Professional Documents
Culture Documents
Executive Editor
Claude Lenfant
Former Director, National Heart, Lung, and Blood Institute
National Institutes of Health
Bethesda, Maryland
Edited by
Zab Mosenifar
CedarsSinai Medical Center
Los Angeles, California, U.S.A.
Practical pulmonary and critical care medicine / edited by Zab Mosenifar, Guy W. Soo Hoo
p. ; cm. -- (Lung biology in health and disease ; v. 213-214)
Includes bibliographical references and index.
Contents: v. 1. Respiratory failure-- v. 2. Disease management.
ISBN-13: 978-0-8493-6663-5 (v. 1 : alk. paper)
ISBN-10: 0-8493-6663-1 (v. 1 : alk. paper)
ISBN-13: 978-0-8247-2597-6 (v. 2 : alk. paper)
ISBN-10: 0-8247-2597-2 (v. 2 : alk. paper)
1. Lungs--Diseases. 2. Respiratory intensive care. I. Mosenifar, Zab, 1951- II. Soo Hoo, Guy W.
III. Series.
[DNLM: 1. Lung Diseases--therapy. 2. Acute Disease--therapy. 3. Critical Care--methods. WF 600
P8947 2006]
RC941.P73 2006
616.2'4--dc22 2005044634
Surely, all of us would certainly concur with the rst sentence of the Preface pre-
pared by Drs. Mosenifar and Soo Hoo: Over the past decade, the pace of pro-
gress in medicine has been astounding. New developments in diagnosis,
management, and therapeutics occur at a breathtaking rate. On the other hand,
one could argue with the time frame, that is, why only the past decade? This is
a signicant point, especially when it comes to pulmonary and critical care. Pul-
monary medicine started to blossom many, many decades ago but especially
during the last 30 years or so. Critical care started its course, to what it is
today, with the development of the oxygen electrode (and later oximetry) and
with the recognition of the value of blood pH measurements in the late 1950s.
At the same time, the use of ventilators and respiratory assist devices became
more common.
However, in reality it matters little when progress began. Today, what
counts is that new developments are being introduced at such a fast pace that pro-
gress may exceed the ability to fully transfer and utilize all this knowledge in the
practice of medicine. Sure enough, tertiary hospitals have the facilities and work
force to adjust to and quickly adapt the changes as they occur. On the other hand,
hospitals which are community based and distant from academic centers may nd
iii
iv Introduction
it more difcult to take full advantage of the newer approaches for the manage-
ment of pulmonary patients and/or critical care situations.
What could be considered a dual standard between tertiary and community
hospitals is well recognized today, and has raised concerns of policy makers,
politicians, and medical leaders. The question is frequently raised about how
fast, and how well, research advances are translated into the practice of medicine,
including pulmonary and critical care medicine.
This monograph, Practical Pulmonary and Critical Care Medicine:
Respiratory Failure and its companion volume titled Practical Pulmonary and
Critical Care Medicine: Disease Management represent important steps to mini-
mize translation difculties. Indeed, they both present a very practical approach
to pulmonary and critical care medicine. The rst volume addresses, and very
carefully describes, the tools to provide the most optimal care and management
of respiratory failure. The second volume examines a number of situationsboth
pulmonary and non-pulmonarythat require the use of these tools. Together,
these volumes will not only enrich the knowledge of the readers, but also will
provide a wealth of practical information that has the potential to positively
impact on the care of their patients.
The editors, Dr. Zab Mosenifar and Dr. Guy W. Soo Hoo, have assembled
contributors with outstanding expertise and a wealth of practical experience,
coming from institutions/environments with large patient populations, and a
wide variety of cases and medical situations. They share their very practical
and real experiences throughout the volumes.
As the Executive Editor of the series of monographs Lung Biology in
Health and Diseases, I am proud to present these two volumes and to express
my most sincere thanks to the editors and the authors.
Claude Lenfant, MD
Gaithersburg, Maryland, U.S.A.
Preface
Over the past decade, the pace of progress in medicine has been astounding. New
developments in diagnosis, management, and therapeutics occur at a breathtaking
rate. A new disease emerges and, aided by the ease of transcontinental travel,
threatens to become the next pandemic. And just as quickly, the causative
agent is identied with diagnostics and therapeutics soon to follow. Old diseases
have their mysteries unraveled, and targeted therapeutics provide hope where
there was once despair. Some conditions come under control in less than a
generations span, while others plod along inexorably to the end, with little
available to alter their course.
Just as the conditions of disease have changed, so have the conduits for
information. The speed and availability of electronic databases now allow access
to vast warehouses of information at the click of a mouse or ick of a stylus. Where
once the resident or houseofcer carried a worn copy of a spiral bound manual,
handheld computers are now the essential accessories. The grand old textbooks
have followed suit, available in versions abbreviated to t le-size limits or
available in their own electronic internet-based versions. This electronic world
not only allows but mandates frequent content changes. Information can be
updated daily and even more frequently if necessary. The online resource is
now predominant in an arena that was once the domain of the print journal.
v
vi Preface
Zab Mosenifar
Guy W. Soo Hoo
Contributors
ix
x Contributors
Silverio Santiago Pulmonary and Critical Care Section, West Los Angeles
Healthcare Center, VA Greater Los Angeles Healthcare System and Geffen
School of Medicine at UCLA, Los Angeles, California, U.S.A.
Nikhil Shah Pulmonary and Critical Care Section, West Los Angeles
Healthcare Center, VA Greater Los Angeles Healthcare System and Geffen
School of Medicine at UCLA, Los Angeles, California, U.S.A.
Guy W. Soo Hoo Pulmonary and Critical Care Section, West Los Angeles
Healthcare Center, VA Greater Los Angeles Healthcare System and Geffen
School of Medicine at UCLA, Los Angeles, California, U.S.A.
Sanjay Vadgama Pulmonary and Critical Care Section, West Los Angeles
Healthcare Center, VA Greater Los Angeles Healthcare System and Geffen
School of Medicine at UCLA, Los Angeles, California, U.S.A.
Steve Y. Wong Mission Hospital, Mission Viejo, Long Beach VA, Long Beach,
California, U.S.A.
This page intentionally left blank
Contents
xiii
xiv Contents
8. Bronchoscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Nikhil Shah, Irawan Susanto, and Silverio Santiago
I. Introduction . . . . 285
II. Specications, Indications, and Contraindications . . . . 286
III. Patient Preparation, Sedation, and Anesthesia . . . . 288
IV. Insertion Techniques . . . . 289
V. Diagnostic Techniques . . . . 289
VI. Post-Procedure Care . . . . 291
VII. Complications . . . . 291
VIII. Specic Diagnostic Indications . . . . 292
IX. Specic Diagnostic and Therapeutic Indications . . . . 297
X. Therapeutic Bronchoscopy:
Indications and Options . . . . 298
XI. Potential Therapeutic Indications . . . . 305
References . . . . 306
xvi Contents
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
1
Oxygen Therapy and Airway Management
I. Introduction
Oxygen (O2) is an elemental gas that is necessary for life in aerobic organisms.
In the absence of O2 (hypoxia), cellular respiration ceases and irreversible
cellular injury and death occur within minutes. At normal atmospheric pressure
and temperature, O2 exists as an odorless, tasteless, and colorless gas. It rep-
resents one-fth of the earths atmosphere by volume (20.96%).
A. Medical Oxygen
Medical grade O2 is manufactured by fractional distillation of liqueed air (1,2).
It is stored as a liquid to reduce the size of the storage container (1 L of liquid O2
produces 860 L of gaseous O2). The liquid O2, stored outside the hospital in
a cryogenic container, is converted to gaseous O2 and delivered to the patient
1
2 Wong and Hess
Table 1 Duration of Flow (in Hours) from an E-Cylinder as a Function of O2 Flow and
the Pressure of Gas Remaining in the Cylinder
500 2.3 1.2 0.8 0.6 0.5 0.4 0.3 0.3 0.3
600 2.8 1.4 0.9 0.7 0.6 0.5 0.4 0.4 0.3 0.3 0.3
700 3.3 1.6 1.1 0.8 0.7 0.5 0.5 0.4 0.4 0.3 0.3 0.3 0.3
800 3.7 1.9 1.2 0.9 0.7 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.3 0.3
900 4.2 2.1 1.4 1.1 0.8 0.7 0.6 0.5 0.5 0.4 0.4 0.4 0.3 0.3 0.3
1000 4.7 2.3 1.6 1.2 0.9 0.8 0.7 0.6 0.5 0.5 0.4 0.4 0.4 0.3 0.3
1100 5.1 2.6 1.7 1.3 1.0 0.9 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.4 0.3
1200 5.6 2.8 1.9 1.4 1.1 0.9 0.8 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.4
1300 6.1 3.0 2.0 1.5 1.2 1.0 0.9 0.8 0.7 0.6 0.6 0.5 0.5 0.4 0.4
1400 6.5 3.3 2.2 1.6 1.3 1.1 0.9 0.8 0.7 0.7 0.6 0.5 0.5 0.5 0.4
1500 7.0 3.5 2.3 1.8 1.4 1.2 1.0 0.9 0.8 0.7 0.6 0.6 0.5 0.5 0.5
1600 7.5 3.7 2.5 1.9 1.5 1.2 1.1 0.9 0.8 0.7 0.7 0.6 0.6 0.5 0.5
1700 7.9 4.0 2.6 2.0 1.6 1.3 1.1 1.0 0.9 0.8 0.7 0.7 0.6 0.6 0.5
1800 8.4 4.2 2.8 2.1 1.7 1.4 1.2 1.1 0.9 0.8 0.8 0.7 0.6 0.6 0.6
1900 8.9 4.4 3.0 2.2 1.8 1.5 1.3 1.1 1.0 0.9 0.8 0.7 0.7 0.6 0.6
2000 9.3 4.7 3.1 2.3 1.9 1.6 1.3 1.2 1.0 0.9 0.8 0.8 0.7 0.7 0.6
Oxygen Therapy and Airway Management 3
Approximate
Usual ow oxygen
Device range concentration Comments
length. The cannula is easily applied and well tolerated by most patients when used
with ows of 6 L/min. It is held in place by an elastic band around the head or,
more commonly, by looping the delivery tubing over the ears and holding it in
place with an adjustable slide under the chin. Oxygen delivery by nasal cannula
produces an FIO2 of approximately 0.24 at 1 L/min to about 0.40 at 6 L/min.
The FIO2 from a nasal cannula is highly variable even if the O2 ow remains con-
stant (Table 3). A number of studies using various methodologies conrm that the
FIO2 resulting from use of a nasal cannula is highly variable and dependent upon
the O2 ow, inspiratory ow, and minute ventilation (2636). The amount of deliv-
ered O2 is reduced if O2 is breathed from only one of the prongs, as may occur if the
prongs are displaced or there is unilateral nasal obstruction (37). There are
Oxygen Therapy and Airway Management 7
nasal
cannula
simple
mask
conicting data regarding the effect of mouth breathing when a nasal cannula is
used (38,39). Although mouth breathing may decrease the FIO2, O2 therapy by
nasal cannula can be effective even with mouth breathing.
The majority of O2 is wasted when it is delivered by nasal cannula. Most of
the O2 ow during the expiratory phase is wasted. It is only during the initial
period of inspiration that the supplemental ow of O2 is needed. It is this rst
part of inspiration that is delivered to the alveoli and participates in gas exchange.
The remainder of the inspired gas lls the anatomic dead space and is exhaled
without contributing in gas exchange. Oxygen-conserving devices conserve O2
by using a reservoir, which provides O2 ow only during the inspiratory phase,
Reservoir O2 cannula (moustache and pendant types): A small reservoir lls with O2
(20 mL for the moustache style and 40 mL for the pendant) during exhalation. At the
beginning of inhalation, a bolus of O2 is drawn from the reservoir. This may allow a
reduction in O2 ow by 50 75%. Reservoir cannulae are not well accepted by patients
due to their appearance.
Transtracheal O2 catheter: This is a small-diameter catheter surgically inserted into the
trachea. It is connected to a small ange and held in place by an adjustable chain. As O2
is continuously delivered directly into the trachea, a reduction in the O2 ow by about
50% is possible. Complications of the catheter placement include infection, bleeding,
and subcutaneous emphysema. The catheter must be cleaned regularly to prevent mucus
accumulation. Catheter obstruction is prevented by instilling saline and inserting a
cleaning wire into the lumen of the catheter.
Demand oxygen conservers: These devices only deliver O2 during the inspiratory phase.
When the patient begins an inspiration, this creates a negative pressure in the supply
tubing and causes a demand valve to open and supply a dose of O2.
valves
reservoir
reservoir
bag
bag
non-rebreathing mask will deliver 100% O2. However, these masks do not provide
an airtight t on the face, and the valves do not provide a perfect seal (45). At ows of
10 to 15 L/min, an FIO2 of 0.6 to 0.8 may be achieved. Moreover, the non-rebreath-
ing masks have a valve over only one of the exhalation ports. This allows inhalation
of room air if the O2 supply ow becomes inadequate.
lower
O2 O2
FIO2
O2 higher O2
FIO2
G. High-Flow Devices
An air-entrainment mask (Fig. 3) consists of a mask, a jet nozzle, and air-entrain-
ment ports (46 49). Oxygen is delivered through the jet nozzle, which increases
its velocity. The high-velocity O2 entrains ambient air into the mask due to the
viscous shearing forces between the gas traveling through the nozzle and
the stagnant ambient air. The FIO2 depends on the nozzle size and the size of
the entrainment ports. Commercially available systems use interchangeable
jets or adjustable entrainment ports. Obstruction of the entrainment port or down-
stream obstruction decreases entrainment and increases FIO2. To deliver a xed
FIO2, the ow to the mask must exceed the peak inspiratory ow of the patient.
This may be difcult to achieve with a higher FIO2 settings (Table 5).
A high-ow O2 delivery system can be used to deliver a precise FIO2 and
any concentration between 0.21 and 1.0. Such a system provides sufcient ow to
meet the inspiratory demands of the patient (50). Because the total ow is set
higher than the inspiratory ow of the patient (typically 30 60 L/min), the
gas should be humidied. An air and an O2 ow meter can be used to deliver
a precise FIO2 (Fig. 4) and a variety of patient interfaces can be used (Fig. 5).
The FIO2 can be calculated from the ow rates of air and O2 (Table 6). A
blender uses pressurized sources of air and O2 to deliver a precise FIO2. Blenders
are compact and convenient but more expensive than using two ow meters.
H. Heliox
Heliox is a gas mixture of helium and oxygen that is clinically useful in some
circumstances due to its low density. Because helium does not support life, it must
always be delivered in a gas mixture containing at least 20% oxygen. There is
an increasing interest in its therapeutic use in patients with obstructive lung
Table 5 FIO2, Minimum Flow Requirements, Outputs, and Entrainment Ratios for an
Air-Entrainment Mask
Source: From Branson RD. The nuts and bolts of increasing arterial oxygenation: devices and
techniques. Respir Care 1993; 38:672 686.
Oxygen Therapy and Airway Management 11
aerosol
mask
tracheostomy
mask
T-piece
diseases, and this has been the source of several reviews (51,52) and meta-ana-
lyses (53,54). Although the role of heliox has been reported benecial in case
series and anecdotal reports, current evidence is insufcient to allow a
recommendation for the use of heliox as a standard therapy for any specic
patient population.
One use of heliox is to reduce resistance with upper airway obstruction
(55,56). An example is post-extubation stridor, where use of heliox has been
reported to be benecial (57). There is also interest in the use of heliox
for acute asthma (58,59). In spontaneously breathing patients with asthma,
heliox decreases PaCO2, increases peak ow, and decreases pulsus paradoxus.
There may be benet related to the combination of heliox with aerosol broncho-
dilator delivery in patients with acute asthma or COPD (60 63). When heliox
(rather than air or oxygen) is used to power the nebulizer, the ow should be
increased by about 50% to assure adequate output from the nebulizer (64). As
demonstrated in several meta-analyses, however, the benet of heliox in the man-
agement of patients with acute asthma has yet to be conclusively demonstrated
(53,54). The role of heliox in the treatment of COPD is unclear (65 67).
COPD is a disease of small airwaysa region of the lungs in which ow is
density independent. Heliox has been reported to decrease work of breathing
in some, but not all, patients with COPD when evaluated just prior to extubation
(68). Benet has been reported for the use of heliox with non-invasive ventilation
in patients with COPD (69 72), and methods to administer heliox with a BiPAP
ventilator have been described (73).
Care must be taken to administer heliox in a safe and effective manner.
To avoid administration of a hypoxic gas mixture, it is recommended that 20%
oxygen/80% helium is mixed with oxygen to provide the desired helium concen-
tration and FIO2. If an FIO2 greater than 0.40 is required, the limited concen-
tration of helium is unlikely to produce clinical benet. For spontaneously
breathing patients, heliox is administered by face mask with a reservoir bag
(Fig. 6). A Y-piece attached to the mask allows concurrent delivery of aero-
solized medications. Sufcient ow is required to minimize contamination of
Oxygen Therapy and Airway Management 13
the heliox with ambient air. This is often at least 12 to 15 L/min and requires 3 to
6 H-size cylinders per day. When using an oxygen-calibrated ow meter for
heliox therapy, it must be remembered that the ow of heliox (80% helium
and 20% oxygen) will be 1.8 times greater than the indicated ow. Heliox admin-
istration during mechanical ventilation can be problematic (74 78). Ventilators
are designed to deliver a mixture of air and oxygen. The density, viscosity, and
thermal conductivity of helium affect the delivered tidal volume and the measure-
ment of exhaled tidal volume. With some ventilators (e.g., Puritan Bennett), no
reliable tidal volume is delivered with heliox, whereas there may be a much
higher delivered tidal volume than desired for other ventilators. One commer-
cially available ventilator has been approved for use with heliox (VIASYS
AVEA, Palm Springs, California, U.S.A.).
A. Indications
Endotracheal intubation is indicated to protect the airway and reduce the risks
of pulmonary aspiration. It also offers a pathway for patients who require prolon-
ged positive pressure ventilation and frequent suctioning. In addition, the ETT
can be used to administer emergency medications when intravenous access is
not available. Most frequently, endotracheal intubation enables the patient to
undergo surgical procedures and allow delivery of inhalational anesthetics.
Outside the operation room, endotracheal intubation typically involves patients
who are in respiratory failure, shock, or cardiopulmonary arrest (79). Despite
the increasing use of non-invasive ventilation, most patients receive mechanical
ventilation invasively via an ETT.
The ETT can be placed either nasally or orally. Nasotracheal intubation is typi-
cally indicated for patients undergoing oral surgeries or when intubation through
the mouth is unsuccessful. It is contraindicated in patients with basilar skull frac-
ture, mid-facial trauma, coagulopathy, nasal polyps, and epistaxis. For long-term
mechanical ventilation, orotracheal intubation is preferred because a nasotracheal
tube may increase the work of breathing as well as the risk of sinusitis. Because
the orotracheal route usually permits placement of a larger ETT, clearance of
secretions may also be easier.
C. Airway Assessment
A thorough airway evaluation should be performed prior to intubation to assess
the degree of difculty in airway management. Careful review of the medical
history and a detailed examination of the anatomic characteristics allow
identication of potential difcult mask ventilation or tracheal intubation.
Anatomic features such as a short muscular neck, receding mandible,
prominent upper incisors, small mouth with a high arched palate, and limited move-
ment of the mandible suggest increased likelihood of potential airway problems (80).
. Class I: the soft palate, uvula, and faucial pillars are all visible.
. Class II: the tonsillar pillars and the base of uvula are obscured by the
base of the tongue.
. Class III: only the soft palate is visible.
. Class IV: soft palate is not visible.
When the pharyngeal class is Mallampati III or IV, a more difcult intubation is
expected. Unfortunately, it is not easy to use the Mallampati classication for
urgent or emergent intubation.
Simple measurement of the thyromental distance, which is the length
between the prominence of the thyroid cartilage and the bony point of the
chin, can be used to aid in predicting difcult intubation. When the thyromental
distance is ,7 cm and the Mallampati class is III or IV, difcult intubation can be
anticipated (83). The performance of the Mallampati test and the measurement of
the thyromental distance should identify most cases of difcult intubation
and allow appropriate preparations.
16 Wong and Hess
Depolarizing Non-depolarizing
Induction agent muscle relaxant muscle relaxant
Source: From Donnelly AJ, Cunningham FE, Baughman VL. Anesthesiology and Critical Care Drug Handbook. Ohio: Lexi-Comp Inc., 2000.
17
18 Wong and Hess
laryngeal inlet view during nasal beroptic laryngoscopy (89). Tamura and
colleagues (90) have also demonstrated that such a maneuver can improve
laryngeal view during laryngoscopy performed by inexperienced physicians.
When the mandibular advancement is combined with the BURP maneuver,
further improvement of laryngeal visualization can be obtained.
Figure 9 Technique for endotracheal intubation. (A) Snifng position. (B) Visualiza-
tion of the glottic opening. (C) Advancement of the endotracheal tube between the
vocal cords under direct visualization. Source: Adapted from Dr. R. Steadman, UCLA.
1. With the manual ventilation bag connected to 100% oxygen, place the
mask over the patients nose and mouth. Assist ventilation as needed to
maintain adequate oxygen saturation.
2. Make sure the patient is completely supine and adjust the bed to a
comfortable height, usually at the level of the practitioners xiphoid
cartilage. Place the patient in a snifng position by putting pads
under the patients head and gently extending the neck. The head
elevation aligns the oral and pharyngeal axes, whereas the neck exten-
sion aligns the oral axis to create a nearly straight view of the glottic
opening. This position maximizes exposure of the glottis and greatly
enhances the chance of a successful intubation. However, in a
patient with suspected or conrmed neck injury, the neck should be
kept in a neutral position.
3. Administer induction agent of choice and insert the oral airway. It is
crucial that the ability to ventilate the patient is conrmed before
any muscle relaxant is given. In a patient at risk of pulmonary
aspiration, pressure should be applied at the cricoid cartilage to
occlude the esophageal lumen, prevent insufation of the stomach,
and minimize regurgitation of the gastric contents (92). This is
known as the Sellicks maneuver. Cricoid pressure should be discon-
tinued only after the endotracheal placement is conrmed and the
ETT cuff is inated.
4. When adequate anesthetic depth is achieved, remove the oral airway.
Hold the laryngoscope with the left hand and insert the blade from
the right side of the patients mouth.
5. Sweep the tongue to the left and advance the blade until the epiglottis
is visualized. If a MacIntosh blade is used, the tip of the blade should
be placed in the vallecula, which is the space between the base of the
tongue and the epiglottis. If a straight blade is used, the tip of the blade
should be passed beneath the laryngeal surface of the epiglottis.
6. With the blade properly positioned, lift the laryngoscope handle in a
forward and upward motion to visualize the glottic opening. To
prevent dental injury, the practitioner should maintain the wrist in a
neutral position and avoid levering the laryngoscope backward.
7. Clear any secretions with the tip of the Yankauer suction catheter.
Once the glottic opening is well exposed, insert the ETT, usually
with a stylet, from the right side of the blade. Advance the ETT
between the vocal cords under direct vision.
8. After the endotrachael tube is advanced to a depth of about 23 cm in
men and 21 cm in women, inate the cuff to obtain a seal in the
presence of 20 to 30 cm H2O positive airway pressure. With the
right hand holding the ETT, remove the stylet with the left hand.
9. Connect the ETT to the oxygen source and conrm the ETT placement
before securing it with adhesive tape.
Oxygen Therapy and Airway Management 21
K. Difcult Airway
When endotracheal intubation is unsuccessful with direct laryngoscopy, it is
crucial to call for help from an experienced intubator. Even though the American
Society of Anesthesiologists (97) provides practice guidelines for management
of the difcult airway (Fig. 11), difcult intubation should be handled by
clinicians who are most experienced in airway management. A practitioner not
accustomed to difcult airway management may have trouble remembering or
following the complicated difcult airway algorithm.
As insertion of the laryngeal mask airway (LMA) (Fig. 12) is easy to learn,
it has been utilized in the difcult airway algorithm as a temporary approach
to ventilation (98). Despite its usefulness, the LMA has two major limitations.
First, it does not protect against aspiration. Second, it may fail to provide
adequate ventilation in patients with high airway resistance or low pulmonary
compliance. Therefore, it is contraindicated in individuals with gastroesophageal
reux, full stomach, pregnancy, obesity, or bronchospasm.
Other techniques of intubation or ventilation, such as beroptic intubation
(99), esophageal tracheal Combitube (97,100), intubating LMA (101), and trans-
tracheal jet ventilation (102), all require extensive practice to achieve prociency
and should not be attempted by anyone unfamiliar with them.
Commercially available kits for cricothyrotomy and retrograde tracheal
intubation can also be utilized in the management of the difcult airway. Both
involve entering the cricothyroid membrane to establish a conduit for ventilation.
In a life-threatening situation, emergency cricothyrotomy should be performed
to provide an airway quickly until tracheostomy can be established.
When extubation is planned in a patient with a known difcult airway, a well-
formulated plan must be established for possible re-intubation. A tube exchanger
can be placed to serve as a guide over which the new ETT can be threaded.
Oxygen Therapy and Airway Management 23
Although this approach does not guarantee a successful re-intubation, the tube
exchanger can be used to insufate oxygen until a denitive airway is secured.
Once the endotracheal position is established, the ETT should be secured so that
the distal end of the ETT is approximately 5 cm above the carina, with the head in
a neutral position (103). Taping the ETT at the 23 cm mark in men and the 21 cm
mark in women seems to reduce the risk of endobronchial intubation (104). Once
the proper position is determined, the centimeter mark on the tube at the level of
the patients lip should be recorded and checked on a regular basis. In addition,
cuff pressure should be monitored closely, as excessive pressure can lead to
tracheal mucosa injury and subsequently tracheal stenosis (105,106). Routine
morning radiographic examination should be used to verify ETT position (107).
Furthermore, meticulous efforts and attention should focus on preventing
unplanned self-extubation, especially in patients who have a lengthy stay in
the intensive care unit (108).
dental injury can result with repeated attempts and increased physical force.
Malpositioning of the ETT is a serious complication and has been reported to
occur more frequently in women after emergency intubation (111). Patients
who are hemodynamically unstable and receiving vasopressor therapy prior to
intubation have the highest mortality rate associated with intubation (110).
IV. Conclusions
References
1. Hess DR, MacIntyre NR, Mishoe SC, Galvin WF, Adams AB, Saposnick AB.
Respiratory care: principles and practice. Philadelphia: WB Saunders, 2001.
2. Branson RD, Hess DR, Chatburn RL. Respiratory care equipment. Philadelphia:
J.B. Lippincott Company, 1999.
3. Dexter F, Reasoner DK. Theoretical assessment of normobaric oxygen therapy to
treat pneumocephalus. Anesthesiology 1996; 84:442 447.
4. Annane D, Troche G, Delisle F, et al. Effects of mechanical ventilation with normo-
baric oxygen therapy on the rate of air removal from cerebral arteries. Crit Care
Med 1994; 22:851 857.
5. Weaver LK, Hopkins RO, Chan KJ, et al. Hyperbaric oxygen for acute carbon
monoxide poisoning. N Engl J Med 2002; 347:1057 1067.
6. Green RJ, Dafoe DC, Rafn TA. Necrotizing fasciitis. Chest 1996; 110:219 229.
7. Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung
disease: a clinical trial. Nocturnal Oxygen Therapy Trial Group. Ann Intern Med
1980; 93:391 398.
8. Moloney ED, Kiely JL, McNicholas WT. Controlled oxygen therapy and carbon
dioxide retention during exacerbations of chronic obstructive pulmonary disease.
Lancet 2001; 357:526 528.
9. Hanson CW, Marshall BE, Frasch HF, Marshall C. Causes of hypercarbia with
oxygen therapy in patients with chronic obstructive pulmonary disease. Crit Care
Med 1996; 24:23 28.
10. Aubier M, Murciano D, Milic-Emili J, et al. Effects of the administration of O2 on
ventilation and blood gases in patients with chronic obstructive pulmonary disease
during acute respiratory failure. Am Rev Respir Dis 1980; 122:747 754.
26 Wong and Hess
11. Crossley DJ, McGuire GP, Barrow PM, Houston PL. Inuence of inspired oxygen
concentration on deadspace, respiratory drive, and PaCO2 in intubated patients with
chronic obstructive pulmonary disease. Crit Care Med 1997; 25:1522 1526.
12. Gomersall CD, Joynt GM, Freebairn RC, Lai CK, Oh TE. Oxygen therapy for hyper-
capnic patients with chronic obstructive pulmonary disease and acute respiratory
failure: a randomized, controlled pilot study. Crit Care Med 2002; 30:113116.
13. Robinson TD, Freiberg DB, Regnis JA, Young IH. The role of hypoventilation and
ventilation-perfusion redistribution in oxygen-induced hypercapnia during acute
exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care
Med 2000; 161:1524 1529.
14. Dick CR, Liu Z, Sassoon CS, Berry RB, Mahutte CK. O2-induced change in
ventilation and ventilatory drive in COPD. Am J Respir Crit Care Med 1997;
155:609 614.
15. Dunn WF, Nelson SB, Hubmayr RD. Oxygen-induced hypercarbia in obstructive
pulmonary disease. Am Rev Respir Dis 1991; 144:526 530.
16. Chien JW, Ciufo R, Novak R, et al. Uncontrolled oxygen administration and
respiratory failure in acute asthma. Chest 2000; 117:728 733.
17. Hedenstierna G, Rothen HU. Atelectasis formation during anesthesia: causes and
measures to prevent it. J Clin Monit 2000; 16:329 335.
18. White AC. The evaluation and management of hypoxemia in the chronic critically
ill patient. Clin Chest Med 2001; 22:123 134.
19. Carvalho CR, Schettino GP, Maranhao B, Bethlem EP. Hyperoxia and lung disease.
Curr Opin Pulm Med 1998; 4:300 304.
20. Ingrassia TS, Ryu JH, Trastek VF, Rosenow EC. Oxygen-exacerbated bleomycin
pulmonary toxicity. Mayo Clin Proc 1991; 66:173 178.
21. Allen J, Zwerdling R, Ehrenkranz R, et al. Statement on the care of the child with
chronic lung disease of infancy and childhood. Am J Respir Crit Care Med 2003;
168:356 396.
22. Jobe AH, Bancalari E. Bronchopulmonary Dysplasia. Am J Respir Crit Care Med
2001; 163:1723 1729.
23. Campbell EJ, Baker MD, Crites-Silver P. Subjective effects of humidication of
oxygen for delivery by nasal cannula. A prospective study. Chest 1988; 93(2):
289 293.
24. Hess D, Figaszewski E, Henry D, Hoffman S, Pino D. Subjective effects of dry
versus humidied low ow oxygen on the upper respiratory tract. Respir Ther
1982(Nov/Dec); 12:71 75.
25. Estey W. Subjective effects of dry versus humidied low ow oxygen. Respir Care
1980; 25:1143 1144.
26. Stausholm K, Rosenberg-Adamsen S, Skriver M, Kehlet H, Rosenberg J.
Comparison of three devices for oxygen administration in the late postoperative
period. Br J Anaesth 1995; 74:607 609.
27. McBrien ME, Sellers WF. A comparison of three variable performance devices for
postoperative oxygen therapy. Anaesthesia 1995; 50:136 138.
28. Nolan KM, Winyard JA, Goldhill DR. Comparison of nasal cannulae with
face mask for oxygen administration to postoperative patients. Br J Anaesth
1993; 70:440 442.
29. Ooi R, Joshi P, Soni N. An evaluation of oxygen delivery using nasal prongs.
Anaesthesia 1992; 47:591 593.
Oxygen Therapy and Airway Management 27
30. Williams AB, Jones PL, Mapleson WW. A comparison of oxygen therapy devices
used in the postoperative recovery period. Anaesthesia 1988; 43:131 135.
31. Redding JS, McAfee DD, Gross CW. Oxygen concentrations received from com-
monly used delivery systems. South Med J 1978; 71:169 172.
32. Schacter EN, Littner MR, Luddy P, Beck GJ. Monitoring of oxygen delivery
systems in clinical practice. Crit Care Med 1980; 8:405 409.
33. Bazuaye EA, Stone TN, Corris PA, Gibson GJ. Variability of inspired oxygen
concentration with nasal cannulas. Thorax 1992; 47:609 611.
34. Gibson RL, Comer PB, Beckham RW, McGraw CP. Actual tracheal oxygen concen-
trations with commonly used oxygen equipment. Anesthesiology 1976; 44:7173.
35. Waldau T, Larsen VH, Bonde J. Evaluation of ve oxygen delivery devices
in spontaneously breathing subjects by oxygraphy. Anaesthesia 1998; 53:256263.
36. Ward JJ, Gracey DR. Arterial oxygen values achieved by COPD patients breathing
oxygen alternately via nasal mask and nasal cannula. Respir Care 1985; 30:
250 255.
37. Hess D, DAgostino D, Magrosky S, Myers C, Shuman J. Effect of nasal cannula
displacement on arterial PO2. Respir Care 1984; 29:21 24.
38. Faireld JE, Goroszeniuk T, Tully AM, Adams AP. Oxygen delivery systemsa
comparison of two devices. Anaesthesia 1991; 46:135 138.
39. Dunlevy CL, Tyl SE. The effect of oral versus nasal breathing on oxygen
concentrations received from nasal cannulas. Respir Care 1992; 37:357 360.
40. Christopher KL. Transtracheal oxygen catheters. Clin Chest Med 2003;
24:489 510.
41. Hoffman LA. Novel strategies for delivering oxygen: reservoir cannula,
demand ow, and transtracheal oxygen administration. Respir Care 1994;
39:363 379.
42. Bliss PL, McCoy RW, Adams AB. Characteristics of demand oxygen delivery
systems: maximum output and setting recommendations. Respir Care 2004;
49:160 165.
43. Jensen AG, Johnson A, Sandstedt S. Rebreathing during oxygen treatment with face
mask. The effect of oxygen ow rates on ventilation. Acta Anaesthesiol Scand
1991; 35:289 292.
44. Milross J, Young IH, Donnelly P. The oxygen delivery characteristics of the
Hudson Oxy-one face mask. Anaesth Intensive Care 1989; 17:180 184.
45. Boumphrey SM, Morris EA, Kinsella SM. 100% inspired oxygen from a Hudson
maska realistic goal? Resuscitation 2003; 57:69 72.
46. Campbell DJ, Faireld MC. The delivery of oxygen by a venturi T piece.
Anaesthesia 1996; 51:558 560.
47. Jeffrey AA, Warren PM. Should we judge a mask by its cover? Thorax 1992;
47:543 546.
48. Hunter J, Olson LG. Performance of the Hudson Multi-Vent oxygen mask. Med J
Aust 1988; 148:444 447.
49. Hill SL, Barnes PK, Hollway T, Tennant R. Fixed performance oxygen masks: an
evaluation. Br Med J (Clin Res Ed) 1984; 288:1261 1263.
50. Foust GN, Potter WA, Wilons MD, Golden EB. Shortcomings of using two jet
nebulizers in tandem with an aerosol face mask for optimal oxygen therapy.
Chest 1991; 99:1346 1351.
28 Wong and Hess
patients with severe chronic obstructive pulmonary disease. Crit Care Med 2000;
28(8):2721 2728.
68. Diehl JL, Mercat A, Guerot E, et al. Helium/oxygen mixture reduces the work
of breathing at the end of the weaning process in patients with severe chronic
obstructive pulmonary disease. Crit Care Med 2003; 31(5):1415 1420.
69. Jolliet P, Tassaux D, Roeseler J, et al. Helium oxygen versus air oxygen nonin-
vasive pressure support in decompensated chronic obstructive disease: a prospec-
tive, multicenter study. Crit Care Med 2003; 31(3):878 884.
70. Jaber S, Fodil R, Carlucci A, et al. Noninvasive ventilation with helium oxygen in
acute exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit
Care Med 2000; 161(4 Pt 1):1191 1200.
71. Jolliet P, Tassaux D, Thouret JM, Chevrolet JC. Benecial effects of helium:oxygen
versus air:oxygen noninvasive pressure support in patients with decompensated
chronic obstructive pulmonary disease. Crit Care Med 1999; 27(11):2422 2429.
