 C was found in 10 of 33 women who reduce the severity of hypotension.
received epidural analgesia, but none of the women in the 6. There have been no adverse maternal or neonatal
nonepidural group had febrile fetuses. In addition, none outcomes associated with epidural-related maternal
of the women in the epidural group had chorioamnionitis. fever.
C H A P T E R 47 / L A B O R A N D D E L I V E R Y 671
[Review]. Cochrane Database Syst Rev Oct. 2004;18(4): 53. Camann WR, Hortvet LA, Hughes N, et al. Maternal
CD000175. temperature regulation during extradural analgesia for
44. Cheek TG, Samuels P, Miller F, et al. Normal saline i.v. fluid labour. Br J Anaesth. 1991;67:565.
load decreases uterine activity in active labour. Br J Anaesth. 54. Vinson DC, Thomas R, Kiser T. Association between epidural
1996;77:632. analgesia during labor and fever. J Fam Pract. 1993;36:
45. Kinsella SM, Pirlet M, Mills MS, et al. Randomized study of 617.
intravenous fluid preload before epidural analgesia during 55. Philip J, Alexander JM, Sharma SK, et al. Epidural anal-
labour. Br J Anaesth. 2000;85:311. gesia during labor and maternal fever. Anesthesiology.
46. Kubli M, Shennan AH, Seed P, et al. A randomised controlled 1999;90:1271.
trial of fluid pre-loading before low dose epidural analgesia 56. Dashe JS, Rogers BB, McIntire DD, et al. Epidural analgesia
for labour. Int J Obstet Anesth. 2003;12:256. and intrapartum fever: Placental findings. Obstet Gynecol.
47. Macaulay BD, Barton MD, Norris MC, et al. Pre-hydration 1999;93:341.
and combined spinal epidural labor analgesia [Abstract]. 57. Negishi C, Lenhardt R, Ozaki M, et al. Opioids inhibit febrile
Anesthesiology. 2000;92(suppl):A83. responses in humans, whereas epidural analgesia does not:
48. Ngan Kee WD, Lee A. Multivariate analysis of factors An explanation for hyperthermia during epidural analgesia.
associated with umbilical arterial pH and standard base Anesthesiology. 2001;94:218.
excess after caesarean section under spinal anaesthesia. 58. Gross JB, Cohen AP, Lang JM, et al. Differences in
Anaesthesia. 2003;58:125. systemic opioid use do not explain increased fever incidence
49. Ngan Kee WD, Khaw KS, Ng FF. Comparison of phenyle- in parturients receiving epidural analgesia. Anesthesiology.
phrine infusion regimens for maintaining maternal blood 2002;97:157.
pressure during spinal anaesthesia for caesarean section. Br 59. Yancey MK, Zhang J, Schwarz J, et al. Labor epidural
J Anaesth. 2004;92:469. analgesia and intrapartum maternal hyperthermia. Obstet
50. Reynolds F, Seed PT. Anaesthesia for caesarean section Gynecol. 2001;98:763.
and neonatal acid-base status: A meta-analysis. Anaesthesia. 60. Macaulay JH, Bond K, Steer PJ. Epidural analgesia in labor
2005;60:636. and fetal hyperthermia. Obstet Gynecol. 1992;80:665.
51. Gaiser R. Neonatal effects of labor analgesia. Int Anesthesiol 61. Lieberman E, Lang JM, Frigoletto F Jr, et al. Epidural
Clin. 2002;40:49. analgesia, intrapartum fever, and neonatal sepsis evaluation.
52. Fusi L, Steer PJ, Maresh MJ, et al. Maternal pyrexia Pediatrics. 1997;99:415.
associated with the use of epidural analgesia in labour.
Lancet. 1989;1:1250.
CHAPTER CESAREAN SECTION
48 Christopher F. James
CASE SUMMARY from the hospital 3 days after her procedure in good
condition.
36-year-old woman, gravida 2, para 1,001,
A
with a 34-week intrauterine pregnancy pre-
sented with a complete placenta previa,
minimal bleeding, and occasional contrac- What Is the Historic Perspective
tions. Over the next 24 hours, contractions
continued despite tocolysis, along with inter- on Cesarean Section?
mittent bleeding and decreased serum hemoglobin levels
from 10.2 g per dL at admission to 9.0 g per dL. A ce- Although cesarean section has been sporadically alluded
sarean section was scheduled and the patient was taken to in ancient times, the accuracy concerning the onset
to the operating room. A combined spinal epidural (CSE) of the procedure remains in question. Roman law, as far
anesthetic was administered, utilizing 12 mg of bupiva- back as the eighth century BC, established that abdominal
caine with dextrose plus 150 g of intrathecal morphine. delivery be performed in a dying or dead woman in an
She was then placed in the left lateral tilt position. Shortly attempt to save the life of the baby or, more commonly,
thereafter the patient complained of nausea and her blood to allow for separate burial for mother and child.1 It
pressure decreased acutely from 120/70 to 85/40 mm Hg. was not until the 19th century that cesarean deliveries
Ten milligrams of intravenous ephedrine were admin- were attempted, not for the sole purpose of saving the
istered immediately, with resolution of the nausea and fetus, but also to save the mother, although the mortality
restoration of blood pressure. A viable male infant weigh- rate approached 100%, and cesarean section was therefore
ing 2,440 g, with Apgar scores of 7 at 1 minute and 9 at reserved for rare circumstances. The most common causes
5 minutes, was delivered without incident, but the pla- of maternal mortality from cesarean section in those early
centa was fragmented. The remainder of the procedure days were from hemorrhage and sepsis. Various surgical
was uneventful until skin closure when the patient de- techniques for abdominal deliveries were subsequently
veloped heavy vaginal bleeding despite increased doses devised to minimize the complications; however, the
of oxytocin and intramuscular prostaglandin F2 (Hema- most common uterine incision todaythe low transverse
bate; Upjohn, Kalamazoo, MI). She underwent immediate incisionwas not reported until the 1920s and was not
surgical reexploration, and an emergency hysterectomy accepted in common practice until the 1950s. With the
was performed. The intraoperative course was eventful evolving and improved surgical and anesthetic techniques
for a marked fluid resuscitation, which included 4 units in the 20th century, along with greater supportive
of packed red blood cells. Approximately 90 minutes care including blood banking, fluid resuscitation, and
later, the patient experienced some abdominal discom- antibiotics, maternal mortality following cesarean section
fort. The spinal block had receded to approximately the markedly decreased, but still remains higher than the
T6-7 dermatome level. She was given 50 g of intra- mortality rate with vaginal delivery. In the most recent
venous fentanyl, and the epidural catheter was dosed report on Confidential Enquiries into Maternal Deaths in
with lidocaine 2% with epinephrine, with resolution of the United Kingdom during the period 2000 to 2002, the
the discomfort. Her postoperative period was significant relative risk of maternal mortality was 3.7 for cesarean
for low grade, disseminated intravascular coagulation, re- sections versus vaginal deliveries2 (see Table 48.1).
quiring more blood products until her hemoglobin and Two of the most striking changes in obstetric anes-
coagulation studies normalized. The epidural catheter thesia practice over the last half century is the increasing
was removed 24 hours later. The patient was discharged incidence of cesarean section deliveries and the marked
673
674 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
TABLE 48.1 Vaginal versus Cesarean Delivery Case Fatality Rates per 100,000 in the United
Kingdom: Years 2000 2002
Delivered Deaths
Total per Direct Death Rate
Type of Delivery Number (105 ) and Indirect (n) Relative 100,000 Risk
Vaginal 1,571 75 48 1.0
Cesarean
Emergency and urgent 212 44 208 4.3
Scheduled and elective 214 29 136 2.8
Total Cesarean 426 73 172 3.7
Adapted, with data from: Why mothers die. The sixth report of the confidential enquiries into maternal deaths
in the United Kingdom, 20002002. Available at: http://www.cemach.org.uk. Accessed November 6, 2006.2
as local anesthesia or infiltration block are seldom used intrathecally, that leads to a total spinal anesthetic versus,
now. These techniques are utilized mainly in extrane- at worst, a high spinal with a spinal local anesthetic dose.
ous circumstances, such as unavailability of anesthesia
for emergent situations and in morbidly obese patients,
where regional and general anesthesia may be quite dif-
ficult. The problem with local anesthesia includes the What Are the Complications
inadequacy of pain relief during an abdominal procedure
and the possibility of local anesthetic toxicity due to the Related to Spinal Anesthesia in
large amount of drug that is usually required to provide the Parturient?
analgesia.
POSTDURAL PUNCTURE
SPINAL ANESTHESIA HEADACHE
The anesthetic of choice for cesarean section today is
A longtime concern with spinal anesthesia has been
spinal anesthesia. The advantages of spinal anesthesia in-
the incidence of postdural puncture headache (PDPH),
clude the rapid onset of the block, diminishing the time
especially in the obstetric population. With the improved
from entry to the operating room to incision as com-
technology of present day spinal needlesmainly, the
pared with epidural anesthesia, which translates to added
pencil-point needles such as the Whitacre, Sprotte, and
efficiency, an important factor in todays hospital environ-
Gertie Marx, as opposed to the older diamond-point or
ment.15 Other advantages include the greater density of
cutting spinal needles such as the Quincke needle and
the block, allowing for more muscle relaxation during the
the smaller gauge needlesthe incidence of PDPH has
procedure, which is obviously preferred among surgeons,
markedly declined to <1% in the obstetric patients, and
along with fewer requirements for systemic supplementa-
therefore has contributed to the popularity of spinal
tion due to pain, the simpler nature of the spinal technique
anesthesia for cesarean section. PDPH has both medical
compared with epidural anesthesia, and fewer overall
and legal implications. In the ASA Closed Claims Study,
complications.15 The two neuraxial complications that
headache was the fourth leading claim in the 1990s (third
can lead to maternal mortality include local anesthetic
in the 1970s and 1980s), consisting of 14% of all obstetric
toxicity and a total spinal. Both of these complications are
anesthesia claims and only surpassed by maternal death,
a result of epidural anesthesia (see Table 48.2). With spinal
newborn brain damage, and maternal nerve injury14 (see
anesthesia, the local anesthetic dose is small, and there
Table 48.3). PDPH invokes a significant morbidity for
is relatively no appreciable systemic absorption from the
postpartum patients because they cannot remain upright
cerebrospinal fluid (CSF). On the other hand, the greater
or ambulate because of the severity and postural nature
dose (fivefold or more) required with an epidural anes-
of the headache, along with its associated symptoms that
thetic, along with the vascular absorption in the epidural
include nausea and vomiting, photophobia, and, more
space, or even a direct vascular injection, especially with
rarely, visual and auditory symptoms such as diplopia,
the engorged epidural veins in the pregnant patient,
tinnitus, and decreased hearing due to cranial nerve (CN)
render an epidural more prone to local anesthetic toxicity.
involvement (CN VI and VIII). There also have been scant
Moreover, it is this large dose of local anesthetic, resulting
reports of intracranial subdural hematomas following
from an epidural attempt which accidentally is injected
dural puncture with neuraxial anesthesia.16 Therefore,
it is imperative for anesthesiologists to strive for a low
TABLE 48.2 Complications of Neuraxial Blocks for headache incidence and treat and carefully follow any
PDPH.
Cesarean Section
TABLE 48.3 American Society of Anesthesiologists (ASA) Closed claims Study: Obstetric Claims, Maternal, 1990s
Cesarean section deliveries (n = 168; 58% of all OB Claims in the 1990s)
General anesthesia: 28% of cesarean section claims
Neuraxial anesthesia: 72% (epidural 42%; spinal 26%)
OB: ALL DELIVERIES (n = 310) NON-OB CASES (n = 3,099)
Severe complications (maternal only)
Maternal nerve damage 20% Patient nerve damage 17%
Maternal death 12% Patient death 36%
Maternal brain damage 6% Patient brain damage 13%
Aspiration pneumonitis 1% Aspiration pneumonitis 2%
Minor complications (39% of OB claims) (7% of non-OB claims)
Headache 14% Headache 2%
Back pain 10% Back pain 1%
Emotional distress 8% Emotional distress 4%
Pain during surgery 7% Pain during surgery 1%
Note that cesarean sections made up 58% of the OB anesthesia claims and the cesarean section rate in the 1990s was 20%22%. The
minor complications with OB claims made up approximately 40% of all the claims as opposed to <10% of general surgical non-OB claims.
OB, obstetric.
Modified from Chadwick. HS. An analysis of obstetric anesthesia cases from the ASA closed claims project database. Int J Obstet Anesth
1996;5:258 and Davies JM. Obstetric anesthesia closed claims-trends over last three decades. ASA News. 2004;68:2.
or without colloid solutions or colloid solutions alone, severe hypotension can lead to fetal acidosis, fetal
have been used to pretreat patients before neuraxial bradycardia, and increased time to sustained respirations
block; however, no technique can eliminate hypotension. in neonates. Alternatively, transient maternal hypotension
Although pretreatment with colloid solutions may show is usually well tolerated, except in situations where
a decreased incidence of hypotension versus crystalloids there may already be compromised to the uteroplacental
in cesarean sections, the increased costs and possible unit or in chronic conditions such as diabetes, chronic
side effects, along with the lack of documented improved hypertension, and collagen vascular diseases.
outcome has precluded the use of colloids in routine
cases. Other studies have even questioned the efficacy
of pretreatment with intravenous fluids versus no bolus
administration before spinal anesthesia for cesarean
section, finding no statistical difference in the incidence
What Are the Complications
of hypotension.1719 Moreover, even with documented Associated with Epidural
increases in blood volume and cardiac output after Anesthesia?
the infusion of 1,500 mL of crystalloid, there was no
significant reduction in the incidence of hypotension.20
Ephedrine, a mixed -adrenergic and -adrenergic With the increasing use of epidural analgesia for the
agonist, has been the vasopressor of choice for the laboring patient, epidural anesthesia for cesarean section
obstetric patient secondary to having the least effect on has increased over the last three decades. Since the
uterine blood flow, as opposed to the pure -adrenergic early 1970s, epidurals for labor have been extended
agonists, such as phenylephrine. However, phenylephrine for use in cesarean sections.22 However, because of
in small doses has been shown to be as safe and the resurgence of spinal anesthesia in obstetrics with
effective as ephedrine and may also produce less fetal the availability of pencil-point needles, de novo epidural
acidosis.21 Phenylephrine in bolus doses of 50 to 100 g anesthesia for elective cesarean sections have markedly
can be administered when there is minimal response decreased. The disadvantages of epidural anesthesia
to ephedrine or in cases of maternal tachycardia that versus spinal anesthesia for operative procedures include
can be worsened by ephedrine. The greater incidence the slower onset of the block, less dense block, and the
occurs in nonlaboring patients versus laboring patients. greater potential for local anesthetic toxicity through an
Unfortunately, with neuraxial blocks, especially to the accidental intravascular injection or total spinal block
extent that is required, that is T4 level, for adequate with an accidental intrathecal injection due to the
surgical anesthesia for abdominal cases, the cardiac larger dose and volume of local anesthetic required with
accelerators fibers to the heart (T1 to T4) are also blocked. epidurals. Despite these factors, there are advantages to
This results in bradycardia, along with hypotension, epidural anesthesia, which include its flexibility, that is,
in contrast to the typical tachycardic response with increasing the duration of the block by further dosing the
an intact sympathetic system, further contributing to epidural catheter if the surgery is prolonged. Prolonged
a decrease in cardiac output. The fetal and neonatal cesarean sections are considered a risk for morbidly
effects of maternal hypotension are dependent on the obese patients and patients with prior multiple cesarean
extent and duration of the hypotension. Prolonged and sections or other abdominal surgeries because they are
CHAPTER 48/CESAREAN SECTION 677
susceptible to technical surgical difficulties, including the neuraxial blocks are commonly blamed for neurologic
possibility of a cesarean hysterectomy. Moreover, with complications; however, most problems are obstetric
epidural anesthesia, the epidural catheter can also be used related, that is, pregnancy and the labor and delivery
for postoperative analgesia. There is also a hemodynamic process. In a recent survey from Sweden, the incidence
advantage over spinal anesthesia, namely a decreased of severe neurologic complications after neuraxial blocks
incidence and degree of hypotension secondary to the was 1 in 25,000 following epidural blocks in the obstetric
slower onset of the epidural block as opposed to the patient versus 1 in 3,600 in all other nonobstetric
faster onset of the spinal block. Obstetric situations that patients.25 Minor neurologic sequelae include headache
may make epidural anesthesia more amenable than spinal (which has been addressed) and backache.
anesthesia include specific cardiac lesions such as mitral
and aortic stenosis and severe preeclampsia.
The most common complication with epidural anes- NEUROLOGIC DEFICITS
thesia is hypotension but, as stated in the preceding text,
the incidence and severity is less than with spinal anesthe- A main concern following neuraxial anesthesia are neuro-
sia due to the nature of the block (Table 48.2). Although logic deficits. Once again, neuraxial blocks are commonly
in theory, PDPH should not occur with epidural anes- blamed for postpartum neurologic injuries; however,
thesia, if the dura matter is penetrated with a typical there are nerve injuries that are common to the obstetric
17-or 18-gauge epidural needle, the incidence of PDPH patient. Palsies of the femoral nerve, lumbosacral trunk,
may be as high as 50% to 80%. The incidence of acci- and more rarely, obturator nerve can occur with com-
dental dural puncture with an epidural needle has been pression of the fetal head at the sacral ala or in the
reported from 0.2% to 3%, depending on the experience pelvis. Moreover, during surgery, retraction can also lead
of the provider. Moreover, the direction of the bevel of to femoral nerve palsy. Also, not commonly associated
spinal needles (diamond-point) and epidural needles con- with cesarean section per se, severe hip flexion during
tribute to the incidence of PDPH. Studies using spinal attempted vaginal delivery can result in femoral nerve
and epidural needles have demonstrated a decreased in- compression. These cases may still require a cesarean
cidence of spinal headaches when the bevel of the needle section if the vaginal delivery attempt failed. Although
is oriented parallel to the longitudinal axis of the verte- lateral femoral cutaneous nerve and common peroneal
bral column, despite the fact that not all dural fibers are nerve palsies usually occur when the patient is in stirrups
longitudinal.23,24 and not during a cesarean section, the lateral femoral
cutaneous nerve can be compressed by the belt used to
secure the patient on the operating room table. Despite
BACKACHE the rarity and potential cause of any neurologic deficit,
diagnostic testing and treatment may be indicated.
Backache is a common problem in the general population
and is obviously exacerbated in pregnancy, secondary
to hormonal changes, that is, relaxin which loosens TRANSIENT NEUROLOGIC
ligaments, and the exaggerated lumbar lordosis from the
gravid uterus. The incidence of back pain in pregnant SYMPTOMS
patients is approximately 40% to 50%, with or without Spinal hyperbaric lidocaine is used for short cases and
regional anesthesia. Unfortunately, in the ASA Closed has been associated with a higher incidence of transient
Claims analysis, back pain made up approximately 10% neurologic symptoms (TNSs) than other local anesthetics.
of the obstetric anesthesia claims (Table 48.3). Although it Symptoms of TNS include pain in the lower extremities
would appear that the larger needle and the larger volume and buttocks, and lasts anywhere from <1 day to as long as
of fluid that are used in epidural anesthesia may render 10 days, but with no evidence of neurologic pathology.26
backaches more common following epidural anesthesia Besides the local anesthetic, the incidence of TNS was
than with spinal anesthesia, backaches after any type of also felt to be due to certain positioning, mainly the
neuraxial blocks are transient and do not have long term or lithotomy position. However, it has been reported in the
serious sequelae. Moreover, some patients with back pain supine position as well, again with a higher incidence with
during pregnancy may also have a radiculopathy, and even hyperbaric lidocaine versus hyperbaric bupivacaine.27
sciatica, that needs to be ascertained and documented However, the incidence among the obstetric patient,
before performing a neuraxial block. especially for cesarean section, appears lower than for the
general surgical population.28 Despite the lower incidence
among obstetric patients, the use of hyperbaric lidocaine
in obstetrics has been questioned.29
What Are the Neurologic
Complications Associated with
Neuraxial Blocks? CAUDA EQUINA SYNDROME
A more severe and possible permanent neurologic deficit
Neurologic complications can be divided into minor and is the cauda equina syndrome, consisting of back pain,
major neurologic sequelae. Unfortunately, in obstetrics, perineal anesthesia, lower extremity sensory and motor
678 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
deficits, bladder, and bowel dysfunction. Cauda equina anesthesia in patients with severe clotting disorders, and
syndrome was reported following accidental spinal ad- in those on certain anticoagulants is discouraged.
ministration in the early 1980s with the old formulation Historically, a neuraxial block in patients with platelet
of 2-chloroprocaine containing the antioxidant, sodium counts <100,000 was discouraged. However, the more re-
bisulfite. More recently, in the early 1990s, the combina- cent consensus is that a platelet count above 75,000 may
tion of a continuous spinal anesthetic with microcatheters be acceptable under certain clinical conditions, barring
(28 and 32 gauge) and hyperbaric lidocaine 5% also re- any clinical signs of bleeding, history of easy bruising or
sulted in a few cases of cauda equina syndrome, which bleeding. An isolated low platelet count, which may be
led the U.S. Food and Drug Administration (FDA) to with- present in up to 8% of all healthy obstetric patients with
draw the microcatheters in 1992. The main cause was no other stigmata, has no predictive value for anesthetic-
postulated to be from pooling of a highly concentrated lo- related sequelae. On the other hand, if the platelet count is
cal anesthetic at the conus of the spinal cord. Fortunately, normal but has acutely fallen, such as in severe preeclamp-
since the mid-1980s, no obstetric cases of cauda equina sia or hemolysis, elevated liver enzymes, and a low platelet
syndrome following neuraxial blocks have been reported. count (HELLP) syndrome, a neuraxial block may not be
indicated, and therefore decisions on neuraxial anesthe-
sia should be individualized, depending on the particular
PROLONGED NEUROLOGIC situation. Unfortunately, there are more obstetric patients
that are being placed on low molecular weight heparin
BLOCK (LMWH) that creates a greater concern for neuraxial
Another occasional scenario following neuraxial anesthe- anesthesia. Neuraxial anesthesia should be delayed for at
sia is a prolonged block. The vast majority of cases simply least 10 to 12 hours after a LMWH dose used for throm-
involve a prolonged duration of the local anesthetic, that boprophylaxis and at least 24 hours for larger, treatment
is, a slow regression of the block, especially in situations doses of LMWH based on the most recent consensus state-
where a cesarean section was preceded by a failed pro- ment from the American Society for Regional Anesthesia
tracted labor and a block that not only required multiple and Pain Medicine.34 Signs and presenting symptoms of
bolus doses, but also if it was lateralized more to one side. a spinal epidural hematoma include new onset numb-
However, despite this more common scenario, a differen- ness, weakness, bowel, and bladder dysfunction, as well
tial diagnosis of a prolonged block includes neurotoxicity as radicular back pain, keeping in mind that severe back
of the local anesthetic, wrong drug administered, trauma pain is not essential for the diagnosis. These symptoms
from either the neuraxial needle or positioning, preexist- can occur within 12 hours of the neuraxial procedure,
ing neurologic disease, and space-occupying lesions such which can make the diagnosis difficult in cases with a
as an epidural hematoma or abscess (see Section Spinal prolonged block. Any suspicion should lead to prompt di-
Epidural Hematoma). Although rare, there have been agnosis through magnetic resonance imaging to perform
case reports of sciatic neuropathy when placing a wedge an emergency, decompressive laminectomy to evacuate
under the right hip for left uterine displacement, con- the hematoma, preferably within 6 hours of any symptoms
to avoid permanent neurologic damage.
sequently causing a left-sided sciatic neuropathy from
compression.
SPINAL INFECTIONS
SPINAL EPIDURAL Spinal infections include meningitis and spinal epidural
HEMATOMA abscess. Fortunately, both meningitis and epidural ab-
scess are rare complications following neuraxial anesthe-
The incidence of epidural hematoma in the obstetric pa- sia in the obstetric patient. Epidural abscess occurs more
tient is rare. The estimated incidence of epidural hematoma commonly through a hematogenous spread from a remote
in the general surgical population is 1 in 150,000 after an infectious site and rarely following neuraxial anesthesia.
epidural anesthetic and 1 in 220,000 after a spinal anes- Moreover, most of the epidural abscess cases occur in un-
thetic.30 In the obstetric patient, a recent review reported healthy patients, such as immunocompromised patients,
only a few cases of epidural hematoma after epidural anes- diabetics, and those with heavy alcohol usage. The onset
thesia, with no cases after spinal anesthesia.31 Moreover, of symptoms of an epidural abscess is more insidious and
most of these obstetric cases had an identifiable coagu- may not manifest until 4 days to 1 week and a half in
lation deficiency. In a retrospective study of more than the postpartum period, in contrast to the more rapid on-
500,000 obstetric cases from the United Kingdom, there set of symptoms from an epidural hematoma. Symptoms
was only 1 case of an epidural hematoma, and in a prospec- include severe lower back pain, fever, and leukocytosis,
tive multicenter series of more than 100,000 obstetric followed by sensory and motor loss of the lower extremi-
cases, no cases of epidural hematoma were reported.32,33 ties. As stated previously, once neurologic symptoms have
Also, in a recent survey from Sweden that differentiated occurred, time is of the essence and magnetic resonance
between obstetric and general surgical patient, the inci- imaging, neurosurgical consultation, and emergency de-
dence of spinal hematoma following epidural anesthesia compressive laminectomy should be performed. On rare
was 1 in 200,000 in the obstetric population versus 1 in situations, antibiotic therapy has been used without the
3,600 in female orthopedic patients.25 Obviously, neuraxial need for surgery.
CHAPTER 48/CESAREAN SECTION 679
meningitis with CSE as was previously reported with spinal With the CSE technique, it is easy to use a small gauge,
anesthesia over 20 years ago. Other complications inher- spinal needle, that is 27-gauge pencil point, as the epidural
ent with the CSE technique include technical difficulty needle itself serves as the ultimate introducer just short of
and failure of the technique, unproven epidural catheter, the dura mater. In contrast, the use of a 27-gauge or smaller
spinal migration of the epidural catheter, and metallic frag- spinal needle for a spinal anesthetic (not for a CSE) can
mentation of the needle with the needle-through-needle be technically more difficult because of the flimsy nature
technique, and increased cost. Like most new techniques, of these needles, especially with tough ligaments and bone
the initial failure rate was reported as high as 10% to 25%. that can easily cause these needles to bend and possibly
The most common failure with the needle-through-needle break. The incidence of PDPH with a CSE technique using
technique is a failed spinal or the inability to obtain CSF. a 27-gauge Whitacre needle is 0.7%.46
The most common cause is the lack of protrusion length
of the spinal needle past the epidural needle, and there-
fore, it is recommended to use manufactured matched
sets that have a protrusion of the spinal needle past the What Minor Complications Can
epidural needle anywhere from 12 to 16 mm. Lateral place-
ment of the epidural needle can lead to a spinal needle Occur with Neuraxial
tangential to the dura mater without piercing it. Despite Anesthesia?
the potential failure rate with the spinal component (or
any of the components) of this technique, the other com-
ponent may still be successful. It is estimated that when
either a spinal or epidural anesthetic is performed for a PAIN DURING CESAREAN
cesarean section, approximately 4% of the time, an alter- SECTION
nate technique is required because of failure. The present
expected failure rate for a CSE is 0.16%, precluding the In the ASA Closed Claims Study, pain during anesthesia
need for general anesthesia.41 In an observational study accounted for 7% (1990s) to 9% (1980s) of all obstetric
from 13 institutions with more than 34,000 procedures for claims; 95% of the cases occurred with cesarean sections,
cesarean section, the failure rate for epidural anesthesia with only 5% related to vaginal deliveries13,14 (Table 48.3).
was 4.3%, spinal anesthesia 2.1%, and CSE 1.6%.36 The All cases occurred with regional anesthesia and none un-
overall regional anesthesia failure rate was 3% and mainly der general anesthesia. One of the inherent problems with
associated with maternal obesity, higher preoperative risk, epidural anesthesia for cesarean section is an inadequate
and block placement later in labor.36 block, especially with an abdominal procedure that re-
Although epidural catheter migration into the sub- quires a high sensory dermatome level along with the
arachnoid space is a theoretical concern with the nature of the procedure, which involves a fair amount
needle-through-needle technique, it is rarely a problem. of surgical manipulation. When comparing epidural ver-
A fresh cadaver study with direct visualization through an sus spinal anesthesia for cesarean section, patients under
epiduroscope failed to produce this problem with repeated epidural anesthesia required more systemic analgesics
attempts using 25-, 26- and 27-gauge spinal needles and and anxiolytics (38% vs. 17%) than those under spinal due
an 18-gauge epidural catheter.42 It was not until five sep- to the inadequacy of the blocks.15 Moreover, in another
arate dural punctures with a 25-gauge spinal needle that study comparing epidural and spinal anesthesia for ce-
the epidural catheter entered the CSF, but only in 5% of sarean section, one third of the patients in the epidural
the attempts. For comparison, after a dural puncture with group and 11% of patients in the spinal group experienced
an 18-gauge epidural needle, the epidural catheter entered pain during the procedure.47 Similarly, when comparing
the CSF in 45% of the attempts under direct vision.41 Al- a CSE versus epidural for cesarean section anesthesia,
though it had been postulated that metallic particles can there was a 22% incidence of pain with epidural anesthe-
be generated with the CSE technique, studies have failed sia compared to none in the CSE group.48 An inadequate
to demonstrate any metal fragments.43,44 block can manifest either as a unilateral block, a bilateral
Another concern with the CSE technique is the inci- block that is not high enough, or an unblocked segment
dence of PDPH with the more widespread use of spinals for or segments. Although the anatomy of the epidural space
the obstetric patient. Studies have failed to demonstrate a may contribute to a nonuniform block secondary to dor-
higher incidence of PDPH with this technique.45 Moreover, sal median folds, connective tissue partitions, and so on,
there may be an increased incidence of accidental dural the main reason for inadequate blocks with epidurals is
puncture with the conventional epidural technique versus overthreading of the epidural catheter and inadequate
the CSE technique. With the occasional uncertainty of the dosing for a particular surgical level. By overthreading an
loss of resistance with an epidural block, the tendency may epidural catheter, it can exit an intervertebral foramen
be to manipulate the needle, that is, advance it a little fur- and thereby cause a unilateral block or a partial bilateral
ther, leading to a dural puncture. On the other hand, with block because part of the anesthetic solution will leak out
the CSE technique, if there is any doubt in the epidural of the vertebral column and nerve root. It is imperative
placement, inserting the spinal needle without any fur- that a solid block extends to at least a T4 sensory der-
ther manipulation of the epidural needle will allow a more matome level (level of the nipples), if not higher. Some
definitive answer, whether CSF is present. However, the patients may require a T1 sensory level during peritoneal
main factor is the type and gauge of the spinal needle. stimulation. Part of the problem may also lie with the
CHAPTER 48/CESAREAN SECTION 681
actual assessment of the block. Loss of sharp pinprick sen- oxytocin, ergonovine, and 15-methyl prostaglandin F2
sation or temperature sensation may not be a useful tool can contribute to nausea and vomiting through different
to access the adequacy of a block; however, loss of touch mechanisms. Minimizing all these factors may ameliorate
sensation may be a better predictor.47,49 As stated earlier, the symptoms, and pretreatment with any antiemetic
most of the cesarean sections today are under spinal, or agent or a combination of agents have resulted in a
even CSE, whereas most of the cesarean sections under decreased incidence of nausea and vomiting compared
epidural anesthesia are a result of a failed vaginal deliv- to providing no pretreatment, but by no means has any
ery, that is dosing a preexisting labor epidural for cesarean particular regimen been very effective.
section, and therefore it behooves the anesthesiologist to
have adequately assessed the labor epidural before dos- Shivering
ing it for an operative procedure and, time permitting,
ensuring an adequate sensory level for the procedure. The incidence of shivering during cesarean section had
been reported to be as high as 50% in the older literature;
however, over the last decade or two, the incidence has de-
MISCELLANEOUS creased with the greater use of spinal anesthesia compared
with epidural anesthesia. As an example, the incidence of
Other forms of discomfort during cesarean section with shivering during cesarean section was 34% with epidu-
neuraxial anesthesia include shivering, nausea and vomit- ral anesthesia compared with 3% with CSE anesthesia48
ing, shoulder pain, and chest discomfort. Pruritus, another (although of unknown etiology with epidural anesthesia).
minor side effect, is seldom present during a cesarean Besides the theoretic compensatory mechanism of shiv-
section; however, it is a common postoperative symptom ering due to the heat loss from cutaneous vasodilatation
resulting from neuraxial opioids, that is morphine. that results from any type of neuraxial block, shivering
resulting from epidural anesthesia is most likely caused
Nausea and Vomiting by the large volume of local anesthetic required for ce-
sarean section. Warming of the local anesthetic solution
There are multiple causes of nausea and vomiting during decreased the incidence, and treatment with a variety of
cesarean section, which include anesthetic-related and drugs has been used with some success, which includes
nonanesthetic-related factors. The incidence of nausea meperidine, diazepam, and clonidine.
and vomiting during cesarean section ranges from as
low as 7% to as high as 42%.50 Obviously, pregnancy Shoulder Pain
in itself is associated with nausea and vomiting due
to hormonal factors, increased progesterone levels, and Shoulder pain, which is referred from the diaphragm, is
decreased motilin levels, which are related to decreased an uncommon symptom. If present, it usually manifests
gastric and small bowel motility, in addition to possibly in the postoperative period, but can occur intraoperatively
contributing to a decreased lower esophageal sphincter and is usually caused from subdiaphragmatic accumula-
tone, with resulting esophageal reflux, and mechanical tion of fluid (i.e., blood, irrigation fluid, or amniotic fluid).
factors, that is, the gravid uterus that may interfere with Shoulder pain may be partially minimized if the patient
normal stomach configuration. The main anesthetic cause is placed head-up to reduce the cephalad accumulation of
of nausea and vomiting, especially immediately after the fluid, which is to be avoided, especially at the time of the
administration of the neuraxial block, is hypotension, uterine incision.
which usually resolves with the correction of the blood
pressure. Theories on the etiology of nausea and vomiting Chest Pain
following a neuraxial block include both central and
peripheral mechanisms that cause cerebral hypoperfusion Chest pain during cesarean section is not an uncommon
and activate the vomiting center in the medulla and scenario and has been associated with other signs and
gastrointestinal hypoperfusion that may activate the symptoms. There are two different scenarios that were
release of serotonin.51 Vagal activity is also accentuated commonly described in the 1980s and early 1990s, and
following a neuraxial block due to the total or near-total included, first, the possibility of venous air, amniotic,
sympathectomy, which can also be a factor in nausea and or thrombotic embolism associated with dyspnea and
vomiting. This vagal activity can be further accentuated precordial Doppler changes and, second, the question of
with the visceral stimulation that occurs in cesarean ischemia with electrocardiogram (ECG) changes, typically
sections, especially with the exteriorization of the uterus. ST-segment depression. Although the type of embolus is
Nausea and vomiting occurring later in the procedure is unknown, it is more likely because of air produced by
usually a manifestation of surgical manipulation, as the the mild pressure gradient between the surgical field and
effects of the neuraxial block have stabilized. Two sets of the heart, especially with exteriorization of the uterus,
pharmacologic agents, including narcotics and uterotonic compounded by the Trendelenburg position and the lax
drugs commonly used in cesarean sections, are also vascularity of the uterus.
contributors to the potentially high incidence of nausea Although the older literature reported a wide range of
and vomiting. Narcotics, whether administered through incidence of venous air embolism during cesarean section,
the neuraxial and/or intravenous route, can be major there have not been any serious sequelae except for one re-
factors of nausea and vomiting. All of the uterotonic drugs, ported case of near-fatal embolism in a patient who made
682 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
a complete recovery.52 Similarly, there had been multiple maternal mortality rates for consecutive 3-year periods,
reports of abnormal ECG changes with ST-segment de- beginning with the 1952 to 1954 term. In their most recent
pression during regional anesthesia for cesarean section. report, 2000 to 2002, cesarean section as a whole had a
Two studies, one utilizing transthoracic two-dimensional 3.7 times relative risk of maternal mortality over vaginal
echocardiograms and another utilizing cardiac enzymes, delivery. Although it would appear intuitive that the ma-
for example, serum creatine kinase (CK) total and isoen- ternal mortality rate from emergency cesarean sections
zyme creatine kinase-myocardial bound (CK-MB) levels, would undoubtedly be greater than vaginal deliveries (rel-
failed to demonstrate any myocardial ischemia associ- ative risk 4.3), surprisingly, the maternal mortality rate
ated with ECG changes and chest pain during cesarean for scheduled and elective cesarean sections had a rel-
section.53,54 Moreover, the type of ECG filtering modality ative risk of 2.8 versus vaginal deliveries (Table 48.1).
used in some of these cases may have falsely promoted Similarly, in the United States, the Pregnancy Mortality
greater degrees of ST-segment changes.55 Although my- Surveillance System at the Center for Disease Control
ocardial ischemia is an unlikely scenario in this setting, has reported on maternal mortality from 1987 to 1997.
the etiology of chest pain during cesarean section re- Eighty-two percent of anesthetic-related deaths occurred
mains unknown. However, in rare cases, these signs and during or shortly after cesarean delivery.10 Compounding
symptoms may represent a cardiovascular event, such as the higher mortality rate with cesarean delivery, the ce-
a myocardial infarction or aortic dissection, as cardiac sarean section rate continues to climb and reached 27.6%
disease was the second most common cause of indirect nationally in 2003.3
deaths in the United Kingdom from 2000 to 2002.2 Anesthesia-related deaths rank seventh among the
In the ASA Closed Claims Project, there has been a causes of maternal mortality in the United States and
decrease in more severe injury claims and an increase United Kingdom, and account for 1.1 to 3.0 deaths
in more minor claims in the 1990s as compared to the per million births, respectively.2,57 In contrast to the
1970s and 1980s, which is attributable to the decreased more comprehensive data from CEMACH, the Pregnancy
use of general anesthesia and increased use of regional Mortality Surveillance System is more limited because
anesthesia13,14 (Table 48.3). of the paucity of data with vital statistics when dealing
with specific causes of death, type of anesthesia, and so
on, for many cases. Despite these limitations, maternal
mortality from general anesthesia appears to be higher
than from regional anesthesia, with the latest relative risk
Why Has the Use of General from the 1991 to 1996 time period of 6.7.11 The number
Anesthesia for Cesarean of deaths from regional anesthesia fell in the mid-1980s,
coincidentally with the removal of 0.75% bupivacaine
Sections Declined? from the obstetric suites, whereas the number of deaths
from general anesthesia remained steady. Although the
The changing practice of obstetric anesthesia has demon- number of direct deaths due to anesthesia have markedly
strated a marked decrease in the percentage of cesarean decreased since 1964 as documented by the CEMACH
sections that are performed under general anesthesia. report, all of the maternal deaths in the 2000 to 2002 report
Even in a large tertiary care hospital with 8,000 to 10,000 occurred as a result of general anesthesia (see Table 48.5).
deliveries per year, general anesthesia was used in only
7.2% of all cesarean sections in 1990 and further decreased Indications for General Anesthesia
to 3.6% in 1995.12 Similarly, in a study from a tertiary care
center in England, the incidence of general anesthesia for Despite the advantages of neuraxial anesthesia for ce-
cesarean section was 76% in 1982 and down to 7.7% in sarean section, there are certain obstetric conditions that
1998.56 In the most recent observational study conducted warrant a general anesthetic (see Table 48.6):
between 1999 and 2002 as part of the Maternal-Fetal
Active hemorrhage from placenta previa
Medicine Units Network that studied more than 34,000
Active hemorrhage from placenta abruption
singleton cesarean deliveries from 13 institutions, 93%
Uterine rupture in a hemodynamically unstable patient
of cases were performed under regional anesthesia and
Persistent fetal bradycardia that does not respond to
7% under general anesthesia36 (Table 48.4). As stated ear-
treatment that may include a prolapsed umbilical cord
lier, the decline of general anesthesia for cesarean section
(see subsequent text for treatment). Conditions that may
can be attributed to the increased incidence of neuraxial
preclude regional anesthesia include the following:
analgesia for laboring patients that can be converted to
Coagulopathy
surgical anesthesia for cesarean delivery, and the higher
Patients on anticoagulants such as LMWH and
incidence of maternal mortality from complications of
Patient refusal
general anesthesia.10,11
The type of delivery in itself has a major impact on the Softer indications for general anesthesia may include
maternal mortality rate. The most extensive documenta- preexisting neurologic disease, cardiac disease (lesion
tion on maternal mortality stems from the confidential en- dependent), previous back surgery with hardware, and
quiries in the United Kingdom (Confidential Enquiry into risk of intraoperative hemorrhage. Cardiac conditions that
Maternal and Child Health [CEMACH]). The reporting may be more amenable to general anesthesia, thereby
began as early as 1928 and was revised in 1952, reporting avoiding afterload reduction from regional anesthesia,
CHAPTER 48/CESAREAN SECTION 683
TABLE 48.5 Estimated Cesarean Sections and Anesthesia-Related Deaths: Comparison of the
1964 1966, 1982 1984, and 2000 2002 Trienniums
Cesarean Section
Note the dramatic increase in cesarean sections from the 1960s to the 2000s and the decrease in anesthetic-
related deaths in the same time period.
a All seven anesthesia-related deaths are attributable to general anesthesia.
Adapted, with data from: Why mothers die. The sixth report of the confidential enquiries into maternal deaths
in the United Kingdom, 20002002.2 Available at: http://www.cemach.org.uk. Accessed November 6, 2006.
include severe aortic or mitral stenosis, right-to-left and, unfortunately, these figures have been consistent
shunts, and asymmetric septal hypertrophy. over the last 20 years with no signs of improvement5860
(see Table 48.7). Factors that contribute to the difficult
obstetric airway include airway edema, obesity, increased
DIFFICULT AIRWAY anterior-posterior chest diameter, breast enlargement,
urgency of the procedure, and possible misapplication of
In the ASA Closed Claims Project, critical events involving
cricoid pressure. Moreover, the lack of nearby personnel
the respiratory system were the leading cause of death
in obstetric and nonobstetric claims.13 However, in the for assistance in emergency situations is a concern in
nonobstetric claims, the leading cause of death was due labor and delivery suites that may be isolated compared to
to a more generalized etiology, such as inadequate ven- the main operating room where extra assistance is usually
tilation, in contrast to the obstetric claims in which the more readily available. Unfortunately, with the decreasing
deaths were attributed to more identifiable mechanisms, incidence of general anesthesia for cesarean section,
such as pulmonary aspiration and airway-related compli- the airway concern is compounded by the potential
cations similar to the reports from the United Kingdom lack of experience with intubations, which obviously
and the United States.13 holds true for staff and trainees. In a British study,
The most feared complication in obstetric anesthesia with a dramatic decrease in cesarean sections under
is the inability to intubate the trachea. The incidence general anesthesia from a high of 76% in 1982 to a
of failed intubations in obstetrics may be as high as low of 7.7% in 1998, the average number of cesarean
eightfold compared to the general surgical population sections under general anesthesia per resident during their
a Coagulation status dependent on laboratory values, such as activated thromboplastin time, prothrombin time,
platelets, fibrinogen, thromboelastogram, etc. For low molecular weight heparin (LMWH), see text.
ATP, autoimmune thrombocytopenia purpura; TTP, thrombotic thrombocytopenia purpura; HELLP, hemolysis,
elevated liver enzymes and a low platelet count; DVT, deep venous thrombosis; PE, pulmonary embolus; LMWH,
low molecular weight heparin.
684 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Note: Obstetric failed intubations are over sevenfold greater than the general surgical population.
a Lyons G. Failed intubation. Anaesthesia. 1985;40:759.
b Lyons G, Macdonald R. Difficult intubation in obstetrics [Letter]. Anaesthesia. 1985;41:1016.
c Samsoon GLT, Young JRB. Difficult tracheal intubation: A retrospective study. Anaesthesia. 1987;47:487.
d Hawthorne L, Wilson R, Lyons G, et al. Failed intubation revisited: 17-yr experience in a teaching maternity unit.
Br J Anaesth. 1996;76:680.
e Bernardo PD, Jenkins JG. Failed tracheal intubation in obstetrics: A 6-year review in a UK region.
Anaesthesia. 2000;55:685.
f Rahman K, Jenkins JG. Failed tracheal intubation in obstetrics: No more frequent but still managed badly.
Anaesthesia. 2005;60:168.
Modified and updated from: James CF. Maternal mortality. Sem Anesth. 192;11:76.
training was 18 in 1982 and down to only 4 in 1998.56 voice prompts. It demonstrated a fair number of omissions
In a more recent study of more than 34,000 cesarean and postulated that a verbal or written checklist could im-
deliveries from 1999 to 2002 in the United States, the only prove patient safety for cesarean delivery.63 Another major
anesthetic-related maternal death was the result of a failed concern with performance is the fatigue factor; in one re-
intubation.36 In the CEMACH maternal mortality report port, 70% of failed intubation during cesarean sections
from the United Kingdom from 2000 to 2002, all of the occurred between 9:00 PM and 8:00 AM, which only made
seven direct deaths from anesthesia were under general up 12% of the cases in a 24-hour period.64
anesthesia and included the inability to intubate the
trachea, pulmonary aspiration, and anaphylaxis.2 With
this declining trend and higher mortality with general Preoperative and Intraoperative Concerns
anesthesia in obstetrics, the question has been posed: Are
anesthesiologists prepared for such cases?.61 Preoperative evaluation of the airway may reveal factors
that correlate with a difficult intubation: Poor visual-
ization of oropharyngeal structures with the patient in
Practice Implications the sitting position, short neck, protruding maxillary in-
Owing to the lack of emergent or catastrophic events in cisors, and a receding mandible.65 Moreover, as opposed
many centers, and thereby the lack of general anesthet- to the nonobstetric patient, an airway evaluation in the
ics for cesarean deliveries, there are multiple ways to obstetric patient on admission may change during the
prepare for these rare situations. Some institutions, espe- course of labor, suggesting that airway reevaluation be
cially large tertiary care centers have developed team performed before a general anesthetic induction.66 To
training, which includes a multidisciplinary approach overcome some of these difficulties, proper patient posi-
involving obstetricians, anesthesiologists, neonatologists, tioning, use of smaller diameter endotracheal tubes, and
and nursing staff for monitoring and evaluating mutual short laryngoscope handles should be used. A difficult
performance through debriefings and other measures and airway cart should be present in all obstetric suites.67
emphasizing communication among the various clinical Although in some countries, the laryngeal mask airway
services, which can be accomplished by multidisciplinary has been used in elective cesarean sections, it does not
rounds.62 The lack of multidisciplinary cooperation was protect the airway from regurgitated material, and there-
a contributing factor in directly-related anesthetic deaths fore, is not recommended for that purpose.68,69 However,
as reported by the latest CEMACH report into maternal the laryngeal mask airway and its variants are invaluable
deaths during the 2000 to 2002 period.2 As far as specific in difficult airway cases because it not only serves as a
anesthetic situations such as an emergent general anes- means of oxygenation and ventilation but also as a con-
thetic, simulation-based training, presently with limited duit for endotracheal intubation.70,71 If a patients airway
availability, could become an invaluable tool as seen in is suspect during labor, the anesthesiologist needs to in-
other industries, mainly the aviation industry. On that form the obstetrician that if an emergent cesarean section
note, a recent pilot study from Australia prepared an is required, an awake intubation may be indicated which
aviation-style checklist system for cesarean sections under usually cannot be done in an expeditious manner. Simi-
general anesthesia using an electronic verbal system with larly, in a morbidly obese patient, a general or regional
CHAPTER 48/CESAREAN SECTION 685
a
Warner, MA, Warner ME, Weber JG. Clinical significance of pulmonary aspiration during the perioperative period Anesthesiology. 1993;78:56.
b
Mendelson CL. The aspiration of stomach contents into the lungs during obstetric anesthesia. Am J Obstet Gynecol. 1946;52:191.
c
Olsson GI, Halle B, Hambraeus-Jonzon K. Aspiration during anaesthesia: A computer-aided study of 185,358 anaesthetics. Acta Anaesth
Scand. 1986;30:84.
d
Soreide E, Bjornestad E, Steen PA. An audit of perioperative aspiration pneumonitis in gynaecological and obstetetric patients. Acta
Anaesthesiol Scand. 1996;40:14.
Modified and updated from James CF. Maternal mortality. Sem Anesth. 1992;11:76.
anesthetic cannot be performed expeditiously. A preex- and mechanical changes, resulting in an increase in in-
isting labor epidural would be the most efficacious way tragastric pressure, decreased lower esophageal sphincter
to proceed, assuming that the epidural has been func- tone, and delayed gastric emptying. Moreover, this prob-
tioning properly and is still in place. It is not uncommon lem is compounded by the urgency of the procedure and
to have these epidurals replaced during the course of the potential difficult airway. Compounding this problem
labor.72 Maternal obesity was a main contributor to failed has been the consumer push for more independent deci-
regional anesthesia for cesarean section.36 To compound sion making, such as birth plans and diet during labor.
this problem, the incidence of cesarean section in the However, the ASA guidelines on obstetric anesthesia do
morbidly obese patients may approach 50% to 60%, and not support the intake of solid foods in labor, but do
the incidence of a difficult tracheal intubation may be as support modest amounts of clear liquids, but only in pa-
high as 25%.72 Along with the difficult airway in obstet- tients without, but not limited to, any of the following
rics, a rapid-sequence induction for a general anesthetic is risk factors such as diabetes, obesity, difficult airway, or
required for a full stomach, resulting in a period of apnea nonreassuring fetal heart rate tracing.67 Even despite the
which is compounded by the limited oxygen reserve in NPO status of laboring patients, in an ultrasonographic
the obstetric patient due to a decreased functional resid- study among term laboring patients, 66% of patients who
ual capacity, along with increased oxygen consumption, had been NPO for 12 to 24 hours had documented partic-
leading to a potentially more rapid and extensive hypoxic ulate matter in their stomachs.77 Besides their effects in
state than the nonobstetric patient.73 An update on the pregnancy, the use of parenteral and neuraxial narcotics
practice guidelines for management of the difficult airway may further impair gastric emptying and intestinal motil-
was published by the ASA in 2003.74 ity. Therefore, all laboring patients should be assumed to
have a full stomach despite their NPO status. The practice
guidelines for preoperative fasting and the use of pharma-
PULMONARY ASPIRATION cologic agents to reduce the risk of pulmonary aspiration
by the ASA pertain to healthy patients undergoing elective
Pulmonary aspiration of gastric contents during general procedures and not to laboring patients.78
anesthesia for cesarean section is the other main con- In nonlaboring patients undergoing elective cesarean
tributing factor to anesthesia-related maternal mortality section, the NPO guidelines should follow the hospital
and is caused by a difficult airway. In 1946, Mendel- anesthetic guidelines and should be at least 6 hours after
son described pulmonary aspiration in obstetric patients, a light meal.67 To demonstrate the difference between
mainly occurring during vaginal delivery and under ether the obstetric and general surgical patient, in elective
anesthesia, with an incidence of 15 per 10,000 cases.75 cesarean section patients who were given tea and toast
Coincidentally, 40 to 50 years later, an extensive re- within 4 hours of surgery had double the gastric volume
port from Sweden and a smaller but more recent one than patients who fasted for 8 hours.79 Pulmonary
from Norway reported that the incidence of pulmonary aspiration prophylaxis with antacids implies prophylaxis
aspiration among cesarean section patients may be up for aspiration pneumonia should aspiration occur, as
to sixfold higher than the general surgical population76 any pharmacologic measure does not prevent aspiration,
(see Table 48.8). The increased incidence of pulmonary per se. Although the incidence of pulmonary aspiration
aspiration in the obstetric patient is a product of hormonal is low and preoperative antacids are not recommended
686 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
transverse uterine incisions in todays obstetric practice general anesthesia versus regional anesthesia.87,88 In a re-
may compound these difficult deliveries and require cent European study, the authors compared their practice
anesthetic assistance for uterine relaxation, such as of the extensive use of general anesthesia for emergency
-agonists, that is, terbutaline or nitroglycerine; in cases cesarean section in 1991 (78%), even with indwelling
under general anesthesia, increased concentrations of the epidural catheters from labor, versus the limited use of
volatile inhalation agents up to 2-MAC values may be general anesthesia in 1997 (12%) utilizing the indwelling
required. catheters for the emergency cesarean delivery; despite a
modest increase in the elapsed time of 13 minutes with
epidural anesthesia versus 8.3 minutes with general anes-
thesia, there was no significant difference in neonatal
What Are Other Obstetric outcome.89
Complications Related to Other surgical concerns with anesthesia implications
during cesarean section include the use of a classical
Cesarean Sections? uterine incision, which usually translates in a greater
intraoperative blood loss compared to low transverse
Although most cesarean hysterectomies are emergency uterine incisions. The classical uterine incision, although
procedures, they can also be nonemergent such as the uncommon, is used for difficult deliveries, such as
ones associated with cervical cancer or previous cesarean multiple gestation, abnormal lies, fetal anomalies, and
sections with large leiomyomas, hydatiform moles, or in- placentas and leiomyomas that may interfere with a low
tractable chronic dysfunctional bleeding. However, most uterine segment incision. Other surgical complications
cesarean hysterectomies are emergent and associated with during cesarean section may include bladder injury and,
uncontrollable hemorrhage, most typically related to pla- more rarely, bowel and ureteral injury. These injuries are
centa accreta, percreta or increta or intractable uterine more common from extensive adhesions after multiple
atony, and less commonly because of infection or uterine previous cesarean sections, other previous abdominal
rupture not amenable to surgical repair. As stated earlier, procedures, or following other treatment modalities such
with the risk of uterine rupture and resulting morbidity as radiation therapy. Although these conditions may result
and possible mortality to the fetus and mother, VBACs are in a greater intraoperative blood loss, the anesthetic
not as commonly attempted today. The incidence of uter- implications in these cases are mainly related to the
ine rupture during attempted VBAC is usually under 1%; prolonged duration of the procedure, which is better
however, under certain circumstances, such as a previous provided by epidural and CSE anesthesia or general
classical or T-shaped uterine incision, and a short interval anesthesia as opposed to single-shot spinal anesthesia.
between deliveries, the uterine rupture rate can be up to
threefold greater and has been reported as high as 9%.85
The usual presenting signs and symptoms of uterine rup-
ture include nonreassuring fetal heart rate pattern with KEY POINTS
decelerations or fetal bradycardia and, less commonly,
abdominal pain, vaginal bleeding, and hypovolemia. The 1. The main changes in obstetric anesthesia practice
present American College of Obstetricians and Gynecol- today include the increased incidence in cesarean
ogists (ACOG) recommendations for attempted VBAC section deliveries, greater use of labor neuraxial
include only one previous cesarean section with a low analgesia, and a lower incidence of general anes-
transverse incision, performed in an institution equipped thesia.
to respond to emergencies with physicians (obstetricians 2. The most common complication of neuraxial anes-
and anesthesiologists) immediately available to provide thesia is maternal hypotension.
emergency care. It is this immediate availability that has 3. Epidural anesthesia for cesarean section today is
rendered VBACs a logistic concern among physicians and usually performed from a preexisting labor epidural,
has placed most of these cases in tertiary care centers and but it has a higher incidence of failure, slower onset
away from small community and rural hospitals. of block, and pain during surgery compared with
A prior cesarean section in a patient with a placenta spinal anesthesia.
previa should be considered at high risk for developing 4. Epidural anesthesia has a greater potential for
placenta accreta, with the risk approaching 25% with local anesthetic toxicity (accidental intravascular
a previous cesarean section and more than 60% with injection) and near-total to total spinal anesthesia
four or more previous cesarean deliveries.86 Traditionally, (accidental intrathecal injection) due to the larger
whether planned or unplanned, the tendency has been dose requirements.
to perform cesarean hysterectomies under general anes- 5. Spinal anesthesia is the most common anesthetic
thesia due to the potential hemodynamic instability with for cesarean section in present day practice due
intractable blood loss commonly seen in these procedures. to its rapid onset and denser block with fewer
However, in multiple studies, although there were more failed blocks; however, the incidence and degree
cases under general anesthesia, the cases performed under of hypotension is greater than epidural anesthesia.
regional anesthesia did not require conversion to general 6. CSE anesthesia is more commonly used today for
anesthesia, but more importantly, intraoperative blood labor analgesia. It is also used for specific situations
loss and transfusion requirements were greater under in cesarean deliveries, such as anticipated longer
688 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
cases, and for certain cardiac patients in whom the 4. Black C, Kaye JA, Jick H. Cesarean delivery in the United
Kingdom: Time trends in the general practice research
block can be titrated to minimize hemodynamic
database. Obstet Gynecol. 2005;106:151.
changes. Complications from CSE are similar to
5. Liu S, Rusen ID, Joseph KS, et al. Recent trends in caesarean
those from spinal and epidural anesthesia. delivery rates and indications for caesarean delivery in
7. Serious complications from neuraxial block for ce- Canada. J Obstet Gynaecol Can. 2004;26:735.
sarean section are similar to any other surgical 6. Meikle SF, Steiner CA, Zhang J, et al. A national estimate of
procedure, but less common than in the general the elective primary cesarean delivery rate. Obstet Gynecol.
surgical population. These include peripheral neuro- 2005;105:751.
logic deficits, spinal epidural hematoma, meningitis, 7. Kalish RB, McCullough L, Gupta M, et al. Intrapartum
epidural abscess, seizures, and cardiopulmonary col- elective cesarean delivery: A previously unrecognized clinical
lapse. entity. Obstet Gynecol. 2004;103:1136.
8. Specific minor complications during cesarean 8. Hawkins JL, Gibbs CP, Orleans M, et al. Obstetric anes-
sections under neuraxial anesthesia are more com- thesia work force survey, 1981 versus 1992. Anesthesiology.
1997;87:135.
mon than the general surgical population. They
9. Stamer UM, Wiese R, Stuber F, et al. Change in anaesthetic
include pain or discomfort, nausea and vomiting,
practice for cesarean section in Germany. Acta Anaesthesiol
shivering, and shoulder and chest pain. Scand. 2005;49:170.
9. Approximately 60% of the closed claims in obstet- 10. Hawkins JL, Koonin LM, Palmer SK, et al. Anesthesia-
ric anesthesia have involved cesarean deliveries. related deaths during obstetric delivery in the United States,
However, most complications following cesarean 19791990. Anesthesiology. 1997;84:277.
sections under neuraxial anesthesia are minor 11. Hawkins JL, Chang J, Callaghan W, et al. Anesthesia-related
(except maternal nerve injury) and not life threat- maternal mortality in the United States, 1991-6. An update
ening such as PDPH, failed block, back pain, and [Abstract]. Anesthesiology. 2002;97:A1046.
emotional distress. 12. Tsen LC, Pitner R, Camann WR. General anesthesia for
10. General anesthesia for cesarean section accounts for cesarean section at a tertiary care hospital 19901995:
most of the maternal morbidity and mortality result- Indications and implications. Int J Obstet Anesth. 1998;7:147.
13. Chadwick HS. An analysis of obstetric anesthesia case from
ing mainly from airway mishaps and/or pulmonary
the ASA closed claims project database. Int J Obstet Anesth.
aspiration of gastric contents.
1996;5:258.
11. The difficult airway in obstetrics occurs from a com- 14. Davies JM. Obstetric anesthesia closed claims-trends over
bination of factors such as, airway edema, obesity, last three decades. ASA Newsl. 2004;68:12.
increased anterior-posterior chest diameter, breast 15. Riley ET, Cohen SE, Macario A, et al. Spinal versus epidural
enlargement, urgency of the procedure, and possi- for cesarean section: A comparison of time, efficiency, costs,
ble misapplication of cricoid pressure. Oxygenation charges and complications. Anesth Analg. 1995;89:709.
is further hampered by the decrease in functional 16. Kayacan N, Arici G, Karsli B, et al. Acute subdural
residual capacity and increased oxygen consump- haematoma after accidental dural puncture during epidural
tion. anaesthesia. Int J Obstet Anesth. 2004;13:47.
12. Although awareness and recall are not as prevalent 17. Rout CC, Rocke DA, Levin J, et al. A reevaluation of the
in the present day practice of obstetric anesthesia, role of crystalloid preload in the prevention of hypotension
associated with spinal anesthesia for elective cesarean
it is still a potential complication especially when
section. Anesthesiology. 1993;79:262.
practitioners use lower anesthetic doses (induction
18. Jackson R, Reid JS, Thorburn J. Volume preloading is not
and maintenance) before delivery of the fetus for essential to prevent spinal-induced hypotension at cesarean
fear of fetal effects. section. Br J Anaesth. 1995;75:262.
13. The anesthesiologist needs to be prepared for 19. Park GE, Hauch MS, Curlin F, et al. The effects of varying vol-
specific situations that can lead to major hemorrhage umes of crystalloid administration before cesarean delivery
such as a uterine rupture from a VBAC attempt or a on maternal hemodynamics and colloid osmotic pressure.
placenta accreta in a patient with a placenta previa Anesth Analg. 1996;83:299.
who has a history of previous cesarean sections that 20. Ueyama H, Yan-Ling H, Tanigami H, et al. Effects of
can lead to an emergency cesarean hysterectomy. crystalloid and colloid preload on blood volume in the
The type of anesthesia, neuraxial versus general, is parturient undergoing spinal anesthesia for elective cesarean
not as important as the execution of resuscitative section. Anesthesiology. 1999;91:1571.
21. Lee A, Warwick D, Kee N, et al. A quantitative, system-
measures.
atic review of randomized controlled trials of ephedrine
versus phenylephrine for the management of hypotension
REFERENCES during spinal anesthesia for cesarean delivery. Anesth Analg.
1. Trolle D. The history of cesarean section. Copenhagen, 2002;94:920.
Denmark: Reizel Booksellers; 1982. 22. Milne M, Lawson JIM. Epidural analgesia for cesarean
2. Lewis G, on behalf of CEMACH Editorial Board. Why section: A review of 182 cases. Br J Anaesth. 1978;45:
mothers die 20002002. Report on confidential enquiries into 1206.
maternal deaths in the United Kingdom. November 12, 2004. 23. Mihic DN. Post-spinal headache and relationship of needle
Available at: http://www.cemach.org.uk/. Accessed: July 27, bevel to longitudinal dural fibers. Reg Anesth. 1985;10:76.
2006. 24. Norris MC, Leighton Bl, Desimone CA. Needle bevel direction
3. Hamilton BE, Martin JA, Ventura SJ, et al. Births: Prelimi- and headache after inadvertent dural puncture. Anesthesiol-
nary data for 2004. Natl Vital Stat Rep. 2005;54:1. ogy. 1989;70:729.
CHAPTER 48/CESAREAN SECTION 689
25. Moen V, Dahlgren N, Irestedt L. Severe neurologic com- spinal-epidural analgesia in labor. Int J Obstet Anesth.
plications after central neuraxial blockades in Sweden 2001;10:168.
19901999. Anesthesiology. 2004;101:950. 47. Russell IF. Levels of anaesthesia and intraoperative pain at
26. Pollock JE, Neal JM, Stephenson CA, et al. Prospective study cesarean section under regional block. Int J Anesth. 1995;
of the incidence of transient radicular irritation in patients 4:71.
undergoing spinal anesthesia. Anesthesiology. 1996;84:1361. 48. Choi DH, Kim JA, Chung IS. Comparison of combined spinal
27. Keld DB, Hein L, Dalgaard M, et al. The incidence of epidural anesthesia and epidural anesthesia for cesarean
transient neurologic symptoms (TNS) after spinal anesthesia section. Acta Anaesthesiol Scand. 2000;44:214.
in patients undergoing surgery in the supine position. 49. Russell IF. The futility of using sharp pinprick (or cold) to
Hyperbaric lidocaine 5% versus hyperbaric bupivacaine assess spinal or epidural anesthesia for cesarean delivery.
0.5%. Acta Anaesthesiol Scand. 2000;44:285. [Letter]. Reg Anesth Pain Med. 2001;26:386.
28. Aouad MT, Siddik SS, Jalbour MI, et al. Does pregnancy 50. Carpenter RL, Caplan PA, Brown DL. Incidence and risk
protect against intrathecal lidocaine-induced transient neu- factors for side effects of spinal anesthesia. Anesthesiology.
rologic symptoms? Anesth Analg. 2001;92:401. 1992;76:906.
29. Schneider MC, Birnbach DJ. Lidocaine neurotoxicity in the 51. Borgeat A, Ekatodramis G, Schenker CA. Postoperative
obstetric patient: Is the water safe? Anesh Analg. 2001;92:287. nausea and vomiting in regional anesthesia. Anesthesiology.
30. Horlocker TT, Wedel DJ. Neuraxial block and low- 2003;98:530.
molecular-weight heparin: Balancing perioperative analge- 52. Epps SN, Robbins AJ, Marx GF. Complete recovery after
sia and thromboprophylaxis. Reg Anesth Pain Med. 1998; near-fatal venous air embolism during cesarean section. Int
23(Supp 2):164. J Obstet Anesth. 1998;7:131.
31. Aramovitz S, Beilin Y. Thrombocytopenia, low molecular 53. Zakowski MI, Ramanathan S, Baratta JB, et al. Electro-
weight heparin, and obstetric anesthesia. Anesthesiol Clin cardiographic changes during cesarean section: A cause for
North America. 2003;21:99. concern? Anesth Analg. 1993;76:162.
32. Scott DB, Hibbard BM. Serious non-fatal complications 54. Ross RM, Baker T. Cardiac enzymes in patients undergoing
associated with extradural block in obstetric practice. Br cesarean section. Can J Anaesth. 1995;42:46.
J Anaesth. 1990;64:537. 55. Camann W, Trunflo GV, Kluger R, et al. Automated ST-
33. Scott DB, Tunstall ME. Serious complications associated segment analysis during cesarean delivery: Effects of ECG
with epidural/spinal blockade in obstetrics: A two-year filtering modality. J Clin Anesth. 1996;8:564.
prospective study. Int J Obstet Anesth. 1995;4:133. 56. Johnson RV, Lyons GR, Wilson RC, et al. Training in
34. Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia obstetric general anaesthesia: A vanishing art? Anaesthesia.
in the anticoagulated patient: Defining the risks (The Second 2000;55:179.
ASRA Consensus Conference on Neuraxial Anesthesia and 57. Hawkins JL. Anesthesia-related maternal mortality. Clin
Anticoagulation). Reg Anesth Pain Med. 2003;28:172. Obstet Gynecol. 2003;46:679.
35. Loo CC, Dahlgren G, Irestedt L. Neurologic complications in 58. Lyons G, MacDonald R. Difficult intubation in obstetrics.
obstetric regional anaesthesia. Int J Obstet Anesth. 2000;9:99. Anaesthesia. 1985;40:1016.
36. Bloom SF, Spong CY, Weiner SJ, et al. Complications of anes- 59. Samsoon GLT, Young JRB. Difficult tracheal intubation:
thesia for cesarean delivery. Obstet Gynecol. 2005;106:281. A retrospective study. Anaesthesia. 1987;42:487.
37. Carp H, Bailey S. The association between meningitis and 60. Rahman K, Jenkins JG. Failed tracheal intubation in
dural puncture in bacteremic rats. Anesthesiology. 1992;76: obstetrics: No more frequent but still managed badly.
667. Anaesthesia. 2005;60:168.
38. Kinirons B, Mimoz O, Lafendi L, et al. Chlorhexidine versus 61. Hawkins JL, Gibbs CP. General anesthesia for cesarean
povidine iodine in preventing colonization of continuous section: Are we really prepared? Ed Int J Obstet Anesth. 1998;
epidural catheters in children: A randomized, controlled 7:145.
trial. Anesthesiology. 2001;94:239. 62. Sachs BP. A 38 year-old woman with fetal loss and
39. Birnbach DJ, Meadows W, Stein DJ, et al. Comparison of hysterectomy. J Am Med Assoc. 2005;294:833.
povidine iodine and DuraPrep, an iodophor-in-isopropyl al- 63. Hart EM, Owen H. Errors and omissions in anesthesia: A
cohol solution for skin disinfection prior to epidural catheter pilot study using a pilots checklist. Anesth Analg. 2005;101:
insertion in parturients. Anesthesiology. 2003;98:164. 246.
40. Rawal N. Single segment combined spinal epidural block for 64. Hawthorne L, Wilson R, Lyons G, et al. Failed intubation
cesarean section. Can Anaesth Soc J. 1986;33:254. revisited: 17-yr experience in a teaching maternity unit. Br J
41. Lyons G. Epidural is outmoded form of regional anaesthesia Anaesth. 1996;76:680.
for elective cesarean section. Int J Obstet Anaesth. 1995;4:34. 65. Rocke DA, Murray WB, Rout CC, et al. Relative risk analysis
42. Holmstrom B, Rawal N, Axelsson K, et al. Risk of catheter mi- of factors associated with difficult intubation in obstetric
gration during combined spinal epidural block-percutaneous anesthesia. Anesthesiology. 1992;77:67.
epiduroscopy study. Anesth Analg. 1995;80:747. 66. Farcon EL, Kim MH, Marx GF. Changing mallampati score
43. Herman N, Molin J, Knape KG. No additional metal par- during labor. Can J Anaesth. 1994;41:50.
ticle formation using the needle-through-needle combined 67. American Society of Anesthesiologists. Practice guidelines
epidural/spinal technique. Acta Anesthesiol Scand. 1996;40: for obstetrical anesthesia. A report by the American Society
227. of Anesthesiologists Task Force on Obstetrical Anesthesia.
44. Holst D, Mollmann M, Schymroszcyk B, et al. No risk Anesthesiology. 1999;90:600.
of metal toxicity in combined spinal-epidural anesthesia. 68. Han TH, Brimacombe J, Lee EJ, et al. The laryngeal mask
Anesth Analg. 1999;88:393. airway is effective (and probably safe) in selected healthy
45. Norris MC, Griece WM, Borkowski M, et al. Complications parturients for elective cesarean section: A prospective study
of labor analgesia: Epidural versus combined spinal epidural of 1067 cases. Can J Anaesth. 2001;48:1117.
techniques. Anesth Analg. 1994;79:529. 69. Barker P, Langton JS, Murphy PJ, et al. Regurgitating
46. Landau R, Ciliberto CF, Goodman SR, et al. Complications of gastric contents during general anesthesia using the
with 25- and 27- gauge Whitacre needles during combined laryngeal mask airway. Br J Anaesth. 1992;69:314.
690 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
70. Hasham FM, Andrews PJD, Juneja MM, et al. The laryngeal 79. Lewis M, Crawford JS. Can one risk fasting the obstetric
mask airway facilitates intubation at cesarean section: A case patient for less than 4 hours? Br J Anaesth. 1987;59:312.
report of difficult intubation. Int J Obstet Anesth. 1993;2:181. 80. Moir DD. Anaesthesia for cesarean section: An evaluation
71. Godley M, Reddy AR. Use of LMA for awake intu- of a method using low concentration of halothane and 50%
bation for caesarean section. Can J Anaesth. 1996;43: oxygen. Br J Anaesth. 1970;42:136.
299. 81. King H, Ashley S, Braithwaite D, et al. Adequacy of general
72. Hood DD, Dewan DM. Anesthetic and obstetric outcome in anesthesia for cesarean section. Anesth Analg. 1993;77:84.
morbidly obese parturients. Anesthesiology. 1993;79:1210. 82. Warren TM, Datta S, Ostheimer GW, et al. Comparison of the
73. Baraka AS, Hanna MT, Jabbour SI, et al. Preoxygenation of maternal and neonatal effects of halothane, enflurane and
pregnant and non-pregnant women in the head up versus isoflurane for cesarean delivery. Anesth Analg. 1983;62:516.
the supine position. Anesth Analg. 1992;75:757. 83. Crawford JS, Burton M, Davies P. Anaesthesia for cesarean
74. American Society of Anesthesiologists. Practice guidelines section: Further refinements of a technique. Br J Anaesth.
for management of the difficult airway: An updated report 1973;45:726.
by the ASA Task Force on management of the difficult 84. Datta S, Ostheimer GW, Weiss JB, et al. Neonatal effects of
airway. Anesthesiology. 2003;98:1269. prolonged anesthetic induction for cesarean section. Obstet
75. Mendelson CL. The aspiration of stomach contents into Gynecol. 1981;58:331.
the lungs during obstetric anesthesia. Am J Obstet Gynecol. 85. ACOG practice bulletin no. 54, vaginal birth after previous
1946;52:191. cesarean section. Obstet Gynecol. 2004;104:203.
76. Olsson GI, Halle B, Hambraeus-Jonzon K. Aspiration 86. Clark SL, Koonings PP, Phelan JP. Placenta previa/accreta
during anaesthesia: A computer-aided study of 185,358 and prior cesarean section. Obstet Gynecol. 1985;66:89.
anaesthetics. Acta Anaesthesiol Scand. 1986;30:84. 87. Chestnut DH, Dewan DM, Redick LF, et al. Anesthetic man-
77. Carp J, Jayaram A, Stoll M. Ultrasound examination of the agement for obstetric hysterectomy: A multi-institutional
stomach contents of parturients. Anesth Analg. 1992;63:665. study. Anesthesiology. 1989;70:607.
78. American Society of Anesthesiologists. Practice guidelines 88. Frederiksen MC, Glassenberg R, Stilka CS. Placenta previa:
for preoperative fasting and the use of pharmacologic A 22-year analysis. Am J Obstet Gynecol. 1999;180:1432.
agents to reduce the risk of pulmonary aspiration: Appli- 89. Bjornstad E, Iversen OE, Raeder J. The impact of increasing
cation to healthy patients undergoing elective procedures. the use of regional anaesthesia for emergency caesarean
Anesthesiology. 1999;90:896. section. Eur J Anaesthesiol. 2004;21:776.