72. Austan F, Polise M. Management of respiratory failure with noninvasive positive
pressure ventilation and heliox adjunct. Heart Lung 2002; 31(3):214 218.
73. Chatmongkolchart S, Kacmarek RM, Hess DR. Heliox delivery with non-
invasive positive pressure ventilation: a laboratory study. Respir Care 2001;
46(3):248 254.
74. Tassaux D, Jolliet P, Thouret JM, Roeseler J, Dorne R, Chevrolet JC. Calibration of
seven ICU ventilators for mechanical ventilation with helium oxygen mixtures.
Am J Respir Crit Care Med 1999; 160(1):22 32.
75. Jolliet P, Tassaux D. Usefulness of helium oxygen mixtures in the treatment of
mechanically ventilated patients. Curr Opin Crit Care 2003; 9(1):45 50.
76. Berkenbosch JW, Grueber RE, Dabbagh O, McKibben AW. Effect of helium
oxygen (heliox) gas mixtures on the function of four pediatric ventilators. Crit
Care Med 2003; 31(7):2052 2058.
77. Devabhaktuni VG, Torres A Jr, Wilson S, Yeh MP. Effect of nitric oxide, peruor-
ocarbon, and heliox on minute volume measurement and ventilator volumes
delivered. Crit Care Med 1999; 27(8):1603 1607.
78. Oppenheim-Eden A, Cohen Y, Weissman C, Pizov R. The effect of helium on ven-
tilator performance: study of ve ventilators and a bedside Pitot tube spirometer.
Chest 2001; 120(2):582 588.
79. Natanson C, Shelhamer JH, Parrillo JE. Intubation of the trachea in the critical care
setting. JAMA 1985; 253:1160 1165.
80. Murrin KR. Intubation prcedure and causes of difcult intubation. In: Latto IP,
Rosen M, eds. Difculties in Tracheal Intubation. London: Baillere Tindall,
1985:75 89.
81. Langeron O, Masso E, Huraux C, et al. Prediction of difcult mask ventilation.
Anesthesiology 2000; 92:1229 1236.
82. Mallampati SR, Gatt SP, Gugino LD, et al. A clinical study to predict difcult
tracheal intubation. A prospective study. Can J Anaesth 1985; 32:429 434.
83. Frerk CM. Predicting difcult intubtion. Anaesthesia 1991; 46:1005 1008.
84. Cormack RS, Lehane J. Difcult tracheal intubation in obstetrics. Anaesthesia
1984; 39:1105 1111.
85. Wilson M, Spiegelhalter D, Robertson J, Lesser P. Predicting difcult intubation.
Br J Anaesth 1988; 61:211 216.
86. Knill RL. Difcult laryngoscopy made easy with a BURP. Can J Anaesth 1993;
40:279 282.
30 Wong and Hess
87. Takahata O, Kubota M, Mamiya K, et al. The efcacy of the BURP maneuver
during a difcult laryngoscopy. Anesth Analg 1997; 84:419 421.
88. Isono S, Tanaka A, Sho Y, et al. Advancement of the mandible improves velophar-
yngeal airway patency. J Appl Physiol 1995; 79:2132 2138.
89. Murashima K, Fukutome T. Effect of jaw-thrust maneuver on the laryngeal inlet.
Anaesthesia 1998; 53:203 204.
90. Tamura M, Ishikawa T, Kato R, et al. Mandibular advancement improves the lar-
yngeal view during direct laryngoscopy performed by inexperienced physicians.
Anesthesiology 2004; 100:598 601.
91. Stoelting RK, Miller RD. Basics of Anesthesia. New York: Churchill Livingstone,
1994.
92. Sellick B. Cricoid pressure to control regurgitation of stomach contents during
induction of anesthesia. Lancet 1961; 2:404 406.
93. Anton WR, Gordon RW, Jordan TM, Posner KL, Cheney FW. A disposable
end-tidal detector to verify endotracheal intubation. Ann Emerg Med 1991;
20:271 275.
94. Sayah AJ, Peacock WF, Overton DT. End-tidal measurement in the detec-
tion of esophageal intubation during cardiac arrest. Ann Emerg Med 1990;
19:857 860.
95. Ornato J, Shipley J, Racht E, et al. Multicenter study of a portable, hand-size,
colorimetric end-tidal carbon dioxide detection device. Ann Emerg Med 1992;
21:518 523.
96. Pollard RJ, Lobato EM. Endotracheal tube location veried reliably by cuff palpa-
tion. Anesth Analg 1995; 81:135 138.
97. Practice guidelines for management of the difcult airway. An updated report by the
American Society of Anesthesiologists Task Force on Management of the difcult
airway. Anesthesiology 2003; 98:1269 1277.
98. Benumof, JL. Laryngeal mask airway and the ASA difcult airway algorithm.
Anesthesiology 1996; 84:686 699.
99. Benumof JL. Management of the difcult adult airway, with special emphasis on
awake tracheal intubation. Anesthesiology 1991; 75:1087 1110.
100. Baskett PJ, Bossaert, Carli P, et al. Guidelines for the advanced management of the
airway and ventilation during resuscitation. Resuscitation 1996; 31:201 230.
101. Baskett PJ, Parr MJ, Nolan JP. The intubating laryngeal mask: results of a multicen-
ter trial with experience of 500 cases. Anaesthesia 1998; 53:1174 1179.
102. Benumof JL, Scheller MS. The importance of transtracheal jet ventilation in the
management of the difcult airway. Anesthesiology 1989; 71:769 778.
103. Henschke CI, Yankelevitz DF, Wand A, et al. Accuracy and efcacy of chest
radiography in the intensive care unit. Radiol Clin North Am 1996; 34:21 31.
104. Owen RL, Cheney FW. Endobronchial intubation: a preventable complication.
Anesthesiology 1987; 67:255 257.
105. Cooper JD, Grillo HC. The evolution of tracheal injury due to ventilatory assistance
through cuffed tubes: a pathologic study. Ann Surg 1969; 169:334 348.
106. Cooper JD, Grillo HC. Analysis of problems related to cuffs on intratracheal tubes.
Chest 1972; 62:21S 27S.
107. Bekemeyer WB, Crapo RO, Calhoon S, et al. Efcacy of chest radiography in a
respiratory intensive care unit. Chest 1985; 88:691 696.
Oxygen Therapy and Airway Management 31
108. Atkins PM, Mion LC, Mendelson W, et al. Characteristics and outcomes of patients
who self-extubate from ventilatory support. Chest 1997; 112:1317 1323.
109. Keane WM, Rowe LD, Denneny JC, et al. Complications of intubation. Ann Otol
Rhinol Laryngol 1982; 91:584 587.
110. Schwartz DE, Matthay MA, Cohen NH. Death and other complications of
emergency airway management in critically ill adults. Anesthesiology 1995;
82:367 376.
111. Schwartz DE, Lieberman JA, Cohen NH. Women are at greater risk for malposi-
tioning of the endotracheal tube after emergent intubation. Crit Care Med 1994;
22:1127 1131.
This page intentionally left blank
2
Non-Invasive Ventilation in Critical Care
I. Introduction
33
34 Soo Hoo
Absolute contraindications
Coma
Respiratory arrest
Cardiac arrest
Any condition requiring immediate intubation
Clinical conditions
Rapidly reversible clinical conditions
COPD exacerbations
Cardiogenic pulmonary edema
After discontinuation of mechanical ventilation
Early extubation
? Post-extubation
? Pneumonia
COPD and pneumonia
Use non-invasive ventilation concomitantly with other therapy
have been in patients with rapidly reversible conditions. The greatest experience
has accrued in these conditions. This virtually eliminates candidates with con-
ditions that may require an extended period (weeks) of ventilatory support or
an illness that may require weeks for resolution. Of course, patients with mild
disease may be candidates for NIMV, but must be evaluated on a case by case
basis. Patients also need to be treated concomitantly with therapy directed at
their underlying disease. NIMV provides an important adjunct to therapy, best
suited for those who require a brief period of support to prevent frank respiratory
failure while allowing the other components of therapy to take effect. These clini-
cal conditions will be examined in detail in the following sections.
Randomized Trials
The rst prospective randomized trial, reported by Bott et al. (15), involved 60
COPD patients and compared NPPV via nasal mask volume-cycled ventilation
plus conventional treatment with conventional treatment administered in a
specialized ward setting. They demonstrated greater improvement in respiratory
acidosis (pH and PaCO2) in treated patients compared with conventional therapy,
along with lower dyspnea scores. Mortality was lower but did not reach statistical
signicance. Those who died were more acidemic (pH 7.31 vs. 7.35) and hyper-
capnic (PaCO2 71 vs. 64 mmHg) on admission than survivors.
Another single-center study reported by Kramer et al. (16), compared nasal
mask bi-level positive airway pressures (BiPAP) plus standard therapy to standard
therapy alone in 31 patients (23 COPD). The intubation rate was markedly reduced
in the BiPAP group, 31% versus 73%, and 9% versus 67% for COPD patients, but
with no difference in mortality. They noted marked improvement in PaCO2, with a
decline in heart rate, respiratory rate, and dyspnea score after just one hour of
BiPAP.
In a multi-center, multi-national study reported by Brochard et al. (17), 85
patients with COPD exacerbations were randomized to either NPPV with face
mask pressure support of 20 cmH2O plus conventional therapy or to conventional
therapy alone. Endotracheal intubation was signicantly lower in the non-invasive
group (26% vs. 74%). The response rates were not different between the participat-
ing institutions, and this treatment modality was demonstrated effective in more
than one institution. The non-invasive group had fewer complications, specically
fewer episodes of nosocomial pneumonia and sepsis, lower mortality rate, and
shorter length of stay. It should be pointed out, though, that the one-hour decline
in PaCO2 noted by others was not evident in this study.
The largest multi-center, prospective, randomized trial, reported by Plant
et al., compared BiPAP using either face or nasal masks and randomized 236
patients (7,18). This occurred in a non-specialist ward setting in contrast to the
previously reported ward and ICU settings. They focused on those with mild
to moderate acidosis with a pH 7.25 as an exclusion criterion. Patients were
treated according to an established protocol, primarily initiated by nursing
staff. The primary outcome measure was the need for intubation, but once
this criterion was met, intubation was at the discretion of the attending physician
and management could continue with NPPV or even standard medical care. The
non-invasive group had a statistically signicantly lower number of patients who
met the need for intubation, 15% compared with 27%, as well as a lower mor-
tality rate, 10% versus 20%. However, there was much less difference in the
actual intubation rate (NPPV, 6% vs. 10%). In a sub-group analysis, the differ-
ence was most evident in those with a pH . 7.30. There was no benet of
NPPV in patients with an initial pH , 7.30.
The methodology of these and other trials have varied widely, ranging from
single-center to multi-center trials, different mask interfaces (nasal, face, or
Non-Invasive Ventilation in Critical Care 37
both), different modes of assisted ventilation (volume vs. pressure), with different
inclusion and exclusion criteria, outcome measures, intubation criteria, and even
cross-over therapy, making consistent comparisons difcult. These studies often
represent narrowly selected groups of patients. In the trial reported by Brochard
et al., only 31% of those evaluated were enrolled in their study. Intubation is
probably the most important outcome measure, but conclusions are tempered
when one outcome is meeting criteria for intubation and the other is intubation.
Some trials enrolled relatively small numbers of patients, and although with sig-
nicant differences, they are limited by potential problems with statistical power.
Meta-analyses may help to provide better insight into the effects of therapy by
pooling patients and ameliorating differences in study design.
Meta-Analyses
Three meta-analyses have been reported, largely focusing on the use of NPPV in
COPD patients, although other causes of respiratory failure are included (19 22).
They have all reached the same general conclusion that NPPV reduces the intu-
bation rate, mortality, and length of stay in COPD patients. Peter et al. analyzed
15 studies, eight involving patients with COPD exacerbations. Focusing on the
reports of COPD patients, they noted a signicant reduction in mortality of
13% (95%CI: 6 21%), need for intubation of 18% (95%CI: 3 33%), and
decrease in hospital length of stay of 5.7 days (95%CI: 1.2 10.1 days).
Keenan et al. restricted their analysis to COPD patients and found a risk reduction
of 10% (95%CI: 5 15%) in mortality, a risk reduction of 28% (95%CI: 15 40%)
for endotracheal intubation, and a decrease in the length of hospital stay by 4.6
days (95%CI: 2.3 6.8 days). They also performed sub-group analysis and
found NPPV had the most impact in those with moderately severe exacerbations
dened by a pH , 7.30. It should be noted that patients in two studies reporting
no benet with NPPV had very mild disease with an average pH of 7.33 + 0.01
and 7.40 + 0.04, respectively. This may have skewed their analysis against those
with less severe disease, as they did not identify any differences in outcome in
those with less severe exacerbations. This conclusion runs contrary to the experi-
ence reported by Plant et al., who found that the greatest benet appeared to be in
those with less severe disease dened by a pH . 7.30.
In summary, the majority of reported data supports the use of NIMV in
COPD. It is particularly effective in patients with hypercapnic respiratory
failure. The optimal thresholds of treatment remain to be dened. There are
reports matching our own experience that COPD patients with an initial pH
range 7.10 can be successfully managed with NIMV, with the best results prob-
ably in patients with an initial pH . 7.25 (moderate severity). It does not add to
the management of patients with mild exacerbations (dened by a pH . 7.35).
When successful, it decreases the intubation rate, length of stay, and mortality
of these patients. In addition to reversing respiratory failure, mortality seems to
be decreased because of fewer episodes of nosocomial pneumonia. There
38 Soo Hoo
exists great heterogeneity in the study design of these trials, and generalization is
tempered by differences in many areas, including threshold for intubation.
total of 40 patients and noted improvement in respiratory rate, heart rate, blood
pressure, PaCO2, and oxygenation after as little as 10 minutes of support. The
intubation rate in CPAP patients was 30%, compared with 60% in control
patients. Bersten et al. compared face mask CPAP at 10 cm H2O in a total of
39 patients with severe pulmonary edema. Their series is noteworthy for the
severity of illness of their patients, with mean APACHE II scores of 20 and arte-
rial blood gases in the CPAP-treated patients of a pH of 7.18 + 0.08, PaCO2 of
58 + 8 mmHg, and PaO2/FIO2 of 138 + 32. Control patients had a similar
severity of disease. None of the CPAP patients required intubation, compared
with 35% of the controls. Lin et al. reported the largest series with CPAP, invol-
ving a total of 100 patients, all with invasive hemodynamic monitoring, treated
with face mask CPAP and titrated over a three-hour period from 2.5 to
12.5 cm H2O. Their results mirrored that of the two prior studies but also demon-
strated a reduction in shunt fraction and improvement in cardiac indices with
CPAP. The intubation rate in CPAP patients was 16%, compared with 36% in
controls.
The improvement in outcome measures did not achieve statistical signi-
cance in these studies, but there were strong trends towards a reduction in intuba-
tion rate and mortality. In a meta-analysis of the three studies, pooled results
demonstrated a risk reduction in intubation of 26% (95%CI: 13 38%), with a
trend towards decreased mortality of 6.6% (95%CI: 23 16%) (32). The duration
of hospitalization was not different, although there was a decrease in the ICU stay
of patients in the study by Bersten et al. by about a day. In another randomized
study in patients with pulmonary edema complicating acute myocardial infarc-
tion, Takeda et al. provided 4 to 10 cm H2O nasal CPAP to 11 patients and
demonstrated a reduction in intubation rate (18% vs. 73%), mortality (6% vs.
64%), and cardiac hemodynamics (pulmonary artery wedge pressure, cardiac
index, stroke volume index) with CPAP.
Given this experience with CPAP, it follows that NPPV would be equally or even
more efcacious in cardiogenic pulmonary edema. This might be expected
because NPPV breaths are augmented and assisted, unlike CPAP breaths.
Work of breathing would be expected to be less with NPPV than CPAP. While
several investigators have included patients with cardiogenic pulmonary
edema, four have focused on the addition of NPPV to conventional therapy in
cardiogenic pulmonary edema (33 35). All used a combination of pressure
cycled ventilation with positive end-expiratory pressure (PEEP), either PSV
with PEEP or BiPAP with bedside or portable ventilators. Three were conducted
at a single institution, with the largest a multi-center study. Demonstration of
benet with NPPV was not as evident as in the other studies that used CPAP.
Masip et al. (33) randomized a total of 40 patients and noted a lower
intubation rate (5% vs. 33%) with more rapid improvement in tachypnea,
40 Soo Hoo
Randomized Trials
Investigations have evaluated NPPV in two basic strategies in the post-extubation
period. One can be designated an early extubation approach in which patients
who do not meet standard criteria for extubation are extubated early in their
course and managed with NPPV. Nava et al. (46) rst reported on this early extu-
bation approach in 50 COPD patients with hypercapnic respiratory failure after
48 hours of intubation. Those unable to tolerate a T-tube weaning trial were ran-
domized to either continued ventilator support or extubation with face mask PSV.
The NPPV patients received an average pressure support of 17.6 + 2.1 cm H2O,
comparable to those continuing with mechanical ventilation. Those treated with
NPPV had greater success at discontinuation of mechanical ventilation (88% vs.
68%) as shorter duration of mechanical ventilation, few ICU days, and reduced
mortality. These patients had no episodes of nosocomial pneumonia compared
with seven in the invasive ventilation group.
In a similar trial, Girault et al. (47) randomized 33 mostly COPD patients to
this early extubation approach if they had met established weaning criteria but
Non-Invasive Ventilation in Critical Care 43
had failed a two-hour T-tube trial. Those randomized to NPPV had been
intubated an average of 4.6 + 1.9 days. By study design, the duration of invasive
ventilation was shorter for the NPPV group, but the NPPV patients actually
required a longer period of ventilatory assistance related to weaning
(11.5 + 5.2 days vs. 3.5 + 1.4 days) than invasively ventilated patients albeit
for a shorter time period each day. No differences were noted in the numbers
of patients eventually able to discontinue mechanical ventilation, duration of
ICU and hospital stay, complications, and mortality. Ferrer et al. (48) also
used this strategy but chose patients who had failed T-tube trials for three con-
secutive days prior to randomization to either BiPAP or continued mechanical
ventilation. In their 43 mostly COPD patients, those randomized to NPPV had
greater success with eventual discontinuation of mechanical ventilation, fewer
episodes of nosocomial pneumonia, fewer ICU days, and improved mortality.
As expected from their study design, the NPPV patients also had a shorter dur-
ation of invasive ventilation. They noted that hypercapnia (PaCO2 . 45 mmHg)
during their T-tube trials was an indicator of prolonged ICU stay and decreased
survival. On the other hand, using a similar approach, Hill et al. (49) reported a
trial involving 21 patients, which constituted less than 10% of their ICU patients
with respiratory failure. While the NPPV-treated patients had fewer days of endo-
tracheal intubation, the re-intubation rate was also higher in this group. All of the
patients in these trials were intubated on average a week or less prior to extuba-
tion and treatment with NPPV.
In conclusion, while this approach is promising, the experience is somewhat
mixed. Nosocomial pneumonia may be averted by this approach, but patients may
still require continued ICU care because of a tenuous respiratory status. The ideal
candidate for this approach remains to be dened but might be an otherwise
improving COPD patient who just requires a small amount of assistance to
regain the balance between ventilatory load and strength. This early extubation
approach seems to be effective up to a week following intubation.
The other approach in this arena involves the use of NPPV to treat post-
extubation respiratory distress and thereby avoid re-intubation and its adverse
effects. The available data casts some doubt on the efcacy of NPPV under
these circumstances. Jiang et al. (50) randomized 93 patients after extubation
to either face mask BiPAP or oxygen therapy. It should be noted that extubation
in 37 of these patients was unplanned (self-extubation). There was a 22% re-intu-
bation rate but no signicant difference between the two treatment groups,
although the re-intubation rate was higher in those receiving BiPAP. Keenan
et al. (51) randomized 81 patients who developed post-extubation respiratory dis-
tress within 48 hours of extubation to either NPPV or standard therapy. Patients
developed respiratory distress on average about 10 + 10 hours following extuba-
tion. No differences were noted in the re-intubation rate, duration of mechanical
ventilation, episodes of pneumonia, duration of ICU or hospital stay, or mortality.
The study patients represented only about 10% of screened patients who met
inclusion criteria. In addition, they identied 24 patients who met inclusion
44 Soo Hoo
criteria but were treated with NPPV outside the study during the study period.
Their re-intubation rate was lower (44%) as opposed to 70% in the study patients.
It was unclear if selection bias may have skewed the outcome of this trial. Sub-
sequently, a multi-center trial involving 221 patients, also did not identify any
difference in the re-intubation rate (48%) with NPPV and noted increased mor-
tality in those assigned to NPPV (25% vs 14%) (52). These ndings temper
the use of NPPV in this group of patients, but sub-group analysis suggests poss-
ible benet in those with COPD.
In summary, the support for NPPV after discontinuation of mechanical
ventilation is inconclusive. It appears to work best in an early extubation strategy
in which patients who do not meet conventional weaning criteria are extubated
and treated with NPPV within a week of intubation. Patients with COPD may
be the sub-group most likely to benet from this approach. The evidence to
support the use of NPPV in patients who meet weaning criteria and then
develop respiratory distress after extubation is disappointing.
D. Pneumonia
Although this has been the primary focus of only one randomized trial, patients
with community-acquired pneumonia constitute a large proportion of study
patients in several other investigations. As in other conditions, patients with pneu-
monia may benet from the improvement in oxygenation and relief of dyspnea
with NIMV, but there may be concerns with hemodynamic instability, secretion
clearance, and refractory respiratory failure. While the gas exchange defects of
pneumonia can be corrected, there may not be the rapid recovery that is character-
istic of the other conditions for which NIMV has demonstrated efcacy.
Confalonieri et al. (53) randomized fty-six consecutive patients with
severe community-acquired pneumonia to either NPPV with PSV (14.8 +
4.7 cmH2O) and PEEP (4.9 + 1.7 cmH2O) plus standard therapy or standard
therapy alone. There was a reduction in the intubation rate and duration of
ICU stay, but the improvement was noted only in those patients with concomitant
COPD. There was no improvement in any outcome measures in patients without
obstructive lung disease. It was unclear if the benet of NPPV may actually have
been due to the effect of NPPV on COPD, as opposed to pneumonia. The COPD
patients had a respiratory acidosis and actually had a higher PaO2/FIO2 than their
non-COPD counterparts.
As noted earlier, CPAP has potential to be of benet in patients with hypoxe-
mic respiratory failure, most notably cardiogenic pulmonary edema. Evaluation of
its efcacy in a larger group of heterogeneous patients with acute hypoxemic res-
piratory failure has not conrmed this benet. Delclaux et al. (54) randomized 123
patients to face mask CPAP (up to 10 cm H2O) in addition to standard therapy.
Sixty-seven of these patients had pneumonia, and as a group, there was improve-
ment in PaO2/FIO2 and symptoms at one hour, but ultimately no differences in
these outcome measures, intubation rate, duration of ICU stay, or mortality.
Non-Invasive Ventilation in Critical Care 45
There were four cardiac arrests in the CPAP group, possibly related to delays in
intubation. There was no separate analysis presented about pneumonia patients.
CPAP may not have been the optimal mode of NIMV. In another prospective
trial of patients with acute hypoxemic respiratory failure, Ferrer et al. (55) random-
ized 105 patients to NPPV with either BiPAP (IPAP 16 + 3 cm H2O;
EPAP 7 + 2 cm H2O) plus standard therapy or standard therapy in a hetero-
geneous group of patients with hypoxemic respiratory failure. Thirty-four patients
had pneumonia without COPD, and they were able to demonstrate a difference in
the intubation rate (26% vs. 73%) with NPPV. NPPV patients also had greater
improvement in hypoxemia and tachypnea, as well as a decrease in the incidence
of septic shock, presumably due to decreased nosocomial pneumonia, decreased
ICU mortality, and 90-day mortality. Although data were presented for the
whole cohort, the benet of NPPV seemed to be restricted to the pneumonia
patients. Of note, there was no difference in the intubation rate for patients with
cardiogenic edema, mirroring the experience reported by Nava et al. Other ran-
domized investigations of NPPV that include pneumonia patients have not had
more than a handful of patients with pneumonia, making it difcult to reach any
meaningful conclusions (56 58).
These ndings urge caution in the use of NPPV in patients with pneumonia.
It is important to note that in analyses of large groups of patients treated with
NPPV, pneumonia has been identied as a predictor of failure with NPPV
(59,60). This concern may lessen as improvements in ventilatory devices
occur and clinicians gain experience and understanding of the limitations of
NPPV. As in other conditions, careful patient selection and close monitoring is
warranted.
The above constitutes the bulk of the favorable experience with NIMV.
These conditions and the primary benets of NIMV are summarized in
Table 2. NIMV has also been used in other conditions, but the evidence to rec-
ommend its routine use is limited. Those conditions that have undergone
single randomized controlled trials are discussed in the following section and
also noted in Table 3. Endorsement of the use of NIMV may be limited to
sub-groups or limited by relatively modest experience, thereby tempering its
widespread use in these conditions.
Outcome variable
Chronic obstructive
pulmonary disease
Cardiogenic pulmonary edema
CPAP 2 +
PSV + 2 2
BiPAP + 2 2 2
CPAP vs. BiPAP (BiPAP) 2 2 2
After discontinuation of
mechanical ventilation
Early extubation
Post-extubation + 2 2 2
Pneumonia
In a group of mostly liver and renal transplant patients, Antonelli et al. (62)
randomized 51 with acute hypoxemic respiratory failure to either NPPV with
pressure support and PEEP plus supportive therapy or supportive therapy
alone. The major causes of respiratory failure were pneumonia, cardiogenic
pulmonary edema, acute respiratory distress syndrome (ARDS), and mucus plug-
ging. They were able to demonstrate greater improvement in oxygenation, intu-
bation rate, incidence of fatal complications, most notably in septic shock,
duration of ICU stay, and mortality. However, the benet was most evident in
those with cardiogenic pulmonary edema, as the benet was not substantially
different from standard therapy in the other sub-groups. In another study of
mostly neutropenic patients with hematologic malignancies or bone marrow
transplants, Hilbert et al. randomized 52 of these patients with febrile hypoxemic
respiratory failure and pulmonary inltrates to NPPV with pressure support
(15 + 2 cm H2O) and PEEP (6 + 1 cm H2O). About half of the patients had a
microbiologic diagnosis of pneumonia. They were able to demonstrate improved
oxygenation, with lower intubation rates, complications, ICU stay, and hospital
mortality. The benet was most pronounced in those with an identied cause
of their pneumonia. It should be noted that NPPV was only administered, on
average, eight hours a day for an average of four days. This indirect measure
of severity would suggest that these patients had a relatively modest case of pneu-
monia as highlighted in the accompanying editorial (63).
In summary, NPPV may benet a very select group of immunocompro-
mised patients. Experience from these single-center studies would suggest that
those with cardiogenic pulmonary edema and an identied cause of pneumonia
may be the best candidates for successful application. Caution should be used
in its application in any other sub-group of immunocompromised patients
because a delay in more appropriate intubation may increase their risk for
adverse events and death.
B. Asthma
Although the treatment of decompensated asthma closely mirrors that of COPD,
there is a striking paucity of reported experience with NIMV in this condition.
There are sufcient differences in the pathophysiology of these two conditions
that may not allow the experience in COPD patients to be translated to asthma
patients. Dynamic hyperination may play more of a role in COPD patients
than in asthma patients, who while subject to dynamic hyperination, are also
subject to increased airway resistance and sudden asphyxia not typically seen
in the COPD patients. Most of the reported experience in asthma is found in
case series, but there has been a randomized evaluation in asthma patients in
an emergency room setting (64). Soroksky et al. randomized 30 patients with
asthma to nasal BiPAP (maximum settings of IPAP 15 cm H2O; EPAP
5 cm H2O) plus standard therapy or standard therapy plus sham BiPAP with
settings of 1 cm H2O. They noted a more rapid and greater improvement in
lung function as well as a lower hospitalization rate. This experience is promis-
ing, but it should serve as the impetus for larger studies. As with other conditions,
this is a reasonable option provided there is careful patient selection and close
monitoring of response to therapy.
48 Soo Hoo
pulmonary edema have been included in many of the prospective studies. Case
series would suggest that patients with ARDS can be successfully treated with
NIMV, but the success seemed to be in those with less severe disease (68).
In their evaluation of hypoxemic respiratory failure, Ferrer et al. (55) enrolled
15 patients with ARDS and found no benet in any outcome variable in this
sub-group. In multi-variate analysis, ARDS was identied as a risk factor for
intubation. This mirrors the nding by Antonelli et al. (60), who identied
ARDS as a risk factor for NPPV failure. There is insufcient data to make any
recommendations on the use of this modality, but the time course of this
illness and many associated complications make it unlikely that it can be ef-
ciently managed with NIMV, except in very mild cases.
G. Other Conditions
There are many other respiratory conditions for which NIMV may be benecial
during episodes of respiratory failure, and they are listed in Table 3. However,
50 Soo Hoo
many of these are reported in case series format and not randomized, controlled
studies. These conditions include cystic brosis, kyphoscoliosis, neuromuscular
disease, decompensated obstructive sleep apnea, and mild Pneumocystis carinii
pneumonia. Except for Pneumocystis carinii pneumonia, most represent acute
decompensation of chronic respiratory failure. With cystic brosis, extensive
experience has accumulated and been reported by Madden et al. (74).
They have treated over 100 patients, and in 90 patients awaiting or undergoing
evaluation for lung transplantation, 38 (42%) have been stabilized with
NIPPV, with 28 having received a lung transplant and the others still on
the list. Some patients were maintained for over a year and a half with
NIMV. Even in a sub-group not composed of lung transplant candidates,
some were able to survive using NIMV for over a year. This experience indi-
cates that NIPPV can help provide a bridge of support until transplantation
and thereby avoid the complications associated with intubation and mechanical
ventilation.
Despite over a decade of experience with NIMV, its use remains limited and
often conned to medical centers with expertise or a special interest in its appli-
cation. Recent surveys on its actual use are telling. In a questionnaire survey,
Doherty and Greenstone (75) found that NIMV was available in 48% of the hos-
pitals surveyed in the United Kingdom. Even then, about 70% of hospitals treated
,20 patients per year. Less than 10% treated .60 patients per year. In another
month long, multi-country survey on ventilator use, over 4000 patients were ven-
tilated for .12 hours (76). Of that group, only 202 (4.7%) were treated with
NIMV, and the use was even less (1%) in a point prevalence study (77). In
another European survey of mostly French institutions, some with familiarity
with NIMV, only 16% of their ICU patients were treated with NIMV (78).
They found a wide range in the use of NIMV in the 42 ICUs surveyed, with
eight ICUs never using NIMV and one ICU using NIMV in 67% of their ICU
patients requiring ventilatory support. Hypercapnia (COPD), post-operative res-
piratory failure, pneumonia, and cardiogenic pulmonary edema were the frequent
conditions treated with NIMV. Therefore, despite the success reported, relatively
few patients with respiratory failure are placed on this treatment. There are
areas of its application that require close attention. Patient selection, mask
leaks, patient synchrony, ventilator efciency, nursing, and respiratory therapy
time are all potential limiting factors in successful application of NIMV.
These were potential stumbling blocks for initial investigators and are issues
that continue to limit the use of NIMV. A better understanding of these factors
not only helps dene the optimal treatment group, but also helps predict its
success.
Non-Invasive Ventilation in Critical Care 51
A. Patient Selection
Patient selection is probably the most important factor to consider prior to initi-
ating NIMV. This is an added element that encompasses all of the clinical con-
ditions being treated. Once it is determined that the patient has a condition readily
treatable with NIMV (i.e., COPD exacerbation, cardiogenic pulmonary edema,
pneumonia, etc.), then a determination needs to be made if other aspects of
their presentation may preclude treatment with NIMV. This is cruial as those des-
tined to fail or only achieve a partial response with NIMV may be better served
with immediate intubation and mechanical ventilation. This spares what may
be ineffective therapy, avoids further progression of respiratory failure, and
limits associated complications from incompletely treated respiratory failure or
intubation under duress. Factors that inuence the success of NIMV can be
grouped into areas that deal with severity of illness, co-morbidities, technical
issues, and patient cooperation.
One of the unanswered questions involving patient selection (and in turn
predicting the success of NIMV) has been the limits imposed by the patients
severity of illness. This applies to both ends of the spectrum. Patients may
have a relatively mild severity of illness, destined to improve with medical
therapy, and not require NIMV. NIMV would be unnecessary and excessive.
There is general agreement that the most severely ill patients are better served
by immediate intubation without a trial of NIMV. Dening the limits of appli-
cation has been elusive. The severity of acidosis correlates closely with the like-
lihood for intubation in those presenting with hypercapnic respiratory acidosis.
Thresholds have been identied based on prospective and retrospective analyses.
A pH of 7.25 has been associated with a .50% intubation rate. It is noteworthy
that a pH of 7.25 has been used to limit enrollment in several NIMV studies. Plant
et al. (79) analyzed their database of patients with respect to pH and PaCO2 as
predictors of failure and meeting criteria for intubation. These two variables
are obviously linked and co-dependent. The pH may be the more important vari-
able. The relative risk of failure was greater with worsening pH for the same level
of hypercapnia than hypercapnia for the same level of pH. The relative risk for
failure with a pH 7.25 and PaCO2 45 was 3.89 for those treated with NIMV
and 9.98 for those treated with standard therapy. This compares to a relative
risk of unity for standard therapy with a pH 7.35, and 0.39 with NIMV. Higher
APACHE II scores indicative of an increased severity of illness have predicted
NIMV failures in several series (59,80).
As medical staff gain more expertise with this modality, the success rate
improves as expected. Carlucci et al. (81) have analyzed their experience with
COPD patients during 208 episodes of acute respiratory failure. While the
failure rate was constant (17.2%) over an eight-year period, patients were being
treated with a greater severity of respiratory acidosis (mean pH: 7.20 + 0.08
vs. 7.25 + 0.07) and with greater successful treatment of the more severe patients
52 Soo Hoo
during the latter periods of their survey compared with the early years. The failures
during the early period of their survey (1992 1996) were very similar in
terms of severity of illness to those successfully treated during the latter years
(1997 1998). One consistent nding in both periods was that successfully
treated patients had a marked improvement with NIMV after one hour of
therapy (pH improvement: 0.06 0.09), while failures had no improvement at
all. The lower limit of effectiveness remains to be dened, but limiting
NIMV to patients with hypercapnic respiratory acidosis to a pH . 7.10 is a
reasonable common sense limitation. It should be noted that there is very little
experience in these severely acidemic patients since most have been excluded
from clinical trials.
B. Co-Morbidities
In addition to severity of illness, co-morbidities also limit effective NIMV. These
are summarized in Table 4. These limitations generally involve issues that render
optimal management difcult without intubation and mechanical ventilation.
These include cardiac instability involved with acute myocardial infarction, ven-
tricular dysrhythmias, shock, GI bleeding, coma, profound lethargy, and status
epilepticus. The main limiting factor revolves around the need to protect the
airway from aspiration or the need to control the airway during treatment. Hemo-
dynamically unstable patients are best managed with intubation, eliminating a
potentially confounding factor in already critically ill patients. Those who
require airway protection or with mechanical upper airway obstruction are
Hemodynamic instability
Acute complicated myocardial infarction
Hemodynamically signicant cardiac dysrrhythmias
Severe septic shock
GI bleeding
Need for airway protection
Coma or any condition with an unstable respiratory drive
Extensive cerebral vascular accident or hemorrhage
Status epilepticus
Potential for life-threatening airway compromise
Head and neck tumors
Any tumor with extrinsic airway compression
Angioedema or anaphylaxis
Generally, the most effective mask is the smallest mask that provides an
adequate t. Patients should be upright, with the head of the bed elevated 458
or more at the time of mask placement. Patients may hold the mask up to their
face, or the mask and attached head straps are loosely placed over the patients
head prior to strapping it in place. Allowing patients to hold the mask also
allows them to become accustomed to the mask. It is important that the head
straps are in place prior to mask placement as this facilitates placement.
For nasal masks, the mask should allow adequate clearance for the nose, but
not cover the vermillion of the upper lip. The soft palate and teeth allow proper
closure of the oronasal cavity during positive pressure and help minimize leaks.
Edentulous patients may have difculty with maintaining proper mask position
and are often better served with a face mask. A face mask should t comfortably
on the mandible, just below the lower vermillion. A mask that encroaches on the
chin is subject to large leaks and ineffective ventilation. The straps should be
tight but allow easy passage of one to two ngers between the face and straps.
Very tight straps are not only uncomfortable, but they increase the likelihood of
developing pressure ulcers over the nose or face. Chin straps are available if
mouth leaks are unavoidable. A small amount of leak may be unavoidable despite
multiple manipulations. Leaks around the nose into the eyes are the most distressing
for patients. Nasal plugs or nasal pillows are not suitable for acutely ill patients. A
mouthpiece is an alternative, although that would require more active cooperation
by the patient than a mask that is anchored to the patient by head straps. Most
masks have the option of bleeding oxygen directly into the mask via oxygen
tubing, but this runs the risk of creating excess pressure in this enclosed space. If
oxygen is necessary, it is better delivered via an adaptor attached to ventilator
tubing. In a bench study assessing the inuence of the site of oxygen delivery
(mask or circuit), with various combinations as to the leak port site (mask,
plateau exhalation value, or circuit) and increasing levels of IPAP and EPAP
(1025 and 510 cm H2O, respectively), the highest concentrations of oxygen
were delivered with oxygen delivered at the mask and the leak in the circuit with
the lowest levels of pressure (83). This might be an issue with some of the older,
smaller bi-level ventilatory assist devices. Newer ventilators are able to blend
oxygen and deliver through the ventilator tubing to allow precise control of FIO2.
Nasal and oronasal (including full face) masks are the primary interfaces
used in NIMV. A summary of their advantages and disadvantages is provided
in Table 5. For most patients, the nasal mask is more comfortable, less obtrusive,
has less dead space, and allows easier communication. The nasal mask does
require a more coordination by the patient. Patients must be able to maintain a
mouth seal and are subject to more mouth leaks. Some patients may be unable
to prevent mouth leaks (edentulous, mouth breathing pattern, pursed lips breath-
ing, and obtundation). The oronasal mask allows for mouth breathing, creates
more difculty with speech and expectoration, and may cause claustrophobia
in some patients. Most experience has been with an oronasal mask that covers
the nose and the mouth. Face masks that cover the whole face including the
Non-Invasive Ventilation in Critical Care 55
eyes are available as an alternative to nasal and oronasal masks, although with
limited experience (84). Helmet masks have also been developed, but likewise
with limited experience (85,86).
It follows that nasal masks are better suited for the less dyspneic patient and
the oronasal face mask better for the more dyspneic patient who is more likely to
be a mouth breather and subject to mouth leaks. Because this comprises the
majority of candidates for NIMV, face mask NIMV is the modality of choice
at our institution. Face mask ventilation is also the predominant interface for
NIMV in reported studies, used in about 70% of treated patients (87). Nasal ven-
tilation would be reserved for those cooperative patients who are unable to toler-
ate the face mask. It has been observed that issues with claustrophobia or mask
discomfort are often outweighed by the relief obtained with unloading of the res-
piratory system.
As might be expected, differences in randomized comparisons of the masks
in acute respiratory failure primarily reect difference related to leaks and efcacy.
In a prospective, randomized trial, Kwok et al. (88) compared nasal and oronasal
masks in 70 patients with acute respiratory failure (primarily pulmonary edema
and COPD). They noted comparable effects on physiologic indices (dyspnea,
comfort, oxygenation, and CO2 reduction) with both masks, but there was
greater mask intolerance with the nasal mask (34% vs. 11%). They also noted
more deaths (11% vs. 6%) and less success in the nasal mask group (49% vs.
66%), but these differences were not statistically different, with the same intuba-
tion rate (23%) in both groups. This mirrors other evaluations in acutely ill
patients, although there have been trends toward more efcient removal of CO2.
In a study of stable patients, there was greater CO2 reduction and higher minute
ventilation with face masks, but better tolerance with nasal masks (89,90).
56 Soo Hoo
D. Ventilators
There has been great diversity in the types of ventilators used to provide venti-
latory support in NIMV. There is experience with both volume- and pressure-
cycled ventilators, large bedside and portable devices, and a variety of ventilatory
modes, including assist control, CPAP, pressure support, proportional assist ven-
tilation (PAV), or some combination of all of these modalities. In the appropriate
setting, all of these can provide effective ventilatory support with relief of
dyspnea, correction of gas exchange abnormalities, and successful prevention
of endotracheal intubation. The success rates are probably comparable, provided
that patients receive an adequate level of support. In this sense, one modality
cannot be endorsed above another, but in reality, pressure support with PEEP
in one survey was used in almost 75% of patients treated with NIMV (78). In
the survey, volume ventilation was used in only 15% of patients. Similarly, the
vast majority of the experience with NIMV has been accumulated with pressure
ventilation with volume ventilation used in only one major randomized trial,
comprising less than 10% of patients analyzed in a recent meta-analysis.
Volume ventilators may pose more problems during NIMV, primarily
because of mask or mouth leaks and potentially wide uctuations in peak
airway pressures. The leaks obviously impede delivery of adequate tidal
volume and can trigger more alarms, which may require more attention from
medical staff. This may occur even with drastic changes in alarm limits. Compen-
sation for the leaks often requires tightening of mask straps, which can contribute
to skin breakdown and pressure sores. High pressures may also lead to gastric dis-
tension. On the other hand, volume ventilators may be better suited for patients
with rapidly changing respiratory system compliance or drive who would require
some minimum minute ventilation. However, despite these potential limitations,
no signicant differences have been demonstrated between the two modes,
although there appears to be better compliance with pressure ventilation (91,92).
Pressure-cycled ventilators, whether provided by larger bedside critical
care ventilator or smaller, more mobile bi-level ventilatory assist devices,
seem to be better adapted for NIMV. The pressure ventilators are more tolerant
of system leaks, and some of the bi-level models have leak compensation
features, which result in fewer alarms requiring medical staff attention. The
bi-level devices are designed specically with NIMV in mind, whereas NIMV
is one of the many options available in the bedside critical care ventilator.
There are signicant cost differences between the two types of ventilators, and
versatility among patients may also inuence the use of one over the other.
Non-Invasive Ventilation in Critical Care 57
a comparison of the key differences between the major types of ventilators used in
NIMV.
As the distinction between ventilators has blurred, it becomes more import-
ant to recognize that different masks are used based on the ventilator used. This
would probably never be an issue except for those unfamiliar with NIMV. The
bi-level devices use a single-hose to deliver ventilatory support with exhalation
of gases usually through a hole in the mask or circuit. This exhalation port
should never be blocked. Likewise, there is also a apper valve that allows deliv-
ery of pressure and entrainment of air. If it is not functional, the mask should be
replaced. A separate mask without any openings in the mask or circuit is used for
those critical care ventilators that can also deliver NIMV. The critical care ven-
tilators use a closed two-hose system (inspiratory and expiratory) and therefore
are not designed to be used with a conguration with a xed leak. This type of
mix-up may result in ineffective ventilation, and there is always the potential
for untoward consequences. Figures 1 and 2 demonstrate the differences in the
masks used with different ventilators and also illustrate the similarities
between the two congurations.
PAV may provide even better patient ventilator synchrony and patient
comfort. It uses an inline pneumotachometer to provide ventilatory assistance
proportional to the patients efforts. It incorporates ow and volume assistance
with each breath, and expiratory cycling occurs with cessation of inspiratory
effort. It is more responsive to changes in patient effort than the previously
Figure 1 Patient supported with face mask non-invasive ventilation using a bedside
bi-level ventilator with a single-hose tubing attached to the mask. The connection has an
exhalation port and a apper valve attached at the interface between the mask and tubing.
Non-Invasive Ventilation in Critical Care 59
Figure 2 Patient supported with face mask non-invasive ventilation using a critical care
ventilator that can be used in patients who are also endotracheally intubated. The mask is
the same as the mask used for the patient in Figure 1, but the head straps and ventilator
tubing attachments are different. The attachment to the mask is of a single-piece construc-
tion, different conguration, without any exhalation ports and is meant to be used with
double-hose tubing.
Initial settings are goal directed to achieve adequate tidal volumes. A goal tidal
volume of 5 to 7 mL/kg is appropriate with additional support added as necessary
to reduce the respiratory rate to 25/min. Appropriate initial settings in most
patients would be an IPAP or pressure support equal to 10 cmH2O and EPAP
or PEEP equal to 5 cm H2O. These pressures are tolerated by most, although
some will require lower pressures. However, it is not recommended that initial
inspiratory pressures be lower than 8 cm H2O or expiratory pressures much
lower than 4 cm H2O. Settings with lower pressures are usually inadequate for
ventilatory support, subjecting the patient to more dyspnea and distress, and
lead to premature termination of therapy. Maximal inspiratory pressures should
be limited to about 20 to 25 cm H2O. Patients do not tolerate these higher press-
ures, and in this range, there is a risk for overcoming the lower esophageal
sphincter pressure with subsequent gastric distension. Changes in settings
60 Soo Hoo
should target the primary respiratory defect. Hypoxemic patients are best treated
with an increase in the PEEP (CPAP) or EPAP with a proportionate increase in
the IPAP effectively maintaining the same tidal volume. Hypercapnic patients
benet most from an increase in tidal volume resulting from increases in pressure
support or IPAP without much change in the EPAP. If using BiPAP, the differ-
ence between IPAP and EPAP represents the level of pressure support supplied
to the patient. A back-up respiratory rate can be set to achieve a minimum
minute ventilation.
NIMV is best delivered in three to four hours blocks with 30 to 60 minutes
breaks. Early in the course of treatment, some patients develop intolerable
dyspnea off NIMV and receive near continuous support. The breaks in therapy
are encouraged to minimize ischemia and necrosis that can develop at the
skin mask interface. This also allows a chance for the patient to clear secretions,
eat, and drink. These breaks can serve as short weaning trials, and the duration
of the break can be extended at any time based on the comfort of the patient. In
this sense, ventilatory support and weaning occur at the same time, and venti-
latory support is continued only as long as absolutely necessary. In some patients,
NIMV is required for most of the rst day (20 hours), but some may sufce with
only eight hours of support. Patients are often very cognizant of the benet with
NIMV, and it is not infrequently that they not only dictate the duration of venti-
latory support, but can also self-wean themselves from NIMV.
Humidication may be an issue if NIMV is administered using one of the
earlier models of the portable bi-level devices. The drying effects of unhumidi-
ed air may dry and thicken airway secretions as well as mucosal surfaces.
Drawing from the experience with CPAP masks in patients with obstructive
sleep apnea, humidication prevents this complication and improves compliance
(99). It is a reasonable addition to patients during NIMV, especially those who
require more than eight hours of daily support. This is not a signicant issue
with the latest models of NIMV ventilators because most have inline humidi-
cation systems.
Treatment of COPD and asthma patients includes bronchodilators. There
may be an advantage of delivering these nebulized bronchodilators inline
during NIMV ventilation. There is potential for more drug to be delivered, and
in one small clinical study, this seemed to translate to more rapid improvement.
Potentially more drug may be able to be delivered with an inline nebulizer.
However, it should be noted that improvement may not be due to improved
drug delivery and distribution, but rather the benets of respiratory support
with NIMV (100,101). While this requires further investigation, there are no
known major adverse effects with this approach. Along the same line, a
helium oxygen mixture used in conjunction with NIMV may also enhance the
benet of NIMV in selected patients. The lower density gas is known to be effec-
tive by reducing the resistive load of breathing in obstructed patients, thereby
decreasing dyspnea and lowering the work of breathing while improving gas
exchange. In a randomized, cross-over study of primarily COPD patients, the
Non-Invasive Ventilation in Critical Care 61
helium oxygen mixture during pressure support NIMV reduced indices of work
of breathing by one-third or more with greater improvement in hypercapnia than
with NIMV (102). These are both important adjuncts in the use of NIMV in res-
piratory failure and have the potential of possibly averting intubation in some
patients. More extensive studies are required to gain a better appreciation of
the role of these adjuncts in management.
to three hours of ventilatory support (1). Patients who eventually fail and require
intubation either have an increase, no change, or a minimal decline in PaCO2.
This can cause somewhat of a conundrum in management. Unless patients
require immediate intubation, a case can be made for some groups of patients
(COPD) with respiratory failure to undergo a trial of NIMV. However, it may
not be clear whether failure to improve represents a failure of NIMV or a
patient who is slow to improve. There must be a time frame that constitutes an
effective trial of NIMV. In some patients, it will be clear that NIMV is ineffective
within a few minutes to an hour of initiation. In others, there may still be uncer-
tainty after three to six hours or longer. One would like to avoid extended trials in
patients destined to fail because problems may be encountered at the time of
intubation, with associated morbidity and mortality. In the two largest studies,
(17, 18) failure of NIMV occurred in the vast majority within 12 and 24 hours
of initiation, in 82% and 61% of patients, respectively.
Moretti et al. (105) identied predictors of late failure (dened as
.48 hours after initiation of NIMV). In their multi-center study of 186 patients,
31 were designated as late failures after an initial response to NIMV. These
patients experienced a relapse on average 8.2 + 2.8 days after initial therapy.
Lower initial functional status scores, hospital complications (pneumonia,
shock, and coma), and lower initial pH (7.22 + 0.08) were predictive factors.
These late failures also had an increased mortality rate (68% vs. 15%). In
summary, a three- to four-hour trial is a reasonable time period to assess the
response to NIMV. There should be some improvement in physiologic
parameters within this time frame, although resolution may require a much
longer period. If patients fail to demonstrate any improvement within this time
frame, intubation may be likely within the initial 12 to 24 hours of presentation.
The greatest success with NIMV has been in COPD and cardiogenic pul-
monary edema. While patients with other conditions can also be treated with
NIMV, they may not respond as favorably, and there may be a lower threshold
for intubation in these conditions. Higher failure rates or slower responses to
therapy (compared with COPD or cardiogenic pulmonary edema) have been
noted in patients with ARDS, pneumonia, and underlying restrictive lung
disease (60,106,107). This information provides additional perspective on the
use of NIMV in these conditions and may inuence not only the decision to
use NIMV, but also the duration of NIMV prior to proceeding to endotracheal
intubation. Table 7 provides a summary of the most useful measures of the
success or failure of NIMV. Of course, each patient must be evaluated individu-
ally, but these parameters should provide a framework for guidance.
The next great hurdle involves the decision to terminate NIMV in favor of
intubation. Reported criteria for intubation have included a pH , 7.20, pH 7.20 to
7.25 on two occasions one hour apart, hypercapnic coma (Glasgow coma scale ,8
and PaCO2 . 60 mm Hg, PaO2 , 45 mm Hg), or cardiopulmonary arrest (18).
Others have used a combination of major or minor clinical criteria (17). Major
criteria have included respiratory arrest, loss of consciousness with respiratory
Non-Invasive Ventilation in Critical Care 63
Severity of illness
Acidosis (pH , 7.25)
Hypercapnia (PaCo2 . 80 and pH ,7.30)
APACHE II (.20)
Level of consciousness
Neurologic score (.4, stuporous, arousal only after vigorous stimulation;
inconsistently follows commands)
Encephalopathy score (.3, major confusion, daytime sleepiness, or agitation)
Glasgow coma score (,8)
Disease conditions
Acute respiratory distress syndrome
Pneumonia
Restrictive lung disease
Predictors of success
Response to brief trial of NIMV (1 3 hr)
Decrease in PaCO2 . 8 mm Hg
Improvement in pH . 0.06
Correction of respiratory acidosis
Time frame for failure requiring intubation
Failure of improvement with NIMV
Within 12 24 hr
Late failures (.48 hr after initiation of non-invasive ventilation)
Admission predictors of failure
Lower functional status (Activity score ,2, dyspnea with light activity)
Initial acidosis (pH 7.22)
Hospital complications (pneumonia, shock, coma)
B. Complications
There are markedly different proles reported for complications associated with
intubation and mechanical ventilation and NIMV (82). The vast majority of these
Non-Invasive Ventilation in Critical Care 65
complications can be attributed to direct injury from the patient interface, and in
the case of NIMV, represent pressure injury and skin necrosis from the mask.
This is by far the most frequent complication, occurring in about 10% of
reports. Some of these lesions can be quite extensive, with erosions of the
nasal bridge to the nasal cartilage or even bone. Gauze or other types of protective
skin dressing have been used to minimize this complication. Softer masks, redu-
cing strap pressure, and padding in key pressure points are other adjustments that
may lessen the incidence of this complication. This underscores the need to
provide breaks off NIMV for patients. Other reported complications are relatively
minor and include problems with nasal congestion, sinusitis, gastric distension,
and eye irritation. Major complications such as barotrauma, nosocomial pneumo-
nia, and sepsis are distinctly uncommon. There have been case reports of unusual
complications such as esophageal perforation, re-opening of an esophageal
pleural stula, and orbital herniation (108 110).
NIMV (113). This was attributed to alarms, mask adjustments, constant monitor-
ing, and equipment issues (nasal masks and portable volume ventilators). This
was likely a reection of the initial learning curve that all personnel must encoun-
ter with this modality. This is clearly a technique that only improves with repeated
administration and experience. Patients treated with NIMV also comprise a
somewhat tenuous group of patients. Intubation and mechanical ventilation
may be perceived as easier, requiring less time and attention, increasing resistance
to its use. However, several controlled trials have subsequently demonstrated that
the time commitment is very comparable to that required for those treated with
conventional therapy undergoing intubation and mechanical ventilation.
Kramer et al. (16) were the rst to address this issue and found an average
greater initial time commitment by respiratory therapists of 56 minutes during
the rst eight hours of therapy, which declined the next eight hours, so that
they actually spent less time in NIMV patients compared with control patients.
This experience mirrors that of Nava et al. (114), who found that the respiratory
therapist spent on average an extra 74 minutes during the initial eight hours of
NIMV, decreasing to equivalent time commitments at the end of 48 hours. In
an analysis of workload by Plant et al. (18), they found that nursing workload
increased by an average of only 26 minutes during the rst eight hours of NIMV,
but by 48 hours the difference was only 21 minutes and not signicant. This
represents a special situation as nursing staff provided the primary respiratory
care for the patients in the study. Other investigators have also reported greater
efciencies over time, evidenced by improving success rates with NIMV (81).
The other systems issue with NIMV involves the optimal location for its
applications. There is a wide range of experience that probably reects differ-
ences in health-care systems and practices, as well as local familiarity with the
technique. Although this has been demonstrated to a feasible technique in an
unmonitored ward setting, with or without specialty expertise, the vast majority
of patients are treated in a monitored setting. This includes the Emergency
Department, where treatment can be initiated, and it can be continued in a
specialized respiratory unit (with continuous monitoring), step-down unit, or
ICU. The need for monitoring stems from the potential need for intubation and
mechanical ventilation in those that fail therapy.
This has obvious implications with respect to health care costs. While
early studies indicate that the resources and costs required for NIMV were com-
parable to that for endotracheal intubation and mechanical ventilation, other
analysis suggest cost savings with this technique (16,114). One evaluation
reected the experience of utilizing NIMV in a ward setting, but another ana-
lyzed its use in critical care units (19,115). These analyses are probably not
applicable to the United States, where the use of NIMV requires a monitored
setting. The other unrecognized element of this analysis is that the health-care
reimbursement system has not yet recognized NIMV as a viable treatment. It
is difcult to code under current guidelines, resulting in suboptimal reimburse-
ment when compared with intubation and mechanical ventilation. In some
Non-Invasive Ventilation in Critical Care 67
centers, the code for CPAP (used to treat obstructive sleep apnea) is used to
code for NIMV. It may be similar in concept but vastly different in personnel
and resource utilization. This serves as a negative incentive for its use and
may explain why this technique seems more prevalent in Europe and Canada
than in the United States.
VII. Conclusion
Although the efcacy of NIMV has been clearly demonstrated and is endorsed by
many, its utilization remains limited. The enthusiasm for treatment is tempered
by some research and practical limitations. Despite well-designed and well-
conducted prospective randomized studies, the actual numbers of patients under-
going study are relatively small. The studies cannot be blinded for obvious
reasons, and the comparison intubation rates in those managed with standard
therapy is inuenced by a treatment bias inherent in this type of study. There
is likely a lower threshold for the intubation of standard therapy patients than
NIMV patients despite the presence of intubation guidelines. Issues with
increased nursing staff and respiratory therapy time have been cited as possible
limiting factors. It is acknowledged that there exists a learning curve with this
technique, and it may be steeper in some institutions or situations. This treatment
does require patience and commitment by all involved. In some circumstances, it
is often easier for the patient to be endotracheally intubated.
On the other hand, once staff has gained sufcient experience with NIMV,
it is an invaluable treatment option. It is ideally suited for patients with COPD
with hypercapnic respiratory failure because this is often a rapidly reversible
condition. Patients with cardiogenic pulmonary edema are probably the other
major group that can be managed with NIMV. It allows dyspnea relief and ven-
tilatory support, preventing further deterioration while allowing other treatments
to take effect. It is intermittently applied, and treatment is discontinued once the
mask is removed. Therefore, short breaks off NIMV serve a dual purpose.
Patients are provided relief from skin and facial ischemia and necrosis, and the
breaks serve as weaning trials. Patients who do not tolerate these breaks can
be placed immediately back on support. This reduces much of the time and uncer-
tainty involved with discontinuing mechanical ventilation in those who are endo-
tracheally intubated. Patients are spared the potential complications and adverse
effects associated with re-intubation. Preservation of upper airway function
reduces the nosocomial/ventilator-associated pneumonia rate. All of these
factors serve to reduce the duration of ventilatory support, ICU stay, and possibly
duration of hospitalization. Reduction in mortality has not been universally
noted. However, the positive experiences call for continued study in the modality,
rening issues regarding patient selection and delivery of ventilatory support. As
it becomes easier to provide NIMV, barriers to its use will disappear, and it will
be the treatment of choice in select patients.
68 Soo Hoo
References
1. Mehta S, Hill NS. Noninvasive ventilation. Am J Respir Crit Care Med 2001;
163(2):540 577.
2. British Thoracic Society Standards of Care Committee. Non-invasive ventilation in
acute respiratory failure. Thorax 2002; 57(3):192 211.
3. International Consensus Conferences in Intensive Care Medicine: noninvasive
positive pressure ventilation in acute Respiratory failure. Am J Respir Crit Care
Med 2001; 163(1):283 291.
4. Severinghaus JW, Astrup P, Murray JF. Blood gas analysis and critical care medi-
cine. Am J Respir Crit Care Med 1998; 157(4 Pt 2):S114 S122.
5. Sandur S, Stoller JK. Pulmonary complications of mechanical ventilation.
Clin Chest Med 1999; 20(2):223247.
6. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med
2002; 165(7):867 903.
7. Ellis ER, Bye PT, Bruderer JW, Sullivan CE. Treatment of respiratory failure
during sleep in patients with neuromuscular disease. Positive-pressure ventilation
through a nose mask. Am Rev Respir Dis 1987; 135(1):148 152.
8. Menzies R, Gibbons W, Goldberg P. Determinants of weaning and survival among
patients with COPD who require mechanical ventilation for acute respiratory
failure. Chest 1989; 95(2):398 405.
9. Carrey Z, Gottfried SB, Levy RD. Ventilatory muscle support in
respiratory failure with nasal positive pressure ventilation. Chest 1990;
97(1):150 158.
10. Belman MJ, Soo Hoo GW, Kuei JH, Shadmehr R. Efcacy of positive vs
negative pressure ventilation in unloading the respiratory muscles. Chest 1990;
98(4):850 856.
11. Brochard L, Isabey D, Piquet J, Amaro P, Mancebo J, Messadi AA,
Brun-Buisson C, Raws A, Lemaire F, Harf A. Reversal of acute exacerbations of
chronic obstructive lung disease by inspiratory assistance with a face mask. N
Engl J Med 1990; 323(22):1523 1530.
12. Meduri GU, Abou-Shala N, Fox RC, Jones CB, Leeper KV, Wunderink RG.
Noninvasive face mask mechanical ventilation in patients with acute hypercapnic
respiratory failure. Chest 1991; 100(2):445 454.
13. Meyer TJ, Hill NS. Noninvasive positive pressure ventilation to treat respiratory
failure. Ann Intern Med 1994; 120(9):760 770.
14. Corrado A, Confalonieri M, Marchese S, Mollica C, Villela G, Gorini M,
Bella Porta R. Iron lung vs mask ventilation in the treatment of acute on chronic
respiratory failure in COPD patients: a multicenter study. Chest 2002;
121(1):189 195.
15. Bott J, Carroll MP, Conway JH, Keilty SE, Ward EM, Brown AM, Paul EA,
Elliot MW, Godfrey RC, Wodzicha JA. Randomised controlled trial of nasal ven-
tilation in acute ventilatory failure due to chronic obstructive airways disease.
Lancet 1993; 341(8860):1555 1557.
16. Kramer N, Meyer TJ, Meharg J, Cece RD, Hill NS. Randomized, prospective trial
of noninvasive positive pressure ventilation in acute respiratory failure. Am J
Respir Crit Care Med 1995; 151(6):1799 1806.
Non-Invasive Ventilation in Critical Care 69
33. Masip J, Betbese AJ, Paez J, Vecilla F, Canizares R, Padro J, Paz MA, de Otero J,
Ballus J. Non-invasive pressure support ventilation versus conventional oxygen
therapy in acute cardiogenic pulmonary oedema: a randomised trial. Lancet
2000; 356(9248):2126 2132.
34. Sharon A, Shpirer I, Kaluski E, Moshkovitz Y, Milovanov O, Polak R,
Blatt A, Simovitz A, Shaham D, Feigenberg Z, et al. High-dose intravenous
isosorbide-dinitrate is safer and better than Bi-PAP ventilation combined with
conventional treatment for severe pulmonary edema. J Am Coll Cardiol 2000;
36(3):832 837.
35. Nava S, Carbone G, DiBattista N, Bellone A, Baiardi P, Cosentini R, Marenco M,
Giostra F, Borasi G, Groff P. Noninvasive ventilation in cardiogenic pulmonary
edema: a multicenter randomized trial. Am J Respir Crit Care Med 2003;
168(12):1432 1437.
36. Levitt MA. A prospective, randomized trial of BiPAP in severe acute congestive
heart failure. J Emerg Med 2001; 21(4):363 369.
37. Chadda K, Annane D, Hart N, Gajdos P, Raphael JC, Lofaso F. Cardiac and respiratory
effects of continuous positive airway pressure and noninvasive ventilation in acute
cardiac pulmonary edema. Crit Care Med 2002; 30(11):24572461.
38. Mehta S, Jay GD, Woolard RH, Hipona RA, Connolly EM, Cimini DM,
Drinkwine JH, Mill NS. Randomized, prospective trial of bilevel versus continuous
positive airway pressure in acute pulmonary edema. Crit Care Med 1997;
25(4):620 628.
39. Liesching TN, Cromier K, Nelson D, Short K, Sucov A, Hill NS. Bilevel noninva-
sive ventilation vs continuous positive airway pressure to treat acute pulmonary
edema. Am J Respir Crit Care Med 2004; 167(7):A864.
40. Pennock BE, Kaplan PD, Carlin BW, Sabangan JS, Magovern JA. Pressure
support ventilation with a simplied ventilatory support system administered
with a nasal mask in patients with respiratory failure. Chest 1991;
100(5):1371 1376.
41. Wysocki M, Tric L, Wolff MA, Gertner J, Millet H, Herman B. Noninvasive
pressure support ventilation in patients with acute respiratory failure. Chest 1993;
103(3):907 913.
42. Epstein SK, Ciubotaru RL, Wong JB. Effect of failed extubation on the outcome of
mechanical ventilation. Chest 1997; 112(1):186 192.
43. Nathan SD, Ishaaya AM, Koerner SK, Belman MJ. Prediction of minimal
pressure support during weaning from mechanical ventilation. Chest 1993;
103(4):1215 1219.
44. Ishaaya AM, Nathan SD, Belman MJ. Work of breathing after extubation. Chest
1995; 107(1):204 209.
45. Vitacca M, Ambrosino N, Clini E, Porta R, Rampulla C, Lanini B, Nava S.
Physiological response to pressure support ventilation delivered before and after
extubation in patients not capable of totally spontaneous autonomous breathing.
Am J Respir Crit Care Med 2001; 164(4):638 641.
46. Nava S, Ambrosino N, Clini E, Prato M, Orlando G, Vitacca M, Brigada P,
Frachia C, Rubini F. Noninvasive mechanical ventilation in the weaning of patients
with respiratory failure due to chronic obstructive pulmonary disease. A random-
ized, controlled trial. Ann Intern Med 1998; 128(9):721 728.
Non-Invasive Ventilation in Critical Care 71
90. Anton A, Tarrega J, Giner J, Guell R, Sanchis J. Acute physiologic effects of nasal
and full-face masks during noninvasive positive-pressure ventilation in patients
with acute exacerbations of chronic obstructive pulmonary disease. Respir Care
2003; 48(10):922 925.
91. Vitacca M, Rubini F, Foglio K, Scalvini S, Nava S, Ambrosino N. Non-invasive
modalities of positive pressure ventilation improve the outcome of acute exacer-
bations in COLD patients. Intensive Care Med 1993; 19(8):450 455.
92. Girault C, Richard JC, Chevron V, Tamion F, Pasquis P, Leroy J, Bonmerchand G.
Comparative physiologic effects of noninvasive assist-control and pressure support
ventilation in acute hypercapnic respiratory failure. Chest 1997; 111(6):1639
1648.
93. Ferguson GT, Gilmartin M. CO2 rebreathing during BiPAP ventilatory assistance.
Am J Respir Crit Care Med 1995; 151(4):1126 1135.
94. Lofaso F, Brochard L, Touchard D, Hang T, Harf A, Isabey D. Evaluation of carbon
dioxide rebreathing during pressure support ventilation with airway management
system (BiPAP) devices. Chest 1995; 108(3):772 778.
95. Nava S, Ambrosino N, Bruschi C, Confalonieri M, Rampulla C. Physiological effects
of ow and pressure triggering during non-invasive mechanical ventilation in patients
with chronic obstructive pulmonary disease. Thorax 1997; 52(3):249254.
96. Hotchkiss JR, Adams AB, Dries DJ, Marini JJ, Crooke PS. Dynamic behavior
during noninvasive ventilation: chaotic support? Am J Respir Crit Care Med
2001; 163(2):374 378.
97. Gay PC, Hess DR, Hill NS. Noninvasive proportional assist ventilation for acute
respiratory insufciency. Comparison with pressure support ventilation. Am J
Respir Crit Care Med 2001; 164(9):1606 1611.
98. Wysocki M, Richard JC, Meshaka P. Noninvasive proportional assist ventilation
compared with noninvasive pressure support ventilation in hypercapnic acute res-
piratory failure. Crit Care Med 2002; 30(2):323 329.
99. Massie CA, Hart RW, Peralez K, Richards GN. Effects of humidication on nasal
symptoms and compliance in sleep apnea patients using continuous positive airway
pressure. Chest 1999; 116(2):403 408.
100. Pollack CV Jr, Fleisch KB, Dowsey K. Treatment of acute bronchospasm with beta-
adrenergic agonist aerosols delivered by a nasal bilevel positive airway pressure
circuit. Ann Emerg Med 1995; 26(5):552 557.
101. Chatmongkolchart S, Schettino GP, Dillman C, Kacmarek RM, Hess DR. In vitro
evaluation of aerosol bronchodilator delivery during noninvasive positive pressure
ventilation: effect of ventilator settings and nebulizer position. Crit Care Med 2002;
30(11):2515 2519.
102. Jaber S, Fodil R, Carlucci A, Boussarsand M, Pigeot J, Lemaire F, Hharf A, Lofuso F,
Isabey D, Brochard L. Non invasive ventilation with helium oxygen in acute
exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care
Med 2000; 161(4 Pt 1): 1191 1200.
103. Nava S, Ceriana P. Causes of failure of noninvasive mechanical ventilation.
Respir Care 2004; 49(3):295 303.
104. Anton A, Guell R, Gomez J, Serrano J, Castellano A, Carrasco JL, Sanchis J.
Predicting the result of noninvasive ventilation in severe acute exacerbations of
patients with chronic airow limitation. Chest 2000; 117(3):828 833.
Non-Invasive Ventilation in Critical Care 75
I. Introduction
In recent years, advances in ventilator technology have given rise to many new
and often confusing terms and acronyms used to describe modes of ventilator
function and patient interface. In an effort to bring order to this new terminology,
a system for understanding modes of mechanical ventilation has been developed
(1) and recently included in textbooks of respiratory care (2).
This new classication scheme enables logical description of new
ventilator modes that function in the same way but are given different brand
names. It also allows for grouping of ventilators in terms of the number of
breathing pattern options each ventilator offers. Despite the simplicity and
logic of this new system for describing ventilator modes, the terms and
abbreviations used in the system are not necessarily those used by physicians
who order mechanical ventilation or by respiratory care practitioners who
implement these orders.
Instead, when new ventilator modes are applied, they are usually ordered
by the terminology the manufacturers coin (and often trademark), which is
reinforced by their own clinical consultants when taught to end-users. In short,
the end-users of newer ventilator systems use a mix of community-accepted
terms and ventilator-specic labels to describe, order, and implement ventilator
mode and breath control settings. To expect otherwise would presume years of
re-educating physicians, nurses, and respiratory care practitioners in the
meaning and application of the new ventilator taxonomy and in the utility of
the new ventilator modes it describes.
77
78 Richards and Mosenifar
Accordingly, the reader is referred to the new system for classifying venti-
lator modes, but herein familiar terms will be used to describe popular ventilator
modes whenever possible and, when not, the manufacturer-labeled modes will be
described in simple terms.
In the context of mechanical ventilation, the breath is the building block of breath
patterns called ventilator modes. A breath is dened as one inspiratory ow event
paired with one expiratory ow event. The areas circumscribed by each of these
ow events (i.e., the integral of ow) are dened as the inspiratory tidal volume
(Vt insp) and the expiratory tidal volume (Vt exp). A zero-ow inspiratory tidal
volume hold may or may not precede the expiratory ow event. A breath may or
may not include a passive expiratory phase that precedes the next breath (Fig. 1).
The ventilator breath can be divided into four phases: the trigger phase, the limit
phase, the cycle phase, and the expiratory phase (2).
80
60
40 Zero Flow
20
.
VL/min 0
20 Passive Exp
40
60
0.5
80 0.5 1.5 sec sec
1.0 sec
sec
IT ET
(1.5 sec) (2.0 sec)
500
V
400 O
L
Vt ml 300
H
200 O
L
100 Vt insp D Vt exp
0
Time-Triggering
Time-Triggering is typically associated with control mode ventilation of the
patient who is apneic due to pathology or pharmacologic control of respiration.
The lapse of the ventilators breath cycle time [60 sec/min divided by set respir-
atory rate (RR)] automatically triggers inspiration.
Patient-Triggering
Patient-triggering occurs when patient inspiratory effort is translated into a
negative pressure signal (i.e., a pressure-trigger) or a negative ow signal
(i.e., a ow-trigger) that activates ventilator gas delivery.
1. Pressure-triggering occurs when a small user-set 1 to 2 cmH2O
pressure drop caused by patient inspiratory effort is communicated
to an upstream demand valve, which then opens to deliver ventilator
gas. Small pressure-trigger settings are more sensitive to patient
inspiratory efforts and make it easier for the patient to trigger ventilator
gas delivery. However, when positive end expiratory pressure (PEEP)
is set and a ventilator system leak causes expiratory baseline pressures
to fall below set PEEP, small pressure-trigger settings can result in
ventilator auto-triggering and ventilatorpatient dyssynchrony (3).
Conversely, large pressure-trigger settings are less sensitive to
patient breathing effort, increase patient trigger work, and delay venti-
lator demand valve activation. Large pressure-trigger settings are
especially problematic where expiratory air trapping results in auto-
PEEP that must be exceeded by patient-generated negative pressures
sufcient to overcome the auto-PEEP plus the user-set trigger pressure
(Fig. 2). Reducing the trigger pressure and setting a low-level PEEP
that approximates the auto-PEEP can decrease patient-trigger effort
and better synchronize it with machine breath delivery (4 7).