CHAPTER PREECLAMPSIA, ECLAMPSIA,
49
AND THE HELLP SYNDROME
CASE SUMMARY
How Are Preeclampsia
24-year-old primigravida presented to her
and Eclampsia Defined?
A
obstetrician at 27 weeks gestation with
swollen hands, face, and ankles, and a mild
headache. On questioning, there were no If we use the American College of Obstetricians and
visual symptoms, but she had malaise for the Gynecologists (ACOG) Practice Bulletin, Diagnosis and
last 24 hours. She denied epigastric or right Management of Preeclampsia and Eclampsia, the defini-
upper quadrant pain. On examination, her blood pressure tions are clear and practical.3 Preeclampsia is a blood
(BP) was 160/100 mm Hg, heart sounds were normal, and pressure of 140 mm Hg systolic or higher, or a diastolic
her chest was clear. She had no hyper-reflexia, but she pressure of 90 mm Hg. . . that occurs after 20 weeks gesta-
had gained 8 lbs in the last 2 weeks. Fetal heart tones were tion in a woman with previously normal blood pressure,
normal. Urine was 4+ positive for protein on dipstick, in the presence of proteinuria of 300 mg in a 24-hour urine
and blood was sent to the laboratory. Her complete blood sample (this is usually 1+ on dipstick). The development of
count (CBC) was normal, but uric acid was elevated. the HELLP syndromehemolysis, elevated liver enzymes
Coagulation studies and liver function tests were normal, and a low platelet countin the presence of hypertension
is sufficient to make a diagnosis of preeclampsia. Severe
but D-dimers were slightly elevated. A decision was made
preeclampsia is diagnosed when any one of the following
to admit her to the labor unit for observation, bedrest,
is present:
and treatment with magnesium sulphate.
This story is familiar to anyone caring for pregnant Systolic BP 160 mm Hg
women, and although the treatment options have changed Diastolic BP of 110 mm Hg or higher on two occasions
somewhat over the years, the etiology and underlying while on bedrest
pathology of preeclampsia is still incompletely known Proteinuria of 5 g in a 24-hour urine specimen (usually
and much debated. The word itself, eclampsia, comes 4+ on dipstick)
from the Greek, meaning lightening strike. This is be- Oliguria of <500 mL in 24 hours
cause the disease, described by Celus (2,000 years ago) Cerebral or visual symptoms
came on suddenlylike a lightening boltand resolved Pulmonary edema or cyanosis
with delivery.1 Later, the disease was known as toxemia Epigastric or right upper quadrant pain
of pregnancy, because some type of poison or toxin was Impaired liver function
thought to be circulating in the mother. By the 19th cen- Thrombocytopeniafetal growth restriction
tury, hypertension and the renal manifestations of the Eclampsia is the new onset [of] grand mal seizures in
disease were recognized, in addition to seizures.2 How a woman with preeclampsia. Using the ACOG definition
do we define preeclampsia in the 21st century? Gener- places anesthesiologists and obstetricians on the same
ally, clinicians divide the disease into mild and severe, page. The patient hereto described has two criteria for
and agree that a seizure in the setting of preeclampsia, severe preeclampsia: Blood pressure of 160 mm Hg and
and in the absence of prior neurologic disease constitutes proteinuria. Immediate treatment includes bedrest, close
eclampsia. However, the textbooks that anesthesiologists monitoring of mother and fetus, and pharmacologic inter-
are likely to have on their shelf yield slightly divergent vention with antihypertensives and magnesium sulphate
definitions and criteria. by infusion.
691
692 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
(PlGF), and soluble fms-like tyrosine kinase 1 (sFlt-1). the same. Others have considered preeclampsia a partial
VEGF and PlGF both promote angiogenesis; sFlt-1, which immune rejection of the fetus, and that complete rejection
is antiangiogenic, binds both VEGF and PlGF and prevents results in abortion. This does not explain the number of
their interaction with endothelial cell receptors, producing multiparas who develop preeclampsia, and the fact that
endothelial dysfunction. sFlt-1 has been found in both having preeclampsia in a first pregnancy is a risk factor
the placentas and blood of preeclamptic women, and for developing the disease in a second pregnancy. There
is increased 5 weeks before the onset of preeclampsia. is, however, evidence that fetal cells and free DNA are
Circulating PlGF is reduced in preeclampsia, probably shed into the maternal circulationand in greater num-
because the increased concentrations of sFlt-1 have bers in the preeclamptic. The first observation of fetal
bound it, further contributing to placental and endothelial cells in the lungs of women who had died of eclampsia
dysfunction. Circulating VEGF is found to be decreased, was made in the 19th century, by Schmorl.17 Hahn and
as one would expect if mopped up by sFlt-1 in some Holzgreve reviewed (and conducted) some of the recent
studies but not in others. PlGF can also be found in studies of fetal cell traffic in the preeclamptic.17 They note
urine; this, too, has been studied as a possible marker. that in pregnancies with donated eggswhere both the
Buhimschi et al. have studied sFlt-1 and PlGF in urine ova and sperm are antigenically different from the preg-
and showed an increase in sFlt-1 and a decrease in nant hostpreeclampsia is vastly increased. Perhaps the
PlGF.13 They also demonstrated a urinary ratio of the initial placental defect in the first trimesterischemia as
two that may be useful as a marker for preeclampsia. the result of faulty vascular developmentleads to this
However, yet another study by Powers et al. found that increase in fetal cellular and DNA traffic, which then
circulating PlGF is not increased in all women with produces the second maternal stage of preeclampsia in
preeclampsia.14 the third trimester. Cotter et al. have studied a small
Traditionally, increased thromboxane and decreased number of RhD-negative women who went on to de-
prostacyclin were thought to be major contributors to velop preeclampsia, and matched them to Rh-negative
the clinical picture, and many textbooks show a seesaw controls.18 Increased amounts of the fetal RhD gene
imbalance of the two, as described by Walsh in 1985.15 was found in the women who developed preeclampsia at
Thromboxane, a potent vasoconstrictor has been found to 15-weeks gestation. This work is significant, not because
be increased in preeclamptics in some studies, but Sibais fetal RhD genes will be a reliable markeronly a small
group, during their low-dose aspirin study, found that percentage of white women are RhD negative, and even
reducing thromboxane levels did not prevent or lessen fewer African or Asian womenbut because it demon-
the severity of preeclampsia.6 Nitric oxide, another potent strates an immune contribution to the disease. All women
vasodilator, has been studied with difficulty because of in the study had circulating fetal RhD genes, but the
its evanescence, and again there is no firm conclusion. amount was increased in those who became preeclamp-
A number of other circulatory and urinary factors, such tic. This tells us that fetal DNA does cross the placenta
as -natriuretic peptide and homocysteine, are under in normal and abnormal pregnancies, and that the mater-
observation, but it may be years yet before a reliable nal immune response to this material may be one of the
marker is discovered, and thus far sFlt-1, PlGF, and VEGF factors that causes the disease. Zhong et al. demonstrated
seem promising. not only an increase in fetal DNA in the preeclamptic,
A curious phenomenon is the marked reduction of but also an increase in maternal DNA.19 The presumed
preeclampsia in smokers. This has been noted since the mechanism for this free DNA of maternal and fetal ori-
1960s, when investigators collected data on smoking, gin is cell death. It is not known whether this cell death
low birth weight, and spontaneous abortion. It seems is causative, or simply another marker for the process
odd that regular exposure to a host of toxins and a underlying preeclampsia.
potent vasoconstrictor such as nicotine would prevent Placentation depends on the invasion of maternal
a hypertensive, vasoactive disease, but such is the case. tissue by cytotrophoblasts. The HLA-G is a molecule
Sibai et al. found this true in a prospective multicenter expressed by extravillous trophoblast cells, and may
study in 1995.9 Zhang et al. looked at a large study group protect these cells from the maternal immune response.
and found a dose-response reduction in preeclampsia Yie et al. have postulated that the reduced expression of
in smokers.16 The physiologic mechanisms behind this this gene could play a role in preeclampsia.20 The HLA-G
observation remain unclear. is found in the placenta and circulation and is almost
exclusively produced by trophoblasts. It was consistently
lower in the first and second trimester in those women
IMMUNOLOGIC who developed preeclampsia versus matched controls.
No statistical difference was seen in third trimester
An immunologic basis for preeclampsia has been postu- levels, although an earlier study by this group20 did
lated since the 1970s. Clearly, pregnancy is an unusual demonstrate a difference. Finding a marked difference
immunologic state in that two different genetic beings, early in pregnancy supports the concept that it is an early
with different human leukocyte antigen (HLA) markers, defect in placental implantation that leads to the later
coexist without mounting an immune response to each maternal manifestation of preeclampsia. Moreover, it is
other. The fact that most preeclampsia occurs in first preg- not known whether this reduced expression of the HLA-G
nancies led investigators to think that a first pregnancy gene is the cause, because it protects trophoblasts from
somehow primed a woman for the next immunologic the maternal immune response, or another marker, but it
assault, provided the paternity of the second fetus was may have predictive value.
694 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
The cytokine TNF- and its variants have been studied be a critical factor.27 There probably is a paternal influ-
in many disease states. It is released by macrophages ence on the development of preeclampsia, but it remains
and mast cells, and is a potent immune modulator undefined at present.
responsible for lymphocyte and interleukin activation.
Schipper et al. focused their attention on TNF in their
study of high-risk women.21 In their cohort study, TNF
levels were monitored throughout pregnancy in 68 women What Is the Clinical Picture
with a history of severe preeclampsia, intrauterine growth
retardation, or hypertension. The levels rose throughout of Preeclampsia?
pregnancy, but no significant differences were noted
between normotensive, hypertensive, or preeclamptic
women in any trimester. This is in contrast to several PREECLAMPSIA
studies that found increased TNF and TNF-receptor
(TNF-r) levels in the third trimester in preeclamptic Preeclampsia can be mild, severe, or complicated by the
women. In the Schipper study, TNF-r levels were higher HELLP syndrome. The disease may originate in the pla-
in the second trimester in women with both preeclampsia centa during the first trimester, but it is not seen in
and intrauterine growth retardation, but not in women the mother until 20 weeks gestation. Patients with mild
affected by preeclampsia without fetal growth restriction. preeclampsia (BP 140/90 mm Hg, proteinuria of more
TNF-r was also high in women with severe preeclampsia. than 300 mg in 24 hours) already have multiorgan system
These varied outcomes may be the result of a small sample disease. Widespread vasoconstriction is manifested in nu-
size, or more likely that there are many mechanisms and merous ways other than hypertension. Roberts describes
influences at work in preeclampsia. autopsy findings consisting of necrosis and hemorrhage
found in the brain mainly as petechial bleeding, in the
liver as areas of infarction, and in the heart as suben-
docardial necrosis.28 The recognition of proteinuria and
GENETICS a presumed glomerular lesion as a sign of preeclampsia
In spite of decades of study, the genetic component dates from the mid-19th century. There is a characteristic
of preeclampsia remains elusive, although observation renal lesionglomerular endotheliosiswhich has been
would suggest at least a partial contribution from an described by many investigators and reviewed by Karu-
individuals genetic makeup to the development of the manchi et al.29 The endothelium of the glomerular capil-
disease. There have been familial cases of preeclampsia, lary is swollen and the lumen almost obliterated, but the
and a subset of women develop the disease in all of their podocytes are normal. The microscopic picture is bland,
pregnancies. Roberts and Cooper, in their review, note a with little increase in cellularity and few red blood cells or
lack of concordance between monozygotic twins, but this casts in the urine. Glomerular filtration, which is normally
could be due to paternal factors.22 There are also different increased in pregnancy, falls and, although creatinine
rates of preeclampsia in different populationssome can rarely rises, uric acid increases. Acute tubular necrosis is
be explained by environment, culture, or diet, but some a rare complication of severe preeclampsia. Proteinuria
are most likely of racial or genetic variation. usually appears with or after hypertension, and resolves
Teasing out the contributions of maternal, paternal, slowly post partum, taking up to 8 weeks to do so. Edema
and fetal genetic material will not be easy. Goodwin and is common, and is caused by relative hypoalbuminemia
Mercer have looked at maternal ethnicity and found the as protein leaks out of the swollen glomerulus and leaky
HELLP syndrome more common in whites, and severe vasculature. As Karumanchi notes, this is not the edema
hypertension more common in African Americans and of congestive heart failure, which is characterized by an
Hispanic women.23 Factor VLeiden and the hereditary thro- underfilled circulation and excessive secretion of renin
mophilias have been associated with preeclampsia, as well and aldosterone. It is like the edema of the nephritic syn-
as other adverse outcomes. Lin and August have done a drome and is relatively overfilled, although the intravas-
meta-analysis of 31 case-controlled studies from 1966 to cular compartment may be contracted. Systemic vascular
2002 and concluded that factor VLeiden is associated with resistance is increased, and there may be poor correlation
a twofold increase in preeclampsia.24 It is anticipated that between central venous pressure and pulmonary capillary
more data will become available when the outcomes of wedge pressure. Oliguria (<0.5 mL of urine/kg/hour, or
the Genetics of Preeclampsia Trial in the United King- <25 mL per hour) is often present and attempts to fill
dom are published.25 This is a multicenter study that the intravascular space, possibly leading to iatrogenic pul-
is collecting genetic material from a large cohort of pa- monary edema. Hypertension and proteinuria remain the
tients, which should yield some insight into the genetics of classical and most easily recognized manifestations of
this complex disease. Evidence that the paternal genetic preeclampsia, although they are only two symptoms of a
contribution plays a role in preeclampsia has been seen. complex vasogenic and inflammatory condition.
Trupin noted in 1996 that a new partner could change
a womans chances of developing preeclampsia.26 The Hemolysis, Elevated Liver Enzymes
contribution that paternal genetic material makes to the and Low Platelet (HELLP) Syndrome
etiology of preeclampsia is not straightforward. Although
evidence suggests as much, it has also been shown that The liver is commonly affected by preeclampsia, with sub-
the time between pregnancies with new partners may also capsular swelling and hemorrhage, as well as increased
C HAPTE R 49/P R E EC L AM P SIA, EC L AM P SIA, AN D TH E H E LLP SYN DROM E 695
enzymes. The swelling produces right upper quadrant treated all preeclamptics, and others reserved treatment
pain, with hemorrhage a rare and life-threatening com- for those with severe preeclampsia or neurologic symp-
plication. Unfortunately, pregnant women sometimes toms. For 30 years, researchers have been looking at ways
present with cholelithiasis and cholecystitis, further com- to prevent seizures in eclampsia. Benzodiazepams, pheny-
plicating the diagnosis. Elevations in serum aspartate toin, and antihypertensives have been studied, along with
aminotransferase, alanine aminotransferase, and lactate magnesium sulphate. The presumed antiseizure mecha-
dehydrogenase are seen. There is often a low level con- nism of magnesium is cerebral vascular dilatation and
sumption of platelets as microthrombi are deposited in central nervous system (CNS) depression. It is also known
many vascular beds. Fibrin degradation products are fre- as a mild antihypertensive and tocolytic, and lowers plasma
quently detected and, in the setting of falling platelets and levels of endothelin.32 Magnesium impairs neuromuscular
worsening disease, are real cause for concern. Fibrinogen transmission and is a potentiator of neuromuscular block-
levels usually remain high or normal until late in the dis- ing agents. One can assess magnesium levels by assessing
ease. True disseminated intravascular coagulation is rare, deep tendon reflexes, which decrease as blood levels of
even in severe preeclamptics, and is usually encountered magnesium rise. The Magpie Trial (MAGnesium Sulphate
in the setting of placental abruption. Preeclamptics with for the Prevention of Eclampsia), conducted from 1998 to
the HELLP syndrome are a particular challenge because 2001, recruited more than 10,000 preeclamptic women in
thrombocytopenia and coagulopathy impact on the choice 33 countries in North and Latin America, Europe, Africa,
of analgesia and anesthesia. the Middle East, and South East Asia. These women were
randomized to treatment with magnesium or placebo.33
The definition of preeclampsia was a diastolic BP of
ECLAMPSIA 90 mm Hg, or a systolic BP of 140 mm Hg or more on
two occasions and proteinuria of 1+ or more. Women
Neurologic manifestations of the disease may be initially who had not delivered and those <24 hours post partum
mild, such as headache and hyperreflexia. There may be were recruited. Those with myasthenia gravis, hepatic
blurred vision and, rarely, temporary loss of visionthat coma with a risk of renal failure, or sensitivity to magne-
is, occipital lobe blindness, of which brain edema is the sium were excluded. Women with oliguria (<25 mL urine
presumed mechanism. Magnetic resonance imaging has per hour) were included, but the dose of drug was halved.
shown a characteristic edema of the posterior white mat- In some centers, an intravenous (IV) regimen was used,
ter, termed posterior leucoencephalopathy, as described and in others an intramuscular (IM) regimen. The loading
by Walker30 and Aagard-Tillery and Belfort.31 In addi- dose was 4 g IV or IM of placebo or magnesium sulphate.
tion to edema and decreased perfusion due to vasospasm, This was followed by an infusion of 1 g per hour IV or
other manifestations such as petechial hemorrhages, mi- 5 g q4h IM for 24 hours. Rescue doses in each treatment
crothombi, and inflammation may present. Neurologic package were available for women who went into seizures.
symptoms are ominous, for they may presage the onset Blood levels of magnesium were not measured, but deep
of eclampsia. Approximately 80% of eclamptics may first tendon reflexes and respiration rates were checked every
present with headache or visual changes, but a significant 30 minutes, and urine output was followed hourly. The
number will have neither antecedent neurologic symp- treatment dose (of placebo or magnesium sulphate) was
toms nor BP persistently elevated above 140/90 mm Hg. halved if deep tendon reflexes were slow, respiratory rate
MacKay et al. in reviewing Centers for Disease Control and fell below 16 per minute, or urine output was <100 mL in
Prevention statistics, found that among women who die 4 hours.
from eclampsia, most (18.8%) have evidence of cerebral Analyses were by intention to treat, and primary
hemorrhage.8 Others die of cerebral edema, cerebral em- outcomes were eclampsia and neonatal death, including
bolus, renal or hepatic failure, HELLP syndrome, and stillbirth. Secondary outcomes were maternal morbid-
disseminated intravascular coagulation. Older women, ityrespiratory depression or arrest, pneumonia, car-
and women developing preeclampsia or eclampsia in the diac arrest, coagulopathy, renal failure, liver failure,
second trimester, are at greatest risk of dying. The over- pulmonary edema, cerebral hemorrhage, with toxicity
all mortality ratio from preeclampsia and eclampsia, in defined as needing rescue with calcium gluconateand
this review of US figures from 1979 to 1992, was 1.5 per a host of side effects including nausea, vomiting, drowsi-
100,000 live births. ness, and muscle weakness. There were 5,055 women
in each group of magnesium sulphate and placebo. The
primary outcome was significantly fewer eclamptic con-
vulsions in the treatment group (0.8%) versus the placebo
How Is Preeclampsia group (1.9%). Maternal death, though not a primary out-
and Eclampsia Prevented come, was 0.2% in the treatment group versus 0.4% in the
placebo group. Surprisingly, the main causes of maternal
and Treated? death were cardiac failure or arrest and stroke, followed
by eclampsia or preeclampsia, renal failure, pulmonary
embolism, infection, hemorrhage, and respiratory failure.
MAGNESIUM SULPHATE It is not clear how, or if, magnesium sulphate lowers the
death rate from these causes. There was no clear differ-
For years, some centers routinely used magnesium sul- ence in neonatal deaths in the treatment or placebo group.
phate prophylaxis and others did not. Some centers Maternal morbidity, as outlined earlier, occurred at the
696 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
same rate in each group. Of equal significance, there was contributes to endothelial damage, then supplementation
no evidence of a tocolytic or antihypertensive effect of with antioxidants, which are free radical scavengers,
magnesium sulphate. Women were treated with antihy- should improve the clinical picture or even prevent the
pertensives as needed. As expected, more side effects were development of preeclampsia. Spinnato and Livingston
seen in the treatment group. Although the mechanisms of reviewed the randomized trials of antioxidants up to
magnesium sulphate are incompletely understood, it has 2004.37 The greatest benefit has been seen with vitamin
clearly proved its value in preventing eclampsia. It is now C and E supplementation in a randomized, blinded,
recommended treatment by the World Health Organiza- placebo-controlled trial by Chappell et al. in 1999.38
tion, although the countries where maternal mortality is Numbers were small (142 in the placebo group and 141
highest have been the slowest to adopt the treatment.34 in the antioxidant group) but results impressive, with a
Sibai is of the opinion that the evidence does not war- 76% reduction in preeclampsia in the treatment group.
rant treatment of mild preeclamptics with magnesium A number of other antioxidants, including selenium, zinc,
sulphate.35 Thus, he suggests that the treatment should coenzyme Q, lycopene, and melatonin have also been
be reserved for those with imminent eclampsia. studied. Rumbold et al. have searched the Cochrane data
base and found seven trials involving 6,082 women, which
were randomized or quasi-randomized.39 Only three trials
were considered to be of high quality. Nonetheless, a
ACETYLSALICYLIC ACID 39% reduction in the risk of preeclampsia occurred if
(ASPIRIN) supplementation with any antioxidant was used. There
was a reduced risk of small-for-gestational-age infants,
Aspirin (acetylsalicylic acid [ASA]) has been prescribed but an increase in preterm births. Caution was advised
in low doses for various indications in pregnancy. in applying these results to the population at large until
Some clinicians use it in women with a repeated better quality studies confirmed the results. Similarly,
history of spontaneous abortion and anticardiolipin calcium supplementation in women with a low dietary
antibodies. There has been interest in using aspirin intake is of benefit, but there is no convincing evidence of
to correct the thromboxane/prostacylin imbalance in reduction in poor outcome.36
preeclampsia. Sibai et al. have conducted at least two
large prospective trials comparing ASA to placebo. In
one multicenter, randomized, double-blind trial of low-
dose aspirin, 2,539 high-risk women were recruited.6 What Are the Anesthetic
High risk was defined as pregestational diabetes, chronic
hypertension, multiple gestation, or a prior pregnancy
Considerations for
complicated by preeclampsia. There was no difference in Preeclampsia and Eclampsia?
preeclampsia rates between the treatment and placebo
groups. Thromboxane levels were lower in the women The only real cure for eclampsia and preeclampsia is
on aspirin, although no difference was seen in clinical delivery. When a preeclamptic woman is admitted to the
outcome between groups. There is some evidence of labor unit, the anesthesiologist should be involved early in
decreased risk of preeclampsia in low-risk patients, but her assessment and management. The obstetric team will
no evidence of improved maternal or fetal outcome. Sibai start treatment with antihypertensives and magnesium
et al. revisited this topic in 2005, and concluded, after sulphate, according to their treatment protocols. The tim-
reviewing numerous trials, that low-dose aspirin is safe ing and mode of delivery should be coordinated between
but has only small-to-moderate benefit in some trials as the obstetric and anesthetic services. Preeclampsia is a
an agent to prevent preeclampsia. This finding was in major cause of premature delivery. If the symptoms are
spite of the fact that a reduction of 19% in the incidence mild, the obstetric team may wish to temporize by treat-
of preeclampsia was reported. Inconsistent definitions of ing the mother and administering corticosteroids for fetal
preeclampsia and timing of administration were cited as lung maturity. If the mother can be stabilized, and the fe-
confounding variables. Their final conclusion was that tus is doing well when assessed by biophysical profile, the
more information is needed, and the decision to use team may be able to add a week to the pregnancy. How-
aspirin should be based on a womans individual risk.36 ever, worsening of the clinical picture for either mother
In this review, heparin and the low molecular weight or fetus may dictate delivery.
heparins showed promise in reducing preeclampsia in The anesthesiologist should make an early assessment
women with thrombophilias. Heparin was therefore of the mothers airway, volume status, and laboratory in-
recommended, although the need for further study in vestigations, especially her platelet count and coagulation
the form of randomized trials was noted. studies. If the plan is to induce labor, an epidural will
alleviate pain, decrease maternal cathecholamine levels,
reduce BP, and provide a regional anesthetic should a
ANTIOXIDANTS cesarean section be necessary for failure to progress or
nonreassuring fetal heart rate. If the team decides that the
Oxidative stress, lipid peroxidation, and the formation best course of action, because of an unripe cervix or pre-
of free radicals have all been proposed as causative maturity, is to move directly to cesarean section, a spinal
agents in preeclampsia. If an excess of free radicals is an option. There has been considerable debate over the
C HAPTE R 49/P R E EC L AM P SIA, EC L AM P SIA, AN D TH E H E LLP SYN DROM E 697
merits of spinal versus epidural anesthesia and the platelet spinal group, but it was of short duration (<1 minute)
level that is safe for regional anesthesia in parturients. and responded to ephedrine. Newborn Apgar scores and
For many years, the standard teaching has been that outcomes were similar in both groups. Ephedrine use
hypotension commonly seen in pregnant women under- was higher, of course, in the spinal group. Santos and
going spinal anesthesia would pose a particular danger to Birnbach, in an editorial, remind readers that this study
the preeclamptic, since such patients often have a depleted contains mainly good news: That a small degree of treat-
intravascular volume (from hypertension and vasocon- able hypotension is preferable to a failed intubation.41
striction). A precipitous fall in BP that might be refractory They advocate spinal anesthesia in severe preeclamptics.
to treatment was feared. However, rapid fluid administra- Aya et al. published a prospective cohort compari-
tion in these women can lead to pulmonary edema due to son in 2003 between severe preeclamptics and normal
a leaky endothelium and low oncotic pressure. parturients presenting for cesarean section under spinal
Absolute contraindications for spinal anesthesia are anesthesia and found less hypotension in the preeclamp-
few and include: tics and less use of ephedrine.42 Critics postulated that this
was not because the relatively hypertensive preeclamptics
Lack of consent
truly had less hypotension, but because they were having
Lack of resuscitative equipment and drugs
smaller babies due to prematurity and intrauterine growth
Significant hypovolemia with no time for correction, or
restriction. Therefore, the decreased hypotension and
ongoing hemorrhage
ephedrine requirements were thought to be the result of
Local infection (at the injection site)
less aortocaval compression. In 2005, Aya et al. published
The impetus to provide regional anesthesia is great, a second study of severe preeclamptics and normoten-
and stems from evidence of decreased maternal mortality sive women who were having preterm babies.43 Severe
documented in closed claims analysis because regional preeclampsia was defined as systolic BP 160 mm Hg
anesthesia was introduced in laboring women. The or a diastolic BP of 110 mm Hg or greater. Nicardipine
particular advantages of spinal anesthesia are its speed was the antihypertensive started, and magnesium sul-
of onsetmaking it a useful technique in emergent phate for seizure prophylaxis was also started if needed.
situationsthe relative ease of placement with a defined In the control group, normotensive women undergoing
end point (flow of cerebral spinal fluid), the smaller, less preterm cesarean section who delivered fetuses of 1,100 g
traumatic needle, and the dense anesthesia it provides. In to 1,900 g were enrolled. A total of 65 preeclamptic pa-
women with thrombocytopenia (often defined as platelet tients and 71 normotensive women with preterm babies
counts 100,000 per mm3 ), there may be greater operator were recruited. Spinal technique and doses were similar
comfort with a small spinal needle and placement through in the two groups. Hypotension leading to treatment with
the avascular dura, rather than a large epidural needle ephedrine was less frequent in the preeclamptic group.
and a catheter that must be threaded into the relatively When hypotension occurred (systolic BP <100 mm Hg),
vascular epidural space. less ephedrine was needed to return the systolic BP to
The concerns regarding hypotension have been ad- normal in the preeclamptic group. Neonatal weights were
dressed in many studies throughout the 1990s and into similar between the two groups, proving that lessened aor-
2005. The results of retrospective and prospective studies tocaval compression was not the cause of the decreased
are reassuring. Spinal anesthesia seems to be safe in this need for ephedrine in the preeclamptic group. Neonatal
setting. Visalyaputra et al. undertook a randomized, multi- outcomes were similar in the two groups. In this ele-
center trial to compare the hemodynamic effects of spinal gant casecohort study, Aya et al. answered their critics
and epidural anesthesia in severe preeclamptics.40 Severe and showed, once again, that spinal anesthesia is safe in
preeclampsia was defined as a systolic BP 160 mm Hg, preeclamptics.
a diastolic BP 110 mm Hg, and proteinuria of 11 mg per The second major controversy in regional anesthe-
dL. Participants were scheduled to have elective or urgent sia for preeclamptics concerns the platelet count that
cesarean sections. They were started on IV magnesium sul- is considered safe for neuraxial techniques. Epidural
phate for seizure prophylaxis and hydralazine to decrease hematomas are rare, can occur spontaneously, and are
diastolic BP to 90 mm Hg. Lactated Ringers solution was a dreaded complication in a young, vital group of pa-
given intravenously at 100 mL per hour, and 500 mL of tients. A low platelet count, generally defined as below
colloid was given before the initiation of regional anes- 100,000 per mm3 , is found in 10% to 25% of preeclamp-
thesia. The epidural group of 47 women received 18 to tics, according to Barton et al.44 Although a decreased
23 mL of 2% lidocaine with epinephrine 1:400,000, fol- platelet count is always cause for concern and usually
lowed by 3 mg of morphine post delivery. The spinal group indicates severe disease, it correlates poorly with the in-
of 53 women received 2.2 mL of 0.5% hyperbaric bupiva- cidence of epidural hematomas. Coagulation studies are
caine with 0.2 mg morphine. In each group, mean arterial usually normal in this setting, although D-dimers may be
pressure was measured every minute for the first 20 min- slightly elevated. Fibrinogen and international normalized
utes, every 2 minutes for the next 10 minutes, and every ratio (INR)/partial thromboplastin time (PTT) are usually
5 minutes thereafter. Women in both groups were placed normal. There are no prospective randomized placebo-
supine with left uterine displacement, and ephedrine was controlled studies on this topic, and therefore we must
given for decreases in mean arterial pressure according rely on retrospective reports. Mandel and Surapaneni re-
to a strict protocol. There was twice as much hypoten- viewed the available studies in 2004 and found a wide
siondefined as systolic BP of <100 mm Hgin the variation in practice.45 In one analysis, 62 patients with
698 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
HELLP syndrome and platelet counts between 19,000 anesthesiologist, and patient proceeded to the operating
to 143,000 per mm3 received spinal or epidural anes- room. After 500 mL of colloid, a single-shot spinal us-
thesia with no neurologic complications. This is a much ing a 25-g, pencil-point needle and 1.5 mL of hyperbaric
lower limit (19,000 per mm3 ) than the traditionally quoted bupivacaine (0.75%) with 10 g of fentanyl and 100 g
80,000 per mm3 , felt to be a safe level for regional anesthe- of preservative-free morphine was performed in the sit-
sia. Thromboelastograms and a platelet function analyzer ting position. The patient was positioned supine with left
have been used, and suggest that platelet function in uterine displacement, and her BP was taken every minute.
preeclamptics with thrombocytopenia may be preserved Prophylactic phenylnephrine and ephedrine were drawn
with counts as low as 60,000 per mm3 . Each individual up, but were not administered initially because her BP
practitioner must decide, within the clinical context he or dropped slowly from 170/95 to 150/90 to 130/80. After
she encounters, what is a safe level of platelets for place- delivery of a female infant and administration of oxy-
ment of a neuraxial technique. Complicating the issue is tocin, her BP was 105/76, and she felt light headed. At
the fact that the very patients who are likely to have throm- this point, 100 g of phenylnephrine was administered,
bocytopenia also present the greatest airway challenges and her symptoms abated as her BP rose to 130/80. The
due to the edema of severe preeclampsia. A stable platelet rest of the operative course was unremarkable. She was
count of 50,000 may be safer than a rapidly falling count kept under close observation on the labor unit and treated
that was 100,000 at 9:00 AM and is only 70,000 at 12 noon. with magnesium sulphate for another 24 hours. She was
A single shot spinal with a small needle may be safer than closely monitored with neurovitals every 2 hours until
struggling with a 17- or 18-g epidural needle and catheter her motor and sensory function returned to normal. Her
in an edematous, restless patient. Careful follow-up of a platelet count slowly increased to 110,000 over the next
postspinal patient with low platelets may be better than 2 days, and she began to diurese. Her BP remained ele-
an unsecured, edematous airway in the setting of an ur- vated until day 5, and her liver enzymes slowly normalized
gent delivery for fetal distress. As published reports seem over the course of a week.
to push the level lower, we continue to have no abso- This is not an unusual case, and the outcome
lute guidelines regarding platelet count. However, when could have been worse. Preeclampsia, especially before
deciding to proceed with regional anesthesia in a throm- 28 weeks, can be a serious threat to a young mothers
bocytopenic patient, it is essential that the rest of the life. Our patient did well with early intervention and
coagulation profile be normal, and that informed consent treatment. If living in a place remote from prenatal and
is obtained. It is useful to remember that pregnancy, even obstetric care, she may have gone on to seize at home and
in the setting of preeclampsia, is a hypercoagulable state, become unresponsive. The natural history of this disease
and that most clotting factors are increased in pregnancy. includes death from cerebral hemorrhage.
To return to our case, our 24-year-old primipara, of In many ways, preeclampsia is a great mimicker with
27 weeks gestation with severe preeclampsia, was admit- a variety of clinical presentations. Women with mild
ted and treated with oral labetalol and IV magnesium preeclampsia in the third trimester generally do well.
sulphate. After 24 hours in the intensive care unit and two Women (and their babies) with severe preeclampsia and
doses of corticosteroid for fetal lung maturity, she was eclampsia in developed nations usually do well with timely
feeling no better. Her BP was 145/90 to 155/90, and al- intervention and treatment, but this is not a benign dis-
though somewhat drowsy from her magnesium sulphate ease. There is significant mortality and morbidity from
infusion, she complained of a frontal headache, blurred this disease, and it is a leading cause of premature deliv-
vision, and right upper quadrant pain. Urine output was a ery in the developed world. In developing nations, severe
scant 25 mL per hour on 100 mL of crystalloid per hour. preeclampsia and eclampsia have a much higher mortal-
She was sent for a fetal ultrasound, which revealed de- ity rate and are worthy causes for aid, intervention, and
creased fetal movement and reduced amniotic fluid. Her education.
CBC was repeated and now showed platelets of 75,000 per
mm3 . Her coagulation studies, including fibrinogen lev-
els, were normal, but she continued to have elevated
D-dimers. Her liver enzymes were now elevated and, be-
KEY POINTS
lieving she had the HELLP syndrome and a fetus who was
not thriving, her obstetrician decided to proceed with a 1. Preeclampsia is a hypertensive disorder of preg-
cesarean delivery. She had been on clear fluids, but had nancy characterized by endothelial damage and
not taken anything by mouth for the last 6 hours. She inflammatory change.
had never had a general or regional anesthetic, but her 2. There is significant morbidity and mortality for
parents had no history of adverse reactions to general mothers and their babies.
anesthesia. On examination, she was edematous, moder- 3. The etiolgy is not fully understood, but abnormal
ately obese (5 ft. 2 in. and 180 lbs), and had a significant placentation, maternal immune response, and ge-
overbite. Head and neck examination revealed normal, netics play a role in the endothelial damage that
intact dentition, a Mallampati III airway, a midline tra- characterizes the disease.
chea, and normal range of neck motion. The neonatal 4. Main treatment goals are to stabilize the mothers
team was called and, after counseling the patient and her BP, prevent seizures, and deliver the baby.
partner regarding the possible outcomes and likely clini- 5. Anesthesia technique and mode of delivery are
cal course for a 27-week fetus, the obstetricians, nurses, determined by maternal and fetal concerns, with
C HAPTE R 49/P R E EC L AM P SIA, EC L AM P SIA, AN D TH E H E LLP SYN DROM E 699
attention to coagulation status, platelet levels, and 12. Levine RJ, Karumanchi SA. Circulating angiogenic factors
in preeclampsia. Clin Obstet Gynecol. 2005;48:372.
end-organ damage.
13. Buhimschi CS, Norwitz ER, Funai E, et al. Urinary an-
6. In spite of tremendous interest and research, there giogenic factors cluster hypertensive disorders and identify
are still many questions and no method to predict women with severe preeclampsia. Am J Obstet Gynecol. 2005;
preeclampsia. 192:734.
7. There may be a role for antioxidants and other mi- 14. Powers RW, Roberts JM, Cooper KM, et al. Maternal serum
cronutrients in lowering the incidence of preeclamp- soluble fms-like tyrosine kinase 1 concentrations are not
sia, but more research is needed. increased in early pregnancy and decrease more slowly
8. A team approach in a center familiar with preeclamp- postpartum in women who develop preeclampsia. Am J
tic and eclamptic patients is ideal. Obstet Gynecol. 2005;193:185.
15. Walsh SW. Preeclampsia: An imbalance in placental prosta-
9. Magnesium sulphate has been proven to prevent
cyclin and thromboxane production. Am J Obstet Gynecol.
seizures and seems to lessen the death rate.
1985;152:335.
10. Spinal anesthesia should not be avoided for concerns 16. Zhang J, Klebanoff MA, Levine RJ, et al. The puzzling
with hypotension. association between smoking and hypertension during
11. Thrombocytopenia will always present a dilemma, pregnancy. Am J Obstet Gynecol. 1999;181:1407.
and platelet levels must be considered in the clinical 17. Hahn S, Holzgreve W. Fetal cells and cell-free fetal DNA
context, and account for the experience of the in maternal blood: New insights into preeclampsia. Hum
operator and risk:benefit analysis. Reprod Update. 2002;8:501.
12. If general anesthesia is indicated, careful preop- 18. Cotter AM, Martin CM, OLeary JJ, et al. Increased fetal
erative assessment of the airway is needed, with RhD gene in the maternal circulation in early pregnancy is
associated with an increased risk of preeclampsia. BJOG.
particular attention to signs of airway edema, in-
2005;112:584.
cluding changes in voice and facial edema. If any
19. Zhong XY, Laivuori H, Livingston JC, et al. Elevation of both
question exists regarding the patency of the airway maternal and fetal extracellular circulating deoxyribonucleic
or ability to secure it, an awake technique should be acid concentrations in the plasma of pregnant women with
used. preeclampsia. Am J Obstet Gynecol. 2001;184:414.
13. Agents capable of acutely lowering BP should be 20. Yie S, Taylor RN, Librach C. Low plasma HLA-G protein
available at induction, and maneuvers to reduce concentrations in early gestation indicate the development
the hypertensive response to intubation should be of preeclampsia later in pregnancy. Am J Obstet Gynecol.
considered (such as the use of predelivery narcotics). 2005;193:204.
14. The effects of magnesium on muscle relaxants and 21. Schipper EJ, Bolte AC, Schalkwijk GC, et al. TNF-receptor
levels in preeclampsiaresults of a longitudinal study in
minimal alveolar concentration (MAC) must be
high risk women. J Matern Fetal Neonatal Med. 2005;18:
remembered.
283.
22. Roberts JM, Cooper DW. Pathogenesis and genetics of
REFERENCES preeclampsia. Lancet. 2001;357:53.
1. Chesley LC. Hypertensive disorders of pregnancy. New York: 23. Goodwin AA, Mercer M. Does maternal race or ethnicity
Appleton-Century-Crofts; 1978. affect the expression of severe preeclampsia? Am J Obstet
2. Roberts JM. Endothelial dysfunction in preeclampsia. Semin Gynecol. 2005;193:973.
Reprod Endocrinol. 1998;16:1. 24. Lin J, August P. Genetic thromobphilias and preeclampsia:
3. American College of Obstetricians and Gynecologists prac- A meta analysis. Obstet Gynecol. 2005;105:182.
tice buletin no. 33. Obstet Gynecol. 2002;99:159. 25. The Genetics of Pre-Eclampsia (GOPEC) Trial. Available at:
4. Cunningham FG, Gant NF, Leveno KJ, et al. Williams http://www.gopec.org/index.html. Accessed May, 2006.
obstetrics, 21st ed. McGraw Hill; 2001. 26. Trupin LS, Simon LP, Eskenazi B. Change in paternity:
5. Ventura SJ, Martin JA, Curtin SC, et al. Births: Final data for A risk factor for preeclampsia in multipara. Epidemiology.
1998. National Center Health Statistics. Natl Vital Stat Rep. 1996;7:240.
2000;48:3. 27. Dekker G, Robillard PY. The birth interval hypothesisdoes
6. Sibai BM, Caritis S, Hauth J, et al. What we have learned it really indicate the end of the primipaternity hypothesis?
about preeclampsia. Semin Perinatol. 2003;27:3. J Reprod Immunol. 2003;59:245.
7. WHO International Collaborative Study of Hypertensive 28. Roberts JM. Endothelial dysfunction in preeclampsia. Semin
Disorders of Pregnancy. Geographic variation in the inci- Reprod Endocrinol. 1998;16:5.
dence of hypertension in pregnancy. Am J Obstet Gynecol. 29. Karumanchi SA, Maynard SE, Stillman IE, et al. Preeclamp-
1988;158:80. sia: A renal perspective. Kidney Int. 2005;67:2101.
8. MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortal- 30. Walker JJ. Preeclampsia. Lancet. 2000;356:1260.
ity from preeclampsia and eclampsia. Obstet Gynecol. 2001; 31. Aagard-Tillery KM, Belfort MA. Eclampsia: Morbidity, mor-
97:533. tality and management. Clin Obstet Gynecol. 2005;48:12.
9. Sibai BM, Gordon T, Thom E, et al. Risk factors for 32. Ramanathan J, Bennet K. Preeclampsia: Fluids, drugs, and
preeclampsia in healthy nulliparous women: A prospective anesthetic management. Anesthesiol Clin North America.
multicenter study. Am J Obstet Gynecol. 1995;172:642. 2003;21:145.
10. Page EW. The relationship between hydatid moles, relative 33. The Magpie Trial Group. Do women with preeclampsia,
ischemia of the gravid uterus and the placental origin of and their babies, benefit from magnesium sulphate? The
eclampsia. Am J Obstet Gynecol. 1939;37:291. Magpie trial: A randomized, placebo-controlled trial. Lancet.
11. Gupta S, Agarwal A, Sharma RK. The role of placental 2002;359:1877.
oxidative stress and lipid peroxidation in preeclampsia. 34. Roberts JM, Villar J, Arulkumaran S. Preventing and treating
Obstet Gynecol Surv. 2005;60:807. eclamptic seizures. Br Med J. 2002;325:609.
700 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
35. Sibai BM. Magnesium sulfate prophylaxis in preeclampsia: 41. Santos AC, Birnbach DJ. Spinal anesthesia for cesarean
Lessons learned from recent trials. Am J Obstet Gynecol. delivery in severly preeclamptic women: Dont throw out the
2004;190:1520. baby with the bathwater. Anesth Analg. 2005;101:859.
36. Sibai BM, Dekker G. Preeclampsia. Lancet. 2005;365:785. 42. Aya AGM, Mangin R, Vialles N, et al. Patients with
37. Spinnato JA, Livingston JC. Prevention of preeclampsia with severe preeclampsia experience less hypotension during
antioxidants: Evidenced from randomized trials. Clin Obstet spinal anesthesia for elective cesarean delivery than healthy
Gynecol. 2005;48:416. parturients: A prospective cohort comparison. Anesth Analg.
38. Chappell LC, Seed PT, Briley AL, et al. Effect of antioxidants 2003;97:867.
on the occurence of preeclampsia in women at increased 43. Aya AGM, Vialles N, Tanoubi I, et al. Spinal anesthesia
risk: A randomized trial. Lancet. 1999;354:810. induced hypotension: A risk comparison between patients
39. Rumbold A, Duley L, Crowther C, et al. Antioxidants with severe preeclampsia and healthy women undergo-
for preventing preeclampsia. Cochrane Database Syst Rev, ing preterm cesarean delivery. Anesth Analg. 2005;101:
Cochrane Libr. 2006;1. 869.
40. Visalyaputra S, Rodanant O, Somboonviboon W, et al. Spinal 44. Barton JR, Witlin AG, Sibai BM. Management of mild
verus epidural anesthesia for cesarean delivery in severe preeclampsia. Clin Obstet Gynecol. 1999;42:455.
preeclampsia: A prospective randomized multicenter study. 45. Mandal NG, Surapaneni S. Regional anesthesia in
Anesth Analg. 2005;101:862. preeclampsia. Drugs. 2004;64:223.
J . I M M U N O LOG I C
A N D I N F EC T I O U S
CHAPTER ANAPHYLACTIC
50
AND ANAPHYLACTOID
REACTIONS
Jerrold H. Levy
A
worker with a history of frank reflux, who death. They first used the word anaphylaxis (ana-
suspects he may have reacted to a sulfa against, prophylaxis- protection) to describe this clini-
drug as a child, presents for knee ligament cal syndrome. Anaphylaxis produces multiorgan system
reconstruction. After application of moni- dysfunction, including shock.2 The target organs include
tors, determination of baseline vital signs, the respiratory, cardiovascular, cutaneous, and gastroin-
and preoxygenation, his anesthetic was induced with testinal systems, all of which contain large quantities
sodium thiopental 350 mg, fentanyl 100 g, and succinyl- of inflammatory cells called mast cells. Because of the
choline 100 mg in a rapid sequence induction utilizing complex effects of the mediators present and the target
cricoid pressure. Endotracheal intubation was easily ac- end organs, the presentation of anaphylaxis is often un-
complished and confirmed, and 1 g of cephalothin was predictable, with variable signs and symptoms.2 In 2004
given for infection prophylaxis while N2 O and isoflurane and 2005, the National Institute of Allergy and Infectious
were initiated. Disease and the Food Allergy and Anaphylaxis Network
The circulating nurse notices red splotchiness on sponsored a multidisciplinary Symposium on the Defini-
the patients thorax. The heart rate increased from tion and Management of Anaphylaxis. This consortium
68 to 92 bpm, and the blood pressure decreased from brought together physicians from various medical spe-
132/78 to 95/60 mm Hg. A repeat blood pressure was cialties that deal with anaphylaxis to review current
60/43 mm Hg, and the patient manifests a suggestion knowledge and to discuss features of a common defi-
of wheezing not noted earlier. The next blood pressure nition, common treatment strategies, and areas in need of
determination cycled without success, and the pulse future research.3,4
oximeter stopped functioning. The electrocardiogram
In 1998, the Joint Task Force on Practice Parameters
(ECG) was sinus tachycardia. A pulse could not be
defined anaphylaxis as an immediate systemic reaction
palpated. The anesthesiologist administered 200 g of
caused by rapid, IgE-mediated immune release of potent
epinephrine. The intravenous (IV) rate was wide open.
mediators from tissue mast cells and basophils.5 The
Another 500 g of epinephrine was administered, and the
most common causes of anaphylaxis depend on the pa-
pulse was again not palpable, with the pulse oximeter
tient population that clinicians manage. For allergists,
reading 100%.
this includes food, certain medications such as antibi-
otics, insect stings including fire ant and hymenoptera,
and environmental antigens. Sampson et al. suggest ana-
phylactic reactions are distinguished from anaphylactoid
How Is Anaphylaxis Defined? reactions, which mimic signs and symptoms of anaphy-
laxis, but are caused by the nonIgE-mediated release
Anaphylaxis was first reported in 1902 by Portier and of potent mediators from mast cells and basophils.3,4
Richet when immunizing animals against jellyfish toxins Unfortunately, these definitions are not useful to most
701
702 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
anesthesiologists, surgeons, or intensivists managing a allergic history may be unknown to the patient. On reex-
patient with acute life-threatening cardiopulmonary col- posure, the antigen binds to bridge two immunospecific
lapse following drug or blood product administration. In IgE antibodies on the surfaces of mast cells and basophils
the perioperative period, certain agents are at an increased to release a complex series of inflammatory molecules
risk to produce anaphylaxis. that can be sufficient to produce acute cardiopulmonary
Agents most often associated with causing anaphy- dysfunction.8 The released mediators produce a symptom
laxis include drugs, blood products, and environmental complex of bronchospasm and upper airway edema in the
antigens such as latex.2 Pharmacologic agents also have respiratory system, vasodilation and increased capillary
the potential to produce predictable and unpredictable permeability in the cardiovascular system, and urticaria
adverse reactions. The most life-threatening form of an in the cutaneous system.2,9 Cardiovascular collapse during
adverse reaction is anaphylaxis; however, the clinical pre- anaphylaxis results from the effects of multiple mediators
sentation of anaphylaxis may represent different immune on the heart and vasculature.10,11 The vasodilation seen
and nonimmune responses.2 As stated in the preceding clinically can result from a spectrum of different medi-
text, there is confusion in the literature about the term, ators that interact with the vascular endothelium and/or
anaphylaxis. On the basis of current concepts, anaphylaxis vascular smooth muscle.2,12
is best defined as: a clinical syndrome characterized by
acute cardiopulmonary collapse following antigen (for-
eign substance) exposure. This chapter will describe the
spectrum of anaphylactic and adverse drug reactions an
anesthesiologist may encounter.
What Are the Mechanisms
of Vasodilatory Shock
in Anaphylaxis?
What Definitions Are Used Vasodilatory shock occurs in anaphylaxis because of mul-
in the Genre of Anaphylaxis? tiple mechanisms, including the excessive activation of
vasodilator mechanisms.12 The increased synthesis of
The term allergy was first described in 1906 by von Pirquet, nitric oxide (NO) contributes to the hypotension and resis-
who suggested that in both immunity and hypersensitivity tance to catecholamines that occur in anaphylactic shock.
reactions, antigens induced changes in reactivity.6 Over NO production is increased because of increased expres-
time, the term, allergy, was used to describe immunoglobu- sion of the inducible form of NO synthase that occurs
lin E (IgE)-mediated allergic disease. It was von Pirquets in vascular smooth muscle cells and endothelial cells.12
intent that the term be considered an uncommitted The mechanisms increasing the inducible NO synthase
biologic response that could eventually lead either to im- are suggested to be cytokines (such as interleukin-1,
munity (a beneficial effect) or allergic disease (a harmful interleukin-6, tumor necrosis factor , and adenosine)
effect).6 and other inflammatory mediators. Increased NO synthe-
The term, atopy, (from the Greek atopos, meaning out sis contributes to vasodilatation in shock. The vasodilating
of place) is also often used to describe IgE-mediated dis- action of NO in vasodilatory shock is mediated mainly by
eases.6 Persons with atopy have a hereditary tendency to the activation of myosin light-chain phosphatase; how-
produce IgE antibodies against common environmental ever, NO may also cause vasodilatation by activating
allergens, and have one or more atopic diseases (e.g., potassium channels in vascular smooth muscle cells.12
allergic rhinitis, asthma, and atopic eczema).6 Some The vascular hyporeactivity to catecholamines that oc-
allergic diseases, including contact dermatitis and hyper- curs in anaphylactic shock can be ameliorated by arginine
sensitivity pneumonitis, develop through other complex vasopressin.12
non-IgE mechanisms, including cell-mediated immune In addition to the increased NO synthesis that acti-
responses, and are considered nonatopic allergic con- vates soluble guanylate cyclase and produces cyclic guano-
ditions. Hypersensitivity reactions are considered un- sine monophosphate (cGMP), prostacyclin synthesis also
toward physiologic events mediated through immune contributes to vasodilation. It activates soluble adenylate
mechanisms. cyclase and produces cyclic adenosine monophosphate
(cAMP), both causing dephosphorylation of myosin, and
hence vasorelaxation. Although multiple mediators, in-
cluding arachidonic acid metabolites and kinins, are
What Is the Pathophysiology responsible for vasodilation, histamine also exhibits a ma-
jor role in acute cardiovascular collapse.12 Stimulation of
of Anaphylaxis? endothelial H1 receptors releases NO and prostacyclin.13
Unfortunately, specific blockade of the target enzyme of
Antigen binding to IgE antibodies causes anaphylaxis.2,7,8 NO pathway may not attenuate vasodilation because of
Prior exposure to the antigen or a substance of similar the other simultaneous mechanisms that also produce
structure is needed to produce sensitization, although an vasodilation.13
C HAPTE R 5 0/ANAP HYL AC TIC AN D ANAP HYL AC TOI D R E AC TION S 703
substance P).18,20,24,25 Different drugs administered dur- more common cause of allergic reactions in the operating
ing the perioperative period degranulate mast cells, but room.30
not basophils, to release histamine in a dose-dependent,
nonimmunologic manner.2,18,25,26 The IV administration
of morphine, atracurium, or vancomycin can release his-
tamine, producing vasodilation, redness, and urticaria How Is Anaphylaxis Managed?
along the vein of administration. This nonimmuno-
logic histamine release has classically been considered
to be the cause of many anaphylactoid reactions. Usu- Most anesthetic drugs and agents administered periop-
ally, the hypotensive effects of histamine release can eratively have been reported in the literature to produce
be treated with temporary pressor support or intravas- anaphylaxis.2 Therefore, a plan for treating anaphylactic
cular volume administration; however, the responses in reactions must be established before the event.2 Airway
different individuals may vary. The newer neuromuscu- maintenance, 100% oxygen administration, intravascular
lar blocking agents (NMBAs), for example, rocuronium volume expansion, and epinephrine are essential to treat
and cisatracurium lack the histamine-releasing effects, the hypotension and hypoxia that result from vasodilation,
but can produce direct vasodilation and false-positive increased capillary permeability, and bronchospasm.2
cutaneous responses that can confound allergy testing A suggested protocol follows for managing anaphylaxis
and interpretation.20 The mechanisms involved in nonim- perioperatively, with representative doses for a 70-kg
munologic histamine release represent the degranulation adult. Therapy must be titrated to the desired effects with
of mast cells, but not basophils, through cellular activa- careful monitoring. Severe reactions need aggressive ther-
tion and stimulation of phospholipase activity in mast apy. The route of administration of epinephrine and the
cells.20,24 dose depend on the patients condition.2 Rapid and timely
intervention, along with common sense, must be utilized
to treat anaphylaxis effectively.2 Reactions may be pro-
tracted with persistent, systemic hypotension, pulmonary
What Agents Are Typically hypertension, and right ventricular dysfunction, lower res-
piratory obstruction, or laryngeal obstruction that persist
Implicated in Anaphylaxis? 5 to 32 hours despite vigorous therapy. Novel therapeutic
approaches for anaphylactic shock and/or right ventricu-
The epidemiology of fatal anaphylaxis suggests risk factors lar failure are under investigation.12,13
for life-threatening reactions such as asthmatic reactions Hypotension is the most common acute manifestation
and shock, both common responses in insect sting and associated with anaphylactic reactions. An important
drug-induced reactions.2,6,16,27 Food allergy is the most mechanism of vascular collapse during anaphylaxis is the
common cause of anaphylaxis outside the hospital.6,27 decrease in systemic vascular resistance. The standard
In children, anaphylaxis that occurs in a non-hospital therapy to reverse vascular collapse is epinephrine, a
setting often results from food allergies. Intraoperative catecholamine with both - and -adrenergic effects.31
anaphylaxis is complicated because patients receive mul- Therapy with epinephrine remains empirical and may not
tiple drugs in a short periodincluding induction drugs, always effectively reverse vasodilation. New information
opioids, antibiotics, neuromuscular blocking drugsand suggests that vasopressin be considered as a potential
are also exposed to other antigens such as latex or drug therapeutic approach for mediator-induced vasodilatory
preservatives.2 Although the incidence of anaphylactic re- shock such as in anaphylactic reactions, especially when
actions in the perioperative setting has been suggested standard therapy is ineffective.12,13,32
to be increasing, most of the information in support of During general anesthesia, patients may have al-
this assumption is from case reports and retrospective tered sympathoadrenergic responses to acute anaphy-
studies. The drugs that present a potentially greater risk lactic shock. In addition, the patient during spinal or
for perioperative anaphylaxis in high-risk patients are the epidural anesthesia may be partially sympathectomized,
polypeptides. thereby requiring earlier intervention with even larger
Any drug administered in the perioperative period doses of epinephrine and other catecholamines.2 Invasive
has the potential to produce some form of adverse monitoring may be helpful when hypotension persists
drug reaction.2 Despite concerns about neuromuscular despite the therapeutic interventions mentioned. When
blocking drugs, agents such as antibiotics, blood products, available, the use of transesophageal echocardiography
and polypeptides (i.e., aprotinin, latex, and protamine) are in an intubated patient can be useful in diagnosing the
well documented to be associated with a more frequent cause of acute or persistent cardiovascular dysfunction.
incidence of reactions. More importantly, if a patient All patients should be admitted to an intensive care unit
has an adverse event in the perioperative period, it is (ICU) following anaphylactic reaction for 24 hours of
important that we follow the current recommendations monitoring because they may develop a recurrence of
by Dhonneur et al. and those we have made to avoid false- manifestations after successful treatment.2,33 On the ba-
positive skin tests to neuromuscular blocking drugs.28,29 sis of the efficacy of vasopressin in vasodilatory shock, it
Given the unreliability of skin testing for neuromuscular should also be considered when anaphylactic shock is not
blocking drug allergy, some have questioned whether responsive to therapy.12 Cardiopulmonary support should
other drugs such as chlorhexidine might actually be a be individualized to the patient.31
C HAPTE R 5 0/ANAP HYL AC TIC AN D ANAP HYL AC TOI D R E AC TION S 705
Bicarbonate
Is Pretreatment for Allergic
Acidemia can develop following shock and may attenuate Patients Effective?
the effect of epinephrine on the heart and systemic
vasculature. In a patient with refractory hypotension and
acidemia, sodium bicarbonate, 0.5 to 1 mEq per kg, should The anesthesia literature suggests that life-threatening
be considered and repeated as dictated by arterial blood hypersensitivity reactions are more likely to occur in
gas values. patients with a history of allergy, atopy, or asthma.2
However, this does not make it mandatory to pretreat
these patients with antihistamines and/or corticosteroids
Airway Evaluation because there are no data in the literature to suggest that
pretreatment is effective for true anaphylactic reactions.
Because marked laryngeal edema can occur, the airway Most of the literature on pretreatment is from studies
should be evaluated before extubation of the trachea. evaluating patients with previous radiocontrast media
Persistent facial edema suggests airway edema. The reactions. These reactions are nonimmunologic. Although
trachea of these patients should remain intubated until attempts to pretreat patients for anaphylactic reactions
the edema subsides. Ruling out significant air leak after to latex are growing in clinical practice, there are no
endotracheal tube cuff deflation and before extubation of data to support this as an effective preventive measure.
the trachea is useful in assessing airway patency. If there is In fact, pretreatment may lull physicians into a false
any question of airway edema, direct laryngoscopy should sense of security. Moreover, even when large doses of
be performed before the trachea is extubated. corticosteroids have been administered, life-threatening
anaphylactic reactions have still occurred.36
Refractory Hypotension
On the basis of the efficacy of arginine vasopressin
in vasodilatory shock, it should also be considered for How Should the Allergic
therapy when anaphylactic shock is unresponsive to
therapy.12,13 Vasopressin may attenuate pathologically in-
Patient Be Managed?
duced vasodilation. Furthermore, adding monitoring such
as echocardiography, preferably transesophageal, should Patients presenting with an allergic history need to
be considered in patients with refractory hypotension to be carefully evaluated. Often, patients will complain of
better determine if the mechanism is one of decreased allergy when, in fact, the reaction was a predictable drug
cardiac function or hypovolemia. side effect. For practical and medicolegal purposes, the
specific class of the offending drug should be avoided
when the history or records are consistent with an
Additional Monitoring allergic reaction, and preservative-free alternatives should
Refractory hypotension despite volume and epinephrine be chosen. The problem occurs whenever multiple drugs
administration requires additional hemodynamic moni- are simultaneously administered or when patients present
toring.31 In critically ill patients, the use of transthoracic with reactions to muscle relaxant due to the risk of
or transesophageal echocardiography for rapid assess- crossreactivity to the biquarternary ammonium ions in the
ment of intraventricular volume and ventricular func- molecule. In this situation, skin testing may be required
tion, and to determine other occult causes of acute to see what the patient can safely be administered.
cardiovascular dysfunction, can be important for accu-
rate assessment of intravascular volume and to better
support therapeutic interventions.31 Following TRALI,
fulminant noncardiogenic pulmonary edema can occur What Agents Are Implicated
after anaphylaxis. This condition requires intravascular in Perioperative Allergic
volume repletion with careful hemodynamic monitor-
ing until the capillary defect improves. It is well known
Reactions?
that colloid volume expansion is not more effective than
crystalloid volume expansion for treating anaphylactic Many agents administered in the perioperative period
shock. have the potential to produce allergic reactions upon
C HAPTE R 5 0/ANAP HYL AC TIC AN D ANAP HYL AC TOI D R E AC TION S 707
reexposure.2 The agents most often reported to cause anaphylaxis.2 Epidemiologic data from France suggest
a perioperative anaphylactic reaction are antibiotics, that neuromuscular blocking drugs are responsible for
blood products, latex, muscle relaxants, and polypep- 62% to 81% of reactions, depending on the time period
tides (protamine or aprotinin).2 A recent epidemiologic evaluated.37 Rocuronium is the neuromuscular block-
study from France reported 789 reactions.37 Anaphy- ing drug most implicated. We and others have re-
laxisdiagnosis based on clinical history, skin tests, ported previously that aminosteroidal compounds, as well
and/or specific IgE assaywas found in 518 cases (66%), as benzylisoquinoline-derived agents, produce positive
and nonimmune reactions found in 271 cases (34%). wheal and flare responses when injected intradermally.
The most common causes of anaphylaxis were NMBAs Estimates of anaphylactic reactions in anesthesia vary,
(n = 306, 58.2%), latex (n = 88, 16.7%), and antibiotics but data suggest that false-positive skin tests may overes-
(n = 79, 15.1%). Rocuronium (n = 132, 43.1%) and suc- timate the incidence of rocuronium-induced anaphylactic
cinylcholine (n = 69, 22.6%) were the most often incrim- reactions. The differences noted in the incidence of reac-
inated NMBAs. The positive predictive value of tryptase tions may reflect the potential for false-positive wheal and
for the diagnosis of anaphylaxis was 92.6%; the negative flare responses. Neuromuscular blocking drugs can also
predictive value was 54.3%.37 produce direct vasodilation. The false-positive skin tests
that were reported to be biopsy-negative for mast cell
degranulation clearly confound interpreting skin tests in
patients who have had life-threatening cardiopulmonary
LATEX collapse.20,29 Dilute solutions of neuromuscular blocking
Latex represents an environmental agent often associated drugs should be used when skin testing for potential
as an important cause of perioperative anaphylaxis.2,38 allergic reactions to these agents; however, the exact con-
Health care workers, children with spina bifida and uro- centration is unclear. Because skin testing procedures
genital abnormalities, and patients with certain food aller- are important in evaluating potential drug allergies, the
gies have also been recognized as individuals at increased threshold for direct vasodilating and false-positive effects
risk for anaphylaxis to latex.2,38 Brown reported a 24% must be determined whenever subjects are skin tested for
incidence of irritant or contact dermatitis and a 12.5% a particular drug.
incidence of latex-specific IgE positivity in anesthesiolo- In recent years, neuromuscular blocking drugs,
gists.39 Of this group, 10% were clinically asymptomatic especially steroid-derived agents, have been reported
although IgE positive. A history of atopy was also a sig- as potentially causative of anaphylactic reactions dur-
nificant risk factor for latex sensitization. Brown suggests ing anesthesia. Data involving neuromuscular blocking
these individuals are in their early stages of sensitization drugs, most recently from France, are largely based on
and perhaps, by avoiding latex exposure, their progres- skin testing; however, earlier studies have reported the
sion to symptomatic disease can be prevented. Patients steroid-derived neuromuscular blocking drugs and other
allergic to both tropical fruits (e.g., bananas, avocados, molecules produce false-positive skin tests (i.e., wheal and
and kiwis) and stone fruits (e.g., fruits with pits) have flare).18,20,26,28 One of the major problems is that anaphy-
also been reported to have antibodies that crossreact with laxis to these agents is rare in the United States but has
latex.38 Multiple attempts are being made to reduce latex been reported more often in Europe.28 Although sugges-
exposure to both health care workers and patients. If la- tions have been made that this is due to underreporting,
tex allergy occurs, strict avoidance of latex from gloves the severity of anaphylaxis and its sequelae that produce
and other sources needs to be considered, following the adverse outcomes clearly makes this unlikely on the basis
recommendations by Holzman (see Table 50.2).40 Be- of the medicolegal climate in the United States. This
cause latex is such a widespread environmental antigen, widely divergent perspective is likely a consequence of
this represents a daunting task. Despite recognizing latex how the diagnosis is made: The recommended threshold
anaphylaxis, other agents (antibiotics, induction agents, test concentrations have not been defined, and therefore
muscle relaxants, nonsteroidal anti-inflammatory drugs, results are unreliable.20,29
protamine, colloid volume expanders, and blood products) Steroid-derived drugs can induce positive wheal and
represent additional etiologic factors often responsible for flare responses independent of mast cell degranulation,
anaphylaxis in surgical patients.2 even at low concentrations, after intradermal injec-
tion.18,20,26,28 This effect is likely due to a direct effect
on the cutaneous vasculature that occurs with most neu-
romuscular blocking drugs at concentrations as low as
NEUROMUSCULAR 105 M with intradermal skin tests. A positive cutaneous
BLOCKING DRUGS reaction without evidence of mast cell degranulation was
noted at small concentrations (100 g per mL) of rocuro-
Neuromuscular blocking drugs have several unique nium in almost all volunteers. Intradermal injections are
molecular features that make them potential allergens. All frequently used to compare the cutaneous effects of anes-
neuromuscular blocking drugs are functionally divalent thetics and other drugs.18,20,26
and capable of crosslinking cell-surface IgE and causing Prick tests are often used for authenticating neuro-
mediator release from mast cells and basophils with- muscular blocking drugs as causative drugs. Dhonneur
out binding or haptenizing to larger carrier molecules. et al. evaluated 30 volunteers by using prick testing, and
Neuromuscular blocking drugs have also been implicated each subject received 10 prick tests (50 L) on both fore-
in epidemiologic studies of anesthetic drug-induced arms.29 They studied the wheal and flare responses to
708 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
Modified from: Holzman RS. Clinical management of latex-allergic children. Anesth Analg. 1997;85:529533.
prick tests with rocuronium and vecuronium by using four and 40% of the subjects had a positive skin reaction to
dilutions (1:1,000, 1:100, 1:10, and 1:1) of rocuronium undiluted rocuronium and vecuronium, respectively.29 To
and vecuronium and two controls (on both forearms). avoid false-positive results, they suggest that prick testing
Wheal and flare were immediately measured and again with rocuronium and vecuronium should be performed in
15 minutes after administration. They noted that 50% subjects who have experienced a hypersensitivity reaction
C HAPTE R 5 0/ANAP HYL AC TIC AN D ANAP HYL AC TOI D R E AC TION S 709
Anaphylaxis preparedness
Yes
Immediate intervention:
Subsequent emergency care that may
Assess airway, breathing, circulation, mentation be necessary depending on response to
epinephrine:
Inject epinephrine Consider:
Placement in recumbent position
Establish airway
O2
IV fluids
Good Consider:
clinical Epinephrine infusion
response? No H1 and H2 antihistamines
Inhaled bronchodilators
Corticosteroids
Yes Glucagon
Vasopressors
Transport to emergency department
or ICU
Observation
Good
Length and setting of observation Yes clinical
must be individualized response?
Provision of epinephrine kits
FIGURE 50.1 Algorithm for the treatment of acute anaphylaxis. ICU, intensive care unit; CPR,
cardiopulmonary resuscitation; ACLS, advanced cardiac life support. (From: Lieberman P, Kemp
SF, Openheimer J, et al. The diagnosis and management of anaphylaxis: An updated practice
parameter. J Allergy Clin Immunology. 2005;115:S483523.)
3. Anaphylaxis may occur immediately upon exposure 7. Secondary treatment of anaphylaxis includes an-
or be delayed for as long as 20 minutes. tihistamines, catecholamines, corticosteroids, and
4. Stopping antigen administration and providing air- vasopressin.
way management, 100% oxygen, epinephrine, and 8. The airway should be evaluated for persistent
volume resuscitation are the mainstays of treating edema before extubation after an anaphylactic
anaphylaxis. episode.
5. Vasopressin should be considered for hypotension 9. All neuromuscular blocking drugs can crosslink IgE
refractory to standard therapy. and cause mediator release.
6. After an anaphylactic event, patients should be 10. An established protocol for treating anaphylaxis
closely observed, as the syndrome may recur. assures the best response to an event.
C HAPTE R 5 0/ANAP HYL AC TIC AN D ANAP HYL AC TOI D R E AC TION S 711
51 Jennifer Janelle
A
jury from a 16-gauge, hollow-bore needle in the health care setting. As early as the 1880s, investi-
left on a tray after placement of a central gators recognized that jaundice was occurring in persons
catheter. The needle was visibly contam- receiving smallpox vaccinations prepared from human
inated with blood. Upon removal of his lymph, and an infectious etiology was suspected. Subse-
gloves, the anesthesiologist found a bleeding quently, outbreaks of jaundice in those vaccinated with
wound on his left thenar eminence. He washed the wound human serum were recognized in the 1930s and 1940s. In
with soap and water, called in a colleague to take over the 1949, Liebowitz et al. recognized occupationally acquired
case, and went to his occupational health department for jaundice in a blood bank nurse who sustained needle
further evaluation and treatment. pricks to her hands in the performance of her work.1 Sub-
The source patient was a homeless man with a his- sequently, multiple similar reports of jaundice in health
tory of intravenous drug abuse. He was infected with care workers (HCWs) were published. It was later recog-
hepatitis C, hepatitis B (HB), and advanced acquired nized that viral hepatitis spread through blood and body
immunodeficiency syndrome (AIDS), and had been ad- fluids was the cause of these infections. In the 1980s, an-
mitted for infective endocarditis. He had been on multi- other infectious agent spread by blood and other bodily
ple antiretroviral agents, but had significant medication fluids was identified, which caused severe abnormalities
noncompliance and had not had medical follow-up for of the immune system with subsequent development of
his human immunodeficiency virus (HIV) infection in unusual infections. This infection became known as the
>6 months. Prior resistance testing showed that his HIV human immunodeficiency virus.
virus was resistant to many antiretroviral drugs. Avoiding exposure to bloodborne pathogens is the
The anesthesiologists records were reviewed. He had primary way to prevent transmission of these infections in
completed the HB vaccine series and had adequate anti- health care settings. Over the years, multiple modifications
body on subsequent testing. He was tested for antibodies have been made in the way health care is provided,
against hepatitis C and HIV and was found to be negative such as the establishment of universal precautions and
for these infections at baseline. Liver function testing at modifications in needles and other sharps in an attempt
baseline was normal. He was offered and accepted pro- to prevent sharps injuries. In addition to these efforts,
phylactic therapy against HIV with a combination of three vaccination against HB and guidelines for management
medications selected by an infectious disease specialist in the event of exposure to bloodborne pathogens are now
based on the source patients prior resistance testing. important parts of workplace safety.
Despite mild nausea, he completed 4 weeks of antiretro- Although most infections from bloodborne pathogens
viral therapy and subsequently tested negative for both have been transmitted from patient to HCW, sometimes
hepatitis C and HIV at 12 weeks, 6 months, and 1 year. the care provider can serve as the source of infection.
Information related to HCWs infected with bloodborne
pathogens will also be covered in this chapter.
TABLE 51.1 Definitions not reliably identify all infected patients, and because
emergencies may not allow time for serologic testing.
Viral Hepatitis: An infectious disease transmitted by If these universal precautions are implemented, no addi-
blood, feces, or other body fluids that affects the tional precautions should be necessary for patients known
liver and leads to abnormalities in liver function; the to be infected with HIV. Universal precautions include the
three main viruses of concern follow: following standards:
HEPATITIS A: A viral infection spread by fecal-oral 1. All HCWs should routinely use appropriate barrier
contact that is typically transient; it is not precautions to prevent skin or mucous membrane
typically acquired in the hospital setting, and exposure when contact with blood or body fluids is ex-
because it is not a bloodborne pathogen, is not pected. Gloves should be worn for contact with blood,
a focus of this chapter body fluids, or mucous membranes or nonintact skin
HEPATITIS B AND HEPATITIS C: Infectious agents that of all patients. Gloves should be changed and hands
can be transmitted in the health care setting washed or cleaned with alcohol-based hand antiseptics
through contact with blood and other body between each patient. Masks and protective eyewear
fluids; these viruses can cause acute or chronic should be used during any procedure likely to gener-
liver disease and cirrhosis, and can potentially ate aerosolized droplets of blood or other body fluids.
lead to hepatic failure or hepatocellular Gowns or aprons should be worn during procedures
carcinoma likely to generate splashes of blood or body fluids.