2. Flow-triggering employs a low level background (or bias) ow of gas
that is measured by ow sensors located at the ventilator circuits gas
inow port and the ventilator circuits gas outow port. When the
patient makes an inspiratory effort, part of this background ow of gas
goes to the patient, causing circuit outow to fall below circuit inow.
This difference in ow translates into a user-set ow signal, usually 1
to 3 lpm, that triggers inspiratory gas delivery to meet machine breath
settings or spontaneous breath demands (Fig. 3). In current generation
ventilators, the user can select ow-triggered or pressure-triggered
mechanisms to enable patient triggering of all breath types. Some studies
80 Richards and Mosenifar
40
30 3cm H2O
20 Auto PEEP*
10
PcmH2O (-5 cm)Actual trigger
0
(-2 cm) Set Trigger threshold; add PEEP
Pressure of 3 cm to restore
-10
trigger threshold to
2cm below baseline.
-20
80
60
40 Air Trapping
20
.
VL/min 0
20
40
60
80
Time (sec)
B. Limit Phase
The limit phase denes the limit to which a user-set inspiratory control, like ow
or pressure, can rise without terminating inspiration (2).
120
Square Wave
Decelerating Wave
60
.
VL/min 0
Air Trapping from
60 prolongation of
I-Time when
Decelerating Flow
120 curve used.
120
Scooped Pressure Curve
due to inadequate flow output
PcmH2O 60 relative to patient demand
Time (sec)
Figure 4 Volume control: square ow versus decelerating ow. Source: Adapted from
Chang DW. Clinical Application of Mechanical Ventilation. 2nd ed. Delmar Thomson
Learning, 2001; 11:304 (Fig. 11.29).
20
200
150
100
50
.
VL/min 0
50
100
150
200 Time (sec)
Figure 5 Effect of rise time control on pressure control breath. Computer rendering of
sketch drawn using common test lung, resistor, and Nellcor Puritan Bennett 840 Ventilator
graphics. Source: Courtesy of Tyco Health Care, Nellcor Puritan Bennett, 2004.
stretch alveolar (PEEP) (Fig. 6). Thus, Vt and PEEP combine to determine
alveolar stretch and related lung damage risks. While it remains controversial
as to which combination of Vt and PEEP provides the best pulmonary gas
exchange with the least risk of ventilator-induced lung injury (21), the pressure
generated by the sum of the set PC pressure and PEEP should be maintained at
no more than 30 35 cmH2O whenever feasible (21 23).
C. Cycle Phase
The cycle phase determines the inspiratory variable that terminates inspiration (2).
120
. 60
VL/min
40 Zero Flow
VT
0
60
PIP = 40 cm H2O 2 Unit Lung Model
PcmH2O
40
Pplat =28cm H2O Raw = (PIP - Pplat ) / Flow
PEEP = C = V / (P -PEEP) VT + PEEP =>
20 5cm H2O ST T plat
PA ~ Pplat
Time (sec)
D. Expiratory Phase
The expiratory phase denes the time it takes for expiratory ow to return to
baseline ow (i.e., active expiration) plus the time in which the variable
known as PEEP or continuous positive airway pressure (CPAP) is applied
Modes of Mechanical Ventilation 85
25
20 18 cm H2O PS
PcmH2O 15
10
80
Max Flow
60 50% Terminal
(60 Lpm) 25% Terminal
40 Flow (15 Lpm) Flow (30 Lpm)
. 20
VL/min 0
20
40
60
80 Time (sec)
FO2 0.3 0.4 0.4 0.5 0.6 0.7 0.7 0.7 0.8 0.9 0.9 1.0
PEEP 5 5 8 8 10 10 12 14 14 14 16 20 24
Pplat (22cmH2O)
400
200
Inspiration
100
Figure 8 Quasi-static pressure volume curve at 10 Lpm ow using test lung and resis-
tor. Computer rendering of sketch drawn using common test lung, resistor, and Nellcor
Puritan Bennett 840 Ventilator graphics. Source: Courtesy of Tyco Health Care.
There are two types of breaths possible during mechanical ventilation, the
machine breath and the spontaneous breath.
A. Machine Breath
The machine breath is a breath in which inspiration is time- or patient-triggered
and cycles into expiration when a user-set Vt or I-Time is delivered by the
machine. Examples of machine breath are discussed subsequently.
600
500
400 V
O
Vtml 300 L
H
200 Vt Vt O
L
100
Insp. Exp. D
80
60
40
V
tc
20 ut
. of
f
Vt hold*
VL/min 0
I-Time I-Time
20
too short too long,
40 for complete now
60 filling exceeding
fill time
80 TIME
Bi-Level Ventilation
In bi-level ventilation (Nellcor-PB 840), like PC ventilation, the breath deliv-
ered is considered a machine breath in which inspiratory ow is either time- or
patient-triggered on and cycled off after the machine delivers a sustained,
pressure-limited breath over a user-set I-Time interval. In the absence of spon-
taneous breathing, the ow and pressure waveform graphics generated during a
bi-level machine breath resemble those generated by a time-cycled PC breath.
The bi-level machine breath differs from a PC breath in that the bi-level inspira-
tory pressure limit is labeled as the High PEEP from which the Low PEEP is
subtracted to determine the ventilating pressure (DP). For example, if the high
PEEP is set at 30 cmH2O and the low PEEP is set at 5 cmH2O, then the
maximum pressure reached is 30 cmH2O with a ventilating pressure (DP) of
25 cmH2O. In PC ventilation, the user-set inspiratory pressure limit is the venti-
lating pressure (DP) and is added to set PEEP to determine PIP. For example, if
the PC pressure limit is set at 30 cmH2O and the PEEP at 5 cmH2O, then the
maximum pressure reached is 35 cmH2O with a ventilating pressure (DP) of
30 cmH2O. The bi-level machine breath is further differentiated from the PC
breath in that it allows for spontaneous breathing atop the user-set inspiratory
pressure time curve; a decrease in sedation is an alleged advantage (40). PS is
also available to augment spontaneous breathing on both inspiratory and expira-
tory phases of the bi-level breath (41) (Fig. 10).
50 Inspiratory Effort
Expiratory
40 Effort
PcmH2O
30 HI PEEP
(30 cmH2O)
20 P=25cmH2O PS (15 cm H2O)
Lo PEEP
10 (5 cm H2O)
SET IT SET ET
80
60
. 40
VL/min
20
0
20
40
60
80
SET IT SET ET
Figure 10 Bi-Level machine breath with spontaneous breathing atop and between
machine breaths. Computer rendering of sketch drawn using common test lung, resistor,
and Nellcor Puritan Bennett 840 Ventilator graphics. Source: Courtesy of Tyco Helath
Care, Nellcor Puritan Bennett, 2004.
90 Richards and Mosenifar
the breath switches to a VC breath in which the pre-set, low level, constant ow is
delivered (and the PC pressure limit overridden) until the target Vt is reached or (in
the event of leak) a VAPS time-limit elapses (46). If the targeted Vt is delivered
before PC ow decays to the constant set ow of VC, then inspiratory ow is
ow-cycled off like a spontaneous PS breath (Fig. 11). However, unlike a spon-
taneous PS breath which must be patient-triggered, subsequent VAPS breaths may
be time-triggered as a function of a user-set RR.
B. Spontaneous Breath
PS Limit Override
25
20
Pressure
15 Support
PcmH2O
Limit
10
5
0
.
VL/min 0
-45
Time (sec)
breathing, during PS, inspiratory ow, inspiratory time, and inspiratory tidal
volume vary breath-to-breath as a function of proximal airway pressure,
patient demand, airway resistance, and pulmonary compliance (47). During
PSV, only the user-set control pressure and the low terminal ow-cycling
threshold distinguish PS inspiration from true spontaneous inspiration. The
main drawback of PSV is that the set control pressure, in the absence of the
clinician, does not vary in response to acute changes in lung mechanics or
breath-to-breath changes in inspiratory effort that reect changing pulmonary
workload. Moreover, during PSV there is no ventilator back-up respiratory rate
to safeguard against bradypnea or apnea (47). Setting ventilator alarms to indicate
inadequacy or excess in monitored volumes and respiratory rate is crucial to safe
patient management during PSV.
up-regulated to match the increased pressure drop across the airway (PS DP
RAW Flow). Conversely, as a patients inspiratory ow decreases, the PS level
is down-regulated to match the decreased pressure drop across the articial
airway. A second control option offered in TC breathing enables the user to set
fractional portions of the calculated TC PS level, thus gradually increasing the
patients resistive work of breathing (48). Patients unable to tolerate these incre-
mental increases in resistive work due to the articial airway may be weanable
only if the airway is removed. If the airway is not removable, then the possibility
of ventilator dependence becomes a consideration (49,50).
Machine and spontaneous breath types, as described earlier, are sequenced into
three traditional breathing patterns or modes of mechanical ventilation:
Assist/Control (A/C), intermittent mandatory ventilation (IMV), and CPAP.
(Text continues on p. 98.)
Table 2 Breath Types and Special Features
Volume control Ubiquitous Machine breath type in which set Vt and ow remain constant while pressure varies in
ventilation response to changing lung mechanics and patient demand. For better control over delivered
Vt, many current generation ventilators auto-compensate for volume loss due to gas
compression and circuit stretch during the VCV breath. Patientventilator dyssynchrony
can occur when patient ow and Vt demands exceed ventilator ow and Vt control settings.
Pressure control Ubiquitous Machine breath type in which control pressure remains constant while ow and Vt vary as a
ventilation function of patient demand and lung mechanics. Tends to better match the ventilatory
pattern of patients with high, variable inspiratory demands than the VCV breath, but at the
risk of excessive Vt delivery. Conversely, acute deterioration in CST and/or RAW may
result in inadequate Vt delivery. Often valued for decreasing ow prole that minimizes
the resistive component of peak airway pressure, thus enabling same tidal volume
ventilation at lower PIPs than typically associated with VC square ow wave breaths.
Pressure control Ubiquitous A PCV breath in which the I-time is set to exceed E-time, thus inverting the physiologic I/E
inverse ratio ratio for purposes of increasing volume hold, mean airway pressure, and PaO2. As set
ventilation I-time exceeds E-time, risks of auto-PEEP and hemodynamic compromise may offset any
improvements in PaO2. Moreover, as set I-time encroaches on neural expiration, patient
agitation and the need for sedation increase.
Bi-level Nellcor Puritan Bennett 840; Resembles a PCV machine breath when spontaneous breathing is absent. But, is more like a
ventilation Drager Evita Series (i.e., distinct ventilator mode during spontaneous breathing, as an active exhalation valve
Evita, Evita 2 Dura, and enables the patient to breathe in an unrestricted manner atop the user-set, inspiratory
Evita 4) pressuretime curve. A decrease in sedation is a manufacturer-alleged advantage. PS is also
available to augment spontaneous breathing atop and between Bilevel machine breaths.
(Continued)
95
Table 2 Breath Types and Special Features (Continued)
96
Breath typea Availabilitya Special features
Airway pressure Nellcor Puritan Bennett 840; Is a variation on the Bi-level breath in which I-time is set to exceed E-time, resulting in a
release Drager Evita Series (i.e., PCIRV breath atop which the patient is able to breath spontaneously. Like PCIRV, APRV
ventilation Evita, Evita 2 Dura, Evita is used to improve oxygenation, but unrestricted spontaneous breathing atop the high
4), Hamilton Galileo Gold Control Pressure time curve is alleged to reduce the patient-ventilator dyssynchrony and
the high cardio-inhibitory mean airway pressures associated with the conventional
PCIRV breath. PS is also available to augment spontaneous breathing atop and between
APRV machine breaths.
Pressure Maquet Siemens Servo 300; The PRVC machine breath is a PCV breath in which the control pressure is automatically
regulated Maquet Siemens Servo i, regulated between breaths in small increments or decrements and in response to changing
volume Adaptive Pressure lung mechanics to achieve a target tidal volume. Thus, the advantage of the PCV breath, a
control Ventilation on the decreasing ow that varies with patient demand and enables same Vt ventilation at lower
Hamilton Galileo; Auto- peak pressures than the VCV square wave breath are combined with the volume guarantee
Flow on the Drager Evita of the VCV breath.
4; VCV on the Nellcor
Puritan Bennett 840
Adaptive Hamilton Galileo A variation on the PRVC machine breath in which the control pressure is up-regulated or
support down-regulated between breaths and in response to changing lung mechanics to achieve a
ventilation targeted VE comprised of multiples of safe RR and Vt combinations. Program breath
97
a
Breath type brand names and their manufacturers from Pilbeam SP, Mechanical Ventilators: General-Use Devices. Pilbeam SP, Cairo JM, eds. Mosbys Respiratory
Care Equipment, 7th ed., St. Louis: Mosby, 2004; 12:391 661.
98 Richards and Mosenifar
A. Assist/Control Ventilation
A/C is a ventilator mode in which the patients entire minute volume is com-
prised of machine breaths delivered in response to small patient-initiated
inspiratory efforts (Assist mode) or at the user-set mechanical respiratory
rate (Control mode). The A/C mode machine breath may be a volume-con-
trolled breath, a pressure-controlled breath, or a pressure-regulated/volume-
targeted breath. However, A/C mode with volume control remains the mode
that physicians most commonly use in the initial ventilator management of
ARDS (23,52). A/C mode has also been used to rest ventilatory muscles in
patients who display signs of intolerance to daily two-hour spontaneous breathing
trials and, when used in this fashion, may result in quicker weaning from mech-
anical ventilation than protocols that gradually reduce levels of ventilator support
(7). When managing patients on the A/C mode of ventilation, the clinician
should remember that the ventilators set RR serves as an apnea safeguard and
should therefore be set at approximately three-fourths the patients assist RR.
It is especially important to set the machine RR only a few breaths below the
assist RR if severe metabolic acidemia is present. Depending on the ventilator,
examples of machine breath types available in the A/C mode of mechanical ven-
tilation include: VC, PC, PRVC, and VAPS.
greatest success with once daily T-piece breathing trials (55) and the other with
progressive PS reductions in CPAP mode (56).
Acknowledgment
References
1. Chatburn RL, Primiano FP. A new system for understanding modes of mechanical
ventilation. Respir Care 2001; 46(6):604 621.
2. Chatburn RL, Volsko TA. Mechanical ventilators. In: Wilkins RL, Stoller JK
Scanlan CL, eds. Egans Fundamentals of Respiratory Care. 8th ed., St. Louis:
Mosby 2003; 39:929 962.
3. Dennison F. Basic ventilator wave form analysis. In: Chang DW, ed. Clinical
Application of Mechanical Ventilators. 2nd ed., Delmar Thomson Learning 2001;
11:264 324.
4. Rossi A, Gottfried SB, Zocchi L, Higgs BD, Lennox S, Calverly PM, Begin P,
Grassino A, Milic-Emili J. Measurement of static compliance of the total respiratory
system in patients with acute respiratory failure. The effect of intrinsic positive and
expiratory pressure. Am Rev Respir Dis 1985; 131(5):672 677.
5. MacIntyre NR, Cheng KG, McConnell R. Applied PEEP during pressure support
reduces the inspiratory threshold of intrinsic PEEP. Chest 1997; 111(1):188 193.
6. Leung P, Jubran A, Tobin M. Comparison of assisted ventilator modes on triggering,
patient effort and dyspnea. Am J Respir Crit Care Med 1997; 155(6):1840 1848.
7. Tobin MJ, Jubran A, Laghi F. Patient ventilator interaction. Am J Respir Crit Care
Med 2001; 163(5):1059 1063.
8. Branson RD, Campbell RS, Davis K, Johnson DJ. Comparison of pressure and ow
triggering systems during continuous positive airway pressure. Chest 1994;
106(2):540 544.
9. Giuliani R, Mascia L, Recchia F, Caracciolo A, Fiore T, Ranieri VM. Patient
ventilator interaction during synchronized intermittent mandatory ventilation.
Effects of ow triggering. Am J Respir Crit Care Med 1995; 151(1):1 9.
10. Sager J, Eiger G, Fuchs B. Ventilator auto-triggering in a patient with tuberculous
bronchopleural stula. Respir Care 2003; 48(5):519 521.
11. Nellcor Puritan Bennett, 840 Ventilator System Operators and Technical Reference
Manual No. 4-070088-00 Rev. C, 1998; TR 12:1 5.
12. Imanaka H, Nishimura M, Takeuchi M, Kimball W, Yahagi N, Kumon K. Auto trig-
gering caused by cardiogenic oscillation during ow-triggered mechanical venti-
lation. Crit Care Med 2000; 28(2):402 407.
13. Davis K, Branson RD, Campbell RS, Porembka DT. Comparison of volume control
and pressure control: is ow wareform the difference? J Trauma Injury Infect Critical
Care 1996; 41(5):808 814.
14. MacIntyre NR, McConnell R, Cheng KC, Sane A. Patient ventilator ow dys-
synchrony: ow-limited versus pressure-limited breaths. Crit Care Med 1997;
25(10):1671 1677.
15. Dick CR, Sassoon CS. Patient-ventilator interactions. Clin Chest Med 1996;
17(3):423 438.
102 Richards and Mosenifar
16. Chiumello D, Pelosi P, Croci M, Bigatello LM, Gattinoni L. The effects of pressur-
ization rate on breathing pattern, work of breathing gas exchange and patient comfort
in pressure support ventilation. Eur Respir J 2001; 18(1):107 114.
17. Bonmarchand G, Chevron V, Menard J, Girault C, Moritz-Berthelot F, Pasquis P,
Leroy J. Effects of pressure ramp slope values on the work of breathing during
pressure support ventilation in restrictive patients. Crit Care Med 1999;
27(4):715 722.
18. Kallet RH, Campbell AR, Alonso JA, Morabito DJ, Mackersie RC. The effects of
pressure control versus volume control assisted ventilation on patient work of breath-
ing in acute lung injury and acute respiratory distress syndrome. Respir Care 2000;
45(9):1085 1096.
19. Tobin MJ. Culmination of an era of research on the acute respiratory distress
syndrome. N Eng J Med 2000; 342(18):1360 1361.
20. Gattinoni L, Vagginelli F, Chiumello D, Taccone P, Carlesso E. Physiologic ration-
ale for ventilator setting in acute lung injury/acute respiratory distress syndrome
patients. Crit Care Med 2003; 31(4):S300 S304.
21. Petrucci N, Iacovelli W. Ventilation with lower tidal volumes versus traditional tidal
volumes in adults for acute lung injury and acute respiratory distress syndrome.
Cochraine Database Syst Rev 2003; (3):CD003844.
22. Tobin MJ. Medical progress: advances in mechanical ventilation. N Eng J Med 2001;
344(26):1986 1996.
23. Thompson BT, Hayden D, Matthay MA, Brower R, Parsons PE. Clinicians
approaches to mechanical ventilation in acute lung injury and ARDS. Chest 2001;
120:1622 1627.
24. Pilbeam SP. Introduction to ventilators. In: Pilbeam SP, Cairo JM, eds. Mosbys
Respiratory Care Equipment. 7th ed. St. Louis: Mosby 2004; 11:317 390.
25. Carmichael LC, Dorinsky PM, Higgins SB, Bernard GR, Dupont WD, Swindell B,
Wheeler AP. Diagnosis and therapy of acute respiratory distress syndrome in adults:
an international survey. J Critical Care 1996; 11(1):9 18.
26. Slutsky AS. Lung injury caused by mechanical ventilation. Chest 1999;
116:9S 15S.
27. Uhlig S. Mechanotransduction in the lung, ventilation-induced lung injury and
mechanotransduction: stretching it too far? AJP Lung Cell Mol Physiol 2002;
282(5):L892 L896.
28. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal
volumes as compared with traditional tidal volumes for acute lung injury and the
acute respiratory distress syndrome. N Engl J Med 2000; 342(18):1301 1308.
29. NIH/NHLBI ARDS Clinical Network: Mechanical ventilation protocol summary,
2000: Available from http://hedwig.mgh.havard.edu/ardsnet/6 ml.html/:1 4.
30. Jonson B, Svantesson C. Elastic pressure volume curves: what information do they
convey? Thorax 1999; 54(1):82 87.
31. Amato MBP, Barbas CSV, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho
G, Kairalla RA, Deheinzelin D, Munon C, Oliveira R et al. Effect of bioprotective
ventilation on mortality in the acute respiratory distress syndrome. N Engl J Med,
1998; 338(6):347 354.
32. Mancini M, Zavala E, Moncebo J, Fernandez C, Babera JA, Rossi A, Roca J,
Rodriguez-Roisin R. Mechanisms of pulmonary gas exchange improvement
Modes of Mechanical Ventilation 103
DEAN R. HESS
I. Introduction
Arterial blood gas (ABG) analysis is commonly regarded as the gold standard
for evaluation of arterial oxygenation and acid-base balance, althought the indi-
cations for ABG remain unclear (2). There are numerous problems associated
with the measurement and use of ABG (Table 1). ABGs are the most frequently
ordered laboratory test in many critical care units. A rate of 4.8 ABG/patient/day
was reported from a surgical ICU, and the presence of an arterial line was most
predictive of the number of ABGs drawn per patient (3). The presence of an arter-
ial line also affects the pattern of transfusion requirements related to phlebotomy
for diagnostic tests in adults (4 7). In one study, patients with an arterial line
were phlebotomized at a rate twice that of patients who did not have a line (5).
Another study reported increases in the number of blood tests (29% increase),
105
106 Hess
Respiratory Monitoring
During Mechanical Ventilation
Ventilator function
Lung mechanics Sedation and disconnect
airway pressure, Neuromuscular leak
flow, volume Blockade occlusion
esophageal pressure humidification
pressure-volume curve FIO2
blood-drawing procedures (30% increase), and the amount of blood volume loss
(44% increase) from patients with arterial lines compared with those without (6).
Several strategies might be used to decrease blood loss because of ABGs
(and other laboratory testing) in critically ill patients (7). Routine-use arterial
lines should be avoided. Blood conservation systems can be used to eliminate
the discard volume from arterial line blood draws (8 11). Multiple diagnostic
tests may be performed on a single sample of blood. Clinically accurate and
precise invasive and non-invasive technology can be substituted for some diag-
nostic tests. Blood sample volume can be reduced, such as by using pediatric
phlebotomy tubes and reduced syringe volume (12).
Pain
Pre-analytical errors: air contamination, heparin dilution, storage, delay before analysis
Analytic errors: accuracy differences between models of blood gas analyzers
Post-analytical errors: transcription errors, delays in reporting results
Infection risk to patient (particularly with arterial catheter)
Infection risk to clinician (particularly with arterial puncture)
Thrombosis and distal embolization (particularly with arterial catheter)
Blood loss (particularly with arterial catheter)
Intermittent information
Cost
Regulation: CLIA-88
Monitoring During Mechanical Ventilation 107
O2 Hb
FO2 Hb
O2 Hb HHb metHb COHb
Oxygen saturation can also be calculated from the measured PO2 using an empirical
equation for the oxyhemoglobin dissociation curve, as is commonly done using soft-
ware in blood gas analyzers. Clinically important errors can result from use of
calculated oxygen saturation in applications such as the shunt equation (2023).
Several errors are related to CO-oximetry. Owing to differences in the light
108 Hess
absorption spectra of fetal hemoglobin (HbF) and adult hemoglobins, the presence
of HbF may result in false elevations of COHb (24). Correction factors for the pre-
sence of HbF have been published (25), but these may not be appropriate under all
conditions (such as transfusion) (26). Intralipid infusions (27) and propofol infu-
sions may result in falsely elevated metHb levels measured by CO-oximetry.
To obtain capillary blood gas samples, a puncture is made with a lancet or similar
device into the cutaneous layer of the skin at a highly vascularized area. To accel-
erate blood ow and reduce the difference between arterial and venous gas press-
ures, vasodilation is achieved by warming or use of a vasoactive cream. Common
sites used for arterialized capillary blood gases are the heel (in neonates), nger,
toe, or earlobe.
Several studies have reported the use of arterialized capillary blood gases in
adults. Pitkin et al. (28) reported good agreement between capillary PO2 and
arterial PO2 (bias 21 mm Hg, limits of agreement 28 to 6 mm Hg). Dar et al.
(29) reported trivial and non-signicant differences between arterial and capil-
lary PO2. In contrast to these reports, Sauty et al. (30) reported a signicant
underestimation (4 mm Hg) of PaO2 by earlobe capillary blood gases with
wide limits of agreement (24 to 13 mm Hg).
Arterialized capillary blood gases produce a mixture of blood from capil-
laries and venules (31). The difference between arterial and venous PO2 is
usually large (about 60 mmHg), but this difference can be reduced by vasodila-
tion. Furthermore, the difference between arterial and venous PO2 decreases with
hypoxemia. Thus, the accuracy of arterialized capillary blood PO2 improves as
the PaO2 decreases. None the less, arterialized capillary PO2 systematically
underestimates the PaO2. Moreover, capillary blood samples are difcult to
obtain anaerobically. Exposure of the blood to room air raises the PO2, and
this may explain the agreement between arterialized capillary PO2 and PaO2
reported in some studies. Arterialized capillary blood gases have also been advo-
cated because they are less painful, but the pain of arterial puncture can be les-
sened by the use of local anesthesia (32).
Samples for blood gas analysis are drawn from the patient, packaged and trans-
ported to the laboratory, and analyzed, and the results are then communicated
back to the patient care unit. This model may result in pre-analytical errors
(e.g., failure to properly transport the sample), post-analytical errors (e.g., tran-
scription errors in reporting the results), and delay in reporting the results.
There has been increasing interest in point-of-care (POC) testing, in which
blood gases are measured at or near the site of patient care (33 40). Several
Monitoring During Mechanical Ventilation 109
studies have reported acceptably accurate blood gas results using POC testing
devices (41 45). Kendall et al. (46) performed a randomized controlled trial
of POC on clinical outcome in an emergency department and reported that
POC testing resulted in signicantly faster decision-making. However, POC
testing did not affect the amount of time spent in the emergency department,
the length of stay in the hospital, admission rates, or mortality.
POC testing devices are portable (some are hand-held), and they typically
require a small volume of blood for testing (a few drops). Blood is introduced into
a single-use disposable cartridge that is introduced into the portable analyzer. The
cartridge chosen determines the tests that will be run (e.g., blood gases, electro-
lytes, hematocrit, glucose, BUN, creatinine, ionized calcium, and others). The
sensor calibrates automatically, after which the blood sample is drawn over the
biosensors. The POC analyzer then calculates, displays, and stores the results.
The POC testing device can communicate with the central laboratory or hospital
information system for reporting and archiving results.
The role of POC testing is evolving and is likely to expand in the future.
Quality control and quality assurance must be appropriately addressed. This is
typically achieved by following the manufacturers recommendations for use
of the device. The costs of the base unit the and the test cartridges must be
balanced against a quicker test turn-around time (potentially resulting in more
rapid treatment and better patient outcomes), the lower overhead (compared to
the central laboratory), and the smaller blood volume required for testing (result-
ing in lower transfusion requirements).
Blood gas monitors measure pH, PCO2, and PO2 without permanently removing
blood from the patient (47). Blood gas monitors use uorescent optodes, which
are optical biosensors. Light is used to activate electrons in a dye to a higher
energy state. When the light exposure ends, the electrons return to their basal
state and light is re-emitted. The lower frequency of re-emitted light is augmented
as the concentration of hydrogen ions or carbon dioxide increases, and the
re-emitted light is quenched as the concentration of oxygen increases. Photo-
sensors are used to quantify the amount of quenching, and a microprocessor is
used to translate the signal into a display of PO2, PCO2, and pH. Various con-
gurations of optodes for intra-arterial blood gas monitors have been developed.
Optode systems are electrically isolated from the patient, they can be miniatur-
ized so that they t through an arterial catheter without affecting other catheter
functions, they are stable over time, they respond rapidly to changes in substrate
concentration, and they are not consumed in the measurement process.
Several approaches can be taken to blood gas monitoring in conjunction
with an arterial catheter. The rst approach uses a probe that passes through
the arterial catheter and resides directly within the arterial lumen (i.e., in vivo
110 Hess
blood gas monitor). With the second approach, the optode system is connected to
the proximal arterial line but does not pass through the catheter. With this
method, when blood gas and pH levels are desired, blood is drawn into a
chamber containing the optodes (i.e., ex vivo blood gas monitor). After analysis,
the blood is ushed back into the artery. Thus, frequent (but not continuous)
blood gas measurements are possible without blood loss. Studies evaluating
these devices (48 64) have generally reported acceptable overall accuracy of
blood gas monitors. Some of these studies, however, report occasional discrepan-
cies between the blood gas monitor and a blood gas analyzer. The reasons for this
may relate to issues such as sensor position and catheter ushing.
One problem with these systems is their cost. Blood gas analyzers are
equally expensive, but they are not dedicated to a single patient. Although manu-
factures are striving to design user-friendly systems that are easy to use, the
amount of technical attention that these systems are likely to require in a busy
ICU remains to be determined. The life of the sensor in a busy ICU also
remains to be seen. The quality control and quality assurance requirements for
clinical and regulatory purposes are unclear. Whether or not the benets of con-
tinuous blood gas and pH measurements will outweigh the costs and technical
support required remains to be seen.
Blood obtained from the central venous circulation or pulmonary artery is used to
assess mixed venous oxygenation. If the catheter is positioned in a distal branch
of the pulmonary artery, rapid aspiration can contaminate the mixed venous
blood with pulmonary capillary blood (65). Although the importance of this
has been questioned (66), it is prudent to use a slow aspiration rate (,1 mL/
30 sec) when blood is sampled from a pulmonary artery catheter.
The mixed venous oxygenation has been commonly thought to represent
tissue oxygenation. However, several points are important in relation to the
ability of mixed venous oxygenation to reect tissue oxygenation. SvO2 (or
PvO2) depends on the oxygen content derived from many vascular beds, and it
thus does not necessarily reect well the SvO2 of any individual vascular bed.
With septic shock, peripheral arteriovenous shunts open, resulting in an SvO2
higher than expected (67). SvO2 will also be falsely elevated (i.e., not reect
tissue oxygenation) in patients with ventricular septal defect or cyanide poison-
ing. During circulatory arrest, large concentrations of CO2 may accumulate in the
venous circulation. With cardiac arrest, very high PvCO2 (venous respiratory
acidosis) has been reported in the presence of normal or low PaCO2 (arterial
respiratory alkalosis) (68,69).
SvO2 can be mathematically derived from the Fick equation:
_ 2
VO
Sv O2 1
DO2
Monitoring During Mechanical Ventilation 111
where VO2 is oxygen consumption and DO2 is oxygen delivery. The relationship
VO2/DO2 is the oxygen utilization coefcient or the oxygen extraction ratio.
SvO2 is thus determined by the balance between VO2 and DO2.
Mixed venous blood is obtained from the pulmonary artery. Oxygen satur-
ation is slightly higher in the inferior vena cava than the superior vena cava due to
the relatively low levels of oxygen extraction relative to perfusion in the renal and
mesenteric vasculature (70). During periods of stress (e.g., shock, hypovolemia,
exercise), however, there is a drop in the oxygen saturation of the inferior vena
cava, so the saturation in the superior vena cava is greater than that of the inferior
vena cava. This occurs as the result of decreased renal and splanchnic perfusion
during these conditions. Use of central venous blood other than that obtained
from the pulmonary artery may be an inadequate reection of true mixed
venous blood. Although mixed venous blood samples should ideally be obtained
from the pulmonary artery, a high correlation between pulmonary shunt calcu-
lations using pulmonary artery samples and central venous samples (e.g., right
atria or superior vena cava) has been reported (70). A peripheral venous blood
gas reects the PO2 and PCO2 of the tissue bed served by that vein. Thus periph-
eral venous blood gases are not useful as a proxy for either an arterial blood gas
sample or a central venous blood gas sample.
The ability to continuously measure mixed venous oxygen saturation via
oximetric methodology became commercially available in the early 1980s.
This system uses a microprocessor, an optical module with light sources and
photodetectors, and a ow-directed pulmonary artery catheter (71 73). Using
beroptics, wavelengths of light between 650 and 1000 nm are pulsed into the
pulmonary artery. This light is reected off red blood cells in the pulmonary
artery and returned to the optical module via another beroptic bundle. SvO2 is
determined by the ratio of transmitted and reected light. Before insertion, the
system is calibrated with an in vitro calibration standard, and this can be
updated periodically by in vivo calibration using a CO-oximetry determined
SvO2. Factors that affect the measurement of SvO2 using this method include
temperature, pH, blood ow velocity, hematocrit, and occlusion of the catheter
tip (e.g., clot or vessel wall).
A number of studies have evaluated the accuracy of SvO2 catheter systems
(74 81). Most of them reported that the three wavelength systems correlate
better with CO-oximeter SvO2 than two wavelength systems and that drift is
less with three wavelength systems. Studies reporting bias and precision analysis
have found greater degrees of inaccuracy than those claimed by the manufacturer.
However, if meticulous attention is paid to care of the catheter system (e.g.,
in vivo calibrations, updating the hematocrit setting), either of these systems
may be acceptable to detect changes or trends. These systems do require more
technical attention than other monitoring systems, such as pulse oximetry or a
conventional pulmonary artery catheter.
Regardless of their technical performance, the clinical usefulness of SvO2
monitoring remains controversial. The results of a large multi-center study
112 Hess
(.10,000 patients in .50 ICUs) showed that targeting SvO2 . 70% did not
affect mortality (82). More recently, Rivers et al. (83) evaluated the efcacy of
early goal-directed therapy before admission to the ICU. Patients who arrived
at an emergency department with severe sepsis or septic shock were randomized
to receive either six hours of early goal-directed therapy or standard therapy
before admission to the ICU. In-hospital mortality was 30.5% in the group
assigned to early goal-directed therapy, compared with 46.5% in the group
assigned to standard therapy. In this study, a central venous catheter capable of
measuring oxygen saturation was used, with a targeted central venous oxygen
saturation 70%.
Continuous pulse oximetry has become a standard of care for critically ill
patients. With pulse oximetry, two wavelengths of light (660 and 940 nm) are
passed through a pulsating vascular bed (84,85). This is accomplished using
two light-emitting diodes (LEDs) and a photodetector. There is some error in
the wavelength of light emitted by the LEDs (+30 nm) that can affect accuracy.
Also, the photodetector is not specic (i.e., if will respond to any wavelength of
light, which can result in interference). Although some of the light emitted from
the LEDs is absorbed by each constituent of the tissue, the only variable absorp-
tion is due to arterial pulsations. This is translated into a plethysmographic
waveform. The ratio of the amplitudes of these two plethysmographic waveforms
is translated into oxygen saturation.
A probe is used to pass light from the LEDs through a pulsating vascular bed.
Disposable and reusable probes are available and equally accurate (86). Pulse oxi-
meter probes can become contaminated with pathogenic bacteria and serve as a
source of nosocomial infection. A protective sheath can be used to allow the
same disposable probe to be used on multiple patients (87). The use of disposable
pulse oximeter probes can become very expensive for the hospital, and it has been
shown that disposable probes can be gas sterilized and recycled (88).
A number of limitations of pulse oximetry should be recognized, appreci-
ated, and understood by everyone who uses pulse oximetry data. Most pulse oxi-
meter errors can be explained as too little signal (e.g., low perfusion, improper
probe placement) or too much noise (e.g., motion, ambient light).
. Accuracy: Many clinical evaluations of the accuracy of pulse oxi-
meters have been published, including a meta-analysis (89). At satur-
ations .80%, the accuracy of pulse oximetry is +4% to +5%.
Below 80%, the accuracy is worse, but the clinical importance of this
is questionable. To appreciate the implications of the limits of accuracy
of pulse oximetry, one must consider the oxyhemoglobin dissociation
curve (Fig. 2). If the pulse oximeter displays a SpO2 of 95%, the true
saturation could be as low as 90% or as high as 100%. If the true
Monitoring During Mechanical Ventilation 113
100
90 pH 7.25
pH 7.40
80
70
SO2 (%)
60
50
40
30
20
10
0
0 20 40 60 80 100 120 140 160
PO2 (mm Hg)
saturation is 90%, the PO2 will be about 60 mmHg. If the true satur-
ation is 100%, however, one does not know how high the PO2 might
be. Also, a shift of the oxyghemoglobin dissociation curve can
change the SpO2, although no change in PaO2 has occurred.