Human Immunodeficiency Virus (HIV): A viral 2. Hands and other skin surfaces should be washed with
infection that causes severe immunodeficiency, with soap immediately should contamination with blood or
subsequent infections and malignancies, that is body fluids occur.
spread through contact with infected blood or body 3. All HCWs should take precautions to prevent injuries
fluids caused by sharp instruments during and after proce-
Advanced AIDS: A subset of patients with HIV infection dures. To prevent needle stick injuries, needles should
will have AIDS; the criteria for AIDS diagnosis is a not be recapped, bent or broken by hand, removed from
CD4 count <200 or the presence of certain disposable syringes, or otherwise manipulated by hand.
opportunistic infections or malignancies such as After they are used, disposable needles or other sharp
Kaposi sarcoma, toxoplasmosis, cytomegalovirus instruments should be placed in puncture-resistant
retinopathy, or colitis containers for disposal. The puncture-resistant con-
Source: A term used in the event of an exposure to a tainers should be located as close as practical to the
bloodborne pathogen to indicate the person who area of use, and should be disposed of when two thirds
was the source of the blood or body fluid full. Large bore, reusable needles should be placed
Exposed: A term used in the event of an exposure to a in a puncture-resistant container for transport to the
bloodborne pathogen to indicate the person in reprocessing area.
contact with blood or other potentially infectious 4. Although saliva has not been implicated in HIV trans-
material mission, equipment such as mouthpieces, resuscitation
Antiretroviral Agent: A drug used to treat HIV infection; bags, or other ventilation devices should be available
there are several different classes of drugs used for to use in areas where the need for resuscitation is
HIV treatment, and patients are treated with a predictable to minimize the need for emergency mouth-
combination of at least two different classes to-mouth resuscitation.
Resistance Testing: Testing (genotype or phenotype) 5. HCWs with exudative skin lesions or weeping dermati-
done on HIV virus to detect the presence of changes tis should refrain from direct patient care and should
in the virus that would predict drug failure not handle patient care equipment until this condition
resolves.
AIDS, acquired immunodeficiency syndrome.
completely, fulminant hepatitis with rapid decompen- approximately 3%. If the source blood is HBeAg-positive,
sation and death, or a chronic infection that leads to the risk increases 20% to 40%.11,12 Precise risk estimates
progressive cirrhosis and eventual liver failure. Chronic for mucocutaneous and other exposures are not available
HBV infection is also associated with increased lifetime but are presumably lower.
risk for hepatocellular carcinoma. Approximately 55% of
adults infected with HBV have no symptoms despite sero-
logic evidence of infection and serve as the main reservoir PREVENTION
for continued HBV transmission.
Passive Immunization
Diagnosis
Prophylaxis against HB infection takes two forms. Passive
The serologic markers associated with HBV infection for immunization in the form of HBIG provides temporary
which there are commercially available assays include protection from HBV. It is typically used as an adjunct to
HBV DNA by polymerase chain reaction (PCR), HB HBV vaccination in those with percutaneous or mucous
surface antigen (HBsAg) and antibody to HB surface membrane exposure to HB. It is used alone in the
antigen (anti-HBs), antibody to HB core antigen (anti- event of percutaneous or mucous membrane exposure
HBcAg), HB e antigen (HBeAg), and antibody to HBeAg to HBV in nonresponders to HBV vaccination. Passive
(anti-HBe). prophylaxis with HBIG should begin as soon as possible
HBsAg is indicative of infection with HBV, and after exposurepreferably within 24 hours.
all persons with confirmed positive HBsAg should be
considered infectious. This marker of infection appears Active Immunization
an average of 30 days from the time of exposure
(range: 6 to 60 days).5,6 It is possible to detect the Active immunoprophylaxis against HB infection is
presence of HB DNA (HBV DNA) in the serum of achieved by vaccination with HB vaccine. This vaccine
an infected person 10 to 20 days before detection of uses HBsAg, and immunity from vaccination results
HBsAg.7 HBsAg will persist in those patients who become in development of anti-HBsAg in the serum at levels
chronically infected with HBV. In those who recover >10 mIU per mL. Although there are two types of HB
from HBV infection, HBsAg is eliminated from the blood vaccines licensed in the United Statesplasma-derived
and anti-HBsAg develops. The presence of anti-HBsAg vaccine (Heptavax-B) and recombinant vaccine (Recom-
typically indicates immunity from HBV infection and bivax HB and Engerix-B)only the recombinant vaccine
can develop after natural infection or after successful is commercially available in the United States. In addi-
vaccination against HBV. In addition, anti-HBs can be tion, there are combination vaccines against both HBV
detected for several months after HB immune globulin and hepatitis A virus. Currently available HB vaccines
(HBIG) administration, but will eventually wane, and are thimerosal-free because of concerns regarding mer-
therefore there will not be indicative of long-standing cury present in thimerosal-containing vaccines.13,14 For
protection. primary vaccination, three intramuscular injections (into
HB core antibody develops at the onset of symptoms the deltoid muscle in adults and children, and into the
or biochemical abnormalities in acute HBV infection, and anterolateral thigh muscle in infants and neonates) are
persists for life. The presence of HB core antibody is given, with the second and third doses 1 and 6 months
not necessarily indicative of immunity because it can be after the initial dose.15
present in those with chronic infection.8 In most people HB vaccine provides virtually complete protection
who recover from natural infection, both anti-HBc and against the acquisition of HBV in persons who develop
anti-HBsAg will be present, whereas those responding to adequate antibody. Routine testing for immunity after
the HBV vaccine will have only anti-HBsAg. In those who vaccination is not needed in most cases for the general
are chronically infected with HBV, HBsAg and anti-HBc public, but it should be done for HCWS who have direct
will persist. patient contact or those with risk of needle stick or
HBeAg can be detected in the serum of persons with sharps injury. Postvaccination testing should be done 1 to
acute or chronic HBV infection and is indicative of high 2 months after the last dose of vaccine. If the HCW is
levels of viral replication and increased risk of infectivity.9 anti-HBs negative 1 to 2 months after the last dose of
Anti-HBeAg is associated with lower levels of virus, but it vaccine, the complete 3-dose vaccination series should be
is possible to revert to HBeAg positivity.10 repeated. Testing for anti-HBs 1 to 2 months after the last
dose of vaccine should be repeated. Failure to respond
to the second vaccination series indicates that the HCW
is a nonresponder to the HB vaccine. In the absence of
RISK OF ACQUIRING HBsAg, this worker should be considered susceptible to
HEPATITIS B IN THE EVENT HBV and should receive HBIG prophylaxis for any known
OF AN EXPOSURE or likely exposure to HBsAg-positive blood. In those that
are known to respond to HBV vaccination, there is no need
The risk of transmission from a single needle stick for booster vaccination at a later date. Even if the HBsAg
exposure varies depending on the HBeAg status of the becomes undetectable over time, protection against HBV
source case. If the source is HBeAg-negative, the risk is exists.
C H A P T E R 51/ V I R A L I N F E C T I O N S 715
TABLE 51.2 Recommendations for Postexposure Prophylaxis after Percutaneous or Mucosal Exposure to Hepatitis B
Virus in an Occupational Setting
a Persons known to have had HBV infection in the past or who are chronically infected do not require HBIG or vaccine.
b Hepatitis B immune globulin (0.06 mL/kg) administered IM.
c Adequate response is anti-HBs of at least 10 mIU/mL after vaccination.
d Revaccination, additional 3-dose series of hepatitis B vaccine administered after the primary series.
e First dose as soon as possible after exposure and the second dose 1 mo later.
f
Testing should be done as soon as possible after exposure.
HBsAg, HB surface antigen; HBIG, hepatitis B immune globulin.
716 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
any information on whether the infection is persistent or period, but the exposed person should not donate blood,
resolved. plasma, organ tissue, or semen. It is also unnecessary
to modify sexual practices or refrain from becoming
pregnant, and exposed women do not have to stop breast-
feeding.20
DIAGNOSIS The exposed persons patient care responsibilities do
The presence of antibody against hepatitis C (anti-HCV) not have to be modified because of exposure to viral
is indicative of infection. The third generation enzyme hepatitis; however, if the individual develops acute in-
immunoassay (EIA) has a sensitivity of approximately fection with HB, reevaluation at that time is indicated.
97% and can detect antibody within 6 to 8 weeks of See the section on What Are Current Recommenda-
exposure.18 tions regarding Health Care Workers Infected with a
Bloodborne Pathogen? for more information.
RISK OF ACQUIRING
HEPATITIS C IN THE EVENT What Are the Relevant
OF AN EXPOSURE Considerations Involving
The risk of hepatitis C acquisition in a nonimmune person Human Immunodeficiency
from a single needlestick exposure from a hepatitis C Virus?
positive source is approximately 1.8%.19,20
HIV is a retrovirus that damages the immune system by
infecting CD4 cells and other cells important for immune
PREVENTION function, leading to the development of characteristic
infections and malignancies. It is spread through blood
There is currently no available vaccine or immunoglobulin and body fluids, and can be transmitted by transfusion
preparation that has been shown effective in the preven- of whole blood, packed red cells, plasma, factor VIII
tion of hepatitis C acquisition. Prevention of exposure to concentrate, factor IX concentrates, and platelets. The
blood or other fluids contaminated with hepatitis C is the likelihood that a person will develop infection with HIV
most effective way to prevent acquisition. after receiving a single-donor blood product that tests
positive for HIV approaches 100%.21,22 Before serologic
testing of blood products for HIV in 1985, 0.04% of
POSTEXPOSURE 1,200,000 blood donations in the United States were
estimated to be HIV-positive.23 Currently, all blood
MANAGEMENT products are screened for HIV, and the current estimated
An individual with documented exposure to hepatitis C risk of HIV transmission through blood that is screened
should be screened for HCV antibodies and have an ALT negative for HIV is estimated to be 1 in 200,000 to 1 in
test done as soon as possible after exposure to exclude 2,000,000 per unit transfused in the United States.24
prior infection. Testing for HCV antibodies and ALT Although routine testing of blood donors has greatly
should be repeated at least once 6 months later. Testing reduced the HIV transmission through blood transfusions,
for HCV RNA and ALT 2 to 4 weeks after exposure, with transmission can still occur if the donor is infected with
referral to a hepatologist for consideration of interferon HIV but has not yet formed antibody to HIV.25 Antibody
treatment, is also done at many institutions because usually develops between 4 and 6 weeks after infection,
there is promising data regarding interferon treatment although antibody formation can take up to 1 year in
early in the infection period. In persons exposed to both someone infected with hepatitis C at the same time as
hepatitis C and HIV in the same event, testing for HIV and HIV exposure.26
hepatitis C should be repeated at 1 year, because delayed
seroconversion has been documented in this situation.
DIAGNOSIS
HIV infection is diagnosed by testing for the presence of
Are Additional Precautions antibodies to HIV. Initial testing is typically done using a
standard, enzyme-linked immunosorbent assay (ELISA)
Necessary following a Health with confirmatory testing by a Western Blot. Rapid tests
Care Workers Exposure are now available for HIV screening, particularly in the
postexposure setting and for those women presenting in
to Hepatitis B or C? labor who have not been HIV-tested. Although these tests
can be used to determine the need for treatment in the
There is no need to take any special precautions to event of HCW exposure, confirmatory tests should also be
prevent secondary transmission during the follow-up done.
C H A P T E R 51/ V I R A L I N F E C T I O N S 717
Is the source material blood or other potentially infectious material (bloody body
fluid) or an instrument contaminated with potentially infectious blood or body fluid?
No
postexposure
Yes No
prophylaxis
needed
No postexposure prophylaxis
Volume needed
Severity
Small Largeseveral drops,
few drops, major blood splash or
limited longer duration of
duration exposure (several Less severe More severe
minutes or more) solid needle, large bore hollow
superficial needle, deep
scratch puncture, e.g.,
visible blood on
device
source patient is assessed according to Figure 51.2. On the and how many agents to use and when to alter treatment,
basis of the results of these algorithms, a decision is made remains empiric.32
regarding the use of antiretroviral agents for postexposure An important goal of postexposure prophylaxis is to
prophylaxis based on Table 51.3. Multiple agents are select a regimen that allows the HCW to be compliant with
available for postexposure treatment, and the decision a 4-week course of treatment. Careful consideration must
regarding which drugs should be used can be complicated. be given to the toxicity profile of the antiretroviral agents
The selection of a treatment regimen, including which chosen, because there are multiple potential side effects
FIGURE 51.2 Algorithm to determine severity of exposure based on status of source patient. HIV,
human immunodeficiency virus; RNA, ribonucleic acid; AIDS, acquired immunodeficiency
syndrome. (From: Adapted from Centers for Disease Control and Prevention. Public health service
guidelines for the management of health-care worker exposures to HIV and recommendations for
postexposure prophylaxis. MMWR. 1988;47(RR-7):1.)
C H A P T E R 51/ V I R A L I N F E C T I O N S 719
TABLE 51.3 Recommended Human Immunodeficiency Virus (HIV) Postexposure Prophylaxis (PEP)
Infection Status of Source
Note in certain situations, such as a source patient with known antiretroviral resistance mutations or prior extensive experience with
antiretroviral agents, expert consultation with a specialist in HIV management should be considered.
Basic 2 drug PEP regimen, combination of 2 nucleoside analogs such as zidovudine and lamivudine (Combivir); Expanded PEP regimen,
typically a combination of 2 nucleoside analogs as above + protease inhibitor such as nelfinavir (Viracept) or non-nucleoside reverse
transcriptase inhibitor such as efavirenz (Sustiva); MM, mucous membrane.
Adapted from Centers for Disease Control and Prevention. Updated U.S. Public Health Service guidelines for the management of occupational
exposure to HIV and recommendations for postexposure prophylaxis. MMWR. 2005;54(RR-9):1.
including nausea, diarrhea, hematologic abnormalities, prophylaxis. Most studies have included AZT, and so
nephrolithiasis, and teratogenicity. Data reveals that 50% far, all postexposure prophylaxis regimens include AZT,
to 90% of HCWs who receive combination drugs for unless modification is needed on the basis of resistance
prophylaxis postexposure (e.g., zidovudine [AZT] plus studies on the source patients HIV strain.
lamivudine with or without a protease inhibitor) reported Given the efficacy of combination regimens for
one or more subjective side effects, and 24% to 36% treatment of HIV using nucleoside reverse transcriptase
had side effects bothersome enough that postexposure inhibitors and protease inhibitors, most experts now
prophylaxis was discontinued early.3335 recommend dual therapy with two nucleosides (e.g., AZT
and lamivudine) for a low-to-moderate risk exposure. For
Rationale a high-risk exposure, most experts would add a third
agenteither a protease inhibitor (such as nelfinavir) or
The rationale for postexposure treatment of HCWs to a non-nucleoside reverse transcriptase inhibitor (such as
prevent HIV infection is based on several lines of efavirenz)to the two nucleoside reverse transcriptase
evidence. Data suggest that systemic HIV infection does inhibitors. These medications should be started as soon
not occur immediately. It may be possible to modify as possible after the exposure, ideally within a few hours
viral replication by treatment with antiretroviral agents rather than days.
during this window of opportunity. Data from studies Figures 51.1 and 51.2 provide an algorithm for
in animal models and humans provide indirect evidence making decisions about the severity of an exposure and
of the efficacy of antiretroviral drugs for postexposure the risk that exists on the basis of the status of the
720 P E R I O P E R AT I V E C L I N I C A L C O N S I D E R AT I O N S
source patients HIV infection. If the source patient has techniques. These include vaginal hysterectomy, major
been exposed to antiretroviral agents in the past and pelvic surgery, and cardiac surgery. They should likewise
may have multiple resistance mutations, the input of an be prohibited from practice if they demonstrate medical
infectious diseases specialist, with experience in caring conditions leading to incompetence, have documented
for patients with HIV infection, should be considered. In untoward events, or refuse to follow guidelines.43
this situation, prior treatment history and resistance tests Hepatitis C and HIV are less readily transmitted. Although
could lead to recommendations for modified prophylactic these providers should double-glove for procedures, they
regimens. The United States Department of Health and should not be excluded from patient care unless they are
Human Services publishes guidelines for postexposure implicated in transmission of infection despite the use of
prophylaxis when drugs change or research suggests adequate precautions.
the need for update. These guidelines are available
at www.aidsinfo.nih.gov/guidelines. Useful information
regarding the potential side effects of these medications
is also available at this website.
KEY POINTS
1. Infections caused by bloodborne pathogens are best
avoided by careful attention to universal precautions.
What Are Current 2. Vaccination against HB is key to the prevention of
this infection in HCWs. In those unvaccinated or with
Recommendations Regarding failure to respond to the vaccine, HBIG is useful in
Health Care Workers Infected the event of exposure to HBV-infected blood.
3. There is presently no effective vaccine against hepati-
with a Bloodborne Pathogen? tis C or HIV.
4. There is no clearly effective management to prevent
Transmission of bloodborne pathogens from HCWs to the acquisition of hepatitis C in the event of exposure,
patients is known to occur. There have been at least 38 but early treatment for this infection is thought to be
outbreaks of health care worker-to-patient transmission helpful.
of HB.36,37 In almost all cases of HB transmission from 5. Postexposure management of HIV includes deter-
provider to patient, the provider was HB antigen-positive mining the need for medications, the number of
(indicative of increased infectivity). medications needed, and careful follow-up for side
Two well-publicized incidences of provider-to-patient effects that may limit compliance with these medica-
transmission of HIV have been documented. In one tions.
instance, DNA sequence analysis linked a Florida dentist 6. HCWs infected with HBV, HCV, and HIV do not
with AIDS to HIV infection in six of his patients.38 In typically need to stop involvement in direct patient
another report, an orthopedic surgeon in France may have care, but may have some specific limitations based on
transmitted HIV to one of his patients during surgery.39 procedures performed.
Despite extensive investigation, no break in infection
control precautions nor clear-cut means of transmission
REFERENCES
was documented in either case.
Reports have also linked HCW transmission of 1. Liebowitz S, Greenwald L, Cohen I, et al. Serum hepatitis in
hepatitis C to patients, including a cardiac surgeon who a blood bank worker. JAMA. 1949;140:1331.
transmitted HCV to at least five patients during valve 2. Centers for Disease Control and Prevention.. Recommen-
replacement surgery.40 An anesthesiologist in Spain may dations for prevention of HIV transmission in health-care
settings. MMWR. 1987;36(Suppl 2S):1S.
have infected >217 patients by injecting himself first with
3. Centers for Disease Control and Prevention. Recommenda-
narcotics, then giving the remainder of the drugs to his
tions for preventing transmission of infection with human
patients.41 T-lymphotrophic virus type III/lymphadenopathy associated
Current guidelines for the management of HCWs virus in the workplace. MMWR. 1985;34:682.
infected with HCV, HBV, or HIV in the United States were 4. Alter MJ, Mast EE. The epidemiology of hepatitis in the
published by The Society for Health Care Epidemiology United States. Gastroenterol Clin North Am. 1994;23:437.
in America (SHEA).42 These recommendations state that 5. Krugman S, Overby LR, Mushahwar IK, et al. Viral hepatitis,
routine testing of health care providers for HB, hepatitis type B: Studies on natural history and prevention re-
C, or HIV is not indicated. Qualified HCWs infected with examined. N Engl J Med. 1979;300:101.
bloodborne pathogens (HB, hepatitis C or HIV) should not 6. Hoofnagle JH, Di Bisceglie AM. Serologic diagnosis of acute
be routinely barred from clinical practice, as long as they and chronic viral hepatitis. Semin Liver Dis. 1991;11:73.
7. Biswas R, Tabor E, Hsia CC, et al. Comparative sensitivity
are compliant with infection control procedures and have
of HBV NATs and HBsAg assays for detection of acute HBV
not been implicated in the transmission of bloodborne
infection. Transfusion. 2003;43:788.
infections. 8. Kao JH, Chen PJ, Lai MY, et al. Acute exacerbations of
Providers who are HBeAg-positive should double- chronic hepatitis B are rarely associated with superinfection
glove for procedures and should not perform the subset of hepatitis B virus. Hepatology. 2001;34(4 Pt 1):817.
of activities linked epidemiologically with increased risk 9. Alter HJ, Seeff LB, Kaplan PM, et al. Type B hepatitis:
of transmission despite the use of good infection control The infectivity of blood positive for e antigen and DNA
C H A P T E R 51/ V I R A L I N F E C T I O N S 721
polymerase after accidental needlestick exposure. N Engl 28. Cardo DM, Culver DH, Ciesielski CA, et al. A case-control
J Med. 1976;295:909. study of HIV seroconversion in health care workers after
10. Shikata T, Karasawa T, Abe K, et al. Hepatitis B e antigen percutaneous exposure. N Engl J Med. 1997;337:1485.
and infectivity of hepatitis B virus. J Infect Dis. 1977;136:571. 29. Centers for Disease Control and Prevention. Public health
11. McMahon BJ, Holck P, Bulkow L, et al. Serologic and clinical service guidelines for the management of health-care worker
outcomes of 1536 Alaska Natives chronically infected with exposures to HIV and recommendations for postexposure
hepatitis B virus. Ann Intern Med. 2001;135:759. prophylaxis. MMWR. 1998;47(RR-7):1.
12. Gerberding JL. Management of occupational exposures to 30. Ciesielski CA, Metler RP. Duration of time between exposure
blood-borne viruses. N Engl J Med. 1995;332:444. and seroconversion in healthcare workers with occupation-
13. Sepkowitz KA. Nosocomial hepatitis and other infections ally acquired infection with human immunodeficiency virus.
transmitted by blood and blood products. In: Mandell GL, Am J Med. 1997;102(Suppl 5B):115.
Bennett JE, Dolin R, eds. Principles and practice of infec- 31. Busch MP, Satten GA. Time course of viremia and antibody
tious diseases, Chapter 295. 5th ed. New York: Churchill seroconversion following human immunodeficiency virus
Livingstone; 2000:3039. exposure. Am J Med. 1997;102(Suppl 5B):S117.
14. Centers for Disease Control and Prevention. Thimerosal in 32. Ridzon R, Gallagher K, Ciesielski C, et al. Simultaneous
vaccines: A joint statement of the American Academy of Pe- transmission of human immunodeficiency virus and hep-
diatrics and the Public Health Service. MMWR. 1999;48:563. atitis C virus from a needle-stick injury. N Engl J Med.
15. Centers for Disease Control and Prevention. Update: Ex- 1997;336:919.
panded availability of thimerosal preservative-free hepatitis 33. Gerberding JL. Occupational exposure to HIV in health care
B vaccine. MMWR. 2000;49(642):651. settings. N Engl J Med. 2003;348:826.
16. Centers for Disease Control and Prevention. Protection 34. Wang SA. The HIV PEP Registry Group. Human immun-
against viral hepatitis: Recommendations of the Immuniza- odeficiency virus (HIV) postexposure prophylaxis (PEP)
tion Practices Advisory Committee (ACIP). MMWR Morb following occupational HIV exposure: Findings from the
Mortal Wkly Rep. 1990;39(RR-2):1. HIV PEP registry. Program and Abstracts of the Infectious
17. Feinstone SM, Kapikian AZ, Purcell RH, et al. Transfusion- Diseases Society of America 35th Annual Meeting. September
associated hepatitis not due to viral hepatitis type A or B. 1316, 1997. [Abstract 482]. Alexandria: Infectious Diseases
N Engl J Med. 1975;292:767. Society of America; 1997:161.
18. Prince AM, Brotman B, Grady GF, et al. Long-incubation 35. Steger KA, Swotinsky R, Snyder S, et al. Recent experience
post-transfusion hepatitis without serological evidence of with post-exposure prophylaxis (PEP) with combination
exposure to hepatitis-B virus. Lancet. 1974;2:241. antiretrovirals for occupational exposure (OE) to HIV.
19. Thomas DL, Lemon SM. Hepatitis C. In: Mandell GL, Program and Abstracts of the Infectious Diseases Society
Bennett JE, Dolin R, eds. Principles and practice of infec- of America 35th Annual Meeting. [Abstract 480]. September
tious diseases, 5th ed. Philadelphia: Churchill Livingstone; 1316, 1997. Alexandria: Infectious Diseases Society of
2000:1736. America; 1997:161.
20. Puro V, Petrosillo N, Ippolito G. Risk of hepatitis C 36. Beekmann R, Fahrner R, Nelson L, et al. Combination
seroconversion after occupational exposures in health care post-exposure prophylaxis (PEP): A prospective study of HIV-
workers. Italian Study Group on Occupational Risk of exposed health care workers (HCW). [Abstract 481]. Program
HIV and Other Bloodborne Infections. Am J Infect Control. and Abstracts of the Infectious Diseases Society of America
1995;23:273. 35th Annual Meeting. September 1316, 1997. Alexandria:
21. US Public Health Service. Updated U.S. Public Health Infectious Diseases Society of America; 1997:161.
Service guidelines for the management of occupational 37. Bell D, Shapiro CN, Chamberland ME, et al. Preventing
exposures to HBV, HCV, and HIV and recommendations bloodborne pathogen transmission from healthcare workers
for postexposure prophylaxis. MMWR. 2001;50(RR-11):1. to patients: The CDC perspective. Surg Clin North Am. 1995;
22. Donegan E, Stuart M, Niland JC, et al. Infection with 75:1189.
the human immunodeficiency virus type 1 (HIV1) among 38. Bell DM. Healthcare worker to patient and patient-to-patient
recipients of antibody-positive blood donations. Ann Intern transmission of bloodborne infections. Fifth Annual Meeting
Med. 1990;113:733. of the Society of Healthcare Epidemiology 16 Supplement.
23. Ward JW, Deppe DA, Samson S, et al. Risk of human San Diego: Society of Healthcare Epidemiology of America;
immunodeficiency virus infection from blood donors who April 6, 1995.
later developed the acquired immunodeficiency syndrome. 39. Ciesielski C, Marianos D, Ou CY, et al. Transmission of
Ann Intern Med. 1987;106:61. human immunodeficiency virus in a dental practice. Ann
24. Ward JW, Grindon AJ, Feorino PM, et al. Laboratory and Intern Med. 1992;116:798.
epidemiologic evaluation of an enzyme immunoassay for 40. Lot F, Seguier J, Fegueux S, et al. Probably transmission of
antibodies to HTLV-III. JAMA. 1986;256:357. HIV from an orthopedic surgeon to a patient in France. Ann
25. Schreiber GB, Bush MP, Keeinman SH, et al. The risk Intern Med. 1999;130:1.
of transfusion-transmitted viral infections: The retrovirus 41. Esteban JI, Gomez J, Martell M, et al. Transmission of
epidemiology donor study. N Engl J Med. 1996;334:1685. hepatitis C by a cardiac surgeon. N Engl J Med. 1996;334:555.
26. Ward JW, Holmberg SD, Allen JR, et al. Transmission of 42. Bosch H. Hepatitis C outbreak astounds Spain. Lancet. 1998;
human immunodeficiency virus (HIV) by blood transfusions 352:1415.
screened as negative for HIV antibody. N Engl J Med. 1988; 43. AIDS/TB Committee of the Society for Healthcare Epi-
318:473. demiology of America. Management of healthcare workers
27. Horsburgh CR Jr, Ou CY, Jason J, et al. Duration of infected with hepatitis B virus, hepatitis C virus, human im-
human immunodeficiency virus infections before detection munodeficiency virus, or other bloodborne pathogens. Infect
of antibody. Lancet. 1989;2(8664):637. Control Hosp Epidemiol. 1997;18:349.
SECTION
3
EQUIPMENT
CHAPTER INVASIVE MONITORING
52
COMPLICATIONS
CASE SUMMARY autopsy reveals a right atrial perforation from the central
venous catheter with a tense hydro-hemopericardium.
67-year old, 75 kg, 5 ft 3 in. woman, status
A
The postoperative radiology report opining the catheter
post multiple bowel procedures, undergoes be withdrawn 5 cm to the level of the right mainstem
an uneventful bowel resection to relieve an bronchus reaches the medical record while the chart is in
obstruction. To enable intraoperative vol- the pathology department.
ume assessment and postoperative hyper-
alimentation, a three-lumen, 20-cm central
venous catheter is placed in the left internal jugular
vein using anatomic landmarks following unsuccessful INTRODUCTION
attempts at a right-sided placement. Pneumothorax is not All invasive monitors have associated complications
seen on postoperative chest radiograph, and the catheter that fall into three main categories: (a) placement,
tip is seen overlying the lateral right atrium. (b) interpretation, and (c) maintenance complications,
On postoperative day 2, the patient experiences an and may be further subdivided into those which occur
acute cardiovascular decline. The code team assesses the early versus late. This chapter examines arterial, central
patient and institutes the Advanced Cardiac Life Support venous, and pulmonary arterial catheterization and their
protocol for pulseless electric activity, using the central respective complications, as well as complications of
venous catheter to give medications and fluid. Subsequent transesophageal echocardiography (TEE).
P A R T I INTRA-ARTERIAL CATHETERS
Arterial catheters are the most commonly placed inva- tions frequent occurrences. Complications range from
sive monitors in the operating room. Major complica- minor to life-threatening, and are based in part on the
tions related to arterial cannulation occur in a relatively anatomic insertion point. Complications may be related
low percentage of patients.1 The prevalence of arterial to placement, interpretation, or maintenance as described
catheterization, however, makes many such complica- in Table 52.1.
725
726 EQUIPMENT
Adapted from: OGrady NP, Alexander M, Dellinger EP, et al. Centers for Disease Control and Prevention. Guidelines for the prevention of
intravascular catheter-related infections. Prevention. MMWR Recomm Rep. 2002;51(RR-10):129.
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 729
is not kept constant. If, for example, the patients bed level, and verifying that it reads zero at that location at
is raised 30 cm while the transducer is kept mounted the time monitoring is begun and each time the position
on an intravenous (IV) pole, then the height difference of the patient or transducer changes.
between the zero reference (midaxillary line) and the
transducer adds the weight of the fluid column in the
pressure monitoring tubing (30 cm H2 O) to the measured ZERO SETTING
pressure. This raises the apparent pressure by the change
in height between transducer and zero reference, that Zeroing the pressure transducer has been described as the
is, 30 cm H2 O 0.7 mm Hg per cm H2 O = 21 mm single, most important step in setting up a pressure mon-
Hg overestimation. The same effect is noted when the itoring system.21 Errors in the zero adjustment result in a
transducer is mounted on the head of the bed and the bed fixed offset (high or low) from the true value, just as with
is moved into a Trendelenburg position, or when the an improper zero reference. Changes in BP will still be ap-
transducer is patient-mounted to the wrist and the bed parent, but the measured values will differ from the actual
is tilted to the side of the transducer. The opposite effect pressures by the offset. Proper zeroing requires opening
occurs if the patients zero reference position is lowered the fluid path of the arterial catheter system to air at the
in relation to the transducer. level of the heart (midaxillary line) in a supine patient, or at
A relevant example of patient positioning is the case the level of the brain in a patient who is in a head-elevated
of the patient who is operated on in a sitting position. position. The zero value may drift over time, especially
There is a large difference (e.g., 30 cm) in height between if there is any moisture in the electric connections, and
the level of the heart and the brain. For patients in a head- therefore, periodically rechecking the zero is advisable.
elevated position, the appropriate zero reference for blood
pressure (BP) is the brain (external auditory meatus) and
not the heart. Failure to zero reference to the brain after SETTING GAIN
sitting a patient up would, in this example, lead to a 21 mm Once zeroed, the next source of error is improper
Hg overestimation of the true cerebral perfusion pressure calibration or gain of the transducer and monitor.
(see Fig. 52.1). Although this is typically taken for granted, the gain
adjustment on the monitor may be off, in which case
Diagnosis recalibration is easily performed. Ideally, calibration is:
Monitoring of a proper zero or zero reference is made by A three-point calibration to verify that the system is
opening a stopcock to air at the desired zero reference linear
Done in the range of pressures expected to be mon-
itored, for example, arterial pressure 200, 100, and
MAP 39 50 mm Hg
Done using a mercury manometer or equivalent as
reference
30 cm
One method is to connect a piece of fluid-filled
MAP 60 extension tubing of known length to the transducer, filling
it with solution and elevating it. If the gain is correct, the
monitor should display a value that is the height (length)
of the tubing 0.7 mm Hg per cm.
1 sec
fn = 1.5 Hz z = 0.11
A1 = 24
A2 = 17
Period = 1.7
FIGURE 52.2 The upper panel shows an arterial pulse waveform with two flushes. The natural
frequency and damping coefficient can be determined from either flush. The lower panel shows the
flush segment enlarged to illustrate the method. The natural frequency (fn ) of the system is estimated
by taking the period of one cycle (period), in this case 1.7 mm, and dividing this into the paper speed,
25 mm/s fn = 25/1.7 = 15 Hz. The damping coefficient is determined by taking the amplitude ratio of
successive peaks of the oscillations, in this case A2 /A1 , = 17/24 = 0.71. (From: Gardner RM. Direct
blood pressure measurementsdynamic response requirements. Anesthesiology. 1981;54:233.)
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 731
fluid path. Utilizing T-connectors is discouraged because Diagnosis is made either by obtaining a brachial cuff
they are compliant and serve as air bubble reservoirs or pressure (generally much less vulnerable to changes in
point of air entrainment during blood sampling if they peripheral resistance) or placing a femoral or axillary
have a needleless access. It is impossible to predict the artery catheter. No preventive measures are known for
relative contributions of each of these factors; therefore, this phenomenon.
if there is a clinically significant discrepancy between
invasive and noninvasive pressures, the flush test provides
a simple way to assess the systems dynamic response
CLAIMS AND LITIGATION
characteristics. This, coupled with zero reference, and The American Society of Anesthesiologist (ASA) Closed
zero and calibration verification assures the accuracy of Claims Project database uses a standardized summary
the system in use. If the test demonstrates inadequate of data collected from a group of professional liability
dynamic response, each of the potential causes should be insurance carriers.26 In the period of time from 1970
ruled out. If dynamic response, zero, and gain are correct, through 2001, only 2.1% of the 6,894 claims were related
consider other causes discussed in the subsequent text. to peripheral catheters, with only 9% of these 140
claims related to arterial catheters. Of these 13 claims,
thrombosis and iliac artery puncture represented the vast
AORTA-RADIAL PRESSURE majority (31% each).
GRADIENTS Of the seven cases involving radial artery catheteri-
zation, two involved problems with retained catheters or
Several reports describe significant disparities between wires, two involved radial nerve damage (associated with
aortic, femoral, and radial artery pressures.24,25 Although multiple punctures), two were associated with arterial oc-
these reports are specific for patients post cardiopul- clusion with resulting hand ischemia, and one resulted in
monary bypass, the analogous circumstances also exist carpal tunnel syndrome from a hematoma. Each of the
in other clinical settings. In both studies, the radial two involving arterial insufficiency occurred in patients
artery pressure underestimated the aortic or femoral with significant risk factors for malperfusion (Raynauds
pressures by at least 10 mm Hg in most patients. disease and severe peripheral vascular disease). Within
The proposed mechanisms for this phenomenon include the remainder of patients, there were two pediatric claims
vasodilatory steal or, conversely, peripheral vasocon- related to femoral arterial catheterization, with one involv-
striction. The appreciation of this phenomenon after ing leg ischemia necessitating amputation and another
cardiopulmonary bypass, or in the setting of hypovolemia that resulted in massive hemorrhage due to iliac artery
(vasoconstriction), sepsis (vasodilation), or vasodilator laceration, causing hypotension, cardiac arrest, and sub-
therapy requires the suspicion that there is a difference. sequent profound neurologic deficits.26
Injuries related to central venous catheters (CVCs) are a large prospective study by Sznajder highlight the
associated with a higher severity of injury than other importance of operator experience.30 In this study of 714
anesthesia-related complications.27 Overall, it is estimated placement attempts, the experienced physicians (50 or
that 15% of patients in whom CVCs are used will suf- more previous catheterizations) were much less likely
fer some form of mechanical, infectious, or thrombotic to fail in their catheterization attempt or to have a
complication.28,29 complication. Although complications and failures were
only about half as common with experienced physicians,
they still occurred with an approximately 6% frequency
with placement at both sites.
What Types of Complications
Can Be Expected from Central Arterial Puncture
Venous Catheters? Arterial puncture is associated with approximately 3% of
subclavian and 4% to 10% of IJV placements, depending
on practitioner experience.3032 Carotid artery puncture
ACUTE COMPLICATIONS is, by far, the most common complication associated
with IJV cannulation. It is usually without sequelae,
Complications related to CVC placement are listed in but may result in a hematoma sufficient to effect
Table 52.3. The complication rates of the two most tracheal compression or a significant extrapleural or
commonly used sites, internal jugular veins (IJVs) and mediastinal hematoma. Stroke has also been reported
the subclavian artery, are significant. The findings of from punctures of the carotid artery as well as from its
732 EQUIPMENT
TABLE 52.3 Complications due to Central Venous 1.7%, pneumothorax in 2.4%, and CVC-associated blood
Catheter Placement stream infection in 16%, which were all significantly
higher than in the group in which ultrasound guidance
Arterial puncture was used (p < 0.001).36 Average access time (skin to
Pneumothorax vein) and number of attempts were also reduced in the
Hemothorax ultrasound group compared with the landmark group
Arrhythmias (p < 0.001). Other studies have found similar results,
Malposition although a learning curve has been noted with the use of
Air embolism ultrasound to facilitate access.3640 However, Augustides
Nerve injury et al. published rates of carotid puncture of 4.2% with
Thoracic duct injury or without ultrasound-assisted needle guidance across
Perforation differences in level of training,40 with carotid puncture
-Hydro- or hemothorax rates of 0% in the hands of experienced attendings.