. Misunderstanding by those who use pulse oximetry: Although pulse
oximetry is commonly used, it may be misunderstood by users.
Several studies (90,91) have reported that physicians and nurses
lacked adequate knowledge of pulse oximetery and made serious
errors in the interpretation of SpO2.
. Differences between devices and probes: The pulse oximeter is
unique in that it requires no user calibration. However, manufacturer-
derived calibration curves programmed into the software vary
from manufacturer to manufacturer, and they can vary among pulse
oximeters of a given manufacturer. Moreover, the output of LEDs
can vary from probe to probe. The accuracy of pulse oximetry has
been shown to vary among devices (91 97). The same pulse oximeter
and probe should ideally be used for each SpO2 determination on a
given patient.
. Penumbra effect: If the nger pulse oximeter probe does not t cor-
rectly, light can be shunted from the LEDs directly to the photodetector
(98). Theoretically, this will cause a falsely low SpO2 if SaO2 . 85%,
and a falsely elevated SpO2 if SaO2 , 85%.
. Dyshemoglobinemias: Because commercially available pulse oxi-
meters use only two wavelengths of light, they are only able to evaluate
O2Hb and HHb. Pulse oximeters assume that COHb and metHb con-
centrations are low. Abnormal elevations of COHb (99 102) and
114 Hess
Motion artifact and low perfusion are common causes of pulse oximetry
errors (123,124). A technique that uses mathematical manipulation of the pulse
oximeter signals to measure and subtract the noise components of motion and
low perfusion is commercially available (Masimo Corporation, Mission Viejo,
California, U.S.) (125). Barker (126) performed a human volunteer experiment
to compare the performances of 20-pulse oximeters during combinations of
motion and hypoxemia. A motorized table produced different hand motions,
and each motion was studied during both room air breathing and hypoxemia.
Pulse oximeters on the non-moving hand were used to provide control
Monitoring During Mechanical Ventilation 115
Figure 3 (Left): In caucasian patients (open circles), a SpO2 92% is reliable in pre-
dicting a PaO2 60 mmHg. (Right): In patients with deeply pigmented skin (lled
circles), a SpO2 95% was required to reliably predict a PaO2 60 mmHg. Source:
Adapted from Ref. 111.
116 Hess
PaCO2, the Pr aCO2 increases. Levy et al. reported an increased mortality when
the Pr aCO2 gap was .20 mm Hg (145).
The gastric tonometer consists of a nasogastric tube with a distal CO2-
permeable silastic balloon. The balloon is lled with 2.53 mL of physiologic
saline solution. An equilibration time of one to two hours allows CO2 in the
gastric lumen to equilibrate with the saline inside the balloon. After discarding
11.5 mL of aspirate from the gastric tube (dead volume), the remaining
11.5 mL is analyzed for PCO2 using a blood gas analyzer. A simultaneous arterial
blood sample is analyzed to determine HCO2 3 . Gastric pHi is calculated from
the HendersonHasselbalch equation, using the PCO2 of the saline from the
gastric balloon and the HCO2 3 of arterial blood. Several studies have reported
substantial preanalytical errors related to air contamination of the saline sample
(147149). This has led to the development of gas tonometry techniques
(146,150153). Another advantage of these newer techniques is that they are
automated and can be used for nearly continuous monitoring. Even with improve-
ments in the technical aspects of gastric tonometry, its use remains controversial
(154,155).
Sublingual PCO2 (PSLCO2) correlates well with gastric intramucosal PCO2
and is relatively easy to perform (156 158). PSLCO2 is measured by using a
CapnoProbe subingual PCO2 measurement system (Nellcor, Pleasanton, CA,
U.S.A.). The CapnoProbe consists of a disposable PCO2 sensor and a battery-
powered handheld instrument. The sensor is a CO2-sensing optode. The optode
has a CO2-permeable capsule lled with a uorescent indicator at the distal
end of an optical ber. The uorescent indicator is excited by light conducted
through the optical ber, with the uorescent emission of the indicator being
monitored by the optical ber. As CO2 enters the capsule, carbonic acid is
formed, resulting in a lower pH. The pH change causes a shift in the uorescence
of the indicator, which is detected through the ber optics. The change in uor-
escence is then converted to a PCO2 via the Henderson Hasselbalch equation.
IX. Capnography
Capnometry is the measurement of CO2 at the airway opening during the venti-
latory cycle (159). Capnography is the graphic waveform display of CO2 as a
function of volume or time. The waveform displayed by a capnograph is called
a capnogram. Most bedside capnometers measure CO2 by infrared absorption,
which uses the absorption peak for CO2 at 4.26 mm. The CO2 in the sample
cell decreases the radiation transmitted to the detector. The increased radiation
transmitted from the reference cell (relative to that from the sample cell) is
detected and used to determine the CO2 concentration. The mass spectrometer
can also be used to measure CO2, in which the sample is aspirated into a
vacuum chamber where the gas is ionized by an electron beam. The charged frag-
ments are accelerated into a dispersion chamber where they are separated
118 Hess
according to mass and charge. When light strikes gas molecules, energy is
absorbed and reemitted at the same wavelength and direction, and a small frac-
tion of the absorbed energy is reemitted at new wavelengths in a phenomenon
known as Raman scattering. Raman scattering results in a red-shifted spectrum,
and the wavelength shift and amount of scattering can be used to measure the
PCO2. A portable non-electronic single-patient-use device is commonly used
to produce a color change (colorimetric end-tidal CO2 detection) in the presence
of exhaled CO2 (i.e., tracheal intubation). The color changes from purple with a
low CO2 concentration to yellow with a CO2 concentration of 2.0% to 5.0%.
Capnometers can be congured as mainstream or sidestream devices. With
the mainstream capnometer, the measurement chamber is placed directly at the
airway. With the sidestream capnometer, gas from the airway is aspirated
through ne bore tubing to the measurement chamber inside the device. Some
devices (e.g., colorimetric capnometers) can only be used as mainstream
devices, whereas other devices (e.g., mass and Raman spectrometers) can only
be used as sidestream devices. Infrared capnometers can be congured as
either mainstream or sidestream devices. There are advantages and disadvantages
of each approach (Table 2).
A new capnography technology, Microstream, has recently been intro-
duced (160). Microstream features low ow rates, reduced dead space, and
lack of moisture-associated occlusion problems. Microstream technology uses
a novel molecular correlation spectroscopy source that operates at room tempera-
ture and emits only CO2 specic radiation. The breath sample is brought to the
measuring cell at a ow of 50 mL/min via a specially designed Microstream
breath sampling circuit using a miniature diaphragm pump.
Carbon dioxide homeostasis is affected by volume of CO2 production
(VCO2), CO2 transport from the tissues to the lungs, and alveolar ventilation.
Conditions that increase VCO2 include fever, activity, sepsis, hyperthyroidism,
Advantages Disadvantages
Mainstream Capnometer
Sensor at patient airway Secretions and humidity block sensor
Fast response (crisp waveform) Sensor heated to prevent condensation
Short lag time (real-time readings) Bulky sensor at patient airway
No sample ow to reduce tidal volume Does not measure N2O
Difcult to use with non-intubated patients
Cleaning and sterilization of reusable sensor
Sidestream Capnometer
No bulky sensors or heaters at airway Secretions block sample tubing
Ability to measure N2O Water trap required
Disposable sample line Slow response to CO2 changes
Can be used with nonintubated patients Sample ow may decrease tidal volume
Monitoring During Mechanical Ventilation 119
trauma and burn injuries, and high carbohydrate intake. Conditions that decrease
VCO2 include hypothyroidism, hypothermia (if shivering is controlled), sedation,
and paralysis. Carbon dioxide from tissue metabolism diffuses into the circula-
tion, producing a mixed venous PCO2 (PvCO2) of about 45 mmHg. The PCO2
of an individual lung unit depends upon the relationship between ventilation
and perfusion (V/Q ) (Fig. 5) (161 167). With no perfusion (dead space;
V/Q 1), the PaCO2 is the same as the inspired PCO2 (i.e., zero). With a
normal V/Q , the PACO2 is the same as the arterial PCO2 (i.e., 40 mm Hg).
With a low V/Q , the PACO2 increases towards the PvCO2 (i.e., 45 mm Hg).
The PACO2, and thus the end-tidal PCO2, must always be between zero and
the PvCO2. With a variety of V/Q throughout the lungs, the end-tidal PCO2
(PETCO2) is normally several mm Hg less than the PaCO2. However, the relation-
ship between the PaCO2 and PETCO2 will vary depending upon the relative
contributions of various V/Q units of the lungs.
The volume-based capnogram is displayed with PCO2 on the ordinate and
volume on the abscissa (Fig. 6). At the beginning of exhalation, PCO2 remains
zero as gas from the anatomic dead space leaves the airway (Phase I). The cap-
nogram then rises sharply as alveolar gas mixes with dead space gas (Phase II).
The capnogram then forms a plateau during most of exhalation (Phase III). Phase
III represents gas ow from alveoli and is thus called the alveolar plateau.
The slope of Phase III is determined by the V/Q status of the lung (163,168
171). The slope of Phase III is increased in patients with airway obstruction
(e.g., COPD, asthma). The PCO2 at end-exhalation is the PETCO2.
Airway dead space volume (anatomic dead space), physiologic dead space
fraction (VD/VT), and VCO2 can be determined from the volume-based capno-
gram (Fig. 6). If VCO2 is known, it is possible to calculate metabolic rate:
_
REE VCO 2 5:52 1440
where REE is resting energy expenditure (kcal/d), VCO2 is in L/min, 5.52 is the
caloric equivalent for CO2, and 1440 is the number of minutes in a day. The
%CO2in
Arterial Blood Alveolar dead space
VD VALV
mixed exhaled PCO2 (PE CO2) can be calculated from VCO2 if the minute ven-
tilation (VE) is known:
PE CO2 _
VCO 2
_
VCO _
2 VE or PE CO2 Pbar
Pbar V_ E
where Pbar is barometric pressure. The VD/VT can then be calculated if the PaCO2
is known:
VD Pa CO2 PE CO2
VT Pa CO2
Increased VD/VT has been associated with mortality in patients with ARDS
(172). Increased VD/VT has also been associated with a lower rate of weaning
from mechanical ventilation (173).
For critical care applications, the time-based capnogram is often displayed
(Fig. 7). Unlike the volume-based capnogram, the time-based capnogram has an
inspiratory segment and an expiratory segment. The PCO2 is zero during the
inspiratory phase. At the beginning of exhalation, PCO2 remains zero as gas
from the anatomic dead space leaves the airway (Phase I). The capnogram
then rises sharply as alveolar gas mixes with dead space gas (Phase II). The
curve then forms an alveolar plateau during most of exhalation (Phase III).
The PCO2 at the end of the alveolar plateau is the PETCO2. Note that Phase I,
Phase II, and Phase III are similar for the time based capnogram and the
volume-based capnogram. The capnogram with airow obstruction is character-
ized by an increased slope of Phase III (Fig. 8). The lack of an alveolar plateau
Monitoring During Mechanical Ventilation 121
end
III exhalation
PCO2
II
I
time
begin
exhalation
Figure 7 Time-based capnogram. I, anatomic dead space; II, the transition from
anatomic dead space to the alveolar plateau; III, the alveolar plateau.
occurs because of the V/Q abnormalities that result from the airow obstruction.
In asthmatic patients with acute bronchospasm, the slope of Phase III has been
shown to correlate with peak expiratory ow rate, and it normalizes with beta-
agonist therapy (170,171).
The PETCO2 represents alveolar PCO2, and the PACO2 is the result of the
V/Q . With a normal V/Q , the PACO2 will approximate the PaCO2. If ventilation
is decreased compared with perfusion, there will be more time for equilibration
between PvCO2 and PACO2, and thus the PACO2 will rise towards PvCO2. With a
high V/Q (i.e., dead space), PACO2 will approach the inspired PCO2 (PICO2),
end
exhalation
III
PCO2
II
I
time
begin
exhalation
which is usually zero. The PETCO2 is a mixture of gas from millions of alveoli
and thus represents the mixture of many potentially different PACO2. Theoreti-
cally, PETCO2 could be as low as the PICO2 (zero) or as high as the PvCO2
(but not higher than this). An increase or decrease in PETCO2 can be the result
of changes in VCO2 and delivery to the lungs (167,174), changes in alveolar ven-
tilation, or an equipment malfunction (Table 3). However, because of homeosta-
sis, compensatory changes may occur so that PETCO2 does not change despite
these changes. For example, if VCO2 increases (such as with fever) and alveolar
ventilation increases proportionately (the normal homeostatic response), then
PETCO2 may not change. Thus, PETCO2 is a non-specic indicator of cardiopul-
monary homeostasis and often does not indicate a specic problem or
abnormality.
If the PaCO2 is measured, the gradient between PaCO2 and PETCO2
(Pa2ETCO2) can be calculated. This gradient is normally small, being less than
5 mm Hg. With dead space-producing disease (high V/Q ), the PETCO2 may be
considerably less than the PaCO2 (Table 4) (174 185). Although shunting may
result in a large gradient between PAO2 and PaO2, it only has a small effect on
the Pa2ETCO2. On occasion, the PETCO2 may be greater than the PaCO2. The
reason for a PETCO2 greater than PaCO2 is not well understood (186) and may
relate to low (but nite) V/Q regions within the lung. Fletcher and Jonson
(162) have reported that PETCO2 is more often greater than PaCO2 when the
tidal volume is high. Tulou and Walsh (187) found that the Pa2ETCO2 decreased
signicantly when measured at maximal exhalation in patients with airway
obstruction. With a larger tidal volume, the greater expiratory time may allow
Pulmonary hypoperfusion
Pulmonary embolism
Cardiac arrest
Positive pressure ventilation (especially PEEP)
High-rate low-tidal volume ventilation
lung units with a low V/Q (and thus a longer time constant) to empty (188). Jones
et al. (189) found PETCO2 greater than PaCO2 during exercise and attributed this
to an increase in PACO2 because of increased CO2 that is emitted into a lung
volume becoming smaller during exhalation.
Stability of the Pa2ETCO2 is necessary if PaCO2 is to be reliably predicted
from PETCO2. Hoffman et al. (178) evaluated PETCO2 in critically ill patients
during changes in mechanical ventilation and found that changes in PETCO2
did not correlate well with changes in PaCO2 and that the trends in PETCO2
were opposite of the trends in PaCO2 in some patients. Raemer et al. (181)
found that the Pa2ETCO2 was too variable during anesthesia to allow precise pre-
diction of PaCO2 from PETCO2. Russell et al. (182 184) found that changes in
PaCO2 and PETCO2 were in opposite directions in a sizeable proportion of
patients. Because of a uctuating and unpredictable Pa2ETCO2, caution must
be used when predicting PaCO2 from PETCO2.
Measurement of PETCO2 is a standard of care to determine proper endotra-
cheal tube position (190 196). Esophageal PCO2 may be high (5%) after
exhaled gas ventilation following inadvertent gastric distention (192), but it
drops to ,2% following six ventilations of the stomach. Ingestion of a carbo-
nated beverage also results in an increased esophageal PCO2, but this rapidly
decreases following 10 to 15 seconds of gastric ventilation (197). Capnography
has also been reported useful to verify feeding tube placement (198).
Typically, the onset of cardiac arrest results in a drop of PETCO2 to zero.
With the initiation of CPR, there is an increase in PETCO2. PETCO2 correlates
with cardiac output (199 202) (i.e., pulmonary blood ow) and coronary per-
fusion pressure (203,204) during CPR. Kalenda (205) observed that PETCO2
decreased when a resuscitator became fatigued and that it increased when resus-
citation was continued with another resuscitator. Garnett et al. (206) reported that
PETCO2 increased immediately in patients who had a return of spontaneous cir-
culation. Similar results were reported by Falk et al. (207) and Sanders et al.
(208). Patients who were resuscitated had a PETCO2 of 15 + 4 mm Hg during
resuscitation, whereas patients who were not resuscitated had a PETCO2 of
only 7 + 5 mm Hg (208,209). Grmec et al. (210) reported that the initial
PETCO2 was greater in patients with asphyxial cardiac arrest than in those with
cardiac arrest due to ventricular brillation or ventricular tachycardia.
124 Hess
The relationship between the Pa2ETCO2 and VD/VT has been reported
(161,162,185,211 213). Fletcher et al. (162) reported that patients with
increased alveolar dead space had an increased slope of the alveolar plateau on
the capnogram. Yamanaka and Sue (185) reported that the Pa2ETCO2 correlated
closely with VD/VT. A common clinical cause of increased dead space is pulmon-
ary embolism, and there has been interest in the use of capnography in this setting
(214 218). In a comparison of capnometry to angiography in the diagnosis of
pulmonary embolism in 44 adult patients with chronic obstructive pulmonary
disease (COPD), Chopin et al. (217) reported a sensitivity of 100% but a speci-
city of only 65% (i.e., a false positive rate of 35%). Although the negative pre-
dictive value was 100%, the positive predictive value was only 74%. The
Pa2ETCO2 is usually increased when pulmonary embolism is present, but the gra-
dient is also increased for a variety of other causes when pulmonary embolism is
not present (e.g., any dead space-producing disease). Hatle (214) reported that
Pa2ETCO2 measured at forced exhalation was more useful in the evaluation of
acute pulmonary embolism. Eriksson et al. (216) and Verschuren et al. (218)
extrapolated phase III of the volumetric capnogram to determine the PCO2 at
15% of the predicted total lung capacity and found that this was useful in the
diagnosis of pulmonary embolism (Fig. 9). With maximal exhalation, the
gradient approached zero in patients with obstructive lung disease but remained
high in patients with pulmonary embolism.
There has been interest in the use of capnometry during weaning from mech-
anical ventilation (219227). Most evaluations of the use of capnometry during
weaning have been conducted on post-operative patients. It can be argued that
the weaning of these patients is usually uncomplicated, that such patients require
little monitoring during weaning, and that extrapolation of these data to difcult-
to-wean patients is not valid. Perhaps more important, hypoventilation is often a
late nding with respiratory muscle fatigue (such as might occur during
weaning). Other clinical ndings such as tachypnea, use of accessory muscles of
breathing, and thoraco-abdominal paradox may be more sensitive indicators than
an increase in PCO2 to indicate fatigue during weaning from mechanical venti-
lation. On the basis of the available evidence, however, routine capnometry
cannot be recommended during weaning from mechanical ventilation. However,
the use of capnometry in long-term weaning unit has been reported useful (228).
Avoidance of hypercapnia remains a common treatment objective in the
care of patients with head injury. Mackersie and Karagianes (229) evaluated
the use of PETCO2 in 36 head-injured patients and reported good correlation
between PETCO2 and PaCO2. Sharma et al. (230) reported a stable Pa2ETCO2
during neuroanesthesia. In contrast to these ndings, others have reported that
PETCO2 is not a useful predictor of PaCO2 in neurosurgical patients (231,232).
Because many head injured patients are young and free of lung disease (at
least early in their hospital course), PETCO2 may be useful to monitor PaCO2.
However, PETCO2 may not be a very good predictor of PaCO2 in these patients
if acute or chronic lung disease is present or if pulmonary blood ow is reduced
secondary to hypotension or pulmonary embolism. Kerr et al. (233) found that
PETCO2 correlated well with PaCO2 in patients without respiratory compli-
cations, but its clinical validity is questionable in patients who have the greatest
need for PETCO2 monitoring (those patients with respiratory failure).
Palmon et al. (234) reported that PCO2 could be more tightly controlled
during transport when PETCO2 was monitored, but their data did not support
routine use of capnometry during short transport times. In trauma patients intu-
bated in the eld, Helm et al. (235) reported tighter control of PCO2 with the
use of capnography. In this study, the incidence of normoventilation was
greater and the incidence of hypoventilation was less with the use of
capnography.
Using volume-based capnography, it is possible to noninvasively measure
cardiac output with the partial CO2 rebreathing technique (NICO, Novametrix,
Wallingford, Connecticut, U.S.) (236,237). VCO2 is calculated on a breath-by-
breath basis, and the differential Fick equation is applied to establish the relation
between VCO2 and cardiac output (Q ):
_
VCO _
2 Q (Cv CO2 Ca CO2 )
126 Hess
where CvCO2 represents the CO2 content in mixed venous blood, and CaCO2 rep-
resents the CO2 content in arterial blood. CO2 rebreathing is performed for
35 seconds every 3 minutes (Fig. 10). Assuming that Q remains constant
during the rebreathing procedure yields the following:
_
DVCO _
2 Q (DCv CO2 DCa CO2 )
where DVCO2 is the change in VCO2 between normal breathing and rebreathing,
DCvCO2 the change in mixed venous carbon dioxide content, and DCaCO2 the
change in arterial carbon dioxide content. If CvCO2 remains constant during
rebreathing, the following equation is used:
_
DVCO _
2 Q (DCa CO2 )
Figure 10 Rebreathing cycle used by NICO to measure cardiac output using the partial
rebreathing technique. Source: Courtesy of Novametrix, Wallingford, Connecticut, U.S.A.
Monitoring During Mechanical Ventilation 127
_
DVCO 2 PCBF (Cc CO2 )
_
PCBF DVCO 2 =(S DPET CO2 )
_ PCBF
Q
_ s =Q
(1 Q _ t)
where Q s/Q
t is the intrapulmonary shunt fraction. The noninvasive method for
estimating Q s/Q t is adapted from Nunns iso-shunt plots, which are a series of con-
tinuous curves indicating the relation between PaO2 and FIO2 for different levels
of shunt. PaO2 is non-invasively determined using a pulse oximeter. There are
several potential limitations of partial rebreathing for the measurement of
cardiac output. In non-paralyzed patients, rebreathing increases respiratory rate,
which reduces the magnitude of the signal and limits the ability to detect
changes in PETCO2 and VCO2. Noise is increased by respiratory pattern irregula-
rities that produce unstable PETCO2 and VCO2 that may impair accuracy.
Additional cardiac output not calculated due to shunt fraction is estimated from
SpO2 and FIO2, which introduces errors. Several studies have reported mixed
results of accuracy of this method for measuring cardiac output (236 245).
VT
PAW Pmus R V_ PEEP
C
where PAW is proximal airway pressure, Pmus the pressure generated by the res-
piratory muscles, C the respiratory system compliance, VT the tidal volume, R the
airways resistance, V the inspiratory ow, and PEEP the end-expiratory alveolar
pressure set on the ventilator. The Equation of Motion predicts that proximal
airway pressure will increase with a higher tidal volume, lower respiratory
system compliance, higher airways resistance, higher inspiratory ow, higher
PEEP, and the presence of auto-PEEP.
Because of the airways resistance, proximal airway pressure will always be
greater than alveolar pressure during inspiration if ow is present. During volume-
controlled ventilation, plateau pressure (Pplat) is obtained by applying an end-
inspiratory breath-hold for 0.5 to 2 seconds, during which pressure equilibrates
throughout the system so that the pressure measured at the proximal airway
approximates the peak alveolar pressure (Fig. 11) (249,250). Pplat is not valid
during active breathing and thus cannot be measured with ventilator modes such
as pressure support ventilation. During pressure-controlled ventiation, the ow
may decrease to zero before the end of the inspiratory phase, in which case PIP
and Pplat are equal.
Pplat is determined by tidal volume, PEEP and respiratory system
compliance: Pplat VT/C PEEP. Pplat indicates a greater risk of alveolar
PIP
resistance
flow
Pplat
compliance
tidal volume
PEEP time
where DP is the pressure applied to the airway above PEEP, e the base of the
natural logarithm, t the elapsed time after initiation of the inspiratory phase,
and t the time constant of the respiratory system.
Incomplete emptying of the lungs occurs if the expiratory phase is termi-
nated prematurely (252). When this occurs, alveolar pressure does not equilibrate
with proximal airway pressure at end-exhalation, and gas trapping results. The
pressure produced by this trapped gas is called auto-PEEP. Auto-PEEP increases
end-expiratory lung volume (hyperination). It is measured by applying an
end-expiratory pause for 0.5 to 2 seconds (Fig. 12). The pressure measured at
the end of this maneuver that is in excess of the PEEP set on the ventilator is
auto-PEEP. For a valid measurement, the patient must be relaxed and breathing
in synchrony with the ventilator. Many patients with COPD contract their
abdominal muscles during exhalation (253 255). This is an important determi-
nant of auto-PEEP in these patients but likely does not produce hyperination.
It has also been shown that the end-expiratory pause method can underestimate
auto-PEEP with complete airway closure during expiration, as may occur
during mechanical ventilation of patients with severe asthma (256).
Auto-PEEP is a function of ventilator settings (tidal volume and expiratory
PIP PIP
pressure
auto PEEP
set PEEP
time
time) and lung function (airways resistance and lung compliance). It can be
expressed mathematically as (247):
VT
auto-PEEP K
C (e x Te 1)
where Kx is the inverse of the expiratory time constant (1/t) and Te is the expira-
tory time. Note that auto-PEEP is increased with increased resistance and com-
pliance (e.g., chronic obstructive lung disease), increased respiratory rate or
increased inspiratory time (both decrease expiratory time), and increased tidal
volume. Clinically, auto-PEEP can be decreased by decreasing minute venti-
lation (rate or tidal volume), increasing expiratory time, or decreasing airways
resistance (e.g., bronchodilator administration). Because set-PEEP may counter-
balance auto-PEEP in patients with ow limitation, auto-PEEP should be meau-
sured with PEEP 0.
Esophageal pressure is measured from a balloon inated with a small
volume of air (,1 mL) placed into the lower esophagus (257). Esophageal
pressure changes reect changes in pleural pressure. However, the absolute eso-
phageal pressure does not reect absolute pleural pressure. Changes in
esophageal pressure can be used to assess respiratory effort and work-of-breath-
ing during spontaneous breathing, to assess chest wall compliance during full
ventilatory support, and to assess auto-PEEP during spontaneous breathing.
If an esophageal balloon is not present, changes in pleural pressure can be esti-
mated by observing the respiratory variability of the central venous pressure
(Fig. 13) (258).
If exhalation is passive, the change in esophageal pressure required to
reverse ow at the proximal airway (i.e., trigger the ventilator) reects the
amount of auto-PEEP. Negative esophageal pressure changes that produce no
ow at the airway indicate failed trigger efforts (Fig. 14). Clinically, this is recog-
nized as a patient respiratory rate that is greater than the trigger rate on the ven-
tilator (observed by inspecting chest wall movement).
The airway pressure waveform is affected by active breathing efforts of the
patient. Patient ventilator dyssynchrony results in an airway pressure waveform
that varies from breath to breath, particularly during volume-controlled ventilation
(250,257). A special form of patient ventilator dyssynchony can occur during
pressure support ventilation, in which the patient actively exhales to terminate
the inspiratory phase (259,260). This is seen as a pressure spike at end-inspiration,
causing the ventilator to pressure-cycle to the expiratory phase (Fig. 15).
During volume-controlled ventilation, the inspiratory ow is that set on the
ventilator. During pressure-controlled or pressure-support ventilation, ow is
determined by the pressure applied across the lungs, airways resistance, and
the time constant (261,262):
_ DP
V (et=t )
R
Monitoring During Mechanical Ventilation 131
Inhalation Exhalation
positive
pressure mm Hg
ventilation 18
Inhalation Exhalation
mm Hg
spontaneous 18 Inspiratory muscle effort
breathing
10
Figure 13 (Top) During positive pressure ventilation in a relaxed patient, the increase
in CVP during the inspiratory phase is determined by chest wall compliance. (Bottom)
During spontaneous breathing, the decrease in CVP during the inspiratory phase is deter-
mined by inpiratory muscle effort. Abbreviation: CVP, central venous pressure.
flow 0
(L/s)
0.8
volume
(L)
0
30
Paw
(cm H2O)
0
20
Peso
(cm H2O) 0
estimation of missed
auto-PEEP trigger effort
Figure 15 With pressure support ventilation, patients may terminate the inspiratory
phase by actively exhaling. This produces an end-inspiratory pressure spike. In this
case, the ow does not decrease to the termination ow, as represented by the dashed
line. Source: Adapted from Ref. 259.
where DP is the sum of the pressure applied to the airway and the pressure gen-
erated by the respiratory muscles, R the airways resistance, t the elapsed time
after initiation of the inspiratory phase, e the base of the natural logarithm, and
t the product of airways resistance and respiratory system compliance (the
time constant of the respiratory system). Expiratory ow is determined by alveo-
lar driving pressure (PALV), airways resistance, the elapsed time since initiation
of exhalation, and the time constant of the respiratory system (248):
_ PALV
V (et=t )
R
Monitoring During Mechanical Ventilation 133
DV VT
CRS
DP Pplat PEEP
PIP Pplat
RI
V_ I
134 Hess
Figure 16 Pressure volume curve during mechanical ventilation. Note that the curve is
nearly linear in the normal condition. In ARDS, the curve demonstrates a lower inection
point and an upper inection point. Abbreviation: ARDS, acute respiratory distress
syndrome.
How much monitoring is needed? This is an important question for both clini-
cians and administrators. Clinicians often want to monitor everything possible,
with a more is better attitude. On the other hand, administrators and
managed care providers become justiably concerned with the costs associated
with monitoring. The presence of many monitors at the bedside can be very
136 Hess
References
7. Corwin HL, Parsonnet KC, Gettinger A. RBC transfusion in the ICU: Is there a
reason? Chest 1995; 108:767 771.
8. Peruzzi WT, Parker MA, Lichtenthal PR, et al. A clinical evaluation of a blood con-
servation device in medical intensive care unit patients. Crit Care Med 1993;
21:501 506.
9. Silver MJ, Jubran H, Stein S, et al. Evaluation of a new blood conserving arterial
line system for intensive care units. Crit Care Med 1993; 21:507 511.
10. Silver MJ, Li YH, Gragg LA, et al. Reduction of blood loss from diagnostic
sampling in critically ill patients using a blood conserving arterial line system.
Chest 1993; 104:1711 1715.
11. Peruzzi WT, Noskin GA, Moen SG, et al. Microbial contamination of blood con-
servation devices during routine use in the critical care setting: results of a prospec-
tive, randomized trial. Crit Care Med 1996; 24:1157 1162.
12. Foulke GE, Harlow DJ. Effective measures for reducing blood loss from diagnostic
laboratory tests in intensive care unit patients. Crit Care Med 1989; 17:1143 1145.
13. Beasley KE, Darin JM, Durbin CG. The effect of respiratory care department man-
agement of a blood gas analyzer on the appropriateness of arterial blood gas utiliz-
ation. Respir Care 1992; 37:343 347.
14. Pilon CS, Leathley M, London R, et al. Practice guideline for arterial blood gas
measurement in the intensive care unit decreases numbers and increases appropri-
ateness of tests. Crit Care Med 1997; 25:1308 1313.
15. Hess D, Agarwal NN. Variability of blood gases, pulse oximeter saturation, and
end-tidal carbon dioxide pressure in stable, mechanically ventilated trauma
patients. J Clin Monit 1992; 8:111 115.
16. Thorson SH, Marini JJ, Pierson DJ, Hudson LD. Variability of arterial blood gas
values in stable patients in the ICU. Chest 1983; 84:14 18.
17. Sasse SA, Chen PA, Mahutte CK. Variability of arterial blood gas values over time
in stable medical ICU patients. Chest 1994; 106:187 193.
18. Tsai YH, Lin MC, Hsieh MJ, et al. Spontaneous variability of arterial oxygenation
in critically ill mechanically ventilated patients. J Clin Monit 1999; 25:37 43.
19. Hess D, Good C, Didyoung R, et al. The validity of assessing arterial blood gases 10
minutes after an FIO2 change in mechanically ventilated patients without chronic
pulmonary disease. Respir Care 1985; 30:1037 1041.
20. Hess D, Elser RC, Agarwal NN. The effects on the pulmonary shunt value of using
measured versus calculated hemoglobin oxygen saturation and of correcting for
the presence of carboxyhemoglobin and methemoglobin. Respir Care 1984;
29:1001 1005.
21. Hess D, Maxwell C, Shefet D. Determination of intrapulmonary shunt: comparison
of an estimated shunt equation and a modied equation with the classic equation.
Respir Care 1987; 32:268 273.
22. Cohn JD, Engler PE. Shunt effect of carboxyhemoglobin. Crit Care Med 1979;
7:54 58.
23. Cane RD, Shapiro BA, Harrison RA, et al. Minimizing errors in intrapulmonary
shunt calculations. Crit Care Med 1980; 8:294 297.
24. Zijlstra WG, Buursma A, Meeuwsen-van der Roest WP. Absorption spectra of
human fetal and adult oxyhemoglobin, de-oxyhemoglobin, carboxyhemoglobin,
and methemoglobin. Clin Chem 1991; 37:1633 1638.
138 Hess
25. Mahoney JJ, Wong RJ, Vreman HJ, Stevenson DK. Fetal hemoglobin of transfused
neonates and spectrophotometric measurements of oxyhemoglobin and carboxy-
hemoglobin. J Clin Monit 1991; 7:154 160.
26. Cornelissen PJH, van Woensel CLM, van Oel WC, de Jong PA. Correction factors
for hemoglobin derivatives in fetal blood, as measured with the IL 282
CO-oximeter. Clin Chem 1983; 29:1555 1556.
27. Sehgal LR, Sehgal HL, Rosen AL, et al. Effect of intralipid on measurements of
total hemoglobin and oxyhemoglobin in whole blood. Crit Care Med 1984;
12:907 909.
28. Pitkin AD, Roberts CM, Wedzicha JA. Arterialized earlobe blood gas analysis:an
underused technique. Thorax 1994; 49:364 366.
29. Dar K, Williams T, Aitken R, et al. Arterial versus capillary sampling for analysing
blood gas pressures. Br Med J 1995; 309:24 25.
30. Sauty A, Uldry C, Debetaz LD, et al. Differences in PO2 and PCO2 between arterial
and arterialized earlobe samples. Eur Respir J 1996; 9:186 189.
31. Hughes JMB. Blood gas estimations from arterialized capillary blood versus arter-
ial puncture: are they different? Eur Respir J 1996; 9:184 185.
32. Giner J, Casan P, Belda J, et al. Pain during arterial puncture. Chest 1996;
110:1443 1445.
33. Harvey MA. Point-of-care laboratory testing in critical care. Am J Crit Care 1999;
8:72 85.
34. Castro HJ, Oropello JM, Halpern N. Point-of-care testing in the intensive care
unit. The intensive care physicians perspective. Am J Clin Pathol 1995;
104:S95 S99.
35. Kost GJ, Ehrmeyer SS, Chernow B, et al. The laboratory-clinical interface. Point-
of-care testing. Chest 1999; 115:1140 1154.
36. Smith BL, Vender JS. Point-of-care testing. Respir Care Clin NA 1995; 1:133 141.
37. Shapiro BA. Point-of-care blood testing and cardiac output measurement in the
intensive care unit. New Horizons 1999; 7:244 252.
38. Zaloga GP, Hill TR, Strickland RA, et al. Bedside blood gas and electrolyte
monitoring in critically ill patients. Crit Care Med 1989; 17:920 925.
39. Zaloga GP, Dudas L, Roberts P, et al. Near-patient blood gas and electrolyte
analyses are accurate when performed by non-laboratory trained individuals.
J Clin Monit 1993; 9:341 346.
40. Hicks JM. Near patient testing: is it here to stay? J Clin Pathol 1996; 49:191 193.
41. Zaloga GP, Roberts PR, Black K, et al. Hand-held blood gas analyzer is accurate in
the critical care setting. Crit Care Med 1996; 24:957 962.
42. MacIntyre NR, Lawlor B, Carsten D, Yetsko D. Accuracy and precision of a
point-or-care blood gas analyzer incorporating optodes. Respir Care 1996;
41:800 804.
43. Wahr JA, Lau W, Tremper KK, et al. Accuracy and precision of a new, portable,
handheld blood gas analyzer, the IRMA. J Clin Monit 1996; 12:317 324.
44. Shelledy DC, Smith WA. A comparison of the I-STAT system and Corning 278 for
the measurement of arterial blood gas values. Respir Care 1997; 42:693 697.
45. Murthy JN, Hicks JM, Soldin SJ. Evaluation of the I-STAT portable clinical
analyzer in a neonatal and pediatric intensive care unit. Clin Biochem 1997;
30:385 389.