-Hydro- or hemomediastinum Complication rates are known to increase with repeat
-Hydro- or hemopericardium punctures, with complications as high as 54% when more
Infection than two punctures are necessary.41 With a trend toward
more frequent use of laryngeal mask airways for general
From: Gravenstein N, ed. Manual of complications during anesthe- anesthesia, it should be noted that the laryngeal mask
sia. Philadelphia: JB Lippincott Co; 1991:271, (Table 7-4). airway has been shown to alter the normal anatomic
location of the IJV with respect to the carotid artery. It
has been found that at the middle and more cephalad
cannulation, occasionally resulting in death.33 Vertebral approach points to the IJV, the overlap of the IJV
artery puncture during CVC placement has similarly been over the common carotid artery rendered a statistically
associated with fatal stroke.34 Hemothorax is reported significant increase of the overlap index (percentage of
following subclavian artery puncture. If CVC placement is carotid overlapped by the IJV), whereas the index at low
required in a patient with a coagulopathy, the subclavian access points was unchanged.42 Rotating a patients head
approach is to be avoided. Unlike with carotid artery <40 degrees also decreases the amount of internal jugular
puncture, it is difficult to apply direct pressure to the carotid overlap during IJV CVC placement.43
subclavian artery or to observe it for hematoma formation. Subclavian artery punctures secondary to jugular
A series of 1,000 IJV placements in patients with venous cannulation, although less common than carotid
coagulopathies resulted in no complications referable punctures, have also been reported.4446 It has been
to a 7% incidence of carotid artery puncture.32 One hypothesized that due to anatomic variations between
patient with a goiter required surgical decompression the right-sided and left-sided arterial structure (the right
of a venous bleeding site.32 These authors suggest that subclavian artery branches from the brachiocephalic
IJV cannulation does not result in severe complications trunk medial to the IJV), this is possibly a right-sided
in patients with a coagulopathy. Despite these findings, phenomenon and may be a consequence of either direct
the external jugular route may be preferable in patients needle puncture or inadvertent advancement of the dilator
with coagulopathy because it avoids the danger of arterial into the subclavian artery.47 Verterbral artery puncture,48
puncture. dissection, and creation of iatrogenic arteriovenous (AV)
Diagnosis of arterial puncture is made by the ap-
fistulae have also been reported with IJV approaches.49
pearance of pulsatile blood, bright red blood, direct
With respect to subclavian artery puncture during
measurement of intraluminal pressure, or a pulsating
a subclavian vein access attempt, one should avoid
needle. The absence of one of these signs does not re-
placing a subclavian catheter lateral to the juncture of
liably exclude an arterial puncture, as highlighted by a
the middle and distal thirds of the clavicle due to the
very bothersome observation where, in a series of 1,021
anatomic location of the subclavian artery behind the
attempted IJV cannulations, 5/43 arterial punctures went
vein at this level. If there is any question about which
unrecognized, and an 8F catheter sheath was placed into
vessel was entered with any central venous catheterization
the carotid artery and one patient died.35
access site, even using ultrasonographic guidance, the
intraluminal pressure should be transduced to identify
Prevention an arterial puncture (not itself a significant complication;
Traditionally, it is taught that the neck should be treated by direct, but not blood flow-obstructing, pressure
palpated to identify the location of the carotid pulse for 5 to 10 minutes) before placing a larger catheter
before IJV puncture. A newer approach is to use an or sheath (a significant complication if placed intra-
ultrasound device to identify the precise location of arterially). This can be easily accomplished with a
the target vein. Ultrasonographic guidance has been disposable length of sterile tubing attached to the access
shown to significantly reduce complications associated needle. The tubing is first held below the level of the
with IJV cannulation. In a prospective, randomized access point to fill it with blood. Then the tubing is
intensive care unit (ICU) study of 900 IJV central lifted above the level of the vessel to verify that the
venous catheterizations, the traditional use of anatomic blood column descends (venous). This method of air
landmarks resulted in puncture of the carotid artery in transduction should not be considered in spontaneously
10.6% of patients, hematoma in 8.4%, hemothorax in breathing patients or patients in whom the Trendelenburg
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 733
position is avoided (risk of air embolism), nor is it reliable catheter tip position. If air is aspirated during IJV place-
when CVCs are placed in extremely hypotensive patients, ment, an immediate chest radiograph is indicated. The
because the central venous pressure (CVP) and arterial use of nitrous oxide should be limited if possible after
pressure may be comparable. any intraoperative subclavian or difficult jugular catheter
placement because:
Pneumothorax 1. Nitrous oxide diffuses into a pneumothorax much more
rapidly than air diffuses out, and therefore the size of
Pneumothorax may occur with as many as 6% of the pneumothorax may double in <10 minutes in the
subclavian vein CVC placements. The incidence following presence of 50% N2 O.50
jugular CVC placement is <0.5% for IJV and 0% for 2. Not all pneumothoraces are evident immediately on
external jugular vein placements. postplacement chest radiograph.
3. The use of intraoperative positive pressure ventilation
Diagnosis can increase the size of a pneumothorax.
Pneumothorax is confirmed by chest radiograph. Ideally, In patients with severe chronic obstructive pulmonary
the film is taken during end expiration and in an upright disease (COPD) or ventilated with high airway pressures,
view. Clinical signs of pneumothorax include tachypnea, the external jugular route is the least likely to result in a
hyperresonance to percussion, decreased breath sounds, pneumothorax.
and, if under tension, contralateral tracheal deviation.
A supine chest radiograph may fail to reveal a small
pneumothorax because the air is anterior with lung behind
Hemothorax
it. When reviewing a supine film for pneumothorax, the Hemothorax is a complication of subclavian artery punc-
medial costophrenic sulcus should be carefully inspected, ture or laceration. It may also be associated with perfora-
because air will tend to collect in that area. If the diagnosis tion of an intrathoracic vein or the vena cava by a needle,
remains unclear, a lateral decubitus film with the affected guidewire, dilator, or catheter.
side up is helpful. Pneumothorax should be suspected
if air is aspirated during needle localization of the vein.
Absence of air aspiration does not, however, preclude the Diagnosis
appearance of a pneumothorax. The index of suspicion Hemothorax is diagnosed by chest radiograph, which
regarding the possibility of pneumothorax should be should always follow subclavian catheter placement.
increased if the clinician is inexperienced (i.e., <50
previous CVC placements by that route), and if placement Treatment
or attempted placement requires multiple needle passes.31
Treatment is by chest tube drainage. If bleeding per-
sists, surgical repair of the laceration is indicated. In
Treatment the case of a catheter-induced venous perforation, the
Treatment of pneumothorax is by tube thoracostomy. catheter should be removed and the effusion drained if it
Awake, spontaneously breathing patients may be treated is symptomatic.
conservatively, with repeated observation if the pneu-
mothorax is minimal (<20%) and asymptomatic. Any Prevention
patient being mechanically ventilated or anticipated to The preferential use of the jugular approach avoids the
require positive pressure ventilation should be treated by subclavian artery and lung. Selecting the right side for
chest tube drainage. If a chest tube is not immediately all CVC placements, regardless of route, makes venous
available, and pneumothorax is symptomatic, a large bore perforations less common because the guidewire, dilator,
IV catheter can be placed in the midclavicular line, typ- and catheter path is more direct, and less likely to result
ically in the second or third intercostal space. If severe in impinging at an acute angle on the innominate vein or
hemodynamic or respiratory compromise is related to superior vena cava (SVC) wall (see Fig. 52.3). Guidewires
CVC placement, the immediate decompression of a sus- with flexible tips are less prone to perforation. It should
pected pneumothorax takes precedence over radiographic be noted that many flexible J-tipped guidewires have
confirmation. an inflexible straight tip at the other end. When rigid
dilators are placed, they should only be inserted deep
Prevention enough to dilate the skin and subcutaneous tissue. There
Prevention of CVC placement-related pneumothorax is by is no additional benefit, and there is increased risk of
choosing the IJV over the subclavian approach. A chest ra- perforation when dilators are inserted further because
diograph following any subclavian catheter placement is a they are more likely to impinge on vessel walls (Fig. 52.3).
priority; this allows early assessment for pneumothorax as
well as verification of catheter tip position. The extremely Arrhythmias
low incidence of pneumothorax following IJV catheter
placement makes a chest radiograph a much lower pri- Arrhythmias are commonly noted during CVC place-
ority, typically taken after leaving the operating room, ment.51 They are the result of mechanical irritation of the
because the radiograph is used primarily for assessing atrium or ventricle and can be caused by either guidewire
734 EQUIPMENT
Clav
3
icie 4
Subclavian
5 vein
1st R
io
Innominate
Suprasternal notch vein
Catheter tip
Sternal angle positioning zone
Superior
vena cava
Right
atrium
Right Left
Xiphoid Inferior
process vena cava
FIGURE 52.3 Central venous access sites. (From Cook. Triple lumen central venous catheter package
insert. Cook, Critical Care; 1986.)
or catheter. Treatment consists of withdrawing the wire bronchus.52 The incidence of a malpositioned catheter is,
or catheter. Mechanically induced arrhythmias resolve in part, a function of the cannulation site selected. The
upon removal of the stimulus. The lengthy guidewire is frequency of malpositioning is considerably higher when
most commonly the culprit. A prospective study demon- the external jugular vein is used than when the IJV is
strated that rigorous control of the depth of guidewire used.53 External jugular and subclavian vein catheters
insertion markedly reduced the incidence of arrhythmias may enter the ipsilateral or contralateral subclavian vein
during pulmonary artery catheter (PAC) introducer inser- or IJVs, or may be positioned below the pericardial
tion from 58% to 15%.51 An additional benefit was that reflection. Right IJV placement appears to be associated
limiting guidewire insertion depth to <22 cm from the with the highest overall success rate for appropriate
right IJV approach virtually abolished the more hemody- intrathoracic positioning.
namically severe, ventricular arrhythmias.51 When evaluating CVCs that appear appropriately
placed on chest radiograph, one must be cognizant that
Malposition CVCs move considerably. This is true for all placement
sites studied. Up to 9.5 cm movement of the CVC occurs
Malposition of CVCs may not necessarily represent a with antecubital CVCs when the arm is adducted or
complication, per se, but clearly predisposes to subsequent abducted,54 up to 2 cm with a subclavian CVC when
complication. Malposition is defined as any catheter that is the shoulder is moved,55 and up to 4 cm with IJV CVCs
outside the catheter tip positioning zone (Fig. 52.3). This when flexing or extending the neck.56,57 This underscores
zone is the SVC from below the level of the first rib to above the need to consider a patients arm, shoulder, and
the pericardial reflection. A useful x-ray referenceto neck position when reviewing a chest radiograph to
reassure that the CVC catheter tip is in the SVC above the assess acceptability of catheter tip position. Further, most
pericardial reflectionis the takeoff of the right mainstem operating room and ICU chest radiographs are done with
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 735
the patient supine, on a hard cassette which causes the (in centimeters) to which the CVC is inserted on the
shoulders, back, and neck to be extended. This CXR basis of the formula:
position tends to place the catheter in a more proximal
position in the SVC than when the patient is not lying on patient height (cm)
RIJ insertion depth (cm) = 2
a radiograph cassette. Long arm CVCs are advanced into 10
the thorax when the arm is adducted, when the patient is Using the right subclavian vein in preference to the left,
turned on his side, or when the arm is laid across the chest. when a subclavian CVC placement is planned
CVC may also be malpositioned in the SVC within the
catheter tip positioning zone. In this case, a malposition
is considered to be one where the catheter tip has an
Air Embolism
unacceptable contact angle with the SVC. Ideally, all This complication is possible any time the pressure in
catheter tips are positioned so that they lie parallel to the a vein open to the atmosphere is below atmospheric.
SVC. This relation is sought to minimize the likelihood This can occur during or after CVC placement with each
for CVC erosion through the SVC. Unfortunately, it is spontaneous inspiration and any time the patient is in a
not always possible to achieve a parallel relation between position where the opening in the vein or break in the CVC
SVC and catheter tip. On the basis of in vitro laboratory tubing is higher than the right atrial hydrostatic pressure.
studies, the maximum angle of incidence between catheter Air embolism may also occur during CVC infusions,
tip and SVC should not exceed 40 degrees, because the especially if pressurized, from air in the infusion bag.
mechanical trauma of a CVC to the SVC is significantly
reduced over that at greater angles (see Table 52.4).58 Left-
sided placement sites are associated with more frequently Diagnosis
malpositioned (i.e., unacceptable impingement angle) Clinical signs of air embolism are nonspecific, and include
CVCs due to the course the catheter must take to reach shortness of breath and tachycardia. Air embolism should
the SVC (Fig. 52.3). be considered in any patient with a CVC who has
sudden cardiovascular collapse. Air embolism obstructs
Diagnosis pulmonary arterioles and causes an increase in dead
space. This manifests as a decrease in end-tidal CO2 , but
Diagnosis is made by chest radiograph. probably requires an embolus of at least 0.1 mL per kg to
be apparent. An average air infusion/entrainment rate of
Treatment 70 to 150 mL per second has been calculated to be fatal.59
This rate is easily attained with a pressurized infusion. It
Malpositioning is corrected by repositioning and securing
is also possible without positive pressure and may occur
the catheter.
with spontaneous inspiration during catheter placement
or disconnection. Ordway calculated the rate of air flow
Prevention through different gauge and length needles (see Fig. 52.4).
The potential for catheter malposition can be minimized Potentially lethal air entrainment rates were found,
by the following: particularly with 14-g and 16-g needles and catheters.
Larger and shorter catheters, such as PAC introducer
Using the right IJV route as a first choice sheaths, have even higher flow/pressure relations.
Using a longer catheter (e.g., 20 cm vs. 15 cm) for left-
sided placements to avoid an acute angle of contact
between the catheter tip and the SVC Treatment
Premeasuring the catheter from the placement site Once air embolism is diagnosed or suspected, it is treated
to the second rib (sternal angle). A useful alterna- by reconnecting or occluding disrupted connections and
tive for right IJV approaches is to limit the depth infusion of air. If the embolus is hemodynamically
TABLE 52.4 Effect on Vessel Perforation of Angle of Incidence Between Catheter Tip
and Simulated Vessel Walla
a
Results of in vitro perforation study with simulated vessel pulsating into catheter tip 80 times per min.
b
p < 0.05 compared with 50
c
p < 0.05 compared with 40
From Gravenstein N, Blackshear R. In vitro evaluation of relative perforating potential of central venous catheters:
Comparison of materials, selected models, number of lumens, and angles of incidence to simulated membrane.
J Clin Monit. 1991;7:1. Used with permission.
736 EQUIPMENT
Pressure versus flow in 5 cm tube air from pressurized infusion bags and tubing before
100
administration is obvious.
I.d. = 0.067
80 Nerve Injury
Flow rate
60
(mL/s)
20
I.d. = 0.047 Perforation
0 I.d. = 0.023
Perforation of a vein or the heart may occur at the time
5 10 15 of CVC placement or any time thereafter. When it occurs
Pressure gradient at the time of placement, it can be the result of needle,
B (cm H2O)
guidewire, dilator, or even the catheter itself.6365 Most
perforations do not occur at the time of placement, but
FIGURE 52.4 Inside diameters (I.d.) of 0.067 in., 0.047 in. and most present within 2 days of catheterization.63 Central
0.023 in. correspond to 14-gauge, 16-gauge, and 20-gauge venous catheter perforation is attributed to the repetitive
needles, respectively. Note the higher flow versus pressure with motion of catheter and/or vena cava or heart in relation
the 5-cm (12A) versus 25-cm (12B) catheter. (From: Ordway CB. to one another, with eventual erosion of the CVC through
Air embolus via CVP catheter without positive pressure: the vessel or chamber wall. Usually the catheter is lodged
Presentation of case and review. Ann Surg. 1974;179:480.) in the perforation site, and therefore most associated
effusions substantially reflect the fluid flowing through
the tip of the catheter, and not blood. Pericardial and
mediastinal effusions result in cardiac tamponade. Those
significant, resuscitation is potentially best accomplished,
outside the mediastinum cause pleural effusions. The
if feasible, with the patient in the Trendelenburg, left
incidence of CVC-related perforations has been estimated
lateral decubitus position.60 If nitrous oxide is in use, it is
to be 0.2%, with a mortality rate of >50%.63 In an autopsy
immediately discontinued to prevent N2 O diffusion into
series, the frequency of CVC-related vascular injury was
the embolus and enlarging it. An attempt should also be
evident when mural thrombi attributed to CVC-related
made to aspirate air from the catheter. Administration of
trauma were noted in 29% of patients with CVCs.64
100% oxygen encourages more rapid reabsorption of the
air embolus.
Diagnosis
The diagnosis of CVC perforation should be considered
Prevention any time there is unexplained cardiovascular collapse
Most reported CVC-related air emboli appear to result in a patient with a CVC in place. If the perforation
from a catheter hub fracture or disrupted connections causes a mediastinal or pericardial effusion, symptoms
(63%), or represent complications of catheter insertion are those of pericardial tamponade, that is, tachycardia,
(21%).61 Prevention is predicated on using a position that increased CVP, hypotension, pulsus paradoxus. If the
places the puncture site in a dependent position during CVC distal lumen is being monitored, loss of respiratory
cannulation, and using only Luer-type connectors. Air fluctuations may also be evident. Perforations into the
embolism along the catheter track after CVC removal is pleural space cause hydrothorax or hemothorax, as well
also a reported complication.62 Thin patients are probably as respiratory compromise which manifests as tachypnea
most vulnerable. Ideally, the patient is in a head-down and hypoxemia. Confirmation of the diagnosis is made
position during CVC removal, followed by application by aspirating from the distal lumen and obtaining
of an occlusive dressing. The advantages to purging either clear or only blood-tinged fluid (pericardium,
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 737
Empiric antibiotics
Suspicion of observe
catheter infection
Yes
Obvious infection of
No Clinical No Non-catheter No Replace catheter
insertion site
sepsis? source by guidewire
by examination
From: Wilson GL, McGregor PJ, Thompson GR. Long-term complications of intravascular cannulation. In: Venus
B, Mallory DL, eds. Problems in critical care: vascular cannulation, Vol 2(2). Philadelphia: JB Lippincott
Co; 1988:298.
CVCs should be removed or replaced: 1-cm zero reference level error induces a 0.7 mm Hg error
in pressure measurement. The absolute error from a CVP
When they are not absolutely necessary (duration of zero reference flaw is the same as for an arterial catheter,
catheterization is directly related to CVC-RBSIs) but the percentage error is much greater because CVP
Within 48 hours if strict aseptic technique cannot be values are so much smaller. The most accurate method of
ensured (i.e., medical emergency) determining the CVP is by inspecting the recorded CVP
With signs of local, site-related infection waveform at end expiration.79
If patient is hemodynamically unstable and CVC-RBSI Diagnosis of bad data is based on confirming that
is suspected the transducer is not zeroed to the proper zero reference
Routine replacement or guidewire changes are not for the patients position and/or that the relation between
recommended. Catheters should not be changed over the transducer and patient has changed. It is helpful to
a guidewire when signs of CVC-related infections are mark the zero reference point on a patients chest so that
present. everyone references to the same point, and to recheck
Evidence has shown that the site of catheterization, the zero of an electronic transducer before any change in
type of catheter, number of lumens, and experience of the therapy based on a CVP change. The zero reference point
operator may all be contributing factors for CVC-RBSI. for CVP is always the right atrial level regardless of patient
Historically, the lowest infection rates occur when using position.
subclavian catheters; however, recent studies employing
team-based risk factor modification have yielded equivo-
cal results with subclavian, IJV, and even femoral access CLAIMS AND LITIGATION
routes.77 Cutdown should be avoided. Antimicrobial or
antiseptic-impregnated catheters have been demonstrated The ASA Closed Claims Project database from 1970
to have significantly lower colonization and CVC-RBSIs through 2002 reveals that 110 out of 6,449 claims were
than nonimpregnated controls. The minimum number of associated with CVC-related injuries. Compared to other
lumens necessary for management of the patient should be claims, CVC-related injuries had a significantly higher
used to minimize potential sources of sterility breaks.66,78 severity of injury and an increased proportion of death
The overall tendency of reported data suggest CVC sep- (47% compared to 29%, p < 0.01).80 The highest mortality
sis is more likely with triple-lumen than single-lumen rates in claims due to complications from CVCs, excluding
catheters; however, there is no consensus. Triple-lumen pulmonary arterial rupture, were hemothorax (16 claims
catheters are handled much more frequently and therefore with 93% mortality), cardiac tamponade (16 claims with
the increased chance of contamination seems intuitive. 81% mortality), and air embolism (4 claims with 75%
mortality). Approximately half of all CVC-related claims
were judged to be possibly preventable, and almost
DATA ACQUISITION 20%including 16 claims for carotid artery puncture
or cannulationwere judged as possibly preventable by
Bad data from CVCs may cause inappropriate fluid use of ultrasonographic guidance or pressure waveform
administration or, conversely, restriction. This is most monitoring. In the carotid artery puncture or cannulation
commonly due to either an error in setting or maintaining category, there were five strokes, four cases of airway
the zero reference, or is a result of using an inaccurate obstruction due to hematoma, and five deaths.
measurement technique. Despite the preventability of certain CVC-related
The influence of a zero error causes a fixed offset that complications, a 2007 survey found that of almost 1,500
is equal to the hydrostatic height difference between the members of the Society of Cardiovascular Anesthesia
right atrial level (phlebostatic axis), which has the surface who responded to a survey, two-thirds still never or
landmark of the fourth intercostal space in the midaxillary almost never use ultrasonography routinely to guide
line in the supine patient. As with any transducer, each CVC placement.81
therapy is helpful in the management of circulatory disor- flotation are also mechanical but not entirely avoidable.
ders, survival benefits have been difficult to demonstrate By changing patient position to slightly head-up with
and, in fact, several studies argue that PA catheterization a right tilt during PAC advancement, there is a lower
increases hospital mortality.8386 incidence of malignant dysrhythmias. Prophylactic ad-
It is easy to advocate that more stringent standards of ministration of lidocaine is not beneficial and, therefore,
training should exist for this relatively invasive procedure. is not recommended.90 The arrhythmias are almost uni-
Iberti et al. reported mean test scores of only 67% formly reversible following advancement of the catheter,
for 496 physicians practicing in 13 US and Canadian or by withdrawal and readvancement. As high as 19%
medical facilities who voluntarily underwent testing for of patients have short runs of ventricular tachycardia
knowledge, understanding, and interpretation of PAC- or persistent premature ventricular contractions that are
derived data.87 hemodynamically significant.90,91 If ectopy persists, me-
chanical irritation is likely, and the catheter should be
removed or repositioned. Case reports of ventricular fib-
ACUTE COMPLICATIONS rillation and death secondary to these arrhythmias exist.
The frequency and potentially malignant nature of the
The large sheath used to insert PACs introduces compli- arrhythmias mandates continuous electrocardiographic
cations exceeding those of simple CVC placement. This is monitoring to diagnose and follow the response of ar-
of particular consequence for the following sequelae. rhythmias to catheter manipulation or therapy during
and after catheter placement.
Air Embolism Up to 3% of patients develop transient bundle branch
block from direct catheter trauma to the His-bundle
Air embolism from PAC introducer sheath placement is during PA catheterization.92 Right bundle branch block
much more likely to cause symptoms or be fatal than dur- is much more common than left. The potential for
ing CVC placement. Because of the large sheath diameter, complete heart block resulting from a new right bundle
there exists a potential route for massive aspiration of air branch block merits consideration each time a PAC is
(Fig. 52.4). Strict use of the Trendelenburg position and placed in a patient with a preexisting left bundle branch
occlusion of the sheath and sideport(s) is a preventive block. Prophylactic pacemaker insertion is, however, not
measure. A variety of PAC introducer sheaths withstand recommended.92 Instead, it is suggested at the time
30 cm of negative pressure without air entrainment, pro- of catheterization that either a functioning external
viding that the valve leaflets are in opposition.88 Covering pacemaker or equipment for transvenous pacemaker
the PAC introducer valve with a locking occlusive cap any insertion be made available. Alternatively, a catheter with
time a PAC is not going through the valve in the introducer pacing capability can be used.91,92 Bundle branch blocks
sheath is a preventive measure. Treatment is supportive resolve spontaneously over a variable period of time,
and aimed at preventing further air entrainment (Trende- which may last up to 24 hours.
lenburg position and/or occlude catheter). If hypotension The PA catheter may also coil, knot, and damage
occurs, the ideal resuscitation position, when feasible, is intracardiac structures, or dislodge an endocardial pacing
probably Trendelenburg and left lateral decubitus. electrode. A transvenous pacing electrode that has been
in place for at least 2 weeks is considered to be
Arterial Puncture relatively secure and unlikely to be dislodged. If resistance
to advancement or removal of a PAC is encountered,
When a sheath is placed intra-arterially, it creates an fluoroscopy assists in identifying the cause and helps
arterial injury of sufficient size that bleeding may be sig- guide further manipulation of the catheter.
nificant, and results in airway compromise and/or necessi-
tates surgical repair. This has led to the recommendation Pulmonary Artery Perforation
that after this complication, elective surgical procedures
requiring heparinization be postponed.89 Clearly, earlier PA perforation is an infrequent but often fatal compli-
identification of an arterial puncture is desirable, and cation of PAC placement and wedge pressure determina-
toward that end, some form of pressure measurement tion.93 The 2003 ASA practice guidelines on pulmonary
through the small gauge access needle or ultrasonographic arterial catheterization94 reported an incidence of PA
visualization of the guidewire in the vein is helpful to rupture of 0.03% to 1.5%. Mortality from this compli-
confirm venous access before placing the dilator sheath cation is reported as 41% to 70%. Pulmonary arterial
assembly. rupture and pseudoaneurysm formation has a reported
female preponderance of 69%.95 Other risk factors for
Arrhythmias PA rupture include pulmonary hypertension, coagulopa-
thy, systemic heparinization, age >60, cardiopulmonary
Arrhythmias occur during most PAC placements. Unlike bypass, hypothermia, peripheral catheter tip location,
the PACs noted with CVC or sheath placement, most multiple wedge pressure determinations, and atypical PA
are ventricular in nature. Arrhythmias related to the waveforms.
guidewire during sheath placement, in large part, can be Although there are other potential means for PA
avoided by limiting the depth of guidewire insertion.51 Ar- rupture, in a retrospective study involving 32,442 PA
rhythmias associated with balloon inflation and catheter catheter placements, 70% of the diagnosed PA ruptures
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 741
were associated with either balloon inflation or catheter catheter migration during bypass.98 Potential compli-
repositioning.96 The importance of keeping wedge deter- cations of catheter withdrawal during cardiopulmonary
minations to a minimum is obvious. bypass include arrhythmias from right ventricular irrita-
Proposed mechanisms of PA perforation include: tion, inability to refloat the PAC, or even venous cannula
Vessel rupture from balloon inflation (most common) obstruction by the PAC balloon.
Eccentric balloon inflation causing an unprotected Regardless of the patient or procedure, all PACs
catheter tip to be pushed into the arterial wall undergo some distal migration as they warm to body
Catheter tip-induced vessel erosion from cardiac pulsa- temperature and soften. If PA diameter decreases, as may
tion or cardiac manipulations during cardiac surgery occur with hypotension, hypovolemia, or increased airway
pressure, a catheter tip may become wedged. Prompt
The common association between this complication identification of a wedged catheter requires a continuous
and cardiac surgery, where many of the predisposing display of the PA pressure waveform. Patients with mitral
factors are present, deserves emphasis.97 valve disease may also be predisposed to undetected
PAC wedging because prominent A and V waves will
Diagnosis mask the otherwise characteristic wedge waveform. In
those patients, extreme care must be taken not to
A high index of suspicion for PA rupture should exist
position the catheter too peripherally or overwedge the
in any patient with a PAC and new onset hemoptysis
balloon. When in doubt about the catheter tip location,
(the hallmark sign), especially if it is temporally re-
it may be withdrawn to the right ventricle and refloated.
lated to PAC balloon inflation. Additional signs include
Because of the high pressures that are readily created
hypoxemia, hypotension, bronchospasm, and pleural ef-
within the PAC balloon, inflation must be done slowly
fusion. During cardiac procedures, catheter perforation
and be accompanied by continuous observation of the
may occur during bypass and may not be evident until
PA pressure waveform. Liquids should not be used to
pulmonary blood flow is reestablished upon termination
inflate the catheter, because they will generate higher
of bypass.
balloon pressures and volumes. The balloon should not
be inflated with >1.5 mL gas. If a wedge position is
Treatment reached with <1 mL inflation, the balloon is deflated
Treatment is based in part on the severity of the symptoms. and the catheter withdrawn until >1 mL, preferably the
In general, it is recommended that the PAC be withdrawn full 1.5 mL, is required to wedge. The balloon should
at least to the main PA. In some instances, the catheter never be left inflated after a measurement is completed,
may be tamponading the perforation, which causes and the syringe should be left over the inflation port to
bleeding to increase upon catheter withdrawal. If this prevent accidental fluid administration. In patients whose
occurs, the catheter should not be withdrawn further, and PA diastolic and wedge pressure correlate, it is suggested
surgical control will likely be necessary. If the patient is to follow the PA diastolic pressure to decrease the number
stable and no further bleeding is noted, close observation of wedge pressure determinations. As has been noted
with continuous monitoring and serial chest radiographs many times, each time a wedge reading is made, a
is indicated.97 If bleeding persists, the airway should conscious assessment of the relative risk (perforation)-
remain protected with an endotracheal tube. If available, a to-benefit ratio should be made.
double-lumen tube is preferable; it allows isolation of the
injured side (usually the right) and isolated application
of positive end-expiratory pressure to the affected lung to
aid in tamponading the hemorrhage. Any anticoagulation
DELAYED COMPLICATIONS
should be reversed. If bleeding persists, surgical control
and repair are necessary.
Perforation
Similar to acute perforations, this may occur any time
Prevention the balloon is inflated or the catheter tip is allowed
Hannan summarized the key to prevention of PA rupture: to remain in a position where it can erode the vessel.
Users of this catheter should expect and anticipate dis- Many spontaneous, tamponaded perforations probably
tal migration of the catheter tip.93 This admonition is occur that remain undiagnosed. On chest radiograph, they
especially true in the setting of cardiac surgery where ini- are difficult to distinguish from pulmonary infarction.
tiation of cardiopulmonary bypass is associated with an Diagnosis and preventive measures are the same as for
average spontaneous distal catheter migration of 5 cm.98 acute perforation.
Upon discontinuation of bypass, this movement places
many catheters in the wedge position. Catheter move- Infarction
ment related to cardiopulmonary bypass results from
the collapse of the right ventricle and manipulation of Obstruction of a branch of the PA may result in an
the heart. Anticoagulation and hypothermia (stiffens the ischemic lesion of a lung segment. One series has
PAC and makes it more prone to perforation) are predis- documented at least a 7% incidence of pulmonary
posing factors.98,99 Johnston et al. recommend routine ischemic lesions with PA catheterization.100 These authors
catheter withdrawal of 5 cm or more to limit distal observed spontaneous wedging from the distal migration
742 EQUIPMENT
of the catheters during the first 12 hours. Follow-up monitored. PACs are prone to the same problems of
chest radiographs and continuous pressure waveform ringing (natural frequency) and damping as are other
monitoring are necessary for diagnosis. Prevention is by catheter transducer systems. Morris found a 31% rate
withdrawing the catheter to a position where 1.5 mL of technical problems with PACs, including mean PAWP
balloon inflation is necessary to obtain a wedge reading. greater than mean pulmonary artery pressure (PAP), no
consistent arterial waveform, inability to aspirate blood
Thrombosis in wedged position, poor dynamic response, damped
tracing, and balloon overinflation.103 Most of these
This frequently occurs with PA catheterization. In one problems can be eliminated by mechanical manipulation.
study, 66% of patients had venographic or autopsy Technical problems, easily corrected, are predominantly
evidence of thrombosis at the site of PAC insertion.101 associated with leaks, air bubbles, inadequate pressure of
Usually, this thrombosis is not clinically evident, but Shah flush solutions, blood in the catheter, and malpositioned
reported several patients in whom the introducer sheath catheters.104 Morris noted clinically important differences
had been left in place for >4 days, whereupon removal ranging from 13 to +22 mm Hg between PAWP
was associated with symptoms of pulmonary embolus.91 measurements made before and after correction of
Preventive measures include use of heparin-bonded PACs technical problems.104 Criteria to evaluate the validity
and minimizing the duration of catheterization as much of PAWP are as follows:
as possible. Of note, as awareness and diagnosis of Mean wedge pressure < mean PAP
heparin-induced thrombocytopenia have increased in Phasic PAWP waveform consistent with an atrial
frequency, heparin-bonded catheters should not be used waveform
in patients with established or suspected heparin-induced Blood easily aspirated in wedged position
thrombocytopenia. Highly oxygenated wedged blood sample
Accurate PAWP measure requires fluid path conti-
Infection nuity between the catheter tip and the left atrium. This
Infection of PACs may occur at time of insertion by means the catheter tip must lie in West zone 3 of the lung
colonization from bacteria via the bloodstream or catheter where pulmonary arterial and venous pressure exceeds
track. Catheter-related sepsis is quite uncommon if the alveolar pressure at all times, that is, below the level of
catheter has been in place <72 hours; however, a catheter the left atrium. Seventy-three percent of PAC catheter tips
contamination rate of 19% is probably typical.102 At end up in the right lower quadrant of the thorax and
highest risk are those patients with catheters in place virtually all are at or below the level of the left atrium.105
for >96 hours and those with a known focus of infection. The flotation method used to position the catheter tip
Diagnosis of catheter-related sepsis requires identification favors ultimate placement in zone 3. In the supine, nor-
of the same organism on catheter segment and peripheral movolemic patient, the catheter tip is almost always in
blood culture without previous isolation from another zone 3. In patients lying in a decubitus or prone position,
site. The multiple channels and sites for infusion and especially left side down, or those in a sitting position, the
monitoring, as well as frequent adjustment of catheter PAC tip is no longer predictably in zone 3.102 Therefore,
insertion depth, all predispose PACs to infection. Strict PAWP values obtained with patients in these positions
aseptic technique during site preparation, and handling should be viewed with suspicion. Placing the patient in
and use of occlusive stopcocks and sterile catheter sleeves the left lateral decubitus position is most likely to result
are helpful adjuncts. An additional potential source of in a PAC that is not in zone 3.
contamination is the thermodilution injectate. Another source of bad data during PAWP monitoring
is positive end-expiratory pressure where alveolar pres-
sure exceeds pulmonary venous pressure and thereby
interrupts the fluid path continuity. Hypovolemia may
PROBLEMS ASSOCIATED have the same effect. The resulting PAWP is greatly
WITH DATA ACQUISITION influenced by alveolar pressure, tending to overestimate
left atrial pressure. When the PAC tip is vertically above
Errors in accuracy of pulmonary artery wedge pressure the left atrium and during administration of positive end-
(PAWP) values may lead to the erroneous diagnosis expiratory pressure, the CVP is actually a better estimate
of hypervolemia or cardiac failure (high PAWP) or of left atrial pressure than is the PAWP in patients with
hypovolemia (low PAWP). As with CVP determination, normal cardiopulmonary function.106
a consistent and proper zero referencing and zeroing A lateral chest radiograph identifies those catheter
of the transducer are extremely important. The zero tips that lie vertically above the level of the left atrium,
reference for PAWP in the supine patient is, as for CVP, the and are therefore most susceptible to the influence of
midaxillary line at the fourth intercostal space. PA catheter alveolar pressure. Another method to confirm that PAWP
transducers are affected by changes in hydrostatic level accurately reflects left atrial pressure is to obtain a blood
between the zero reference point and patient as with specimen while the PAC is wedged. This is important
other catheters (0.7 mm Hg per cm height difference). whenever wedge pressure is the central datum for clinical
Numerically small errors in the absolute value give large decision-making.104 A valid pulmonary venous specimen
percentage errors because of the relatively low pressures is obtained after clearing the PA dead space beyond the
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 743
balloon by withdrawing at least 15 mL of blood with the output results. The effect of small errors in injectate vol-
catheter tip in the wedged position. A true PA wedged ume on cardiac output determination are much less when
blood gas paired with a systemic arterial one satisfies the 10 mL versus 3 mL injectate volumes are used. A change in
following criteria:104 IV fluid administration rate just before or during cardiac
output determination will have an effect similar to that of
Wedge PO2 at least 19 mm Hg greater than systemic
changing the injectate volume. Steady state infusions do
Wedge PCO2 at least 11 mm Hg less than systemic
not pose a problem.108
Wedge pH at least 0.8 units higher than systemic
Continuous cardiac output catheters require a vari- As with peripheral pulse oximeters, PACs with con-
able period of time to determine the cardiac output. They tinuous oximetry cannot distinguish between different
do not report single bolus-derived calculations, but rather forms of hemoglobin. Carboxyhemoglobin is registered
a series of averaged cardiac outputs, depending on the as 90% oxygenated hemoglobin and 10% desaturated
users set preferences. Modern catheters are fitted with a hemoglobin, and therefore the oximeter may overestimate
heated filament, which allows automatic thermodilution the effective saturation, whereas the presence of methe-
measurement by briefly intermittently heating the blood moglobin will result in readings that trend toward 85%,
and measuring the resultant thermodilution trace. All regardless of the true saturation. When methylene blue
thermodilution type cardiac output determinations can is administered, a short-lived reduction in saturation es-
be dramatically affected by the presence of intracardiac timations is registered. Errors in determination of mixed
shunts, depending on the level and the direction of the venous oximetry may also be related to catheter position
shunt. Similarly, if the distal thermistor is engaging tissue (partially occluded catheter tip, overwedged, catheter in
rather than blood (i.e., positioned against a vessel wall), atrium or right ventricle).
the calculated CO may be erroneous.