Monitoring During Mechanical Ventilation 139
46. Kendall J, Reeves B, Clancy M. Point of care testing: randomised controlled trial of
clincial outcome. Br Med J 1998; 316:1052 1057.
47. Shapiro BA. In-vivo monitoring of arterial blood gases and pH. Respir Care 1992;
37:165 169.
48. Peruzzi WT, Shapiro BA. Blood gas monitors. Respir Care Clin NA 1995;
1:143 156.
49. Shapiro BA. Clinical and economic performance criteria for intraarterial and extra-
arterial blood gas monitors, with comparison with in vitro testing. Am J Clin Pathol
1995; 104:S100 S106.
50. Shapiro BA, Cane RD, Chomka CM, Bandala LE, Peruzzi WT. Preliminary evalu-
ation of an intra-arterial blood gas system in dogs and humans. Crit Care Med 1989;
17:455 460.
51. Zimmerman J, Dellinger RP. Initial evaluation of a new intra-arterial blood gas
system in humans. Crit Care Med 1993; 21:495 500.
52. Shapiro BA, Mahutte CK, Cane RD, et al. Clinical performance of a blood gas
monitor. A prospective, multicenter trial. Crit Care Med 1993; 21:497 494.
53. Mahutte CK, Sasse SA, Chen PA, et al. Performance of a patient-dedicated, on-
demand blood gas monitor in medical ICU patients. Am J Respir Crit Care Med
1994; 150:865 869.
54. Haller M, Kilger E, Breigel J, et al. Continuous intra-arterial blood gas and pH
monitoring in critically ill patients with severe respiratory failure: a prospective,
criterion standard study. Crit Care Med 1994; 22:580 586.
55. Venkatesh B, Clutton TH, Hendry SP. A multiparameter sensor for continous intra-
arterail blood gas monitoring: a prospective evaluation. Crit Care Med 1994;
22:588 594.
56. Larson CP, Vender J, Leiver A. Multisite evaluation of a continuous intraarterial
blood gas monitoring system. Anesthesiology 1994; 81:543 552.
57. Venkatesh B, Clutton-Brock TH, Hendry SP. Evaluation of the Paratrend 7 intra-
vascular blood gas monitor during cardiac surgery: comparison with the C4000
in-line blood gas monitor during cardiopulmonary bypass. J Cardiovasc Vasc
Anesth 1995; 9:412 419.
58. Kurahashi K, Hirose Y, Yamada H, et al. Intra-arterial blood gas moni-
toring system: more accurate values can be obtained. J Clin Monit 1996;
12:141 147.
59. Franklin ML, Peruzzi WT, Moen SG, Shapiro BA. Evaluation of an on-demand,
ex vivo bedside blood gas monitor on pulmonary artery blood gas determination.
Anesth Analg 1996; 83:500 504.
60. Weiss IK, Fink S, Edmunds S, et al. Continuous arterial gas monitoring: initial
experience with the Paratrend 7 in children. Intensive Care Med 1996; 22:1414
1417.
61. Abraham E, Gallagher TJ, Fink S. Clinical evaluation of a multiparameter intra-
arterial blood-gas sensor. Intensive Care Med 1996; 22:507 513.
62. McKinnley BA, Parmley CL. Clinical trial of an ex vivo arterial blood gas monitor.
J Crit Care 1998; 13:190 197.
63. Bailey PL, McJames SW, Cluff NL, et al. Evaluation in volunteers of the Via V-
ABG automated bedside blood gas, chemistry, and hematocrit monitor. J Clin
Monit Comput 1998; 14:339 346.
140 Hess
64. Weiss IK, Fink S, Harrison R, et al. Clinical use of continuous arterial
blood gas monitoring in the pediatric intensive care unit. Pediatrics 1999;
103:440 445.
65. Suter PM, Lindauer JM, Fairley HB, Schlobohm RM. Errors in data derived from
pulmonary artery blood gas values. Crit Care Med 1975; 3:175 181.
66. Mihm F, Feeley TW, Rosenthal M, Rafn TA. The lack of effect of variable blood
withdrawal rates on the measurement of mixed venous oxygen saturation. Chest
1980; 78:452 455.
67. Nelson RD, Wilkins RL, Jacobsen WK, et al. Supranormal PvO2 in the presence of
tissue hypoxia: A case report. Respir Care 1983; 28:191 194.
68. Nowak RM, Martin GB, Carden DL, et al. Selective venous hypercarbia during
human CPR: implications regarding blood ow. Ann Emerg Med 1987; 16:527
530.
69. Weil MH, Rakow EC, Trevion R, et al. Difference in acid-base state between
venous and arterial blood during cardiopulmonary resuscitation. N Engl J Med
1986; 1986:153 156.
70. Tahvanainen J, Meretoja O, Nikki P. Can central venous blood replace mixed
venous blood samples? Crit Care Med 1982; 10:758 761.
71. Paulus DA. Invasive monitoring of respiratory gas exchange: continuous measure-
ment of mixed venous oxygen saturation. Respir Care 1987; 32:535 543.
72. Baraka A, Baroody M, Haroun S, et al. Continuous venous oximetry during cardi-
opulmonary bypass: inuence of temperature changes, perfusion ow, and hemato-
crit levels. J Cardiothorac Anesth 1990; 4:35 38.
73. Cariou A, Monchi M, Dhainaut J. Continuous cardiac output and mixed venous
oxygen saturation monitoring. J Crit Care 1998; 13:198 213.
74. Rouby JJ, Poete R, Bodin L, et al. Three mixed venous saturation catheters in
patients with circulatory shock and respiratory failure. Chest 1990; 98:954 958.
75. Gettinger A, DeTraglia MC, Glass DD. In vivo comparison of two mixed venous
saturation catheters. Anesthesiology 1987; 66:373 375.
76. Karis JH, Lumb PD. Clinical evaluation of the Edwards Laboratories and Oximetrix
mixed venous oxygen saturation catheters. J Cardiothorac Anesth 1988;
2:440 444.
77. Reinhart K, Moser N, Rudolph T, et al. Accuracy of two mixed venous saturation
catheters during long-term use in critically ill patients. Anesthesiology 1988;
69:769 773.
78. Hecker BR, Brown DL, Wilson D. A comparison of two pulmonary artery mixed
venous oxygen saturation catheters during the changing conditions of cardiac
surgery. J Cardiothorac Anesth 1989; 3:269 275.
79. VanWoerkens ECSM, Trouwborst A, Tenbrinck R. Accuracy of mixed venous sat-
uration catheter during acutely induced changes in hematocrit in humans. Crit Care
Med 1991; 19:1025 1029.
80. Scuderi PE, Bowton DL, Meredith JW, et al. A comparison of three pulmonary
artery oximetry catheters in intensive care unit patients. Chest 1992; 102:896 905.
81. Scuderi PE, MacGregor DA, Bowton DL, et al. A laboratory comparison of three
pulmonary artery oximetry catheters. Anesthesiology 1994; 81:245 253.
82. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic therapy
in critically ill patients. N Engl J Med 1995; 333:1025 1032.
Monitoring During Mechanical Ventilation 141
105. Scott DM, Cooper MG. Spurious pulse oximetry with intrauterine methylene blue
injection. Anaesth Intens Care 1991; 19:267 284.
106. Scheller MS, Unger RJ, Kelner MJ. Effects of intravenously administered dyes on
pulse oximeter readings. Anesthesiology 1986; 65:550 552.
107. Cote CJ, Goldstein EA, Fuchsman WH, et al. The effect of nail polish on pulse oxi-
metry. Anesth Analg 1989; 67:683 686.
108. Veyckemans F, Baele P, Guillaume JE, et al. Hyperbilirubinemia does not interfere
with hemoglobin saturation measured by pulse oximetry. Anesthesiology 1989;
70:118 122.
109. Chelluri L, Snyder JV, Bird JR. Accuracy of pulse oximetry in patients with hyper-
bilirubinemia. Respir Care 1991; 36:1383 1386.
110. Zeballos RJ, Wiesman IM. Reliability of noninvasive oximetry in black subjects
during exercise and hypoxia. Am Rev Respir Dis 1991; 144:1240 1244.
111. Jubran A, Tobin MJ. Reliability of pulse oximetry in titrating supplemental oxygen
therapy in ventilator-dependent patients. Chest 1990; 97:1420 1425.
112. Cahan C, Decker MJ, Hoekje PL, et al. Agreement between noninvasive oximetric
values for oxygen saturation. Chest 1990; 97:814 819.
113. Severinghaus JW, Koh SO. Effect of anemia on pulse oximeter accuracy at low
saturation. J Clin Monit 1990; 6:85 88.
114. Lee S, Tremper KK, Barker SJ. Effects of anemia on pulse oximetry and continuous
mixed venous hemoglobin saturation monitoring in dogs. Anesthesiology 1991;
75:118 122.
115. Fluck RR, Schroeder C, Frani G, et al. Does ambient light affect the accuracy of
pulse oximetry? Respir Care 2003; 48:677 680.
116. Sami HM, Kleinman BS, Lonchyna VA. Central venous pulsations associated with
a falsely low oxygen saturation measured by pulse oximetry. J Clin Monit 1991;
7:309 312.
117. Wilkins CJ, Moores M, Hanning CD. Comparison of pulse oximeters:
effects of vasoconstriction and venous engorgement. Br J Anaesth 1989; 62:439
444.
118. Stewart KG, Rowbottom SJ. Inaccuracy of pulse oximetry in patients with severe
tricuspid regurgitation. Anaesthesia 1991; 46:668 670.
119. Chemello PD, Nelson SR, Wolford LM. Finger injury resulting from pulse
oximeter probe during orthognathic surgery. Oral Surg Oral Med Oral Pathol
1990; 69:161 163.
120. Murphy KG, Secunda JA, Rockoff MA. Severe burns from a pulse oximeter.
Anesthesiology 1990; 73:350 352.
121. Stogner SW, Owens MW, Baethge BA. Cutaneous necrosis and pulse oximetry.
Cutis 1191; 48:235 137.
122. Bashein G, Syrovy G. Burns associated with pulse oximetry during magnetic
resonance imaging. Anesthesiology 1991; 75:382 383.
123. Plummer JL, Ilsley AH, Fronsko RRL, et al. Identication of movement
artefact by the Nellcor N-200 and N-3000 pulse oximeters. J Clin Monit 1997;
13:109 113.
124. Gehring H, Nornberger C, Matz H, et al. The effects of motion artifact and low per-
fusion on the performance of a new generation of pulse oximeters in volunteers
undergoing hypoxemia. Respir Care 2002; 47:48 60.
Monitoring During Mechanical Ventilation 143
125. Goldman JM, Petterson MT, Kopotic RJ, et al. Masimo signal extraction pulse oxi-
metry. J Clin Monit 2000; 16:475 483.
126. Barker SJ. Motion-Resistant pulse oximetry: a comparison of new and old
models. Anesth Analg 2002; 95:967 972.
127. Durbin CG, Rostow SK. More reliable oximetry reduces the frequency of arterial
blood gas analyses and hastens oxygen weaning after cardiac surgery: a prospec-
tive, randomized trial of the clinical impact of a new technology. Crit Care Med
2002; 30:1734 1740.
128. Moller JT, Pedersen T, Rasmussen LS, et al. Randomized evaluation of pulse oxi-
metry in 20,802 patients. I. Design, demography, pulse oximetry failure rate, and
overall complication rate. Anesthesiology 1993; 78:436 444.
129. Moller JT, Johannessen NW, Espersen K, et al. Randomized evaluation of pulse
oximetry in 20,802 patients. II. Perioperative events and postoperative compli-
cations. Anesthesiology 1993; 78:445 453.
130. Hartert TV, Wheeler AP, Sheller JR. Use of pulse oximetry to recognize severity of
airow obstruction in obstructive airway disease: correlation with pulsus para-
doxus. Chest 1999; 115:475 481.
131. Gys T, Hubens A, Neels H, et al. Prognostic value of gastric intramucosal pH in
surgical intensive care patients. Crit Care Med 1988; 16:1222 1224.
132. Landow L, Phillips DA, Heard SO, et al. Gastric tonometry and venous oximetry in
cardiac surgery patients. Crit Care Med 1991; 19:1226 1233.
133. Hartmann M, Montgomery A, Jonsson K, et al. Tissue oxygenation in hemorrhagic
shock measured as transcutaneous oxgyen tension, subcutaneous oxygen
tension, and gastrointestinal intramucosal pH in pigs. Crit Care Med 1991;
19:205 210.
134. Silverman HJ, Tuma P. Gastric tonometry in patients with sepsis. Effects of dobu-
tamine infusions and packed red blood cell transfusions. Crit Care Med 1192;
102:184 188.
135. Gutierrez G, Bismar H, Danzker DR, et al. Comparison of gastric intramucosal pH
with measures of oxygen transport and consumption in critically ill patients. Crit
Care Med 1992; 20:451 457.
136. Gutierrez G, Palizas F, Doglio G, et al. Gastric intramucosal pH as a therapeutic
index of tissue oxygenation in critically ill patients. Lancet 1992; 339:195 199.
137. Roumen RMH, Vreugde JPC, Goris RJA. Gastric tonometry in multiple patient
patients. J Trauma 1994; 36:313 316.
138. Mohsenifar Z, Hay A, Hay J, et al. Gastric intramucosal pH as a predictor of success
or failure in weaning patients from mechanical ventilation. Ann Intern Med 1993;
119:794 798.
139. Marik PE. Gastric intramucosal pH. A better predictor of multiorgan dysfunction
syndrome and death than oxygen-derived variables in patients with sepsis. Chest
1993; 104:225 229.
140. Knichwitz G, Van Aken H, Brussel T. Gastrointestinal monitoring using measure-
ment of intramucosal PCO2. Anesth Analg 1998; 87:134 142.
141. Keenan SP, Guyatt GH, Sibbald WJ, et al. How to use articles about diagnostic
technology: Gastric tonometry. Crit Care Med 1999; 27:1726 1731.
142. Fiddian-Green RG. Associations between intramucosal acidosis in the gut and
organ failure. Crit Care Med 1993; 21:S103 S107.
144 Hess
180. Poppius H, Korhonen O, Viljanen AA, et al. Arterial to end-tidal CO2 difference in
respiratory disease. Scand J Resp Dis 1975; 56:254 262.
181. Raemer DB, Francis D, Philip JH, et al. Variation in PCO2 between arterial blood
and peak expired gas during anesthesia. Anesth Analg 1983; 62:1065 1069.
182. Russell GB, Graybeal JM, Strout JC. Stability of arterial to end-tidal carbon dioxide
gradients during postoperative cardiorespiratory support. Can J Anaesth 1990;
37:560 566.
183. Russell GB, Graybeal JM. Reliability of the arterial to end-tidal carbon dioxide gra-
dient in mechanically ventilated patients with multisystem trauma. J Trauma 1994;
36:317 322.
184. Russell GB, Graybeal JM. The arterial to end-tidal carbon dioxide difference in
neurosurgical patients during craniotomy. Anesth Analg 1995; 81:806 810.
185. Yamanaka MK, Sue DY. Comparison of arterial-end-tidal PCO2 difference and
dead space/tidal volume ratio in respiratory failure. Chest 1987; 92:832 835.
186. Piiper J. Blood-gas equilibrium of carbon dioxide in the lungs: a continuing contro-
versy. J Appl Physiol 1986; 60:1 8.
187. Tulou PP, Walsh PW. Measurement of alveolar carbon dioxide tension at maximal
expiration as an estimate of arterial carbon dioxide tension in patients with airway
obstruction. Am Rev Respir Dis 1970; 102:921 926.
188. Bhavani-Shankar K, Kumar AY, Moseley HS, et al. Terminology and the
current limitations of time capnography: a brief review. J Clin Monit 1995;
11:175 82.
189. Jones NL, Robertson DG, Kane JW. Difference between end-tidal and arterial
PCO2 in exercise. J Appl Physiol 1979; 47:954 960.
190. Birmingham PK, Cheney FW, Ward RJ. Esophageal intubation: a review of detec-
tion of techniques. Anesth Analg 1986; 65:886 891.
191. Guggenberger H, Lenz G, Federle R. Early detection of inadvertent oesophageal
intubation: pulse oximetry vs. capnography. Acta Anaesthesiol Scand 1989;
33:112 115.
192. Holland R, Webb RK, Runciman WB. The Australian Incident Monitoring Study.
Oesophageal intubation: an analysis of 2000 incident reports. Anaesth Intensive
Care 1993; 21:608 610.
193. Linko K, Paloheimo M, Tammisto T. Capnography for detection of accidental
oesophageal intubation. Acta Anaesthesiol Scand 1983; 27:199 202.
194. Murray IP, Modell JH. Early detection of endotracheal tube accidents by monitor-
ing carbon dioxide concentration in respiratory care. Anesthesiology 1983;
59:344 346.
195. Szekely SM, Webb RK, Williamson JA, et al. Problems related to the endo-
tracheal tube: an analysis of 2000 incident reports. Anaesth Intens Care 1993;
21:611 616.
196. Vaghadia H, Jenkins LC, Ford RW. Comparison of end-tidal carbon dioxide,
oxygen saturation and clinical signs for the detection of oesophageal intubation.
Can J Anaesth 1989; 36:560 564.
197. Garnett AR, Gervin CA, Gervin AS. Capnographic waveforms in esophageal
intubation: effect of carbonated beverages. Ann Emerg Med 1989; 18:387 390.
198. Araujo-Preza CE, Melhado ME, Gutierrez FJ, et al. Use of capnometry to verify
feeding tube placement. Crit Care Med 2002; 30:2255 2259.
Monitoring During Mechanical Ventilation 147
199. Bosman RJ, Stoutenbeek CP, Zandstra DF. Non-invasive pulmonary blood ow
measurement by means of CO2 analysis of expiratory gases. Intensive Care Med
1991; 17:98 102.
200. Isserles SA, Breen PH. Can changes in end-tidal PCO2 measure changes in cardiac
output? Anesth Analg 1991; 73:808 814.
201. Ornato JP, Garnett AR, Glauser FL. Relationship between cardiac output and the
end-tidal carbon dioxide tension. Ann Emerg Med 1990; 19:1104 1106.
202. Weil MH, Bisera J, Trevino RP, et al. Cardiac output and end-tidal carbon dioxide.
Crit Care Med 1985; 13:907 909.
203. Sanders AB, Atlas M, Ewy GA, et al. Expired PCO2 as an index of coronary per-
fusion pressure. Am J Emerg Med 1985; 3:147 149.
204. Lewis LM, Stothert J, Standeven J, et al. Correlation of end-tidal CO2 to cerebral
perfusion during CPR. Ann Emerg Med 1992; 21:1131 1134.
205. Kalenda Z. The capnogram as a guide to the efcacy of cardiac massage. Resusci-
tation 1978; 6:259 263.
206. Garnett AR, Ornato JP, Gonzalez ER, Johnson EB. End-tidal carbon dioxide
monitoring during cardiopulmonary resuscitation. J Am Med Assoc 1987;
257:512 515.
207. Falk JL, Rackow EC, Weil MH. End-tidal carbon dioxide concentration during
cardiopulmonary resuscitation. N Engl J Med 1988; 318:607 611.
208. Sanders AB, Kern KB, Otto CW, et al. End-tidal carbon dioxide monitoring during
cardiopulmonary resuscitation. A prognostic indicator for survival. J Am Med
Assoc 1989; 262:1347 1351.
209. Levine RL, Wayne MA, Miller CC. End-tidal carbon dioxide and outcome of out-
of-hospital cardiac arrest. N Engl J Med 1997; 337:301 306.
210. Grmec S, Lah K, Tusek-Bunc K. Difference in end-tidal CO2 between asphyxia
cardiac arrest and ventricular brillation/pulseless ventricular tachycardia
cardiac arrest in the prehospital setting. Crit Care 2003; 7:R139 R144.
211. Julian DG, Travis DM, Robin ED, et al. Effect of pulmonary artery occlusion upon
end-tidal CO2 tension. J Appl Physiol 1960; 15:87 91.
212. Bermudez J, Lichtiger M. Increases in arterial to end-tidal CO2 tension differences
after cardiopulmonary bypass. Anesth Analg 1987; 66:690 692.
213. Poppius H, Korhonen O, Viljanen AA, et al. Arterial to end-tidal CO2 difference in
respiratory disease. Scand J Resp Dis 1975; 56:254 262.
214. Hatle L, Rokeseth R. The arterial to end-expiratory carbon dioxide tension gradient
in acute pulmonary embolism and other cardiopulmonary diseases. Chest 1974;
66:352 357.
215. Nutter DO, Massumi RA. The arterial-alveolar carbon dioxide tension gradient in
diagnosis of pulmonary embolus. Dis Chest 1966; 50:380 387.
216. Eriksson L, Wollmer P, Olsson C, et al. Diagnosis of pulmonary embolism based
upon alveolar dead space analysis. Chest 1989; 96:357 362.
217. Chopin C, Fesard P, Mangalaboyi J, et al. Use of capnography in diagnosis of pul-
monary embolism during acute respiratory failure of chronic obstructive pulmonary
disease. Crit Care Med 1990; 18:353 357.
218. Verschuren F, Liistro G, Coffeng R, et al. Volumetric capnography as a scree-
ning test for pulmonary embolism in the emergency department. Chest 2004;
125:841 850.
148 Hess
219. Healey CJ, Fedullo AJ, Swinburne AJ, Wahl GW. Comparison of noninvasive
measurements of carbon dioxide tension during withdrawal from mechanical ven-
tilation. Crit Care Med 1987; 15:764 767.
220. Weinger MB, Brimm JE. End-tidal carbon dioxide as a measure of arterial carbon
dioxide during intermittent mandatory ventilation. J Clin Monit 1987; 3:73 79.
221. Niehoff J, DelGuercio C, LaMorte W, et al. Efcacy of pulse oximetry and capno-
metry in postoperative ventilatory weaning. Crit Care Med 1988; 16:701 705.
222. Smith RA, Novak RA, Venus B. End-tidal CO2 monitoring utility during weaning
from mechanical ventilation. Respir Care 1989; 34:972 975.
223. Thrush DN, Mentis SW, Downs JB. Weaning with end-tidal CO2 and pulse oxime-
try. J Clin Anesth 1991; 3:456 460.
224. Withington DE, Ramsay JG, Saoud AT, et al. Weaning from ventilation after car-
diopulmonary bypass:evaluation of a non-invasive technique. Can J Anaesth 1991;
38:15 19.
225. Hess D, Schlottag A, Levin B, et al. An evaluation of the usefulness of end-tidal
PCO2 to aid weaning from mechanical ventilation following cardiac surgery.
Respir Care 1991; 36:837 843.
226. Morley TF, Giaimo J, Maroszan E, et al. Use of capnography for assessment of
the adequacy of alveolar ventilation during weaning from mechanical ventilation.
Am Rev Respir Dis 1993; 148:339 344.
227. Saura P, Blanch L, Lucangelo U, et al. Use of capnometry to detect hypercapnic
episodes during weaning from mechanical ventilation. Intensive Care Med 1966;
22:374 381.
228. Dasgupta A, Rice R, Mascha E, et al. Four-year experience with a unit for long-term
ventilation (respiratory special care unit) at the Cleveland Clinic Foundation. Chest
1999; 116:447 455.
229. Mackersie RC, Karagianes TG. Use of end-tidal carbon dioxide tension for
monitoring induced hypocapnia in head-injured patients. Crit Care Med 1990;
18:764 765.
230. Sharma SK, McGuire GP, Cruise CJ. Stability of the arterial to end-tidal carbon
dioxide difference during anaesthesia for prolonged neurosurgical procedures.
Can J Anaesth 1995; 42:498 503.
231. Isert P. Control of carbon dioxide levels during neuroanaesthesia: current practice
and an appraisal of our reliance upon capnography. Anaesth Intensive Care 1994;
22:435 441.
232. Christensen MA, Bloom J, Sutton KR. Comparing arterial and end-tidal carbon
dioxide values in hyperventilated neurosurgical patients. Am J Crit Care 1995;
4:116 121.
233. Kerr ME, Zwmpsky J, Sereika S, et al. Relationship between arterial carbon dioxide
and end-tidal carbon dioxide in mechanically ventilated adults with severe head
trauma. Crit Care Med 1996; 24:785 790.
234. Palmon SC, Liu M, Moore LE, et al. Capnography facilitates tight control of ven-
tilation during transport. Crit Care Med 1996; 24:608 611.
235. Helm M, Schuster R, Hauke J, et al. Tight control of prehospital ventilation by cap-
nography in major trauma patients. Br J Anaesth 2003; 90:327 332.
236. Tachibana K, Imanaka H, Takeuchi M, et al. Noninvasive cardiac output measure-
ment using partial carbon dioxide rebreathing is less accurate at settings of reduced
Monitoring During Mechanical Ventilation 149
254. Lessard MR, Lofaso F, Brochard L. Expiratory muscle activity increases intrinsic
positive end-expiratory pressure independently of dynamic hyperination in
mechanically ventilated patients. Am J Respir Crit Care Med 1995; 151:562 569.
255. Ninane V, Yerault J, de Troyer A. Intrinsic PEEP in patients with chronic obstruc-
tive pulmonary disease. Role of expiratory muscles. Am Rev Respir Dis 1993;
148:1037 1042.
256. Leatherman JW, Ravenscraft SA. Low measured auto-positive end-expiratory
pressure during mechanical ventilation of patients with severe asthma: hidden
auto-positive end-expiratory pressure. Crit Care Med 1996; 24:541 546.
257. Truwit JD, Marini JJ. Evaluation of thoracic mechanics in the ventilated patient.
Part II: applied mechanics. J Crit Care 1988; 3:199 213.
258. Chieveley-Williams S, Dinner L, Puddicombe A, et al. Central venous and bladder
pressure reect transdiaphragmatic pressure during pressure support ventilation.
Chest 2002; 121:533 538.
259. Parthasarathy S, Jubran A, Tobin MJ. Cycling of inspiratory and expiratory muscle
groups with the ventilator in airow limitation. Am J Respir Crit Care Med 1998;
158:1471 1478.
260. Jubran A, Van de Graaff WB, Tobin MJ. Variability of patient-ventilator interaction
with pressure support ventilation in patients with chronic obstructive pulmonary
disease. Am J Respir Crit Care Med 1995; 152:129 136.
261. Marini JJ, Crooke PS, Truwit JD. Determinants and limits of pressure-preset
ventilation: a mathematical model. J Appl Physiol 1989; 67:1081 1092.
262. Burke WC, Crooke PS, Marcy TW, et al. Comparison of mathematical and mech-
anical models of pressure-controlled ventilation. J Appl Physiol 1993; 74:922 933.
263. Kress JP, OConnor MF, Schmidt GA. Clinical examination reliably detects intrin-
sic positive end-expiratory pressure in critically ill, mechanically ventilated
patients. Am J Respir Crit Care Med 1999; 159:290 294.
264. Hess D, Tabor T. Comparison of six methods to calculate airway resistance during
mechanical ventilation. J Clin Monit 1993; 9:275 282.
265. Marini JJ, Ravenscraft SA. Mean airway pressure: physiologic determinants and
clinical importance. Part 1: physiologic determinants and measurements. Crit
Care Med 1992; 20:1461 1472.
266. Marini JJ, Ravenscraft SA. Mean airway pressure: physiologic determinants and
clinical importance. Part 2: clinical implications. Crit Care Med 1992;
20:1604 1616.
267. Primiano FP, Chatburn RL, Lough MD. Mean airway pressure: theoretical con-
siderations. Crit Care Med 1982; 10:378 383.
268. Brochard L. Respiratory pressure-volume curves. In: Tobin MJ, ed. Principles and
Practice of Intensive Care Monitoring. New York: McGraw-Hill, 1998.
269. Lu Q, Vieira SRR, Richecoeur J, et al. A simple automated method for measuring
pressure volume curves during mechanical ventilation. Am J Respir Crit Care Med
1999; 159:275 282.
270. Servillo G, Svantesson C, Beydon L, et al. Pressure volume curves in acute respir-
atory failure: automated low ow ination versus occlusion. Am J Respir Crit Care
Med 1997; 155:1629 1656.
271. Bigatello LM, Hurford WE, Pesenti A. Ventilatory management of severe acute
respiratory failure for Y2K. Anesthesiology 1999; 91:1567 70.
Monitoring During Mechanical Ventilation 151
272. OKeefe GE, Gentilello LM, Erford S, et al. Imprecision in lower inection point
estimation from static pressure-volume curves in patients at risk for acute respirat-
ory distress syndrome. J Trauma 1998; 44:1064 1068.
273. Harris RS, Hess DR, Venegas JG. An objective analysis of the pressure volume
curve in the acute respiratory distress syndrome. Am J Respir Crit Care Med.
2000; 161:432 439.
274. Venegas JG, Harris RS, Simon BA. A comprehensive equation for pulmonary
pressure volume curve. J Appl Physiol 1998; 84:389 395.
275. Ranieri VM, Brienza N, Santostasi S, et al. Impairments of lung and chest wall
mechanics in patients with acute respiratory distress syndrome. Am J Respir Crit
Care Med 1997; 156:1082 1091.
276. Mergoni M, Martelli A, Volpi A, et al. Impact of positive end-expiratory pressure
on chest wall and lung pressure volume curve in acute respiratory failure. Am J
Respir Crit Care Med 1997; 156:846 854.
277. Hickling KG. The pressure-volume curve is greatly modied by recruitment. Am J
Respir Crit Care Med 1998; 158:194 202.
278. Hickling KG. Best compliance during a decremental, but not incremental, positive
end-expiratory pressure trial is related to open-lung positive end-expiratory
pressure: a mathematical model of acute respiratory distress syndrome lungs.
Am J Respir Crit Care Med 2001; 163:69 78.
279. Morrison WE, Haas EC, Shaffner DH, et al. Noise, stress, and annoyance in a pedi-
atric intensive care unit. Crit Care Med 2003; 31:113 119.
280. Chambrin MC. Alarms in the intensive care unit: how can the number of false
alarms be reduced? Critical Care 2001; 5:184 188.
281. Lawless ST. Crying wolf: false alarms in a pediatric intensive care unit. Crit Care
Med 1994; 22:981 985.
282. Tsein CL, Fackler JC. Poor prognosis for existing monitors in the intensive care
unit. Crit Care Med 1997; 25:614 619.
283. Bentt LR, Santora TA, Laverle BJ, et al. Accuracy and utility of pulse oximetry in
the surgical intensive care unit. Curr Surg 1990; 47:267 268.
284. Kestin IG, Miller BR, Lockhart CH. Auditory alarms during anesthesia monitoring.
Anesthesiology 1988; 69:106 109.
285. Kellerman AL, Cofer CA, Joseph S, et al. Impact of portable pulse oximetry on
arterial blood gas test ordering in an urban emergency department. Ann Emerg
Med 1991; 20:130 134.
286. Orlando R. Continuous mixed venous oximetry in critically ill patients. High-tech
cost effectiveness. Arch Surg 1986; 121:470 471.
287. Pearson KS, Gomez MN, Moyers JR, et al. A cost/benet analysis of randomized
invasive monitoring for patients undergoing cardiac surgery. Anesth Analg 1989;
69:336 341.
288. Jastremski MS, Chelluri L, Beney KM, et al. Analysis of the effects of continuous
on-line monitoring of mixed venous oxygen saturation on patient outcome and cost
effectiveness. Crit Care Med 1990; 17:148 153.
This page intentionally left blank
5
Weaning from Mechanical Ventilation
I. Introduction
153
154 Hackner, Dass, and Lewis
for Weaning from Prolonged Mechanical Ventilation. The fth section encom-
passes topics of comfort, analgesia, sedation, and delirium during weaning,
which have great impact on duration of ventilation and patient satisfaction.
The sixth and nal section, Goals of Therapy: Conversations about Ventilation,
provides a bioethical rubric for initiation or terminating ventilation and weaning.
Figure 1 Relationship between the tension-time index of the diaphragm, predicted time
to task failure, and duration of trial of spontaneous breathing in weaning failure patients.
Note progressive increment in tension-time index of the diaphragm as the weaning trial
continued. Overt diaphragm contractile fatigue would likely have ensued if the trial had
continued for another 13 minutes. Source: Adapted from Refs. 14,15.
The publication of over 400 putative weaning predictors has given way to
increasing skepticism regarding the clinical utility of such parameters (28).
Indeed, the purported benets of weaning parameters have recently been high-
lighted as a myth in an editorial entitled Weaning the patient or weaning
old-fashioned ideas(29). Over 30 years of publications in this area have not
yielded strongly reproducible results. Furthermore, the study groups have been
small, and the data have generated unhelpful likelihood ratios (LR) and low pre-
dictive power (28). It has been suggested that physicians a priori clinical prob-
abilities of success of weaning may affect the selection of patients and in turn the
accuracy of the tests (28,29).
Weaning from Mechanical Ventilation 161
A. Weaning Parameters
Traditional weaning parameters have entailed indications of gas exchange and
indices of chest mechanics or breathing pattern. Gas exchange criteria have
spanned dead space calculations to hypoxemia ratios. In general, the minimal
acceptable PaO2 is 60 mmHg on a FIO2 of 0.4 and a PEEP less than 5 cmH2O.
Other parameters of oxygenation that can be used include PaO2/PAO2
(.0.35), A2a O2 (,350 mmHg), or PaO2/FIO2 (.200) (30). Although ade-
quate arterial oxygenation is essential to initiate weaning, the predictive value
of these indices for the outcome of the weaning trial is usually low (31). Strict
adherence to these criteria should not preclude liberating a patient from mechan-
ical ventilation, if deemed clinically appropriate.
Historically, indices of chest mechanics and respiratory pattern during
spontaneous breathing have included vital capacity (VC) greater than
10 15 mL/kg, spontaneous tidal volume (VT . 5 mL/kg), respiratory rate
(RR , 35/ min), minute ventilation (VE , 10 15 L/min), maximum voluntary
ventilation (MVV) at least twice VE, and maximal inspiratory mouth pressure
(PImax using, 215 to 230 cmH2O) (28,32,33). VC reects not only lung and
chest wall mechanics, but also respiratory muscle strength, albeit a non-specic
marker particularly of the latter (see chapter 7, vol. 2 on neuromuscular respirat-
ory failure). Pooled data for RR , 38 breaths/min has been analyzed with a LR
of 1.1, indicating unchanged post-test probability and thus an unhelpful measure
(28,34,35). By contrast, a RR . 38 breaths/min yielded a LR of 0.32, indicating
a somewhat reduced probability of weaning success (28,34,35). A high VE may
reect an increased dead space, load, and work of breathing. PImax is a measure
of global inspiratory muscle strength and is limited by the need for patient
cooperation, coordination, inuences of lung volume, and wide inter- and
intra-subject variability (see chapter 7, vol. 2 on neuromuscular respiratory
failure). While a one-way valve has been utilized to circumvent some of these
hindrances, its use has been limited (36). (This valve permits only expiration,
while the most negative pressure generated by inspiration against an occluded
airway over a 20-second period is recorded as PImax.) Table 2 depicts sensitivity
and specicity for several of these traditional indices.
Newer studies have attempted to improve the robustness of weaning tests
by either grouping variables or by seeking more specic physiological correlates.
Several integrative weaning indices and unique parameters will be briey
discussed. Yang and Tobin developed the commonly employed rapid shallow
breathing index (RSBI) after discovering that patients who failed weaning devel-
oped tachypnea and decreasing VT almost immediately upon discontinuation of
ventilatory support (37). The RSBI was measured after one minute as the ratio
of frequency to VT while breathing spontaneously (33). Using 105 (breaths/
min)/L as the threshold, the sensitivity was 97% for predicting weaning
success, whereas the specicity was 64%, indicating that 36% with an
RSBI , 105 failed the weaning trial (i.e., false-positives) (33). While pooled
162 Hackner, Dass, and Lewis
data from six studies yielded similar sensitivities for RSBI, with a range of 87
97%, specicities were generally lower (mean: 47%; range: 33 60%) (28). Fur-
thermore, sensitivity for RSBI decreased with increased duration of mechanical
ventilation (33), whereas specicity was inuenced by disease state (38). For
example, specicity of RSBI in patients with COPD was 65%, while in patients
with neurologic disease or acute respiratory failure of varying etiology, the speci-
city was 26% and 28%, respectively (38). An analysis of extubation failure by
Epstein (39) showed that when weaning failed, despite an RSBI ,100 (i.e., false-
positive test), the mechanism was secondary to other comorbid processes and not
the underlying respiratory disease alone. RSBI has been reported to be higher in
females and patients with smaller endotracheal tubes, suggesting that the index
threshold needs appropriate adjustment to specic patient populations and con-
ditions (40). A threshold of 130 has been suggested as a more appropriate cut
point in elderly patients (41). Krieger and Isber suggested that stability of the
RSBI when measured serially from the onset of the weaning trial and for the sub-
sequent two to three hours gains sensitivity and specicity.