P A R T IV TRANSESOPHAGEAL ECHOCARDIOGRAPHY
Echocardiography is a relatively innocuous diagnostic with intraoperative cardiac arrest and may directly guide
tool and uses ultrasonography at a range of approximately specific, potentially lifesaving therapy.114
20,000 Hz to insonate target tissue by electrically stimulat- Prudence dictates a careful evaluation of benefits
ing piezoelectric crystals. Ultrasonography in this range versus potential risks and complications whenever an in-
may produce mildly elevated local temperatures. Although vasive procedure is considered for either diagnosis or
high frequency ultrasonography is used in a variety of cir- management purposes. It is well established that TEE
cumstances because of its ability to cause damage (e.g., changes the course of many different operative pro-
lithotripsy), direct local toxicity to human tissue does cedures. In a retrospective series of more than 1,000
not occur from diagnostic ultrasonography. Transtho- congenital cardiac surgery patients, TEE was found to
racic echocardiography (TTE) carries virtually no risk, have a major impact in 13.8% of cases, more frequently
but yields inferior examinations to TEE in most circum- during redo surgeries, with a minor complication rate of
stances, including (but not limited to) the following:112 <1% and no major complications.115 In a similar retro-
spective review of one centers experience with more than
Localization and characterization of cardiac and para- 1,200 intraoperative TEEs performed in a 5-year period
cardiac masses on adult cardiac surgical patients, new information was
Establishing the diagnosis of aortic dissection found before bypass in 15% of patients, directly affecting
Diagnosing direct and indirect sources of cardioem- surgery in 14% of the patients. Similarly, new information
bolism was found after bypass in 6% of the patients, resulting in
Evaluating mitral prostheses a change in surgery or hemodynamic management in 4%
Endocarditis-related complications of aortic prostheses of the total.116 Although it may be argued that complex
Visualizing the left atrial appendage procedures are more likely to have additional findings,
and therefore more potential benefits, a series examining
Additionally, TTE is impractical during the course of TEE in routine coronary artery bypass grafting procedures
many intrathoracic and intra-abdominal procedures. TEE demonstrated new findings (prebypass and postbypass) in
has been used clinically in numerous scenarios outside 13% of coronary artery revascularization patients, and
of the cardiac surgical arena, such as in the intensive altered the surgical plan in 5.5% of patients.117 On the ba-
care setting and select intraoperative major vascular sis of these and other outcomes, the ASA in conjunction
procedures.113 TEE has been shown to have utility in with the Society of Cardiovascular Anesthesiologists have
providing additional diagnostic information in patients published guidelines for use of perioperative TEE.118
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 745
ASPIRATION
What Types of Complications
Aspiration may occur as a consequence of TEE examina-
Occur During Transesophageal tion in the sedated patient with an unprotected airway.
Echocardiography? More commonly, however, aspiration is likely to occur
subclinically due to alterations in gastrointestinal (GI)
Complications related to TEE are caused by direct trauma motility after TEE examination. TEE during cardiac sur-
from the probe due to placement, leaving the probe in gical procedures has been associated with a 7.8 times
position for prolonged duration, maneuvers necessary to increased risk of developing postoperative dysphagia in a
perform a complete examination, or related to sedation study comparing patients with and those without intra-
during performance of the procedure. Complications operative echocardiographic examinations,122 and likely
related to placement vary to a great degree depending predisposes to aspiration as well.
on the circumstances during which TEE is performed,
and frequently are related to the level of consciousness of
the patient and whether or not the airway is controlled. COMPRESSION
OF STRUCTURES
DISCOMFORT Compression of local structures has also been reported
following placement of TEE probes, although this compli-
Discomfort applies largely to the conscious or sedated cation is more common in the pediatric population due to
patient undergoing TEE. TEE can be performed safely size mismatch of TEE probes with small infants. Virtually
and comfortably using topical local anesthetic only; any mediastinal structure must be considered potentially
however, most patients prefer some degree of sedation. It at risk. The literature is replete with case reports involving
is necessary to suppress the gag reflex at least to a degree in TEE compression of the PA, innominate artery, tracheo-
the conscious patient, particularly for probe introduction. broncheal tree, pulmonary venous confluence, and even
After the probe is passed into the upper esophagus, it implicating TEE with recurrent laryngeal nerve palsy after
is a bit less stimulating. Transgastric maneuvers, and cardiovascular surgery.123127
in particular deep transgastric maneuvers, are less well
tolerated in patients under mild or no sedation. Severe
odynophagia has been reported at a rate of 0.1% of ARRHYTHMIAS AND
patients in one large series of 7,200 intraoperative TEE
examinations, and dental injury occurred at a rate of MYOCARDIAL ISCHEMIA
0.03%.119 For the patient undergoing TEE during general
Arrhythmias and myocardial ischemia have both been re-
anesthesia, placement may elicit a sympathetic response
ported as a complication of TEE examination,128,129 likely
not unlike laryngoscopy; therefore, caution should be
due to the stress response induced by the examinations be-
exercised, and the probe should only be placed when
ing performed under conscious sedation in these patients.
the physician is comfortable with the depth of anesthesia.
complications with TEE examinations include a splenic 6. Murphy GS, Szokol JW, Marymont JH, et al. Retrograde
laceration and esophageal dissection.133,134 Fortunately, blood flow in the brachial and axillary arteries during rou-
these appear to be relatively unique reports. tine radial arterial catheter flushing. Anesthesiology. 2006;
105:492.
7. Chang C, Dughi J, Shitabata P, et al. Air embolism and the
radial arterial line. Crit Care Med. 1988;16:141.
8. Prian GW, Wright GB, Rumack CM, et al. Apparent
KEY POINTS cerebral embolization after temporal artery catheterization.
ARTERIAL CATHETERS J Pediatr. 1978;93:115.
9. Adler DC, Bryan-Brown CW. Use of the axillary artery for
1. Low incidence of complications intravascular monitoring. Crit Care Med. 1973;1:148.
2. Zero reference to the brain if in a head-elevated 10. Wilkins RG. Radial artery cannulation and ischaemic
position damage: A review. Anaesthesia. 1985;40:896.
3. Avoid temporal and right axillary artery catheters to 11. Bedford RF. Radial arterial function following percuta-
decrease CNS embolization risk neous cannulation with 18- and 20-gauge catheters. Anes-
4. Femoral arterial punctures proximal to the inguinal thesiology. 1977;47:37.
ligament may bleed into the retroperitoneum 12. Lambert D, Martin C, Bantz P, et al. Comparison of
thrombogenic risk between teflon and polyethylene in
prolonged catheterization of the radial artery. Ann Fr Anesth
CENTRAL VENOUS CATHETERS Reanim. 1991;10:255.
13. Kim JM, Arakawa K, Bliss J. Arterial cannulation: Factors
1. The jugular route is safer than subclavian.
in the development of occlusion. Anesth Analg. 1975;54:836.
2. Right-sided catheters are less prone to complications. 14. Weiss BM, Gattiker RI. Complications during and following
3. Catheter tip should be above pericardial reflection radial artery cannulation: A prospective study. Intensive
(R mainstem bronchus origin) and parallel or <40- Care Med. 1986;12:424.
degree angle to SVC. 15. Ryan JF, Raines J, Dalton BC, et al. Arterial dynamics of
4. Catheter-related sepsis is a common source of mor- radial artery cannulation. Anesth Analg. 1973;52:1017.
bidity. 16. Cheng EY, Lauer KK, Stommel KA, et al. Evaluation of
5. Full barrier technique is required during catheter the palmer circulation by pulse oximetry. J Clin Monit.
placement. 1989;5:1.
6. Ultrasonography-guided jugular vein cannulation is 17. Mangano DT, Hickey RF. Ischemic injury following uncom-
plicated radial artery cannulation. Anesth Analg. 1979;58:55.
safer and more successful than the traditional land-
18. Band JD, Maki DG. Infections caused by arterial catheters
mark technique. used for hemodynamic monitoring. Am J Med. 1979;67:
735.
PULMONARY ARTERY CATHETERS 19. Walrath JM, Abbott NK, Caplan E, et al. Stopcock: Bacterial
contamination in invasive monitoring systems. Heart Lung.
1. All catheters migrate distally after initial placement. 1979;8:100.
2. PA perforation is the most serious complication. 20. OGrady NP, Alexander M, Dellinger EP, et al. Centers
3. PA perforation is overwhelmingly a complication of for Disease Control and Prevention. Guidelines for the
balloon inflation. prevention of intravascular catheter-related infections.
4. Accuracy of bolus-thermodilution measurement is MMWR Recomm Rep. 2002;51(RR-10):1.
very much technique dependent. 21. Gardner RM, Hollingsworth KW. Optimizing the electro-
cardiogram and pressure monitoring. Crit Care Med. 1986;
14:651.
TRANSESOPHAGEAL ECHOCARDIOGRAPHY 22. Shinozaki T, Deane RS, Mazuzan JE. The dynamic re-
sponses of liquid-filled catheter systems for direct mea-
1. Risk of complications to the alimentary tract must be
surements of blood pressure. Anesthesiology. 1980;53:
considered. 498.
2. Improves intraoperative diagnosis and intervention 23. Gardner RM. Direct blood pressure measurement
(especially in cardiac procedures). dynamic response requirements. Anesthesiology. 1981;34:
227.
REFERENCES 24. Stern DH, Gerson JI, Allen FB, et al. Can we trust
the direct radial artery pressure immediately following
1. Slogoff S, Keats AS, Arlund BS. On the safety of radial cardiopulmonary bypass? Anesthesiology. 1985;62:557.
artery cannulation. Anesthesiology. 1983;59:42. 25. Gallagher JD, Moore RA, McNicholas KW. Comparison of
2. Mandel MA, Dauchot PJ. Radial artery cannulation in 1,000 radial and femoral arterial blood pressures in children after
patients: Precautions and complications. J Hand Surg [Br]. cardiopulmonary bypass. J Clin Monit. 1985;1:168.
1977;2:482. 26. Liau DW. Injuries and liability related to peripheral
3. Hughes P, Scott C, Bodenham A. Ultrasonography of the catheters: A closed claims analysis. ASA Newsl. 2006;70:
femoral vessels in the groin: Implications for vascular 1113,16.
access. Anaesthesia. 2000;55:1198. 27. Domino KB, Bowdle TA, Posner KL, et al. Injuries and
4. Downs JB, Rackstein AD, Klein EF, et al. Hazards of radial liability related to central vascular catheters: A closed claims
artery catheterization. Anesthesiology. 1973;38:283. analysis. Anesthesiology. 2004;100:1411.
5. Lowenstein E, Little JW III, Lo HH. Prevention of cerebral 28. Mansfield PF, Hohn DC, Fornage BD, et al. Complications
embolization from flushing radial artery cannulae. N Engl and failures of subclavian-vein catheterization. N Engl J
J Med. 1971;284:1414. Med. 1994;331:1735.
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 747
29. Merrer J, De Jonghe B, Golliot F, et al. French Catheter 48. Morgan RN, Morrell DF. Internal jugular catheterization.
Study Group in Intensive Care. Complications of femoral A review of a potentially lethal hazard. Anaesthesia. 1981;
and subclavian venous catheterization in critically ill 36:512.
patients: A randomized controlled trial. JAMA. 2001;286; 49. Finlay DJ, Sanchez LA, Sicard GA. Subclavian artery in-
700. jury, vertebral artery dissection, and arteriovenous fistulae
30. Sznajder JI, Zveibil FR, Bitterman H, et al. Central vein following attempt at central line placement. Ann Vasc Surg.
catheterization: Failure and complication rates by three 2002;16:774.
percutaneous approaches. Arch Intern Med. 1986;146:259. 50. Eger EI, Saidman LJ. Hazards of nitrous oxide anesthesia
31. Schwartz AJ, Jobes DR, Greenhow DE, et al. Carotid in bowel obstruction and pneumothorax. Anesthesiology.
artery puncture with internal jugular cannulation using 1964;26:61.
the Seldinger technique: Incidence, recognition, treatment, 51. Royster RL, Johnston WE, Gravlee GP, et al. Arrhythmias
and prevention. Anesthesiology. 1979;51:S160. during venous cannulation prior to pulmonary artery
32. Goldfarb G, Lebrec D. Percutaneous cannulation of the catheter insertion. Anesth Analg. 1985;64:1214.
internal jugular vein in patients with coagulopathies: An 52. Caruso LJ, Gravenstein N, Layon AJ, et al. A better landmark
experience based on 1,000 attempts. Anesthesiology. 1982; for positioning a central venous catheter. J Clin Monit
56:321. Comput. 2002;17:331.
33. Reuber M, Dunkley LA, Turton EP, et al. Stroke after 53. Belani KG, Buckley JJ, Gordon JR, et al. Percutaneous
internal jugular venous cannulation. Acta Neurol Scand. cervical central venous line placement: A comparison of
2002;105:215. the internal and external jugular vein routes. Anesth Analg.
34. Sloan MA, Mueller JD, Adelman LS, et al. Complications of 1980;59:40.
venous catheterization. Neurology. 1992;41:1092. 54. Kalso E, Rosenberg PH, Vuorialho M, et al. How much do
35. Jobes DR, Schwartz AJ, Greenhow DE, et al. Safer jugular arm movements displace cubital central venous catheters?
vein cannulation: Recognition of arterial puncture and pref- Acta Anesth Scand. 1982;26:354.
erential use of the external jugular route. Anesthesiology. 55. Dijk B, Bakker PM. Appraisal of the dislocation of central
1983;59:353. venous catheter tips using subclavian and arm veins.
36. Karakitsos D, Labropoulos N, DeGroot E, et al. Real- Anaesthetist. 1977;26:138.
time ultrasound-guided catheterisation of the internal 56. Curelaru VI, Lindder LE, Gustavvson B. Displacement of
jugular vein: A prospective comparison with the landmark catheters inserted through the internal jugular vein with
technique in critical care patients. Crit Care. 2006;10:R162. neck flexion and extension: A preliminary study. Intensive
37. Hind D, Calvert N, McWilliams R, et al. Ultrasounic locating Care Med. 1980;6:179.
devices for central venous cannulation: Meta-analysis. Br 57. Lee DS, Kuhn J, Schaffer MJ, et al. Migration of tips of
Med J. 2003;327:361. central venous catheters in seated patients. Anesth Analg.
38. Randolph AG, Cook DJ, Gonzales CA, et al. Ultrasound 1984;63:949.
guidance for placement of central venous catheters: A meta- 58. Gravenstein N, Blackshear R. In-vitro evaluation of relative
analysis of the literature. Crit Care Med. 1996;24:2053. perforating potential of central venous catheters: Compar-
39. Denys BG, Uretsky BF. Anatomical variations of internal ison of materials, selected models, number of lumens, and
jugular vein location: Impact on central venous access. Crit angles of incidence to simulated membrane. J Clin Monit.
Care Med. 1991;19:1516. 1991;7:1.
40. Augoustides JG, Horak J, Ochroch AE, et al. A random- 59. Ordway CB. Air embolism via CVP catheter without positive
ized controlled clinical trial of real-time needle-guided pressure: Presentation of case and review. Ann Surg.
ultrasound for internal jugular venous cannulation in a 1973;179:479.
large university anesthesia department. J Cardiothorac Vasc 60. Durant TM, Long J, Oppenheimer MJ. Pulmonary (venous)
Anesth. 2005;19:310. air embolism. Am Heart J. 1947;33:269.
41. Eisen LA, Narasimhan M, Berger JS, et al. Mechanical 61. Kashuk JL. Air embolism after central venous catheteriza-
complications of central venous catheters. J Intensive Care tion. Surg Gynecol Obstet. 1984;159:249.
Med. 2006;21:40. 62. Roberts S, Johnson M, Davies S. Near-fatal air embolism:
42. Takeyama K, Kobayashi H, Suzuki T. Optimal puncture Fibrin sheath as the portal of air entry. South Med J.
site of the right internal jugular vein after laryngeal mask 2003;96:1036.
airway placement. Anesthesiology. 2005;103:1136. 63. Aldridge HE, Jay A. Central venous catheters and heart
43. Sulek CA, Blas ML, Lobato EB. A randomized study of perforation. Can Med Assoc J. 1986;135:1082.
left versus right internal jugular vein cannulation in adults. 64. Ducatman BS, McMillan JC, Edward WD. Catheter-induced
J Clin Anesth. 2000;12:142. lesions of the right side of the hearta one year prospective
44. Huddy SP, McEwan A, Sabbat J, et al. Giant false aneurysm study of 141 autopsies. JAMA. 1985;253:791.
of the subclavian artery. An unusual complication of 65. Kapadia CB, Heard SO, Yeston NS. Delayed recognition of
internal jugular venous cannulation. Anaesthesia. 1989;44: vascular complications caused by central venous catheters.
588. J Clin Monit. 1988;4:267.
45. Powell H, Smallman JM, Morgan M. Internal jugular 66. Pemberton LB, Lyman B, Lander V, et al. Sepsis from triple
catheterisation. Case report of a potentially fatal hazard. vs. single-lumen catheters during total parenteral nutrition
Anaesthesia. 1990;45:458. in surgical or critically ill patients. Arch Surg. 1986;121:
46. Jain U, Shah KB, Belusko RJ, et al. Subclavian artery lac- 591.
eration and acute hemothorax on attempted internal jugu- 67. Elliott TSJ, Moss HA, Tepps SE, et al. Novel approach
lar vein cannulation. J Cardiothorac Vasc Anesth. 1991;5: to investigate a source of microbial contamination of
608. central venous catheters. Eur J Clin Microbiol Infect Dis.
47. Kulvatunyou N, Heard SO, Bankey PE. A subclavian artery 1997;16:210.
injury, secondary to internal jugular vein cannulation, 68. Jeske C, Raedler C, von Goedecke A, et al. Early iden-
is a predictable right-sided phenomenon. Anesth Analg. tification of bacteria leading to central venous catheter
2002;95:564. contamination. Anesth Analg. 2003;97:940.
748 EQUIPMENT
69. Digiovine B, Chenoweth C, Watts C, et al. The attributable 89. McEnany MT, Austen GW. Life threatening hemorrhage
mortality and costs of primary nosocomial bloodstream from inadvertent cervical arteriotomy. Ann Thorac Surg.
infections in the intensive care unit. Am J Respir Crit Care 1977;24:234.
Med. 1999;160:976. 90. Salmenpera M, Peltola K, Rosenberg P. Does prophylactic
70. Rello J, Ochagavia A, Sabanes E, et al. Evaluation of lidocaine control cardiac arrhythmias associated with pul-
outcome of intravenous catheter-related infections in crit- monary artery catheterization? Anesthesiology. 1982;56:210.
ically ill patients. Am J Respir Crit Care Med. 2000;162: 91. Shah KB, Rao TLK, Laughlin S, et al. A review of pulmonary
1027. artery catheterization in 6,245 patients. Anesthesiology.
71. Dimick JB, Pelz RK, Consunji R, et al. Increased resource 1984;61:27.
use associated with catheter-related bloodstream infection 92. Sprung CL, Elser B, Schein RMH, et al. Risk of right bundle-
in the surgical intensive care unit. Arch Surg. 2001;136:229. branch block and complete heart block during pulmonary
72. Mermel LA. Prevention of intravascular catheter-related artery catheterization. Crit Care Med. 1989;17:1.
infections. Ann Intern Med. 2000;133:229. 93. Hannan AT, Brown M, Bigman O. Pulmonary artery
73. OGrady NP, Alexander M, Dellinger EP, et al. Healthcare catheter-induced hemorrhage. Chest. 1984;85:128.
Infection Control Practices Advisory Committee. Guide- 94. American Society of Anesthesiologists Task Force on Pul-
lines for the prevention of intravascular catheter-related monary Artery Catheterization. Practice guidelines for pul-
infections. Infect Control Hosp Epidemiol. 2002;23:759. monary artery catheterization. An updated report by the
74. Crnich CJ, Maki DG. The promise of novel technology for American Society of Anesthesiologists task force on pul-
the prevention of intravascular device-related bloodstream monary artery catheterization. Anesthesiology. 2003;99:988.
infection. I. Pathogenesis and short-term devices. Clin Infect 95. Poplausky MR, Rozenblit G, Rundback JH, et al. Swan-
Dis. 2002;34:1232. Ganz catheter-induced pulmonary artery pseudoaneurysm
75. Garland JS, Alex CP, Mueller CD, et al. A randomized trial formation: Three case reports and a review of the literature.
comparing povidone-iodine to a chlorhexidine gluconate- Chest. 2001;120:2105.
impregnated dressing for prevention of central venous 96. Kearney TJ, Shabot MM. Pulmonary artery rupture associ-
catheter infections in neonates. Pediatrics. 2001;107:1431. ated with the Swan-Ganz catheter. Chest. 1995;108:1349.
76. Maki DG, Ringer M, Alvarado CJ. Prospective randomised 97. Muller BJ, Gallucci A. Pulmonary artery catheter-induced
trial of povidone-iodine, alcohol, and chlorhexidine for pulmonary artery rupture in patients undergoing cardiac
prevention of infection associated with central venous and surgery. Can Anesth Soc J. 1985;32:258.
arterial catheters. Lancet. 1991;338(8763):339. 98. Johnston WE, Royster RL, Choplin RH, et al. Pulmonary
77. Deshpande KS, Hatem C, Ulrich HL, et al. The incidence artery catheter migration during cardiac surgery. Anesthe-
of infectious complications of central venous catheters at siology. 1986;64:258.
the subclavian, internal jugular, and femoral sites in an 99. Cohen JA, Gravenstein N, Blackshear RH, et al. More
intensive care unit population. Crit Care Med. 2005;33:13. frequent perforation by pulmonary artery catheters during
78. Lee RB, Buckner M, Sharp KW. Do multi-lumen catheters hypothermia. Anesth Analg. 1989;68:555.
increase central venous catheter sepsis compared to single- 100. Foote GA, Schabel SI, Hodges M. Pulmonary complications
lumen catheters? J Trauma. 1988;28:1472. of the flow-directed balloon-tipped catheter. N Engl J Med.
79. Verweij J, Kester A, Stroes W, et al. Comparison of three 1974;290:927.
methods for measuring central venous pressure. Crit Care 101. Chastre J, Cornud F, Bouchama A, et al. Thrombosis
Med. 1986;14:288. as a complication of pulmonary-artery catheterization
80. Domino KB, Bowdle TA, Posner KL, et al. Injuries and via the internal jugular vein: Prospective evaluation by
liability related to central vascular catheters: A closed claims phlebography. N Engl J Med. 1982;306:278.
analysis. Anesthesiology. 2004;100:1411. 102. Michel L, Marsh HM, McMichan JC. Infection of pulmonary
81. Bailey PL, Blance LG, Eaton MP, et al. A survey of the use artery catheters in critically ill patients. JAMA. 1981;
of ultrasound during central venous catheterization. Anesth 245:1032.
Analg. 2007;104:491. 103. Morris AH, Chapman RH, Gardner RM. Frequency of
82. Squara P, Bennett D, Perret C. Pulmonary artery catheter: wedge pressure errors in the ICU. Crit Care Med. 1985;13:
Does the problem lie in the users? Chest. 2002;121:2009. 705.
83. Pulmonary Artery Catheter Consensus Conference. Consen- 104. Morris AH, Chapman RH. Wedge pressure confirmation
sus statement. Crit Care Med. 1997;25:910. by aspiration of pulmonary capillary blood. Crit Care Med.
84. Parker MM, Peruzzi W. Pulmonary artery catheters in 1985;13:756.
sepsis/septic shock. New Horiz. 1997;5:228. 105. Kronberg GM, Quan SF, Schlobohm RM. Anatomic loca-
85. Ivanov RI, Allen J, Sandham JD, et al. Pulmonary artery tions of the tips of pulmonary-artery catheters in supine
catheterization: A narrative and systematic critique of patients. Anesthesiology. 1979;51:467.
randomized trials and recommendations for the future. 106. Rajacich N, Burchard KW, Faysal MH, et al. Central venous
New Horiz. 1997;5:268. and pulmonary capillary wedge pressure as estimates of left
86. Sibbald W, Keenan S. Show me the evidence: A critical atrial pressure: Effects of positive end-expiratory pressure
appraisal of the Pulmonary Artery Catheter Consensus and catheter tip malposition. Crit Care Med. 1989;17:7.
conference and other musings on how critical practitioners 107. Levett JM, Replogle RL. Thermodilution cardiac output:
need to improve the way to conduct business. Crit Care Med. A critical analysis and review of the literature. J Surg Res.
1997;25:2060. 1979;27:392.
87. Iberti TJ, Fischer EP, Leibowitz AB, et al. Pulmonary 108. Pearl RG, Rosenthal MH, Nielson L, et al. Effect of injectate
Artery Catheter Study Group. A multicenter study of volume and temperature on thermodilution cardiac output
physicians knowledge of the pulmonary artery catheter. determination. Anesthesiology. 1986;64:798.
JAMA. 1990;264:2928. 109. Elkayam U, Berkley R, Azen S. Cardiac output by ther-
88. Conahan TJ, Barberii JK, Calkins JM. Valve competency modilution technique: Effect of injectates volume and
in pulmonary artery catheter introducers. Anesthesiology. temperature on accuracy and reproducibility in the crit-
1983;58:189. ically ill patient. Chest. 1983;84:418.
C H A P T E R 5 2 / I N V A S I V E M O N I T O R I N G C O M P L I C AT I O N S 749
110. American Edwards Laboratories. Models 9520 and 9520A. 123. Preisman S, Yusim Y, Mishali D, et al. Compression of
Cardiac output computer operations and troubleshooting the pulmonary artery during transesophageal echocardio-
manual. 1983. graphy in a pediatric cardiac patient. Anesth Analg. 2003;
111. Stevens JH, Raffin TA, Mihm FG, et al. Thermodilution 96:85.
cardiac output measurementeffects of the respiratory 124. Frommelt PC, Stuth EA. Transesophageal echocardio-
cycle on its reproducibility. JAMA. 1985;253:2240. graphic in total anomalous pulmonary venous drainage:
112. Khandheria BK, Seward JB, Tajik AJ. Transesophageal Hypotension caused by compression of the pulmonary
echocardiography. Mayo Clin Proc. 1994;69:856. venous confluence during probe passage. J Am Soc Echocar-
113. Joseph MX, Disney PJ, Da Costa R, et al. Transthoracic diogr. 1994;7:652.
echocardiography to identify or exclude cardiac cause of 125. Janelle GM, Lobato EB, Tang YS. An unusual complication
shock. Chest. 2004;126:1592. of transesophageal echocardiography. J Cardiothorac Vasc
114. Memtsoudis SG, Rosenberger P, Loffler M, et al. The Anesth. 1999;13:233.
usefulness of transesophageal echocardiography during 126. Arima H, Sobue K, Tanaka S, et al. Airway obstruction
intraoperative cardiac arrest in noncardiac surgery. Anesth associated with transesophageal echocardiography in a
Analg. 2006;102:1653. patient with a giant aortic pseudoaneurysm. Anesth Analg.
115. Randolph GR, Hagler DJ, Connolly HM, et al. Intraopera- 2002;95:558.
tive transesophageal echocardiography during surgery for 127. Kawahito S, Kitahata H, Kimura H, et al. Recurrent
congenital heart defects. J Thorac Cardiovasc Surg. 2002; laryngeal nerve palsy after cardiovascular surgery: Rela-
124:1176. tionship to the placement of a transesophageal echocar-
116. Click RL, Abel MD, Schaff HV. Intraoperative trans- diographic probe. J Cardiothorac Vasc Anesth. 1999;13:
esophageal echocardiography: 5-year prospective review 528.
of impact on surgical management. Mayo Clin Proc. 2000; 128. Goland S, Shimoni S, Attali M, et al. Fatal ventricular
75:241. arrhythmia as a complication of transesophageal echocar-
117. Qaddoura FE, Abel MD, Mecklenburg KL, et al. Role diography. Eur J Echocardiogr. 2005;6:151.
of intraoperative transesophageal echocardiography in 129. Sovova E, Marek D, Lukl J, et al. Acute ischemia as a
patients having coronary artery bypass graft surgery. Ann complication by transoesophageal echocardiography. Eur J
Thorac Surg. 2004;78:1586. Echocardiogr. 2005;6:228.
118. American Society of Anesthesiologists. Practice guide- 130. Min JK, Spencer KT, Furlong KT, et al. Clinical features
lines for perioperative transesophageal echocardiography. of complications from transesophageal echocardiography:
A report by the American Society of Anesthesiologists A single-center case series of 10,000 consecutive examina-
and the Society of Cardiovascular Anesthesiologists Task tions. J Am Soc Echocardiogr. 2005;18:925.
Force on Transesophageal Echocardiography. Anesthesiol- 131. Greene MA, Alexander JA, Knauf DG, et al. Endoscopic
ogy. 1996;84:986. evaluation of the esophagus in infants and children im-
119. Kallmeyer IJ, Collard CD, Fox JA, et al. The safety of mediately following intraoperative use of transesophageal
intraoperative transesophageal echocardiography: A case echocardiography. Chest. 1999;116:1247.
series of 7200 cardiac surgical patients. Anesth Analg. 132. Lennon MJ, Gibbs NM, Weightman WM, et al. Trans-
2001;92:1126. esophageal echocardiography-related gastrointestinal com-
120. Yamamoto H, Fujimura N, Namiki A. Swelling of the plications in cardiac surgical patients. J Cardiothorac Vasc
tongue after intraoperative monitoring by transesophageal Anesth. 2005;19:141.
echocardiography. Masui. 2001;50:1250. 133. Olenchock SA Jr, Lukaszczyk JJ, Reed J III, et al. Splenic
121. Liu JH, Hartnick CJ, Rutter MJ, et al. Subglottic stenosis injury after intraoperative transesophageal echocardiogra-
associated with transesophageal echocardiography. Int J phy. Ann Thorac Surg. 2001;72:2141.
Pediatr Otorhinolaryngol Sep. 2000;15:55. 134. El-Chami MF, Martin RP, Lerakis S. Esophageal dissection
122. Rousou JA, Tighe DA, Garb JL, et al. Risk of dysphagia complicating transesophageal echocardiogramthe lesson
after transesophageal echocardiography during cardiac to be learned: Do not force the issue. J Am Soc Echocardiogr.
operations. Ann Thorac Surg. 2000;69:486. 2006;19:579, e5e7.
CHAPTER CARDIOPULMONARY BYPASS
A
combined procedure: Aortic valve replace- ing heparin reversal with protamine, a dry surgical field
ment, multivessel coronary artery bypass before chest closure, and minimal chest tube drainage
graft (CABG), and a left carotid endarterec- upon arrival in the ICU, was followed by a series of postop-
tomy (CEA). The patient has a history of erative complications. Complications included left-sided
longstanding hypertension, tobacco abuse, weakness upon awakening, new onset atrial fibrillation on
chronic obstructive pulmonary disease (COPD), diffuse postoperative day (POD) 2, and an increased creatinine to
coronary artery disease with 90% left main stenosis, 2.6 mg per dL.
moderate/severe aortic stenosis (51 mm Hg gradient; aor- Before discharge to a facility for continued rehabilita-
tic valve area: 1.0 cm2 ), preserved left ventricular (LV) tion of his cerebrovascular accident (CVA) symptoms, the
function (ejection fraction >50% and moderate LV hy- patients creatinine returned to baseline and, following
pertrophy), and peripheral vascular disease (severe left rate control intervention with a calcium channel blocker,
internal carotid artery stenosis and a 9-cm infrarenal ab- atrial fibrillation resolved by POD 6.
dominal aortic aneurism). Current medications include
metoprolol, lisinopril, aspirin, and subcutaneous heparin.
On the day of surgery, vital signs were: Pulse 67 bpm,
blood pressure 101/63 mm Hg, respiratory rate 20 per INTRODUCTION
minute, SpO2 97% to 98% with 2 L of O2 nasal can-
nula, hematocrit (Hct) 31%, and creatinine 1.6 mg per dL. Complications secondary to CPB are varied, as they oc-
Following induction with etomidate, cisatracurium, and cur in virtually every organ system for numerous reasons.
fentanyl, a right internal jugular central venous catheter Anticipated mechanical problems associated with initia-
(CVC) using ultrasonographic guidance and a trans- tion and/or maintenance of CPB (e.g., obstructed venous
esophageal echocardiography (TEE) probe were placed. return, malpositioning of aortic cannula, air embolism,
TEE examination confirmed the preoperative studies. The aortic dissection, oxygenator failure, arterial pump fail-
maintenance anesthetic was sevoflurane, O2 -air mixture, ure, and total electric failure) can yield catastrophic
cisatracurium, and fentanyl. A bolus, followed by an results. These anticipated problems can largely be avoided
infusion of aminocaproic acid was initiated following with vigilant observation and monitoring by the anesthe-
placement of the CVC. Surgically, an initial left CEA, siologist, perfusionist, and surgeon. Their coordinated
utilizing procedural heparin and a shunt during carotid efforts are required for successful management of these
clamping, was followed by a 120-minute aortic cross- intraoperative problems. The incidence, diagnosis, and
clamp and a total cardiopulmonary bypass (CPB) time management of these critical events during CPB have
of 180 minutes for a four-vessel CABG with a left inter- been reviewed thoroughly.1,2
nal mammary artery (LIMA) to left anterior descending Other complications that may not be specific to CPB
(LAD) artery, and the placement of a 23-mm porcine aor- or cardiac surgery, but reported to occur with similar
tic valve. A seemingly uneventful intraoperative course incidences as other surgical procedures in the perioper-
that included the administration of 2 units of packed red ative period, are described elsewhere in this monograph.
blood cells (RBCs) to maintain Hct >25% during CPB Perhaps a noteworthy exception is that of injury to the
moderate hypothermia (32 C) and a maximum warm- ulnar nerve and brachial plexus3,4 where the incidence in
ing temperature of 37 C during CPB, maintenance of cardiac surgery (2% to 38%) exceeds that of noncardiac
higher pump flows and mean arterial pressure (MAP) surgery (0.02% to 0.06%). Apart from the controversy over
between 60 to 80 mm Hg, 250 mL urine during CPB, ideal patient positioning, multiple mechanisms have been
750
C H A P T E R 5 3 / C A R D I O P U L M O N A R Y BY PA S S 751
Endothelium
CPB-pump CPB-pump
and/or
Extracorporeal Extracorporeal
subendothelial tissue
circulation circulation
(surgical wound)
Activated
platelets C5a and C3
Prothrombin
PAI-1
Thrombin tPA
Fibrinogen Fibrin
Plasminogen Plasmin
Inflammation
(activated cytokines)
Coagulation Fibrinolysis/bleeding (activated monocytes)
postulated for increased incidence during cardiac surgery The resulting thrombin formation and platelet activation
(e.g., excessive sternal retraction, internal mammary har- produce a generalized procoagulant state. A reactive
vesting, duration of CPB and surgery, hypothermia, injury endothelium responding to input from circulating cy-
to the first rib or to the internal jugular vein). Again, vigi- tokines, having been generated by both hemostatic and
lance on the part of the entire cardiac team with regard to inflammatory pathways, further enhances thrombin acti-
patient positioning and to factors pertaining to the above- vation and platelet adhesion. CPB also activates and alters
mentioned contributing mechanisms can be advocated to the fibrinolytic pathway. Under normal circumstances,
minimize this complication. this pathway is involved in limiting coagulation and pro-
Unpredictable or unexpected complications as a re- moting healing; however, during CPB, this pathway may
sult of CPB, however, can result from the interplay of the be responsible for excessive blood loss. Lastly, the acti-
patients pathophysiology and the normal physiologic re- vation of coagulation and fibrinolytic cascades, as well
sponse to CPB. This chapter focuses on the complications as the direct contact with the perfusion circuit, serve to
arising from the pathophysiologic responses to bypass. It activate enzyme cascades and monocytes involved in in-
is in this area that the anesthesiologist, by anticipating flammation. These inflammatory pathways may already
potential adverse events, may significantly impact patient be primed by the patients underlying pathophysiol-
management decisions. Therefore, a clear understanding ogy, and in turn can exacerbate the activation of both
of the physiologic response to CPB is required. thrombogenic and fibrinolytic pathways. The activation
of thrombogenic, fibrinolytic, and inflammatory pathways
does not occur in a linear manner, but rather, all these
pathways appear to be stimulated simultaneously (see
What Are the Hemostatic Fig. 53.1).
to maintain a vasodilated state but also to inhibit platelet in apoptosis of endothelium, the dead cells then becoming
activation. Endothelial production of heparin sulfate, chemotactic for platelets and procoagulant. Evidence
thrombomodulin, and protein C also act as endogenous for all these endothelial changes has been observed in
anticoagulant factors on the venous side by their action on experimental models of CPB and may eventually explain
platelet adhesion/activation, activation of protein C, and the variety of responses of patients to similar stimuli.
cleavage of coagulation factors Va and VIIIa, respectively. Endothelial cells are also activated by cytokines
An intact epithelium is therefore required for normal involved with inflammation, including C5a, IL-1, and
platelet action on both venous and arterial sides of the tumor necrosis factor (TNF).18,19 Epithelial activation
vasculature. through inflammatory mediators triggers the endothelium
During damage or trauma, exposure of platelets to to express procoagulant and fibrinolytic proteins and
arterial subendothelial collagen and vWF causes rapid ac- also activates the endothelial expression of membrane
tivation of platelets and adhesion on endothelial surfaces adhesion molecules specific to the adhesion and eventual
because of low affinity binding between vWF monomers transmigration of neutrophils and monocytes.20
and glycoprotein I (GPI) of platelet membranes. This Endothelial cells are sensitive to hypoxia, temperature
binding causes the platelet to roll along the endothelial changes, circulating cytokines, endotoxin, cholesterol,
surface and further activate GPII and GPIII proteins on nicotine, surgical manipulation, and hemodynamic sheer
the platelet surface to firmly bind the platelet to vWF stress. Events related to CPB, the surgical process itself, or
multimers and fibrinogen. Activated platelets also pro- reperfusion of previously injured tissue can therefore glob-
vide a negatively charged phospholipid surface on which ally stimulate endothelium to become more procoagulant.
coagulation reactions can occur and receptors for coag- Preexisting influences such as hypertension, smoking, di-
ulation factors Xa, IXa, and Va. On the venous side, the abetes, or hyperlipidemia also disrupt endothelial cells.
circulating thrombin generated through both extrinsic Therefore the hypertensive patient with a history of smok-
and intrinsic coagulation may be the most likely stimu- ing, hyperlipidemia, and type 2 diabetes probably enters
lus to activate platelets. Heparin increases the sensitivity the operating suite with an activated endothelium that,
of platelets to soluble agonists, inhibits binding to vWF, when exposed to perturbations resulting from cardiac
and modestly increases template bleeding times. As CPB surgery and/or CPB, can result in marked global activa-
continues, activated complement, plasmin, hypothermia, tion of thrombin formation, platelet consumption, and
platelet activating factor, interleukin 6 (IL-6), cathepsin activation of fibrinolytic pathways.7
G, serotonin, epinephrine and other agonists also activate
platelets and contribute to their loss and dysfunction.
Overall, the function of the platelet mass is reduced,
and bleeding times are prolonged to approximately twice-
INFLAMMATION
normal values for several hours following the conclusion Inflammatory cascades involve complement activation by
of CPB and heparinprotamine binding.15 both the classic and alternative pathways.20 The classic
pathway is activated by the surface contact in the per-
fusion system or by antigenantibody complexes. The
EFFECTS OF ENDOTHELIUM alternative pathway also forms C3 convertase, and is acti-
vated by polysaccharides from bacteria or other cell mem-
The serum enzymes involved in the thrombogenic and branes; it is actually the predominant pathway mediating
fibrinolytic cascade pathways do not exist in isolation; inflammation following CPB. Both pathways culminate in
they are constantly exposed to and interact with reac- the activation of complement factor C5a. Factor C5a, a
tive peptides involved in inflammation, with platelets and potent chemotactic agent that initiates leukocyte aggrega-
other blood cells, and with the endothelium. Under normal tion and activation, has been implicated in mediating the
conditions, the endothelium plays a protective role by con- systemic inflammatory response syndrome (SIRS) associ-
taining and secreting endogenous anticoagulants, such as ated with CPB. At the end of CPB, the classic pathway is
thrombomodulin, tPA, and heparin-like molecules. The activated a second time by the heparinprotamine com-
endothelium also protects against high velocity stress in plex.21 Increased C3a and C5a levels seem to be associated
small arterioles by producing relaxing factors, namely with cardiac and pulmonary dysfunction.
NO and prostacyclin. Therefore, under normal circum- The activation of complement, coagulation, fibrinol-
stances, the endothelium plays a protective and limiting ysis, and platelets also activates blood cells involved
role for thrombogenesis. What then is altered in the car- in mediating an inflammatory response.21 During CPB,
diac patient undergoing CPB for coronary artery graft or neutrophils release neutral proteases, lysosomal en-
valve surgery? zymes, myeloperoxidase, hydrogen peroxide, hydroxyl
Diffuse endothelial activation or damage occurs radicals, hypochlorous acid, acid hydrolases, and collage-
during CPB and cardiac surgery, and has probably already nases.21 These compounds amplify further inflammatory
occurred as a result of preexisting disease. During injury, or coagulation cascades and white blood cell activation.
the expression of endothelial anticoagulant proteins is Neutrophils in this scheme play a major role in ischemia-
downregulated and that of procoagulant peptides are reperfusion injuries and are responsible for much of the
upregulated.7 Upon injury, endothelial cells lose their inflammatory response associated with CPB.20 Monocyte
tight connectivity, allowing serum proteins to be exposed activation during CPB produces the release of various
to subendothelial connective tissue. Finally, injury results cytokines (including IL-1, IL-6, and IL-8) and TNF. The
754 EQUIPMENT
resulting activation of an inflammatory response has been heparin inhibits coagulation at the end of the cascade after
implicated in end-organ injury or dysfunction. Further- most coagulation factors have been converted to active
more, the number of B and T lymphocytes and the enzymes that can influence neutrophil and monocyte
responsiveness of lymphocytes to mitogens and other ago- activation or inflammatory cascades. Therefore, many of
nists are reduced during the first week after CPB, thereby the hemostatic or inflammatory complications attributed
increasing the likelihood of infection.21,22 to CPB could also be due to a direct effect of heparin
and/or heparinprotamine complexes.24,25
In certain individuals, heparin can be associated with
an allergic response, which can range from mast cell re-
ISCHEMIC-REPERFUSION lease of histamine following bolus doses to devastating
INJURY heparin-induced thrombocytopenia (HIT). In those pa-
tients who are prone to HIT, various other anticoagulant
Diffuse arteriosclerosis and/or coronary artery disease set regimens have been used.2630 The use of direct thrombin
up conditions of limited flow and O2 delivery to end inhibitors has produced acceptable outcomes in cardiac
organs with a potential for ischemic injury. The reper- surgery with and without CPB. Their anticoagulant effect,
fusion of ischemic tissue following aortic crossclamping, which can be monitored by activated clotting time (ACT)
nonpulsatile CPB, and/or resultant coronary artery re- and/or activated partial thromboplastin time (aPTT), oc-
anastomoses delivers platelets, inflammatory blood cells curs immediately and, although these compounds have
and enzymes involved in both inflammatory and hemo- no established antidote, they have relatively short half-
static cascades to tissue or endothelium that is already lives (30 to 60 minutes). Hemofiltration can hasten the
in a primed or proinflammatory state. Subsequent activa- anticoagulant reversal. The direct thrombin inhibitors
tion of inflammation, with ensuing release of vasoactive currently used clinically include recombinant-hirudin,
cytokines, produces microcirculatory vasodilation that Argatroban, a synthetic L-arginine, and Bivalirudin, a
manifests itself as hypotension and hypoperfusion of end semisynthetic hirulog. Low molecular weight heparins
organs.22 (nadroparin, dalteparin, and enoxaparin) are other alter-
Short periods of ischemia (i.e., preconditioning) natives to heparin.29,30 Their anticoagulative effect is due
have been shown to provide a protective effect against to conformational changes of antithrombin and heparin
subsequent longer ischemic periods in a variety of cofactor II.
organs, including the heart, liver, lung, kidney, and Low molecular weight heparins affect factor IIa
brain. Ischemic preconditioning is a cellular response (thrombin) much less than factor Xa compared to
to injury that is mediated by adenosine and NO. ATP- unfractionated heparin. The indications for this strategy
sensitive potassium (K+ -ATP) channels in cellular and in CPB are currently unclear because cross-reactivity in
mitochondrial membranes are important in mediating patients with HIT has been reported to be up to 90%,
the effects of preconditioning. Blockade of these ion the half-life of these molecules is longer than that of
channels in animal studies of reperfusion injury negate heparin, and there are no coagulation assays to monitor
the protective effects of preconditioning.23 their activity.
Regarding the CPB circuit, the strategy is to create
surfaces that will not activate the intrinsic system. The
use of surface-bound heparin (bound with either covalent
What Are the Current Strategies or ionic bonds) is one approach. Others have employed
to Minimize Coagulation, circuits with surface-modifying additives (SMA), aimed at
reducing contact activation. These SMA are copolymers
Bleeding, and Systemic that can be either added directly to the base polymer
Inflammation Associated with resins before processing the circuit components or coated
onto the blood-contact surfaces. A recent clinical trial
CPB? by Defraigne et al. showed that the platelet count in the
SMA group decreased less, and there was a reduction in
the thromboglobulin plasma level. The decrease in blood
COAGULATION loss, however, did not reach a statistically significant
level, except in the group of patients taking aspirin
Heparin is the drug of choice for anticoagulation during preoperatively.31,32
extracorporeal circulation because of its immediate onset
of action, its reversibility, and its ability to inhibit
coagulation throughout the CPB circuit.24,25 Heparin acts PLATELET PRESERVATION
by binding to antithrombin III and thereby accelerates
antithrombin III binding to thrombin.25 The goal of Currently, the mainstay strategy to deal with platelet
providing adequate anticoagulation is to overwhelm the loss and dysfunction during CPB is platelet transfusion.
system with high doses of heparin, causing a putative sink A newer strategy that focuses on the preservation of
for the antithrombin IIIthrombin complexes. Heparin platelets is known as platelet anesthesia.33,34 Platelet acti-
also has direct effects on platelets and other enzymes in the vation during CPB can be suppressed by agents such as
coagulation and inflammatory pathways. Furthermore, tirofiban and eptifibatide, which are inhibitors of platelet
C H A P T E R 5 3 / C A R D I O P U L M O N A R Y BY PA S S 755
aggregation. These two agents inhibit platelet aggrega- did not report significant adverse events, except for the
tion by reversibly binding to the platelet receptor GP possibility of anaphylactic reactions to repeated doses,
IIb/IIIa, thereby preventing the binding of fibrinogen. In- which were manifested mainly as severe hypotension.38,39
hibition of platelet aggregation occurs in a dose- and Two postmarketing observational studies have reported
concentration-dependent manner. These drugs are admin- adverse outcomes associated with aprotinin use, mainly
istered in combination with heparin to prevent ischemic renal dysfunction.40,41 Both studies used propensity sta-
complications in patients with acute coronary syndrome. tistical scores to adjust for differences between patient
Their use in platelet preservation during cardiac surgery is groups (i.e., preexisting disease profiles). The reasons for
a novel approach and one awaiting clinical outcome trials. the aprotininassociated increase in renal dysfunction
Both tirofiban and eptifibatide, when used in combination are not well understood; it is possible that the antifib-
with NO, showed successful preservation of platelets in rinolytic action favors formation of microemboli, which
primates.33,34 Using NO alone for this purpose showed then are responsible for end-organ ischemia and damage,
partial protection.35 NO stimulates guanylate cyclase to or that aprotinin directly inhibits formation of protec-
increase the levels of cyclic guanosine 3 , 5 monophos- tive cytokines at the end-organ level. Similarly, while the
phate (cGMP), which inhibits platelet aggregation and studies by Mangano et al. and Karouti et al. seem to in-
adhesion, as well as cGMP-inhibited cyclic adenosine dicate that TXA or EACA are safer alternatives in terms
3 , 5 monophosphate(CAMP) phosphodiesterase activity. of end-organ damage, this also remains to be definitely
Inhibition of this phosphodiesterase increases platelet demonstrated.40,41
cAMP, which decreases cytosolic calcium ion and in- Although it is possible that these observational stud-
hibits basic platelet reactions (e.g., aggregation, adhesion, ies, by their very nature of not being randomized,
secretion of dense and -granule contents, and release of double-blinded controlled studies, could have resulted in a
acid hydrolases).34 Since NO and platelet aggregation in- disproportionate number of patients at risk for renal fail-
hibitors work through different mechanisms, a synergistic ure from other causes in aprotinin treatment groups.40,41
effect may be expected. However, preliminary studies in It is also possible that premarketing clinical trials, given
primates have not shown this synergistic effect. Regard- their strict inclusion and exclusion criteria, could have
ing the reversal of effects, the actions of NO will dissipate resulted in the omission of patients at risk for develop-
upon its termination, whereas the <2-hour half-life of the ing renal complications. A retrospective review of 67,000
platelet aggregation inhibitors will coincide with the dura- patients also suggests that aprotinin is associated with in-
tion of most CPB procedures. Preliminary studies report creased death, renal failure, heart failure, and stroke.38,39
the return of normal bleeding times 30 to 60 minutes after In light of these new studies, the U.S. Food and Drug
the administration of protamine.33,34 Administration (FDA) modified its prescribing informa-
tion with more focused indications, more warnings about
adverse events, and new contraindications.39 As of De-
cember 15, 2006, recommendations and considerations
BLEEDING for physicians include the following:
Exclusive of good surgical technique, perioperative blood Aprotinin use should be limited to only those patients
loss has mainly been controlled by the antifibrinolytics at increased risk for blood loss and blood transfusion in
tranexamic acid (TXA), -aminocaproic acid (EACA), and association with CPB in the course of coronary artery
aprotinin.13,36 EACA and TXA are both synthetic lysine bypass grafting. Aprotinin should be limited to those
analogs that inhibit plasmin-related fibrinolysis and plas- situations where the clinical benefit of blood loss is
minogen binding to fibrin with subsequent activation to essential to the medical management of the patients,
plasmin.13 Aprotinin, a nonspecific serine protease in- and the benefit outweighs the risk.
hibitor, affects fibrinolysis at multiple points. Aprotinin Patients should be carefully monitored for the occur-
inhibits tPA release by competing with plasmin for recep- rence of toxicity, particularly of the kidney, heart, or
tor binding sites and by inhibiting kallikrein activation. central nervous system or for signs of anaphylactic re-
Other factors that may contribute to decreased blood actions. Physicians should be aware of new product
loss include inhibition of serine proteases involved in labeling that warns of increased risk for renal dys-
the clotting cascade and preservation of platelet func- function, potential increase in postoperative dialysis,
tion.13 Additionally, aprotinin has been shown to exert stronger warnings regarding anaphylactic reactions to
anti-inflammatory properties by the inhibition of enzymes aprotinin, and a contradicted use in patients with prior
involved in inflammatory cascades. Therefore, because of exposure to aprotinin or aprotinin-containing products
its multiple sites of action, aprotinin gained popularity (i.e., fibrin glues such as Tissucol or Tisseel). (Tissucol
as an agent of choice for patients at high risk for post- or Tisseel, Baxter International, Deerfield, IL).
operative bleeding. In the late 1980s, with the rising cost Serious and unexpected adverse events should be
of blood transfusion and the increasing risk of acquired reported to the drug manufacturer and/or the FDA
infection from allogenic blood (e.g., human immunodefi- MedWatch program.
ciency virus [HIV], hepatitis C), the established efficacy of Patient counseling should include the indications for
aprotinin prompted its use in a wider range of patients. use of aprotinin (e.g., high risk of intraoperative bleed-
While aprotinins efficacy in controlling blood loss is ing), risks associated with aprotinin (e.g., anaphylaxis
unquestioned,13 controversy exists as to its safety pro- and renal dysfunction), and preexisting factors that in-
file.37 Initial premarketing clinical trials with aprotinin crease a patients chance for developing complications
756 EQUIPMENT
in association with aprotinin use (e.g., prior heart significantly improve tasks involving the visual-spatial
surgery, prior treatment with aprotinin, known drug domain.46 A phase III trial of 3,099 patients in 205
allergies, known kidney disease). centers showed that, although pexelizumab only had a
tendency to decrease mortality by 18% at 30 days in
The more recent multicenter, observational study by
CABG-only patients, it significantly decreased mortality
Mangano et al.37 which reviewed the 5-year mortality
or postoperative MI in those patients undergoing CABG
after CABG surgery in more than 1,000 patients reveals an
plus AVR. All patients were considered high risk and
expected versus observed 5-year mortality in aprotinin-
had at least one of the following risk factors: Diabetes,
treated patients of 1.48. This excess mortality was not
prior CABG, urgent need for surgery, female gender, prior
found in aminocaproic or TXAtreated patients whose
history of neurologic event, prior history of congestive
5-year mortality was similar to control.37 The authors
heart failure (CHF), prior history of more than two MI
conclude that the use of aprotinin does not seem prudent
or recent MI within the last 30 days. Pexelizumab de-
given the excess mortality and safer alternatives.
creased mortality by 28% in those patients with two or
more concomitant risk factors (67% of the total patient
population).47 These results seem to indicate the follow-
INFLAMMATION/ ing: The morbidity associated with CPB is multifactorial;
REPERFUSION INJURY anti-inflammatory treatment may benefit some groups
of patients more than others; and multiple interventions
Hypotension and low organ perfusion is a result of may be needed to address the complex pathophysiologic
vasodilation associated with the systemic inflammatory response to CPB.
response. Maintenance of cardiac output (CO) and end-
organ perfusion pressure is accomplished by the in-
travascular administration of fluids, inotropes, and other
pharmacologic agents. Ischemia produces a proinflam- ON-PUMP VERSUS
matory state in end organs that can be attenuated by OFF-PUMP
preconditioning. Treatment modalities that attenuate this
inflammatory response or mimic preconditioning are be- During CPB, venous blood from the right auricle drains
ing pursued as adjuncts to traditional therapy. Adenosine by gravity into an external reservoir and is then pumped
is a mediator in preconditioning, and infusions of adeno- through an oxygenator back into the systemic circulation
sine or adenosine-1 receptor agonists have been advocated generally through a cannula placed in the ascending aorta.
as interventions to ischemic reperfusion injury. Opiates, By placing a clamp across the aorta proximal to the
particularly morphine, seem to mimic preconditioning in cannula, the heart and lungs are effectively removed from
in vitro and in vivo models. Inhalation agents such as the circulation. Because the CPB apparatus assumes the
isoflurane also can mimic preconditioning effects. There- functions of the heart and lungs, the heart is emptied
fore, the effects of reperfusion injury may be attenuated and rendered electrically and mechanically silent with
by deliberately choosing specific anesthetic and analgesic cardioplegia. CPB entails the extracorporeal transit of
agents.23 blood, nonpulsatile blood flow, hemodilution, and some
Therapeutic approaches to attenuate systemic in- degree of hypothermia.
flammation have included aprotinin, steroids, and inhi- OPCAB has been proposed as a means by which
bition of complement protein C5. Aprotinin, as discussed the proinflammatory and procoagulant effects of the
previously, affects inflammation by inhibiting proteases bypass circuit can be circumvented. OPCAB reduces the
involved in the complement cascade.7,42 The effects of cor- inflammatory response associated with CPB22 but does
ticosteriods on SIRS in cardiac surgical patients remains not affect other inflammatory stimuli such as surgical
unsettled.7,43 Studies have shown that glucocorticoids trauma, reperfusion injury, and thrombin activation.
blunt the release of cytokines involved in inflammation Studies comparing OPCAB with CPB-assisted CABG have
(e.g., TNF, IL-6, IL-8) leading to improved hemodynamic shown modest reductions in coagulation or inflammation
stability. However, glucocorticoid therapy results in im- with OPCAB. Conflicting data exist regarding the effects
munosuppression and hyperglycemia that may exacerbate of OPCAB on complement activation or IL-6 production.
postoperative complications. Preclinical animal studies of OPCAB seems to reduce the generation of circulating
pexelizumab, a monoclonal antibody against complement IL-8, a potent stimulus for neutrophil migration and
protein C5, have shown that pexelizumab decreases the myocardial ischemic injury. Outcome measures of death,
amount of myocardial damage associated with ischemia stroke, or MI suggest a trend toward the beneficial
and reperfusion.44 Despite success with C5 antibody ther- effect of OPCAB, though meta-analysis of comparable
apy in myocardial infarction (MI) patients, clinical trials trials has not shown a significant effect of OPCAB when
in CABG patients have been equivocal. Phase II trials compared with CPB.7 Other analyses of OPCAB seem
of pexelizumab bolus plus infusion to patients under- to indicate reduced morbidity and mortality associated
going CABG or CABG/aortic valve replacement (AVR) with reduced bleeding, making OPCAB a viable option
showed an improvement in primary outcomes at 4 or for those patients at high risk for perioperative bleeding
30 days (MI and/or mortality), but not at 90 days in CABG- or in whom blood transfusions are not an option (e.g.,
only patients.45 While pexelizumab also did not improve Jehovahs Witnesses).48 The beneficial effect of OPCAB
the overall incidence of neurocognitive damage, it did on end-organ function is still being determined. It is
C H A P T E R 5 3 / C A R D I O P U L M O N A R Y BY PA S S 757
TABLE 53.1 Coronary Artery Bypass Graft Surgery with Cardiopulmonary Bypass
(CPB)(On-Pump) versus Coronary Artery Bypass Surgery (Off-Pump)
evident that complications resulting from CABG are infections and postoperative mortality.51 Nondiabetic pa-
multifactorial and not only because of the effects of tients who are exposed to the stress of cardiac surgery have
extracorporeal perfusion. Table 53.1 illustrates several of also been shown to benefit from tight glycemic control.53
the differences and similarities of OPCAB and CABG with The direct relation between fasting glucose levels and the
CPB (on-pump). Although extracorporeal circulation, risk of sustaining a cardiovascular event and the rela-
along with its concomitant physical characteristics, is tion between admission blood glucose levels following an
eliminated with the OPCAB technique, several of the acute MI and long-term mortality has established hyper-
currently used protocols (e.g., heparin and protamine glycemia as an independent predictor of cardiovascular
administration, aortic and beating heart manipulation) risk.51,57 These studies have also established the benefit
remain in place. These similarities of technique may of aggressive treatment with insulin and resultant tight
obscure identification of additional benefits from the glucose control.58 On the basis of these clinical studies,
avoidance of CPB. a reasonable recommendation for perioperative glycemic
control is maintenance of blood glucose at 150 mg per
dL before CPB and at 180 mg per dL during CPB.58 Ini-
tiation of an insulin drip in the operating room is a very
TIGHT GLYCEMIC CONTROL useful tool to help effect and maintain tight perioperative
Hyperglycemia, a common occurrence during CPB in glucose control.
both nondiabetic as well as diabetic patients, is associ- Mechanisms for improved outcomes following tight
ated with adverse outcomes including: increased length glycemic control have been suggested to be the result
of intensive care unit (ICU)/hospital stay, increased inci- of an attenuated inflammatory response, decreased en-
dence of dysrhythmias, depressed myocardial contractile dothelial injury, decreased circulating levels of noxious
function, recurrent ischemia, need for postoperative in- factors including cytokines and free O2 radicals, and/or
otropic support, increased incidence of wound infection, decreased metabolic demand on reperfused tissue. Acute
and increased mortality.4955 Tight glycemic control has and chronic hyperglycemia reduces the preconditioning
been associated with a decreased incidence in the above effects of anesthetics through mechanisms involving the
outcomes. Surprisingly, tight glucose control has failed inhibition of potassium channels that regulate vascular
to improve neurologic or neurobehavioral outcomes in perfusion and/or myocyte contractility. Sulfonurea-based
patients without diabetes mellitus undergoing CABG.55 oral hypoglycemic agents, by virtue of their action on
Whether the etiology of hyperglycemia is due to the stress potassium channels, also seem to further negate pre-
response alone or in conjunction with the release of proin- conditioning.59 On the basis of these observations and
flammatory cytokines, catecholamines, glucagon, growth inferences, it may be prudent to withhold sulfonurea-
hormone, and glucocorticoids, or is the result of insulin based hypoglycemic agents (i.e., glyburide) 24 to 72 hours
resistance in the diabetic or prediabetic patient, or the before cardiac surgery and, if appropriate, initiate insulin
influence of various cardioplegia solutions containing dex- therapy. Metformin, an oral hypoglycemic agent that low-
trose, hyperglycemia has been shown to impact cellular ers glucose levels by sensitizing target tissues to insulin,
components regulating metabolism,51 immune cell acti- inhibits hepatic glucose production, and increases pe-
vation and circulating cytokines,56 and myocardial ion ripheral glucose uptake, may be an appropriate agent
channels.57 Clinical studies addressing diabetic cardiac to continue in the perioperative period.58 The interplay
surgical patients have suggested that tight glucose control of the many cellular and humoral systems with glucose
(125 to 200 mg per dL) in the perioperative period is op- and preexisting pathophysiology remains an active area
timal and results in a lower incidence of sternal wound of investigation.
758 EQUIPMENT
In the assessment and management of aortic cannulation, definitive recommendations. Hypothermia causes reduc-
the following recommendations have been proposed to tion in cerebral O2 consumption that attenuates the phys-
minimize ascending aortic manipulation.7175 iologic impact of reductions in both perfusion pressure
and Hct, thereby extending the safe period of low flow
Avoid partial aortic occlusion cross-clamp (i.e., use a CPB and circulatory arrest.60 Even small temperature dif-
single-crossclamp technique). ferences have important effects. As little as 2 C decrease in
Utilize internal mammary artery for proximal bypass temperature has protective effects,64 whereas hyperther-
graft anastomosis. mia significantly worsens outcome.60,76,77 This concept
Cannulate alternative sites (i.e., axillary artery, innomi- is clinically important during rewarming, when cerebral
nate artery or distal aortic arch cannulation). temperatures can reach their highest point (sometimes
Modify aortic cannula (e.g., deploy an intra-aortic filter exceeding 39 C) and the risk of embolism is the greatest.
or low velocity jetting profile). This has led to a change in practice and the avoidance of
Convert to off-pump CABG with Y-graft anastomosis high brain temperatures during the rewarming phase of
for notouch technique. CPB. Additionally, randomized trials have shown a benefit
Replace the ascending aorta under circulatory arrest in of rewarming from mild hypothermia to a temperature of
the case of severe atherosclerosis. 34 C, as opposed to a body temperature of 37 C.78
Finally, the prevention of blood loss and minimizing
Another approach that reaches class IIb status for the return of cardiotomy suction blood to the patient
neuroprotection during CPB is maintaining higher are further recommendations to minimize brain hypo-
limits for minimal intraoperative MAP (i.e., greater than perfusion, the activation of inflammatory or coagulation
lower target of 50 mm Hg). A MAP of 50 mm Hg is cascades, and/or the generation of emboli. Purging the
commonly viewed as the minimal acceptable arterial arterial filter to the venous reservoir as opposed to
blood pressure during CPB. However, prior pathology the cardiotomy suction reservoir also decreases the
such as diabetes mellitus, hypertension, or prior stroke likelihood of brain injury for similar reasons.79 There are
can impair cerebral autoregulation and brain perfusion at also accumulating data to suggest that monitoring and
lower mean blood pressures. Retrospective studies suggest optimizing cerebral regional O2 saturation to maintain
that a MAP >50 mm Hg during CPB may benefit patients it within 25% of resting baseline decreases major organ
at risk for neurologic complications due to advanced age morbidity (including stroke) and length of stay.67 Even
or aortic atherosclerosis.71 The optimal MAP target for simple techniques can have an important impact in
such patients remains to be defined. preserving brain function after CPB.
An almost universal approach during CPB is the use of
mild hypothermia. Although the limitations of the current
studies preclude definite conclusions based on the existing CARDIAC COMPLICATIONS
data, a meta-analysis by Res et al.76 concluded that there
is no evidence of neurologic protection with hypother- CABG surgery is associated with a mortality rate of 3%
mic CPB. The limitations of the existing data preclude to 5%, and in certain high-risk groups (i.e., advanced age,
760 EQUIPMENT
prior CABG) the mortality can exceed 15%. The primary CO (e.g., cardiac index 2 L/minute/m2 ) despite normal
cause of this mortality and associated morbidity is my- or high LV filling pressure for prolonged periods (e.g.,
ocardial stunning and/or necrosis. Myocardial stunning80 >30 minutes), one is dealing with a condition that has
describes mechanical dysfunction that persists after reper- been referred to as low CO syndrome.91 This syndrome,
fusion despite relatively normal coronary flow and the ab- the result of myocardial calcium overload and oxidative
sence of irreversible cellular damage (i.e., necrosis). This stress, has a reported incidence of approximately 9%, an
state of transient postischemic dysfunction results from associated approximately 20-fold increase in mortality,
the reperfusion-induced release of reactive O2 species and prolonged hospitalization. It is associated with ele-
that can exacerbate an already disrupted myocardium. vated pulmonary artery and pulmonary capillary wedge
These ionic imbalances result in calcium overload and pressures (e.g., surrogates of high LV filling), low SvO2
a dysfunctional myocardium. Cardiac surgery with CPB (60% to 65%), metabolic/lactic acidosis, low urine output,
subjects the myocardium to this potential injury as a con- and high dose inotropes. Preoperative risk factors in-
sequence of intraoperative events such as the periods of clude poor LV contractile function (EF <20%), advanced
ischemia/reperfusion and/or inadequate protection during age (>70 years), redo surgery, emergency surgery, dia-
the application of the aortic cross-clamp, coronary em- betes mellitus, female gender, recent MI, and left main
bolism, or direct trauma. Although it is difficult to assign or three-vessel disease.92 Another common clinical sce-
the precise limits of myocardial enzyme elevations that in- nario, in contrast to the low output state, is the high
dicate significant cellular necrosis (i.e., MI) from the levels output-low pressure state, the result of weaning from
of biomarkers indicative of subclinical cell damage in the CPB (e.g., warming effects, electrolyte shifts), periopera-
postoperative cardiac surgical patient, several studies have tive medications (e.g., angiontensin-converting enzyme
shown a strong correlation between creatinine kinase-MB [ACE] inhibitors, phosphodiesterase III inhibitors), or
(CK-MB) levels following CABG and short-term and long- neuroendocrine imbalances. These instances of post-CPB
term mortality.81,82 In the GUARDIAN trial,81 a CK-MB instability can be evaluated and resolved by utilizing
>10 times the upper level of normal (ULN) was asso- the hemodynamic relationships noted in Table 53.4 and
ciated with a significant increase in 6-month mortality treating accordingly, for example, by increasing systemic
that was similar to that observed in patients with spon- vascular resistance (SVR).
taneous MI (9%). In this trial, in which 2,918 patients Myocardial irritability increases during weaning from
underwent CABG, the distribution of CK-MB 5 ULN, CPB. Multiple factors may contribute to dysrhythmoge-
10 ULN, and 20 ULN was approximately 40%, 20%, nesis during this critical intraoperative period and in-
and 8%, respectively. The American College of Cardiol- clude hypothermia, hypoxemia, electrolyte and acidbase
ogy guidelines define an MI following a CABG as any imbalance, coronary air embolism, inadequate myocar-
CK-MB release >10 times ULN or CK-MB >5 times the dial protection during aortic cross-clamp/CPB, surgical
ULN with new Q waves on electrocardiogram (ECG).83 trauma to the myocardium and conduction system, inad-
Specific levels of troponin I and T, recognized as the equate revascularization, and inotropic intervention.
more sensitive and specific markers of myocardial necro- Ventricular tachydysrhythmias (extrasystoles, sus-
sis, have not been clearly associated with similar patient tained or nonsustained ventricular tachycardia [VT] and
outcomes. Risk factors associated with perioperative MI ventricular fibrillation [VF] occur in up to 50% of
in patients undergoing CABG include urgent procedure, cardiac surgical patients, with 10% to 30% requiring
redo operations, advanced age (>65 yrs), female gender,
diabetes, recent angina, left main or three-vessel disease,
and chronic renal insufficiency. In addition to advances in TABLE 53.4 Hemodynamic Relationships
surgical techniques and the use of cardioplegic solutions,
preventative approaches have included the perioperative Relation Normal Values
use of -blockers, aspirin, and statins, as well as intraop-
erative Hct control (25%) and pharmacologic therapies BP = CO SVR
(i.e., preconditioning agents, sodium-hydrogen exchang- CO = HR SV 57 L/min
ers, anti-inflammatory agents) directed at mechanisms CI = CO/BSA 2.44.2 L/min/m2
that mediate myocardial injury/necrosis in response to SV = (CO/HR) 1000 50110 mL/beat
ischemia/reperfusion and SIRS.84 90 Although many of SI = SV/BSA 4070 mL/beat/m2
these therapies have the potential ability to protect the SVR = (MAPRAP)/CO 80 7001,600 dynes-sec/cm5
myocardium from injury during cardiac surgery and CPB, PVR = (MPAPLAP)/CO 80 20150 dynes-sec/cm5
none have been approved for this indication.
As noted previously, following the removal of the aor- BP, blood pressure; CO, cardiac output; SVR, systemic vascular
resistance; HR, heart rate; SV, stroke volume; CI, cardiac index; BSA,
tic cross-clamp or weaning from CPB, cardiac dysfunction
body surface area; SI, stroke index; MAP, mean arterial pressure;
may be the result of myocardial stunning or myocardial RAP, right atrial pressure; PVR, pulmonary vascular resistance; MPAP,
necrosis/infarction. Although this state of low CO and/or mean pulmonary artery pressure; LAP, left atrial pressure.
hypotension may be attributed to other physiologic pa- General assumptions if not directly measured: RAP CVP; LAP
rameters that contribute to hemodynamic stability (i.e., CVP.
preload, afterload, regular heart rate and rhythm), if phar- BSA = Ht (in cm)0.725 Wt (in kg)0.425 71.84 104 (Dubois
macologic or mechanical interventions (e.g., intra-aortic Formula).
balloon pump [IABP]) are needed to maintain an adequate BSA = (Ht (in cm) + Wt (in kg)60)/100 (Jacobson Formula).
C H A P T E R 5 3 / C A R D I O P U L M O N A R Y BY PA S S 761
cardioversion/defibrillation or lidocaine following removal an approximately 5% incidence. Most likely the result of
of the aortic cross-clamp during weaning from CPB.93 a reentry mechanism,99 the onset of atrial fibrillation at
The resolution of these dysrhythmias, or interventions 24 to 60 hours following surgery has a peak incidence
to treat the dysrhythmogenic factors noted in the pre- in the second to third POD. Preoperative risk factors in-
ceding text, are crucial for establishing hemodynamic clude advanced age (the strongest predictor, >65 years),
stability, whether it is through the return of normal sinus prior history of atrial fibrillation/flutter or COPD, sys-
rhythm or the establishment of a paced rhythm. Con- temic hypertension, aortic atherosclerosis, male gender,
tinued hemodynamic instability should prompt a search elevated C-reactive protein, and withdrawal of -blockers
for and treatment of reversible causes, beginning with or ACE inhibitors.97,100,101 Intraoperative risk factors in-
ischemia. The inspection and possible revision of anas- clude prolonged CPB time, valve surgery, and combined
tomoses, the placement of additional anastomoses, and valve/CABG procedures.9698 Bicaval venous cannulation,
the detection and treatment of coronary air embolism found to be another risk factor,97 may be a surrogate, be-
are critical first steps for the cardiac team. The detec- cause it is generally employed in mitral valve procedures.
tion of air embolism can involve direct observation of Inadequate myocardial protection during the ischemic pe-
bubbles in a graft or ST-segment elevation in inferior riod of the aortic cross-clamp may initiate or contribute
leads that represent the anterior ostium and the right to this or any dysrhythmia. Although its natural course is
coronary artery. Specific treatment maneuvers to pro- self-limiting, the treatments for this dysrhythmia in clini-
mote collateral flow, force emboli distally, and accelerate cal practice are varied and include pharmacologic agents
the dissolution of the air bubble include increasing CO (e.g., digoxin, calcium channel antagonists, -blockers,
through CPB pump flow or inotropic support, raising the amiodarone), overdrive pacing, and cardioversion. Hy-
level of circulating O2 , and initiating the infusion of coro- potension and/or CHF, although generally well tolerated,
nary vasodilators (nitroglycerin). Consideration of LIMA may arise from a rapid ventricular response/tachycardia
or free-artery conduit spasm should prompt vasodilator or loss of an atrial kick that can account for up to
therapy (i.e., nitrates/NO-promoter, phosphodiesterase III 40% of the ventricular filling during diastole. Observa-
inhibitor, or calcium channel antagonist). Should such tional and randomized control trials have indicated that
maneuvers prove inadequate, additional interventions preoperative and postoperative -blockers and ACE in-
may include a return to CPB, continued CPB, or the hibitors, postoperative potassium supplementation, and
utilization of ventricular assist devices (VADs) (e.g., IABP,