A number of integrative indices have been reported. The compliance, rate,
oxygen, and pressure (CROP) index incorporates gas exchange, chest wall mech-
anics, and respiratory strength.
CROP index [Dynamic compliance PImax (PaO2/PAO2)]/RR (33).
This index had a lower sensitivity and specicity than the RSBI (Table 2). A
weaning index based on ventilatory endurance and efciency of gas exchange
by Jabour et al. (42) yielded high predictive value; however, it is cumbersome
to calculate and was based on a limited number of patients. With airway occlu-
sion, the ratio of the pressure generated with the rst breath (PI) to the pressure
Attempts at standardization of the weaning process have been touted for over a
decade (58). Early reports focused on the best practice techniques in weaning
from ventilation. In one prospective, randomized trial of weaning by intermittent
mandatory ventilation (IMV), outcomes of time to extubation were similar to T-
tube (59). The authors concluded that protocols involving both approaches may
be applied to clinically stable patients with success. Despite some early experi-
ence in acute surgical populations with IMV trials, weaning protocols involving
T-piece trials or pressure support ventilation now appear to be superior to inter-
mittent mandatory and are preferred for acute medical and surgical patients (60).
However, for post-ICU long term weaning cohorts, protocols involving synchro-
nized IMV plus pressure support have enjoyed success (61). Similar approaches
have achieved promising results in certain special populations, such as spinal
cord injury patients, with expected long duration of weaning (62).
B. Practice Variation
Personal preference and variability of weaning practice has frequently been
reported, and several randomized clinical trials have produced conicting
results regarding the best technique for carrying out the weaning process (63).
Work done by the Spanish Lung Failure Collaborative Group in the 1980s
demonstrated wide variation in regional weaning practice as well as reliance
on AC and SIMV techniques during weaningapproaches linked to long dur-
ations of weaning (64). Esteban et al.s analysis of international ventilation prac-
tices found remarkable similarity in approaches to maintenance ventilation but
wide practice variation around weaning (65). Their group and others have
demonstrated the superiority of not one but several approaches over AC or
Weaning from Mechanical Ventilation 165
SIMV, including daily and multiple daily spontaneous T-tube trials and pressure
support weaning (66).
Some groups have applied multi-disciplinary processes aimed at the
leading factors prolonging ventilator stay, such as ventilator-associated pneumo-
nia (67). In other cases, technology for monitoring weaning progress has also
been promoted. In one series, esophageal manometry was used adjunctively to
keep patients on protocol (68). No differences in steps to weaning were observed;
however, the monitoring technology appeared to give clinicians a greater con-
dence with staying on protocol without an increase in protocol intolerance. In
short, while QI tools, technology, and protocols may reduce variability in prac-
tice, a disciplined process and not a particular approach appears to be the
salient message.
C. To Wean or to Extubate?
Although improvements in time to wean have modestly improved on protocols,
overall duration of ventilation has markedly decreased on some protocols (69).
One explanation is that protocols enable practitioners to identify patients ready
for rapid extubation without weaning. Saura et al. (69) studied the clinical con-
sequences of a protocol that enabled direct extubation without a weaning tech-
nique, which was applied in 10% of their cohort and resulted in decreased
duration of mechanical ventilation and ICU stay. Safe, direct extubation, and
not weaning methods per se, appeared to be the factor linked to timely extubation.
Similar rapid extubation strategies have been applied with nurse-directed proto-
cols following cardiac surgery, and such protocols under various specialist direc-
tion are now standard of care (70).
Several protocols are available that have been shown in well-designed trials to
efciently wean most patients. After a spontaneous breathing trial, patients
who appear to be doing well by clinical criteria or weaning parameters may be
extubated, placed on t-tube, or ventilated with low-level pressure support.
Usually the level of pressure support is 7 cmH2O. Trials on t-tube or pressure
Weaning from Mechanical Ventilation 167
support may last 30 to 120 minutes. Little information is known about the optimal
duration of spontaneous breathing trials, and these may safely last from 30 to 120
minutes (83). Many advocate a exible approach of spontaneous trials of 30
minutes or low-level pressure support (84).
Where patients do not tolerate immediate low level pressure support or t-tube,
a pressure wean or titration can take place. A pressure support titration can establish
the minimal level to provide an adequate respiratory rate, tidal volume, and minute
ventilation. Then titration can be followed by gradual reduction by 2 cmH2O every
one to two hours. Patients weaned on consecutive days can often return to the last
pressure support level tolerated. Two randomized studies found that, in difcult-to-
wean patients, synchronized intermittent mandatory ventilation may be even more
effective (85,86). Other studies indicate that patients with prolonged ventilation be
placed on a chronic weaning protocol, which usually involves slower, graded IMV
reductions followed by pressure support reductions.
Non-invasive ventilation has been touted as a means to improve weaning
success, and studies have shown improvements in mortality (87). Unfortunately,
weaning failures or the duration of mechanical support related to weaning were
unimproved (88). One subgroup, chronic obstructive lung disease patients,
appeared to demonstrate the greatest mortality reduction (88). In all, given the
mixed results on ventilation, the Cochrane group still considers the data on
non-invasive ventilation insufcient to recommend its routine use in weaning.
Additional detail is described in the AHRQ technology assessment (89).
Automated, adaptive support ventilation techniques have been reported to
facilitate fewer ventilator manipulations but produced no improvement in out-
comes (90). In summary, reduction of practice variation at a single site
through the use of protocols may produce reductions in duration of ventilation,
despite differences in practitioners and details of protocols.
Expert and evidence-based recommendations for acute weaning protocols
have been summarized by Ely et al. in Table 3. The group provides guidance on
essential content, development strategies, and implementation of protocol.
The long-term acute care (LTAC) setting may be the best setting for protocolized
approaches. Scheinhorn et al. (61) reported signicantly shorter time to
weaning than benchmarks and historical controls through therapist-implemented
patient-specic (TIPS) weaning protocols. Scheinhorns protocol, trialed in over
250 patients and 9000 ventilator days at Barlow Respiratory Hospital and
Research Center, emphasizes adherence with a rigid set of steps, after assessing
for patient-specic issues with readiness and tolerance screens. While retaining
patient-specic adaptations, such as undiagnosed hypoxemia (with blood gases),
the overall emphasis is a rigid codes of procedure and a mantra: assess-
ment ! intervention ! assessment. This fusion of process engineering and
clinical judgment allows the expertise of the physicians to be present at the
168 Hackner, Dass, and Lewis
VI. Sedation
Analgesia
Pain assessment with an appropriate approach on a regular basis
Patients who can communicate: Encourage subjective (visual analog) scale
Patients who cannot communicate: observe pain-related behaviors and the change in
these parameters following analgesia
Establish therapeutic plan and communicate to all caregivers
Opioid analgesics: fentanyl, hydromorphone, and morphine are the recommended
agents
Scheduled opioid doses or a continuous infusion is preferred over an as needed
regimen
Fentanyl is preferred for a rapid onset of analgesia in acutely distressed patients
Fentanyl or hydromorphone are preferred for patients with hemodynamic instability or
renal insufciency
Morphine and hydromorphone are preferred for intermittent therapy because of their
longer duration of effect
NSAIDs or acetaminophen may be used as adjuncts to opioids in selected patients
Ketorolac therapy should be limited to a maximum of ve days, with close monitoring
for the development of renal insufciency or gastrointestinal bleeding
Other NSAIDs may be used via the enteral route in appropriate patients
Sedation
Sedation of agitated critically ill patients should be started only after
providing adequate analgesia and treating reversible physiological causes
A sedation goal or endpoint should be established and regularly redened for each
patient
Regular assessment and response to therapy should be systematically documented
The use of a validated sedation assessment scale (SAS, MAAS, or VICS) is
recommended
After seven days, consider the potential for opioid, benzodiazepine, and propofol
withdrawals
Doses should be tapered systematically to prevent withdrawal symptoms
Delirium
Haloperidol is the preferred agent for the treatment of delirium in critically ill
patients
Patients should be monitored for electrocardiographic changes when receiving
haloperidol
(QT interval prolongation and arrhythmias)
Sleep
Sleep promotion should include optimization of the environment and
non-pharmacologic methods to promote relaxation with adjunctive
use of hypnotics
Is patient comfortable
or at goal?
Consider Continuous
Infusion Versus
Nonpharmacologic Intravenous Push
Treatment
Fentanyl,
Assess pain / analgesia
Hydromorphone,
(Visual Analog Scale)
Morphine
(317 167 hr), and the lengths of ICU stay (19.1 9.9 days) and total stay without
a change in mortality (92).
Similarly, one should avoid partial opioid agonists, such as butorphanol, bupre-
norphine, and nalbuphine.
In addition to opiate or benzodiazepine withdrawal, clinicians should be
vigilant about alcohol withdrawal. Overlooking the diagnosis of alcohol withdra-
wal and empiric treatment with neuroleptics has been shown to lead to adverse
clinical outcomes (117). Vigilance for withdrawal signs and symptoms may
avoid costly clinical misadventures, prolonged ventilation in an attempt to re-
sedate, or potentially increased mortality.
I. Delirium
Delirium in the ICU has been associated with fewer median days alive and without
mechanical ventilation (118). Delirium is also linked to higher incidence of cogni-
tive impairment at hospital discharge and higher six-month mortality, even after
adjusting for relevant covariates (118). The condition is also associated with 39%
higher ICU costs (94). Long-term neuropsychological impairment is increasingly
recognized following mechanical ventilation (119). Severely ill patients, elderly
patients, children, and psychotic patients may become more confused with benzo-
diazepine sedatives. Other risk factors for delirium are listed in Table 6.
The need for reliable interdisciplinary measures has led to increasing use of
tools such as the Confusion Assessment Method (CAM-ICU) (120) For its treat-
ment, haloperidol is generally well-tolerated and has anxiolytic properties. In
patients with prior cardiac disease, the benets and risks of alternatives should
be weighed. In general, patients on prolonged neuroleptics should be monitored
for behavioral changes and for electrocardiographic abnormalities. Routine
assessment and treatment of delirium, along with sedation and analgesia,
should take place for the ventilated patient.
Age over 70 yr
Alcohol abuse (within a month)
Analgesics or sedatives (benzodiazepines or narcotics)
BUN/creatinine ratio .18
Cardiogenic or septic shock
Central venous catheters
Drug overdose or illicit drug use (within week)
History of congestive heart failure
History of stroke, epilepsy
HIV infection
Hypo- or hyperglycemia
Hypo- or hypernatremia
Hypo- or hyperthermia (fever)
Hypo- or hyperthyroidism
Liver disease (T-bilirubin .2 mg/dL)
Malnutrition
Physical restraints (including posey vest)
Prior history of deperession
Rectal or bladder catheters
Renal failure (Cr . 2 mg/dL)
Transfer from a nursing home
Tube feeding or parenteral nutrition
Visual or hearing impairment
study suggests that the vast majority of patients in the inpatient setting do wish to
discuss advance directives or goals of therapy (122). Some may have thought
carefully about the possibilities but will usually not have had a prior experience
with mechanical ventilation (Fig. 5). Moreover, studies report a dissociation
between surrogate statements of wishes and those of the patient, especially in
cases where a patient wishes not to be intubated/resuscitated (123).
One approach to addressing goals is to use a structured approach to discuss-
ing goals and values with the patient to elicit specic values regarding venti-
lation. In a structured approach, the clinician explores the patients health
status (baseline), minimal acceptable health status/outcome, willingness to
sustain burdensome therapy, and expressed wishes regarding chances of
success or duration of therapeutic trials. Then ventilation plans, intubations, extu-
bations, and sedation strategies are adapted to the matrix of values (Table 7).
Advance directives can be recommended based on a patients goals and values
and may change as clinical probabilities or feelings about burden or duration
of therapy changes.
The rst concept in laying out goals is the burden of therapy. Prior to initi-
ating mechanical ventilation, a patient should be willing to sustain the burdens of
176 Hackner, Dass, and Lewis
Table 7 Matrix for Addressing Goals and Values for the Ventilated Patient
Acknowledgment
Special thanks to Lawrence Maldonado, MD, for his assistance with the Goals
of Therapy: Conversations About Ventilation section and Table 7.
References
24. Girault C, Breton L, Richard JC, et al. Mechanical effects of airway humidication
devices in difcult to wean patients. Crit Care Med 2003; 31:1306 1311.
25. Talpers SS, Romberger DJ, Bunce SB, Pingleton SK. Nutritionally associated
increased carbon dioxide production: excess total calories vs high proportion of
carbohydrate calories. Chest 1992; 102:551 555.
26. Oh TE, Bhatt S, Lin ES, Hutchinson RC, Low JM. Plasma catecholamines and
oxygen consumption during weaning from mechanical ventilation. Intensive Care
Med 1991; 17:199 203.
27. Lemaire F, Teboul JL, Cinotti L, et al. Acute left ventricular dysfunction during
unsuccessful weaning from mechanical ventilation. Anesthesiology 1988;
69:171 179.
28. Meade M, Guyatt G, Cook D, et al. Predicting success in weaning from mechanical
ventilation. Chest 2001; 120(suppl 6):400S 424S.
29. Hess DR. Liberation from mechanical ventilation: weaning the patient or weaning
old-fashioned ideas? Crit Care Med 2002; 9:2154 2155.
30. Lessard MR, Brochard LJ. Weaning from ventilatory support. Clin Chest Med
1996; 17(3):475 489.
31. Wright PE, Marini JJ, Bernard GR. In vitro versus in vivo comparison of endotra-
cheal tube airow resistance. Am Rev Respir Dis 1989; 140:10 16.
32. Sahn SA, Lakshminarayan S. Bedside criteria for discontinuation of mechanical
ventilation. Chest 1973; 63:1002 1005.
33. Yang KL, Tobin MJ. A prospective study of indexes predicting the outcome of trials
of weaning from mechanical ventilation. N Engl J Med 1991; 324(21):14451450.
34. Stroetz RW, Hubmayr RD. Tidal volume maintenance during weaning with
pressure support. Am J Respir Crit Care Med 1995; 152:1034 1040.
35. Rivera L, Weissman C. Dynamic ventilatory characteristics during weaning in post-
operative critically ill patients. Anesth Analg 1997; 84:1250 1255.
36. Marini JJ, Smith TC, Lamb V. Estimation of inspiratory muscle strength in
mechanically ventilated patients: the measurement of maximal inspiratory pressure.
J Crit Care 1986; 1:32 38.
37. Tobin MJ, Perez W, Guenther SM, et al. The pattern of breathing during successful
and unsuccessful trials of weaning from mechanical ventilation. Am Rev Respir Dis
1986; 134(6):1111 1118.
38. Vallverdu I, Calaf N, Subirana M, Net A, Benito S, Mancebo J. Clinical character-
istics, respiratory functional parameters, and outcome of a two-hour T-piece trial in
patients weaning from mechanical ventilation. Am J Respir Crit Care Med 1998;
158:1855 1862.
39. Epstein SK. Etiology of extubation failure and the predictive value of the rapid
shallow breathing index. Am J Respir Crit Care Med 1995; 152(2):545 549.
40. Epstein SK, Ciubotaru RL. Inuence of gender and endotracheal tube size on
preextubation breathing pattern. Am J Respir Crit Care Med 1996; 154(6 Pt
1):1647 1652.
41. Krieger BP, Isber J. Serial measurements of the rapid-shallow-breathing index as
a predictor of weaning outcome in elderly medical patients. Chest 1997; 11:10291034.
42. Jabour ER, Rabil DM, Truwit JD, Rochester DF. Evaluation of a new weaning
index based on ventilatory endurance and the efciency of gas exchange. Am
Rev Respir Dis 1991; 144:531 537.
Weaning from Mechanical Ventilation 181
59. Tomlinson JR, Miller KS, Lorch DG, Smith L, Reines HD, Sahn SA. A prospective
comparison of IMV and T-piece weaning from mechanical ventilation. Chest 1989;
96(2):348 352.
60. Dries DJ. Weaning from mechanical ventilation. J Trauma 1997; 43(2):372
384.
61. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M, Wallace WA. Outcomes in post-
ICU mechanical ventilation: a therapist-implemented weaning protocol. Chest
2001; 119(1):236 242.
62. Thomas E, Paulson SS. Protocol for weaning the SCI patient. SCI Nurs 1994;
11(2):42 45.
63. Kollef MH, Horst HM, Prang L, Brock WA. Reducing the duration of mechanical
ventilation: three examples of change in the intensive care unit. New Horiz 1998;
6(1):52 60.
64. Esteban A, Alia I, Ibanez J, Benito S, Tobin MJ. Modes of mechanical ventilation
and weaning: a national survey of Spanish hospitals. Spanish Lung Failure Colla-
borative Group Chest 1982; 106:1188 1193.
65. Esteban A, Anzueto A, Alia I, et al. How is mechanical ventilation employed in the
intensive care unit? An international utilization review. Am J Respir Crit Care Med
2000; 161(5):1450 1458.
66. Esteban A, Frutos F, Tobin MJ, et al. A comparison of four methods of weaning
patients from mechanical ventilation. Spanish Lung Failure Collaborative Group.
N Engl J Med 1995; 332(6):345 350.
67. Joiner GA, Salisbury D, Bollin GE. Utilizing quality assurance as a tool for redu-
cing the risk of nosocomial ventilator-associated pneumonia. Am J Med Qual 1996;
11(2):100 103.
68. Gluck EH, Barkoviak MJ, Balk RA, Casey LC, Silver MR, Bone RC. Medical
effectiveness of esophageal balloon pressure manometry in weaning patients
from mechanical ventilation. Crit Care Med 1995; 23(3):504 509.
69. Saura P, Blanch L, Mestre J, Valles J, Artigas A, Fernandez R. Clinical conse-
quences of the implementation of a weaning protocol. Intensive Care Med 1996;
22(10):1052 1056.
70. Davies N. Nurse-initiated extubation following cardiac surgery. Intensive Crit Care
Nurs 1997; 13(2):77 79.
71. Kollef MH, Shapiro SD, Silver P, et al. A randomized, controlled trial of protocol-
directed versus physician-directed weaning from mechanical ventilation. Crit Care
Med 1997; 25(4):567 574.
72. Wood G, MacLeod B, Moffatt S. Weaning from mechanical ventilation: physician-
directed vs a respiratory-therapist-directed protocol. Respir Care 1995;
40(3):219 224.
73. Hynes-Gay P, Brindley A, Chu W, et al. Implementation of an RN/RT-initiated
extubation protocol. Dynamics (Pembroke, Ont) 2000; 11(4):12 16.
74. Horst HM, Mouro D, Hall-Jenssens RA, Pamukov N. Decrease in ventilation time
with a standardized weaning process. Arch Surg 1998; 133(5):483 488.
75. Hospital cuts time on ventilators, ICU LOS. Health Benchmarks 1999;
6(5):57 59.
76. Hospital cuts length of stay for vent patients by 34%. Hosp Case Manag 1999;
7(9):152 156.
Weaning from Mechanical Ventilation 183
77. Randolph AG, Wypij D, Venkataraman ST, et al. Effect of mechanical ventilator
weaning protocols on respiratory outcomes in infants and children: a randomized
controlled trial.[see comment]. J Am Med Assoc 2002; 288(20):2561 2568.
78. Ely EW, Meade MO, Haponik EF, et al. Mechanical ventilator weaning protocols
driven by nonphysician health-care professionals: evidence-based clinical practice
guidelines. Chest 2001; 120(suppl 6):454S 463S.
79. Ely EW, Bennett PA, Bowton DL, Murphy SM, Florance AM, Haponik EF. Large
scale implementation of a respiratory therapist-driven protocol for ventilator
weaning. Am J Respir Crit Care Med 1999; 159(2):439 446.
80. Calhoun CJ, Specht NL. Standardizing the weaning process. AACN Clin Issues Crit
Care Nurs 1991 Aug 2002;398 404.
81. Krishnan JA, Moore D, Robeson C, Rand CS, Fessler HE. A prospective, controlled
trial of a protocol-based strategy to discontinue mechanical ventilation. Am J
Respir Crit Care Med 2004; 169(6):673 678.
82. Tobin MJ. Of principles and protocols and weaning. Am J Respir Crit Care Med
2004; 169(6):661 662.
83. Esteban A, Alia I, Tobin MJ, et al. Effect of spontaneous breathing trial duration on
outcome of attempts to discontinue mechanical ventilation. Spanish Lung Failure
Collaborative Group. Am J Respir Crit Care Med 1999; 159(2):512 518.
84. Frutos-Vivar F, Esteban A. When to wean from a ventilator: an evidence-based
strategy. Cleve Clin J of Med 1997; 70(5):389 393.
85. Alia I, Esteban A. Weaning from mechanical ventilation. Crit Care 2000 2004;
72 80.
86. Frutos-Vivar F, Esteban A. When to wean from a ventilator: an evidence-based
strategy. Cleve Clin J Med 2003; 70(5):389, 392 393, 397.
87. Girou E, Schortgen F, Delclaux C, et al. Association of noninvasive ventilation with
nosocomial infections and survival in critically ill patients. J Am Med Assoc 2000;
284(18):2361 2367.
88. Burns K, Adhikari N, Meade M. Noninvasive positive pressure ventilation as a
weaning strategy for intubated adults with respiratory failure. Cochrane Database
Systemat Rev 2004; 1, 2004.
89. Agency for Healthcare Research and Quality. Criteria for weaning from mechanical
ventilation. Evidence Report/Technology Assessment No. 23. [AHRQ Publication
No. 01-E010], 2003.
90. Petter AH, Chiolero RL, Cassina T, Chassot PG, Muller XM, Revelly JP. Automatic
respirator/weaning with adaptive support ventilation: the effect on duration of
endotracheal intubation and patient management. Anesth Analg 2003;
97(6):1743 1750.
91. Gluck E. Program and abstracts of Weaning96, Weaning from Prolonged
Mechanical Ventilation. 1996. Palm Springs, FL.
92. Woods JC, Mion LC, Connor JT, et al. Severe agitation among ventilated medical
intensive care unit patients: frequency, characteristics and outcomes. Intensive Care
Med 2004; 14.
93. Jacobi JPFB, Fraser GLPF, Coursin DBM, Riker RRM. Clinical practice guidelines
for the sustained use of sedatives and analgesics in the critically ill adult. Society of
Critical Care Medicine and the American Society of Health-System Pharmacists I,
ed. Crit Care Med 30(1), 119 141.
184 Hackner, Dass, and Lewis
94. Milbrandt EB, Deppen S, Harrison PL, et al. Costs associated with delirium in
mechanically ventilated patients. Crit Care Med 2004; 1932:955 962.
95. Ely EW, Gautam S, Margolin R, et al. The impact of delirium in the
intensive care unit on hospital length of stay. Intensive Care Med 2001;
27(12):1892 1900.
96. Milbrandt EB, Deppen S, Harrison PL, et al. Costs associated with delirium in
mechanically ventilated patients. Crit Care Med 2004; 32(4):955 962.
97. Cheng E. The cost of sedating and paralyzing the critically ill patient. Crit Care Clin
1995; 11(4):1005 1019.
98. Barrientos-Vega R, Mar Sanchez-Soria M, Morales-Garcia C, Robas-Gomez A,
Cuena B, yensa-Rincon A. Prolonged sedation of critically ill patients with
midazolam or propofol: impact on weaning and costs [see comment]. Crit Care
Med 1997; 25(1):33 40.
99. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative
infusions in critically ill patients undergoing mechanical ventilation. N Engl J
Med 2000; 342(20):1471 1477.
100. ACP J Club. Daily screening and breathing trials hastened the discontinuation of
mechanical ventilation. ACP J Club 1997; v126:74-Jun.
101. Mascia MF, Koch M, Medicis JJ. Pharmacoeconomic impact of rational use guide-
lines on the provision of analgesia, sedation, and neuromuscular blockade in critical
care. Crit Care Med 2000; 28(7):2300 2306.
102. Krachman SL DGCG. Sleep in the intensive care unit. Chest 1995; 107:
1713 1720.
103. Rothaar RC, Epstein SK. Extubation failure: magnitude of the problem, impact on
outcomes, and prevention. Curr Opin Crit Care 2003; 9(1):59 66.
104. Ferguson J GDPK. Dimensions of pain and analgesic administration associated
with coronary artery bypass in an Australian intensive care unit. J Adv Nurs
1997; 26:1065 1072.
105. NHS Centre for Reviews and Dissemination. Using and understanding
sedation scoring systems. Database of Abstracts of Reviews of Effectiveness
2004:2001.
106. Gamper G, Willeit M, Sterz F, et al. Life after death: posttraumatic stress disorder in
survivors of cardiac arrestprevalence, associated factors, and the inuence of
sedation and analgesia. Criti Care Med 2004; 378 383.
107. NHS Centre for Reviews and Dissemination. A lack of evidence of superiority of
propofol versus midazolam for sedation in mechanically ventilated critically ill
patients. Database of Abstracts of Reviews of Effectiveness 2004 2001.
108. Durbin CG. Sedation of the agitated, critically ill patient without an articial
airway. Critical Care Clinics 1995; 11(4):913 936.
109. Grounds RM, Lalor JM, Lumley J, Royston D, Morgan M. Propofol infusion for
sedation in the intensive care unit: preliminary report. Br Med J (Clin Res Ed)
1987; 294(6569):397 400.
110. Devlin JW, Holbrook AM, Fuller HD. The effect of ICU sedation guidelines and
pharmacist interventions on clinical outcomes and drug cost. Ann Pharmacother
1997; 31(6):689 695.
111. Jastremski CA. Pressure relief bedding to prevent pressure ulcer development in
critical care. J Crit Care 2002; 17(2):122 125.
Weaning from Mechanical Ventilation 185
112. Helman DL Jr, Sherner JH III, Fitzpatrick TM, Callender ME, Shorr AF. Effect of
standardized orders and provider education on head-of-bed positioning in mechani-
cally ventilated patients. Crit Care Med 2003; 31(9):2285 2290.
113. Berenholtz SM, Milanovich S, Faircloth A, et al. Improving care for the ventilated
patient. Jt Comm J Qual Saf 2004; 30(4):195 204.
114. Pitimana-aree S, Forrest D, Brown G, Anis A, Wang XH, Dodek P. Implementation
of a clinical practice guideline for stress ulcer prophylaxis increases appropriate-
ness and decreases cost of care. Intensive Care Med 1998; 24(3):217 223.
115. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative infu-
sions in critically ill patients undergoing mechanical ventilation. N Engl J Med
2000; 342:1471 1477.
116. Cammarano WB, Pittet JF, Weitz S, Schlobohm RM, Marks JD. Acute withdrawal
syndrome related to the administration of analgesic and sedative medications in
adult intensive care unit patients. Crit Care Med 1998; 676 684.
117. Mayo-Smith MF, Beecher LH, Fischer TL, et al. Management of alcohol with-
drawal delirium: an evidence-based practice guideline. Arch Intern Med
164(13):1405 12, 2004.
118. Ely EW, Shintani A, Truman B, et al. Delirium as a predictor of mortality in
mechanically ventilated patients in the intensive care unit. J Am Med Assoc
2004 Apr 14 291; 1753 1762.
119. Jackson J, Hart R, Gordon Seal. Six-month neuropsychological outcome of medical
intensive care unit patients. Criti Care Med 2003; 31:1226 1234.
120. Ely EW, Margolin R, Francis J, et al. Evaluation of delirium in critically ill patients:
validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-
ICU). Crit Care Med 2001; 29(7):1370 1379.
121. Travaline JM, Silverman HJ. Discussions with outpatients with chronic obstructive
pulmonary disease regarding mechanical ventilation as life-sustaining therapy.
Southern Med. Journal 1995; 88(10):1034 1038.
122. Reilly BM, Magnussen CR, Ross J, Ash J, Papa L, Wagner M. Can we talk?
Inpatient discussions about advance directives in a community hospital. Attending
physicians attitudes, their inpatients wishes, and reported experience [see
comment]. Arch Intern Med 1994; 154(20):2299 2308.
123. Sonnenblick M, Friedlander Y, Steinberg A. Dissociation between the wishes
of terminally ill parents and decisions by their offspring [see comment]. J Am
Geriatrics Soc 1993; 41(6):599 604.
124. Miller RG. Examining the evidence about treatment in ALS/MND [Review] [12
refs]. Amyotrophic Lateral Sclerosis Other Motor Neuron Disorders 2001;
2(1):3 7.
125. Simonds AK. Nasal ventilation in progressive neuromuscular disease: experience
in adults and adolescents [Review] [28 refs]. Monaldi Arch Chest Dis 2000;
55(3):237 241.
This page intentionally left blank
6
Prolonged Mechanical Ventilation
I. Introduction
187
188 Epstein and Nevins
The etiology of acute respiratory failure likely plays a role in determining which
patients go on to require prolonged mechanical ventilation. In a study of 183
patients with chronic obstructive pulmonary disease (COPD) and acute respirat-
ory failure admitted to a medical ICU, 10.4% remained mechanical ventilated
21 days after intubation (26). In contrast, in the same ICU, the relative risk of
remaining ventilated at day 21 was twice as high in patients intubated
for acute lung injury (21 of 107) (27). In a prospective study of 23 long-term
acute care hospitals, 1411 patients were admitted after a median of 25 days
(range: 0 273 days) of invasive mechanical ventilation at the referring acute
care hospital (28). The average age was 72 years, and 50% were women.
Seventy percent of patients were smokers, averaging 57 pack years. In this
cohort, 43% had COPD, 54% coronary artery disease or congestive heart
failure, and 20% neurologic disease. Among patients with respiratory failure
190 Epstein and Nevins
secondary to medical illness (61% of the cohort), the precipitating etiology was
bacterial pneumonia (34%), sepsis (21%), acute neurological disease (20%),
acute exacerbation of COPD (17%), congestive heart failure (14%), and
aspiration pneumonia (13%). In 39% of cases, respiratory failure followed a sur-
gical procedure (44% cardiovascular and 22% gastrointestinal). Despite the
advanced age of this cohort, other investigators have not found older patients
to be at higher risk for prolonged mechanical ventilation compared with
younger patients, after controlling for confounding factors (29,30).
Prior to 1990, nearly all patients requiring prolonged mechanical ventilation were
cared for in the ICU, with 14% to 25% eventually being discharged home
while still requiring mechanical ventilation (14,16). Subsequently, there was a
proliferation of centers specializing in the care and weaning of patients with pro-
longed mechanical ventilation. These have taken the form of specialized units
within acute care hospitals and freestanding LTAC or post-acute care hospitals.
In a 10-year follow-up study, the percentage of patients receiving prolonged
mechanical ventilation in an acute hospital setting decreased (from 62% to
18%), while the number of patients in post-acute care (from 22% to 46%) and
long-term care (from 2% to 24%) facilities increased (15).
One principle used to support the expansion of these centers is that patients
have recovered from their acute illness prior to transfer and therefore require a
lower intensity of care. As an example, nurse-to-patient stafng ratios are typi-
cally lower in chronic ventilator facilities than in ICUs (31). This may have
important implications because appropriate nursing care appears crucial for
favorable weaning outcomes in COPD patients on invasive mechanical venti-
lation in the acute care setting (32). In a European investigation, the duration
of mechanical ventilation increased from 7.3 to 38.2 days when the index of
nursing (a measure of the effective workforce of nurses as dened by both
stafng ratios and nursing qualications) fell. Indeed, the average duration of
mechanical ventilation decreased to 9.9 days when the initial index of nursing
was restored. Additionally, patients are now transferred earlier and with a rela-
tively high severity of illness (33). As an example, Scheinhorn et al. (34)
originally reported on 421 patients admitted to their regional weaning center,
observing a mean duration of mechanical ventilation of 49 days prior to
LTAC transfer (34). In subsequent reports, the authors noted a signicant trend
toward shorter duration of pre-admission mechanical ventilation (median:
33 days) and higher admission severity of illness scores (median APS of
APACHE III: 32) (33). A recent preliminary update totaling 1600 patients
described a continued, rising trend in severity of illness scores in their patients
that had become comparable to those reported at the time of ICU admission
Prolonged Mechanical Ventilation 191
patients are transferred (and die) in an outside acute care facility. There are no
randomized controlled trials comparing outcomes for patients cared for in
acute ICU versus those managed in an LTAC. Gracey et al. (4,37) compared
patients managed in their acute care ICU at the Mayo Clinic to similar patients
requiring prolonged mechanical ventilation managed in their chronic
ventilator-dependent unit (CVDU). Patients treated in the CVDU had a signi-
cantly lower mortality rate (9.8% vs. 28.6%, P 0.001) after excluding patients
with multi-organ failure to control for discrepancies between the two populations.
Mortality rates range from 7% to 61% in single-center reports depending on
the setting (acute care hospital vs. LTAC) and the denition of prolonged mechan-
ical ventilation (Table 1). In a large multi-center trial of more than 1400 patients
admitted to 23 LTACS around the United States, Scheinhorn et al. (28) noted a
26% mortality rate. Some investigators have focused on more long-term outcomes,
such as one-year survival, because of the limitations noted above. Indeed, one-year
survival ranges from just 23% to 38%, likely a reection of underlying comorbid
conditions, an ongoing debilitated state, and vulnerability to recurrent acute
disease experienced by short-term survivors of prolonged mechanical venti-
lation (21,33). Other studies found that long-term survival was correlated with
younger age and shorter pre-CVDU duration of mechanical ventilation (38).
Patients who fail extubation are more likely to require prolonged
mechanical ventilation. As an example, 12 additional days of invasive mecha-
nical ventilation were required for a cohort of medical patients who needed rein-
tubation after planned extubation (43). The need for reintubation also identies
patients more likely to need tracheostomy and eventual transfer to a post-acute
care facility (43 45).
Weaning success rates vary widely depending on the denition of success,
the population studied, and the admission criteria applied by the unit under
investigation. When looking at weaning success at hospital discharge, rates
range from 25% to 82% in single-center studies (Table 1). In a prospective
multi-center trial, 54% of 1411 patients (73% of hospital survivors) were charac-
terized as successfully liberated from mechanical ventilation at the time of
discharge from 23 long-term acute care units (28). Twenty-seven percent of
survivors (20% of all patients admitted) remained ventilator-dependent at the
time of hospital discharge.
A. Systemic Factors
As noted above, patients with prolonged mechanical ventilation have persistently
high severity illness (as assessed by APACHE acute physiology scores) and
Prolonged Mechanical Ventilation
Table 1 Outcomes for Patients Requiring Prolonged Mechanical Ventilation
ICU care
Sivak (150) 1980 ICU 15 14 7 66
Morganroth et al. (129) 1984 ICU 11 30 27 73
Spicher and white (5) 1987 ICU 245 10 61 ?
Gracey et al. (4) 1992 ICU 104 29 42 53
Non-ICU care
Indihar (41) 1991 PRCU 171 Median 36 33 34
Cordasco et al. (151) 1991 Non-ICU 99 ? 25 25
Gracey et al. (4) 1992 CVDU 61 21 9.5 82
Scheinhorn et al. (34) 1994 RWC 421 Mean 49 28 53
Nava et al. (123) 1994 IICU 42 21 29 36
Gracey et al. (37) 1995 CVDU 132 21 9.8 78
Latriano et al. (59) 1996 RCF 224 Mean 23 50 47
Gluck (42) 1996 LTVF 72 21 38 27 (at 42 days)
Gracey et al. (38) 1997 CVDU 206 21 8 70
Scheinhorn et al. (33) 1997 RWC 1123 Median 33 29 56a
(Continued)
193
194
Table 1 Outcomes for Patients Requiring Prolonged Mechanical Ventilation (Continued)
Mechanical factors
Increased work of breathing
Reduced respiratory muscle capacity
Critical illness polyneuropathy
Steroid myopathy
Disuse myopathy
Isolated phrenic nerve/diaphragmatic injury (e.g., post-cardiac surgery)
Imbalance between increased work of breathing and respiratory muscle capacity
Metabolic/systemic factors
Chronic comorbid conditions
(e.g., malignancy, chronic obstructive pulmonary disease, immunosuppression)
Overall severity of illness
Non-pulmonary organ failure
Poor nutritional status
Iatrogenic Factors
Imposed work of breathing from tracheotomy tubes
Upper airway obstruction (e.g., tracheal stenosis)
Recurrent aspiration
Infection (e.g., pneumonia, sepsis)
Psychological factors
Sedation
Delirium
Depression
Anxiety
Process of care factors
Absence of weaning protocols
Inadequate nursing stafng
Insufcient physician experience
frequently have important comorbid conditions (e.g., COPD) that likely contrib-
ute to persistent weaning failure. Indeed, a large proportion of patients transferred
to an LTAC with prolonged mechanical ventilation ultimately require transfer
back to acute care (46,47). In one study, of 97 patients (71 with prolonged mech-
anical ventilation) transferred to LTACs, 23% required readmission to acute
care within 30 days of transfer (47). Patients with prolonged mechanical venti-
lation frequently have non-pulmonary organ failure, the presence of which is
associated with poor outcome. In one study of 52 patients with prolonged mech-
anical ventilation and requiring hemodialysis, none were successfully weaned
and only three survived (48). Chao et al. (49) studied 63 patients with severe
renal dysfunction (creatinine .2.5 mg/dL) at the time of transfer to their
regional weaning center. Forty of these patients were on renal replacement
therapy at transfer, and hemodialysis was initiated in another 10 while at the
196 Epstein and Nevins
LTAC. Patients with severe renal dysfunction were less likely to be successfully
weaned (13% vs. 56%), and only four patients requiring renal replacement
therapy were liberated from the ventilator.
Cardiac disease often limits liberation from mechanical ventilation in the
acute setting. Myocardial ischemia (often occult), left ventricular dysfunction,
and pulmonary edema have all been documented during trials of spontaneous
breathing (50 55). Systematic examinations of cardiac factors have not
been published in patients requiring prolonged mechanical ventilation. In a pre-
liminary report, successful diuresis and weight loss were associated with weaning
success in patients transferred to a regional weaning center (56).
Poor nutritional status may contribute to prolonged mechanical ventilation
by adversely affecting respiratory muscle function and the ventilatory response
to gas exchange abnormalities, and by predisposing to infection (57). Indeed,
several studies of prolonged mechanical ventilation have noted an association
between hypoalbuminemia and weaning failure (58,59).
B. Mechanical Factors
An extensive body of research has identied key mechanical factors related to
weaning failure in the acute setting. These factors include abnormalities in respir-
atory drive, increased work of breathing (resistive, elastic, intrinsic PEEP),
decreased respiratory muscle capacity, cardiovascular factors (ischemia, pulmon-
ary edema), and psychological factors.
Numerous studies performed in acute hospital settings have shown patients
who fail weaning trials will usually have increased, not diminished, central res-
piratory drive (60 63). Nevertheless, there are some patients with central
nervous system disturbances or who are heavily sedated in whom weaning and
extubation may be delayed (64 66). In the large majority of patients who fail
weaning trials, the problem lies in an imbalance between the inspiratory load
and the capacity of the respiratory system (60,67,68). Patients failing weaning
trials often develop a pattern of rapid, shallow breathing in the face of increased
load (63). This pattern of breathing is seen within minutes of disconnection from
the ventilator suggesting that diaphragmatic fatigue is not the primary mechan-
ism involved, at least initially (69). Nevertheless, studies demonstrate that
weaning failure in the acute setting is associated with a tension time index
(TTI) of .0.15, which is thought to represent the threshold beyond which respir-
atory muscle fatigue ensues (60,68,70). As in patients with shorter durations of
ventilation, patients with a mean of 20 days who failed a trial of spontaneous
breathing also developed or maintained values of the TTI in the fatigue
zone (67). This possibility that fatigue might occur was further suggested by
Cohen et al. (71), who noted EMG evidence suggestive of high-frequency
diaphragmatic fatigue during failed weaning trials. Yet, in a study of eight
weaning success and 11 weaning failure patients, Laghi et al. (72) measured
twitch transdiaphragmatic pressure before and 30 minutes after spontaneous
Prolonged Mechanical Ventilation 197
breathing trials lasting up to 60 minutes. The weaning failure patients did not
develop low-frequency fatigue of the diaphragm despite considerable diaphrag-
matic weakness, and greater load and diaphragmatic effort. The absence of
fatigue may have resulted from increased recruitment of ribcage and expiratory
muscles or because of prompt reinstitution of ventilatory support (e.g., before
fatigue could ensue). The issue of fatigue is not trivial, as some studies in
normals indicate that it can take more than 24 hours to recover full diaphragmatic
function (73).
There is an emerging literature examining mechanical factors limiting
weaning in patients with prolonged mechanical ventilation. Appendini et al.
(74) monitored respiratory mechanics and diaphragmatic effort [using
pressure time product (PTP) and diaphragmatic TTI] in eight patients with
COPD requiring prolonged mechanical ventilation. Respiratory muscle strength
(maximal inspiratory, pleural, and transdiaphragmatic pressures) was found com-
patible with successful weaning but was countered by excessive load (increased
PEEPi, pulmonary resistance, and PTP) (74). Gluck (42) noted that patients with
prolonged mechanical ventilation demonstrated a pattern of rapid shallow breath-
ing, increased resistance, increased dead space, elevated dead space, and a
decreased compliance. Abnormalities in mechanics may not be overtly detect-
able. As an example, Reinoso et al. (75), using interrupter mechanics, detected
supramaximal ow transients after shutter valve release during passive
expiration, indicating occult airways disease in six of 25 patients tested.
Purro et al. (76) similarly evaluated 39 patients who had been intubated for
more than three weeks, 28 with COPD and 11 who had undergone cardiac surgery
[post-cardiac surgery (PCS)] complicated by diaphragmatic dysfunction. Control
groups were made up of nine patients with severe, stable COPD (maintained
with tracheotomies but not ventilator-dependent), and 11 PCS patients who
had been successfully liberated from mechanical ventilation within 48 hours of
their surgery. Eight of 28 COPD patients tolerated unsupported breathing for
60 minutes (WS-COPD), while the remaining 20 COPD patients (VD-COPD)
and all 11 PCS patients (VD-PCS) failed (mean time to failure: 40 min) and
needed reinstitution of ventilatory support. When compared with the stable
COPD and WS-COPD patients, the VD-COPD group demonstrated reduced
tidal volume, minute ventilation, maximal transdiaphragmatic pressure, and
maximal inspiratory pressure. The VD-COPD group demonstrated an increase
in central respiratory frequency, airway occlusion pressure at 0.1 seconds
(P0.1), PEEPi, and pulmonary resistance. Similarly, when compared with the
stable successfully extubated PCS patients, the VD-PCS group had reduced
tidal volume and maximum inspiratory pressure (MIP), while demonstrating
higher respiratory frequency and P0.1. Initially no differences were noted in the
mean respiratory rates between the stable COPD, WS-COPD, and VD-COPD
groups. In contrast, analysis of esophageal pressure tracings revealed untriggered
breaths (trigger asynchrony) in eight of 20 VD-COPD patients, resulting in a
central respiratory frequency 40% higher than measured from inspiratory
198 Epstein and Nevins
Figure 1 Plots of respiratory duty cycle (Ti/Ttot) and respiratory muscle effort
(mean diaphragmatic pressure generated per breath divided by maximal diaphragmatic
pressure, Pdi/Pdimax) for VD, W, and stable patients with either COPD or PCS. The
product of the duty cycle and respiratory muscle effort is the TTI. Ventilator dependent
patients had higher TTI values, usually greater that the 0.15 threshold thought to represent
a fatiguing load on the respiratory system. Abbreviations: COPD, chronic obstructive
pulmonary disease; PCS, post-cardiac surgery; TTI; tension time index; VD, ventilator-
dependent; W, weaned. Source: Adapted from Ref. 76.
Prolonged Mechanical Ventilation 199
C. Iatrogenic Factors
Tracheotomies have been shown to reduce the work of breathing imposed by an
articial airway when compared with endotracheal tubes, though the impact on
weaning success remains unclear (96). Whereas in the acute setting a majority
of patients are ventilated through oral or nasal endotracheal tubes, most patients
admitted to a chronic ventilator facility have tracheotomies in place (21,41).
Indeed, in a large multi-center trial, 95% transferred to 23 LTACs came with a
tracheotomy (28). Nevertheless, abnormalities of the upper airway resulting
from complications of the articial airway can contribute to ventilator depen-
dence. Ten percent of patients with prolonged mechanical ventilation will
develop tracheal injury despite the use of articial airways with low-pressure,
high-volume cuffs (97,98). Tracheal injury at or above the level of the tracheal
tube is typically clinically silent until the time of extubation or decannulation.
Rumbak et al. (99) identied 37 of 756 (5%) patients on prolonged mechanical
ventilation who had evidence for distal tracheal obstruction contributing to ven-
tilator dependence. All patients had transient elevations in peak airway pressures,
200 Epstein and Nevins
and difculty passing a suction catheter was uniformly present. Bypassing the
obstruction using either a longer tracheal tube or tracheal stent allowed 35 of
the 37 patients to be successfully liberated from mechanical ventilation within
one week.
Tracheostomies can also contribute to swallowing dysfunction and aspira-
tion in patients on prolonged mechanical ventilation. Indeed, 50% of 83 patients
had evidence of aspiration when studied using videoouroscopic tapes of modied
barium swallows. The majority of aspiration events were not accompanied by
clinical evidence of respiratory distress (100). In another investigation of 35
patients, 83% had swallowing dysfunction by videoouroscopy (101). Using scin-
tigraphy to identify aspiration, Schonhofer et al. (102) found a somewhat lower
rate of aspiration (30%) among patients requiring prolonged mechanical venti-
lation. Aspiration may contribute to weaning failure by compromising lung func-
tion, increasing respiratory secretions, predisposing to ventilator-associated
pneumonia, and possibly through compromising oral intake of nutrition.
In the acute setting, increased duration of mechanical ventilation has
been associated with increased risk of complications, including nosocomial
infection. In their prospective observation study, Scheinhorn et al. noted that
acquired infection was common: urinary tract infection (32%), lower respiratory
tract infection (28%), Clostridium difcile infection (18%), and central line infec-
tion (12%) (Fig. 2). Importantly, those who acquired lower respiratory tract infec-
tion had a longer length of stay, took longer to wean, were less likely to be
More than 50 distinct weaning predictors have been studied in an effort to foretell
the outcome of weaning trials in the acute ICU setting (116,117). The rationale
for using objective physiologic parameters is strong as prediction based on clini-
cal gestalt alone is relatively inaccurate (118,119). More importantly, precise pre-
dictors would enable physicians to postpone weaning in patients likely to fail.
Alternatively, predictors could be used to better identify patients ready to initiate
weaning trials with a goal of hastening the process of liberation, thereby reducing
the duration of mechanical ventilation. The latter goal is based on the association
between increased duration of ventilation and complications (ventilator-associ-
ated pneumonia, airway injury, barotrauma, sinusitis, thromboembolism, and
gastrointestinal bleeding), which further increase duration of stay, costs, and mor-
tality (120,121). Recent analyses and evidence-based guidelines indicate that
weaning predictors are insufciently accurate to adequately inform decision-
making in the acute ICU (117,122).
Studies in patients requiring .21 days of mechanical ventilation have
shown that respiratory mechanics used to predict weaning outcomes in acute
patients may have some value among patients requiring prolonged mechanical
ventilation (76,123). Nava et al. observed that lower maximal inspiratory
pressure and higher central drive as measured by the airway occlusion pressure
were predictive of weaning success among patients with COPD (76,123).
Although the frequency tidal volume ratio was also signicantly different in
patients with weaning success when compared with those of weaning failure,
the mean values in the latter group were well below the threshold originally
described by Yang and Tobin (69). As suggested by recent work, patients with
COPD may be unable to increase their respiratory rates signicantly because
of the mechanical constraints of expiratory-ow limitation and hyperination.
In this setting, a lower f/Vt threshold may yield more accurate predictions
(124). Indeed, Scheinhorn et al. (125) used the frequency-to-tidal volume ratio
to accelerate weaning progress through a therapist-implement protocol (125).
Weaning success was nearly twice as likely when the f/Vt was ,80 compared
with .120 (breaths/L)/min (Fig. 3) (126). Failure to appreciate respiratory
efforts that do not trigger the ventilator or result in inspiratory ow
may lead to an underestimation of the respiratory rate and therefore the f/Vt
ratio (76).
Indeed, Chao et al. (127) noted an association between patient ventilator
trigger asynchrony and weaning failure in patients requiring prolonged mechan-
ical. Nineteen of the 200 patients screened were noted to have trigger asynchrony.
Interventions such as altering trigger sensitivity, changing to ow triggering,
and increasing external PEEP were unsuccessful in eliminating the trigger
asynchrony. Only decreasing the level of pressure support was successful in
Prolonged Mechanical Ventilation 203
Figure 3 Percentage of patients with weaning success based on the frequency to tidal
volume ratio determined prior to the initiation of weaning efforts. Source: Adapted
from Ref. 126.
IX. Conclusion
References
1. Nevins ML, Epstein SK. Weaning from prolonged mechanical ventilation. Clin
Chest Med 2001; 22:13 33.
2. Arabi Y, Venkatesh S, Haddad S, Al Shimemeri A, Al Malik S. A prospective study
of prolonged stay in the intensive care unit: predictors and impact on resource
utilization. Int J Qual Health Care 2002; 14:403 410.
3. Gillespie DJ, Marsh HM, Divertie MB, Meadows JA III. Clinical outcome of res-
piratory failure in patients requiring prolonged (greater than 24 hours) mechanical
ventilation. Chest 1986; 90:364 369.
4. Gracey DR, Viggiano RW, Naessens JM, Hubmayr RD, Silverstein MD, Koenig
GE. Outcomes of patients admitted to a chronic ventilator-dependent unit in an
acute-care hospital. Mayo Clin Proc 1992; 67:131 136.
5. Spicher JE, White DP. Outcome and function following prolonged mechanical
ventilation. Arch Intern Med 1987; 147:421 425.
6. Scheinhorn DJ, Hassenpug M, Artinian BM, LaBree L, Catlin JL. Predictors of
weaning after 6 weeks of mechanical ventilation. Chest 1995; 107:500 505.
7. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M. Liberation from prolonged
mechanical ventilation. Crit Care Clin 2002; 18:569 595.
208 Epstein and Nevins
25. Chatila W, Kreimer DT, Criner GJ. Quality of life in survivors of prolonged
mechanical ventilatory support. Crit Care Med 2001; 29:737 742.
26. Nevins ML, Epstein SK. Predictors of outcome for patients with COPD requiring
invasive mechanical ventilation. Chest 2001; 119:1840 1849.
27. Zilberberg MD, Epstein SK. Acute lung injury in the medical ICU: comorbid con-
ditions, age, etiology, and hospital outcome. Am J Respir Crit Care Med 1998;
157:1159 1164.
28. Scheinhorn D, Chao D, Stearn-Hassenpug M, Doig G, Epstein S, Knight E,
Petrak R, Pitt E, Votto J. Ventilator-dependent survivors of catastrophic illness: a
multicenter outcomes stuy. Am J Respir Crit Care Med 2003; 167:A458.
29. Ely EW, Evans GW, Haponik EF. Mechanical ventilation in a cohort of elderly
patients admitted to an intensive care unit. Ann Intern Med 1999; 131:96 104.
30. Esteban A, Anzueto A, Frutos-Vivar F, Alia I, Ely EW, Brochard L, Stewart TE,
Apezteguia C, Tobin MJ, Nightingale P, Matamis D, Pimentel J, Abroug F.
Outcome of older patients receiving mechanical ventilation. Intensive Care Med
2004; 30:639 646.
31. Elpern EH, Silver MR, Rosen RL, Bone RC. The noninvasive respiratory care unit.
Patterns of use and nancial implications. Chest 1991; 99:205 208.
32. Thorens JB, Kaelin RM, Jolliet P, Chevrolet JC. Inuence of the quality of nursing
on the duration of weaning from mechanical ventilation in patients with chronic
obstructive pulmonary disease. Crit Care Med 1995; 23:1807 1815.
33. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M, LaBree LD, Heltsley DJ. Post-
ICU mechanical ventilation: treatment of 1,123 patients at a regional weaning
center. Chest 1997; 111:1654 1659.
34. Scheinhorn DJ, Artinian BM, Catlin JL. Weaning from prolonged mechanical
ventilation. The experience at a regional weaning center. Chest 1994;
105:534 539.
35. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M. Post-ICU mechanical ventilation-
treatment of 1575 patients over 11 years at a regional weaning center. Am J Respir
Crit Care Med 2000; 161:A793.
36. Sirio CA, Shepardson LB, Rotondi AJ, Cooper GS, Angus DC, Harper DL,
Rosenthal GE. Community-wide assessment of intensive care outcomes using a
physiologically based prognostic measure: implications for critical care delivery
from Cleveland Health Quality Choice. Chest 1999; 115:793 801.
37. Gracey DR, Naessens JM, Viggiano RW, Koenig GE, Silverstein MD, Hubmayr
RD. Outcome of patients cared for in a ventilator-dependent unit in a general
hospital. Chest 1995; 107:494 499.
38. Gracey DR, Hardy DC, Naessens JM, Silverstein MD, Hubmayr RD. The Mayo
Ventilator-Dependent Rehabilitation Unit: a 5-year experience. Mayo Clin Proc
1997; 72:13 19.
39. Bach PB, Carson SS, Leff A. Outcomes and resource utilization for patients with
prolonged critical illness managed by university-based or community-based
subspecialists. Am J Respir Crit Care Med 1998; 158:1410 1415.
40. Bagley PH, Cooney E. A community-based regional ventilator weaning unit:
development and outcomes. Chest 1997; 111:1024 1029.
41. Indihar FJ. A 10-year report of patients in a prolonged respiratory care unit. Minn
Med 1991; 74:23 27.
210 Epstein and Nevins
42. Gluck EH. Predicting eventual success or failure to wean in patients receiving
long-term mechanical ventilation. Chest 1996; 110:1018 1024.
43. Epstein SK, Ciubotaru RL, Wong JB. Effect of failed extubation on the outcome of
mechanical ventilation. Chest 1997; 112:186 192.
44. Epstein SK, Ciubotaru RL. Independent effects of etiology of failure and time to
reintubation on outcome for patients failing extubation. Am J Respir Crit Care
Med 1998; 158:489 493.
45. Daley BJ, Garcia-Perez F, Ross SE. Reintubation as an outcome predictor in trauma
patients. Chest 1996; 110:1577 1580.
46. Chan M, Mehta R, Vasishtha N. Ventilator care in a nursing home. Am J Respir Crit
Care Med 1999; 159:A374.
47. Nasraway SA, Button GJ, Rand WM, Hudson-Jinks T, Gustafson M. Survivors of
catastrophic illness: outcome after direct transfer from intensive care to extended
care facilities. Crit Care Med 2000; 28:19 25.
48. Tafreshi M, Schneider R, Rosen M. Outcome of patients who require long-
term mechanical ventilation and hemodialysis (abstract). Chest 1995;
108(suppl):134S.
49. Chao DC, Scheinhorn DJ, Stearn-Hassenpug M. Impact of renal dysfunction on
weaning from prolonged mechanical ventilation. Crit Care (London) 1997;
1:101 104.
50. Hurford WE, Lynch KE, Strauss HW, Lowenstein E, Zapol WM. Myocardial
perfusion as assessed by thallium-201 scintigraphy during the discontinuation of
mechanical ventilation in ventilator-dependent patients. Anesthesiology 1991;
74:1007 1016.
51. Hurford WE, Favorito F. Association of myocardial ischemia with failure to wean
from mechanical ventilation. Crit Care Med 1995; 23:1475 1480.
52. Chatila W, Ani S, Guaglianone D, Jacob B, Amoateng-Adjepong Y, Manthous CA.
Cardiac ischemia during weaning from mechanical ventilation. Chest 1996;
109:1577 1583.
53. Abalos A, Leibowitz AB, Distefano D, Halpern N, Iberti TJ. Myocardial ischemia
during the weaning period. Am J Crit Care 1992; 1:32 36.
54. Jubran A, Mathru M, Dries D, Tobin MJ. Continuous recordings of mixed venous
oxygen saturation during weaning from mechanical ventilation and the ramica-
tions thereof. Am J Respir Crit Care Med 1998; 158:1763 1769.
55. Lemaire F, Teboul JL, Cinotti L, Giotto G, Abrouk F, Steg G, Macquin-Mavier I,
Zapol WM. Acute left ventricular dysfunction during unsuccessful weaning from
mechanical ventilation. Anesthesiology 1988; 69:171 179.
56. Scheinhorn D. Increase in serum albumin and decrease in body weight correlate
with weaning from prolonged mechanical ventilation (abstract). Am Rev Respir
Dis 1992; 145:A522.
57. Kreimer D, Berger B, Rahim P. Nutritional status of chronic ventilator dependent
patients on admission to a ventilator rehabilitation unit (abstract). Am J Respir
Crit Care Med 1998; 157:A307.
58. Dasgupta A, Rice R, Mascha E, Litaker D, Stoller JK. Four-year experience with a
unit for long-term ventilation (respiratory special care unit) at the Cleveland Clinic
Foundation. Chest 1999; 116:447 455.
Prolonged Mechanical Ventilation 211
59. Latriano B, McCauley P, Astiz ME, Greenbaum D, Rackow EC. Non-ICU care
of hemodynamically stable mechanically ventilated patients. Chest 1996; 109:
1591 1596.
60. Jubran A, Tobin MJ. Pathophysiologic basis of acute respiratory distress in patients
who fail a trial of weaning from mechanical ventilation. Am J Respir Crit Care Med
1997; 155:906 915.
61. Pourriat JL, Lamberto C, Hoang PH, Fournier JL, Vasseur B. Diaphragmatic
fatigue and breathing pattern during weaning from mechanical ventilation in
COPD patients. Chest 1986; 90:703 707.
62. Sassoon CS, Mahutte CK. Airway occlusion pressure and breathing pattern as pre-
dictors of weaning outcome. Am Rev Respir Dis 1993; 148:860 866.
63. Tobin MJ, Perez W, Guenther SM, Semmes BJ, Mador MJ, Allen SJ, Lodato
RF, Dantzker DR. The pattern of breathing during successful and unsuccessful
trials of weaning from mechanical ventilation. Am Rev Respir Dis 1986;
134:1111 1118.
64. Coplin WM, Pierson DJ, Cooley KD, Newell DW, Rubenfeld GD. Implications
of extubation delay in brain-injured patients meeting standard weaning criteria.
Am J Respir Crit Care Med 2000; 161:1530 1536.
65. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative
infusions in critically ill patients undergoing mechanical ventilation. N Engl J
Med 2000; 342:1471 1477.
66. Brook AD, Ahrens TS, Schaiff R, Prentice D, Sherman G, Shannon W, Kollef MH.
Effect of a nursing-implemented sedation protocol on the duration of mechanical
ventilation. Crit Care Med 1999; 27:2609 2615.
67. Capdevila X, Perrigault PF, Ramonatxo M, Roustan JP, Peray P, dAthis F,
Prefaut C. Changes in breathing pattern and respiratory muscle performance par-
ameters during difcult weaning. Crit Care Med 1998; 26:79 87.
68. Vassilakopoulos T, Zakynthinos S, Roussos C. The tension time index and the
frequency/tidal volume ratio are the major pathophysiologic determinants of
weaning failure and success. Am J Respir Crit Care Med 1998; 158:378 385.
69. Yang KL, Tobin MJ. A prospective study of indexes predicting the outcome
of trials of weaning from mechanical ventilation. N Engl J Med 1991; 324:
1445 1450.
70. Bellemare F, Grassino A. Effect of pressure and timing of contraction on human
diaphragm fatigue. J Appl Physiol 1982; 53:1190 1195.
71. Cohen CA, Zagelbaum G, Gross D, Roussos C, Macklem PT. Clinical manifes-
tations of inspiratory muscle fatigue. Am J Med 1982; 73:308 316.
72. Laghi F, Cattapan SE, Jubran A, Parthasarathy S, Warshawsky P, Choi YS, Tobin
MJ. Is weaning failure caused by low-frequency fatigue of the diaphragm? Am J
Respir Crit Care Med 2003; 167:120 127.
73. Laghi F, DAlfonso N, Tobin MJ. Pattern of recovery from diaphragmatic fatigue
over 24 hours. J Appl Physiol 1995; 79:539 546.
74. Appendini L, Purro A, Patessio A, Zanaboni S, Carone M, Spada E, Donner CF,
Rossi A. Partitioning of inspiratory muscle workload and pressure assistance
in ventilator-dependent COPD patients. Am J Respir Crit Care Med 1996;
154:1301 1309.
212 Epstein and Nevins
75. Reinoso MA, Gracey DR, Hubmayr RD. Interrupter mechanics of patients
admitted to a chronic ventilator dependency unit. Am Rev Respir Dis 1993;
148:127 131.
76. Purro A, Appendini L, De Gaetano A, Gudjonsdottir M, Donner CF, Rossi A.
Physiologic determinants of ventilator dependence in long-term mechanically ven-
tilated patients. Am J Respir Crit Care Med 2000; 161:1115 1123.
77. Roussos CS, Macklem PT. Diaphragmatic fatigue in man. J Appl Physiol 1977;
43:189 197.
78. Capdevila X, Lopez S, Bernard N, Rabischong E, Ramonatxo M, Martinazzo G,
Prefaut C. Effects of controlled mechanical ventilation on respiratory muscle
contractile properties in rabbits. Intensive Care Med 2003; 29:103 110.
79. Powers SK, Shanely RA, Coombes JS, Koesterer TJ, McKenzie M, Van Gammeren
D, Cicale M, Dodd SL. Mechanical ventilation results in progressive contractile
dysfunction in the diaphragm. J Appl Physiol 2002; 92:1851 1858.
80. Radell PJ, Remahl S, Nichols DG, Eriksson LI. Effects of prolonged mechanical
ventilation and inactivity on piglet diaphragm function. Intensive Care Med
2002; 28:358 364.
81. Sassoon CS, Caiozzo VJ, Manka A, Sieck GC. Altered diaphragm contractile
properties with controlled mechanical ventilation. J Appl Physiol 2002;
92:2585 2595.
82. Yang L, Luo J, Bourdon J, Lin MC, Gottfried SB, Petrof BJ. Controlled mechanical
ventilation leads to remodeling of the rat diaphragm. Am J Respir Crit Care Med
2002; 166:1135 1140.
83. Anzueto A, Peters JI, Tobin MJ, de los Santos R, Seidenfeld JJ, Moore G, Cox WJ,
Coalson JJ. Effects of prolonged controlled mechanical ventilation on diaphrag-
matic function in healthy adult baboons. Crit Care Med 1997; 25:1187 1190.
84. Hanson P, Dive A, Brucher JM, Bisteau M, Dangoisse M, Deltombe T. Acute
corticosteroid myopathy in intensive care patients. Muscle Nerve 1997;
20:1371 1380.
85. Hirano M, Ott BR, Raps EC, Minetti C, Lennihan L, Libbey NP, Bonilla E,
Hays AP. Acute quadriplegic myopathy: a complication of treatment with steroids,
nondepolarizing blocking agents, or both. Neurology 1992; 42:2082 2087.
86. Grifn D, Fairman N, Coursin D, Rawsthorne L, Grossman JE. Acute myopathy
during treatment of status asthmaticus with corticosteroids and steroidal muscle
relaxants. Chest 1992; 102:510 514.
87. Segredo V, Caldwell JE, Matthay MA, Sharma ML, Gruenke LD, Miller RD.
Persistent paralysis in critically ill patients after long-term administration of
vecuronium. N Engl J Med 1992; 327:524 528.
88. Wijdicks EF, Litchy WJ, Harrison BA, Gracey DR. The clinical spectrum of critical
illness polyneuropathy. Mayo Clin Proc 1994; 69:955 959.
89. Bolton CF, GrandMaison F, Parkes A, Shkrum M. Needle electromyography of the
diaphragm. Muscle Nerve 1992; 15:678 681.
90. Bolton CF, Gilbert JJ, Hahn AF, Sibbald WJ. Polyneuropathy in critically ill
patients. J Neurol Neurosurg Psychiatry 1984; 47:1223 1231.
91. Leijten FS, Harinck-de Weerd JE, Poortvliet DC, de Weerd AW. The role of
polyneuropathy in motor convalescence after prolonged mechanical ventilation.
JAMA 1995; 274:1221 1225.
Prolonged Mechanical Ventilation 213
of Chest Physicians, the American Association for Respiratory Care, and the
American College of Critical Care Medicine. Chest 2001; 120:375S 395S.
123. Nava S, Rubini F, Zanotti E, Ambrosino N, Bruschi C, Vitacca M, Fracchia C,
Rampulla C. Survival and prediction of successful ventilator weaning in COPD
patients requiring mechanical ventilation for more than 21 days. Eur Respir J
1994; 7:1645 1652.
124. Alvisi R, Volta CA, Righini ER, Capuzzo M, Ragazzi R, Verri M, Candini G,
Gritti G, Milic-Emili J. Predictors of weaning outcome in chronic obstructive
pulmonary disease patients. Eur Respir J 2000; 15:656 662.
125. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M, Wallace WA. Outcomes in post-
ICU mechanical ventilation: a therapist-implemented weaning protocol. Chest
2001; 119:236 242.
126. Scheinhorn DJ, Chao DC, Stearn-Hassenpug M. Utility of RSBI in a therapist-
implemented protocol for weaning from prolonged mechanical ventilation in the
post-ICU setting. Am J Respir Crit Care Med 2000; 161:A791.
127. Chao DC, Scheinhorn DJ, Stearn-Hassenpug M. Patient ventilator trigger
asynchrony in prolonged mechanical ventilation. Chest 1997; 112:1592 1599.
128. Hendra KP, Bonis PA, Joyce-Brady M. Development and prospective validation of
a model for predicting weaning in chronic ventilator dependent patients. BMC
Pulm Med 2003; 3:3.
129. Morganroth ML, Morganroth JL, Nett LM, Petty TL. Criteria for weaning from
prolonged mechanical ventilation. Arch Intern Med 1984; 144:1012 1016.
130. Esteban A, Frutos F, Tobin MJ, et al. A comparison of four methods of weaning
patients from mechanical ventilation. Spanish Lung Failure Collaborative Group.
N Engl J Med 1995; 332:345 350.
131. Vitacca M, Vianello A, Colombo D, Clini E, Porta R, Bianchi L, Arcaro G, Vitale
G, Guffanti E, Lo Coco A, Ambrosino N. Comparison of two methods for weaning
patients with chronic obstructive pulmonary disease requiring mechanical venti-
lation for more than 15 days. Am J Respir Crit Care Med 2001; 164:225 230.
132. Cohen IL, Bari N, Strosberg MA, Weinberg PF, Wacksman RM, Millstein BH,
Fein IA. Reduction of duration and cost of mechanical ventilation in an intensive
care unit by use of a ventilatory management team. Crit Care Med 1991; 19:
1278 1284.
133. Saura P, Blanch L, Mestre J, Valles J, Artigas A, Fernandez R. Clinical conse-
quences of the implementation of a weaning protocol. Intensive Care Med 1996;
22:1052 1056.
134. Randolph AG, Wypij D, Venkataraman ST, Hanson JH, Gedeit RG, Meert KL,
Luckett PM, Forbes P, Lilley M, Thompson J, Cheifetz IM, Hibberd P, Wetzel
R, Cox PN, Arnold JH. Effect of mechanical ventilator weaning protocols on res-
piratory outcomes in infants and children: a randomized controlled trial. JAMA
2002; 288:2561 2568.
135. Schultz TR, Lin RJ, Watzman HM, Durning SM, Hales R, Woodson A, Francis B,
Tyler L, Napoli L, Godinez RI. Weaning children from mechanical ventilation: a
prospective randomized trial of protocol-directed versus physician-directed
weaning. Respir Care 2001; 46:772 782.
136. Namen AM, Ely EW, Tatter SB, Case LD, Lucia MA, Smith A, Landry S, Wilson
JA, Glazier SS, Branch CL, Kelly DL, Bowton DL, Haponik EF. Predictors of
216 Epstein and Nevins
152. Scalise PJ, Gerardi DA, Wollschlager CM, Votto JJ. A regional weaning center for
patients requiring mechanical ventilation: an 18-month experience. Conn Med
1997; 61:387 389.
153. Gracey DR, Hardy DC, Koenig GE. The chronic ventilator-dependent unit: a lower-
cost alternative to intensive care. Mayo Clin Proc 2000; 75:445 449.
154. Schonhofer B, Euteneuer S, Nava S, Suchi S, Kohler D. Survival of mechanically
ventilated patients admitted to a specialised weaning centre. Intensive Care Med
2002; 28:908 916.
This page intentionally left blank
7
Procedures in the Intensive Care Unit
I. Introduction
This chapter is divided into four sections that focus on major practical principles
pertaining to the management of patients with arterial catheters, central venous cath-
eters, pulmonary artery catheters, and chest tubes. Each section carefully reviews the
indications, equipment, techniques, and complications of these procedures.
219
220 Vadgama, Au, and Kamangar
B. Indications
The indications for arterial cannulation are shown in Table 1. The appropriate use
limits discomfort caused to the patient and avoids unnecessary complications.
C. Equipment
The equipment used varies depending on the artery to be cannulated and the insti-
tutions preference. Commonly used sites for cannulation are the radial, femoral,
axillary, and dorsalis pedis artery. The equipment includes a 19- or 20-gauge
Teon catheter over a guidewire, 1% lidocaine solution without epinephrine,
and 3.0 or 4.0 silk suture. Monitoring equipment consists of the monitor with
amplier, electronic monitoring equipment, oscilloscope display screen and
recorder, constant ush device, transducer, and uid lled non-compliant
tubing with stopcocks. Accurate and reproducible data is dependent upon the
absence of technical problems. Management of the most common problems is
discussed in the forthcoming section.
D. Techniques
As stated earlier, there are many potential sites for the insertion of arterial cath-
eters. The radial and femoral arteries are the most commonly used. When choos-
ing the ideal site for catheter insertion, several factors have to be considered.
These include hand dominance, patency of collateral circulation, ease for
nursing care, and absence of infection.
The radial artery remains the most frequently cannulated artery. Concerns about
collateral blood ow stem from studies which have demonstrated incomplete
palmar arch anastomosis in 3% to 6% (3) of specimens, that with inadequate
or absent ulnar ow in 12% (4). The modied Allens test, rst described in
1929 (5), has been used to test for collateral ulnar ow. The patients hand is elev-
ated with clenched st, while both radial and ulnar arteries are compressed by the
examiner. This allows the blood to be drained from the hand. Once pallor is pro-
duced, the hand is lowered and the st opened. One artery is released and the time
for color to return to the palm is noted. The procedure is then repeated with the
other artery. Color should return to the hand within seconds, indicating a patent
ulnar artery and intact supercial arch. There remains some controversy over the
validity of Allens test. When compared with Doppler examination, it has been
shown to have a sensitivity of 87% and specicity of only 18% (6). In the
absence of peripheral vascular disease, Allens test does not appear to be a pre-
dictor of ischemia of the hand during or after cannulation (7). Doppler ultrasound
has been shown to be a safe and rapid method to localize the radial artery in
difcult cases (8,9).
Once the radial artery is located either by palpation or Doppler ultrasound,
the wrist is placed in hyperextension. The skin is prepped and draped using sterile
technique. Using a 25-gauge needle, approximately 0.5 mL of the 1% lidocaine is
inltrated down to the periosteum on both sides of the radial artery. A 20-gauge
non-tapered Teon catheter over needle is advanced into the artery at a 308 to 458
angle to the skin, approximately 3 to 5 cm from the proximal wrist crease. When
blood return is noted in the hub, the needle is held in place. The angle is decreased
to 258 and the catheter advanced into the artery. When the needle is removed,
correct position is conrmed by blood return. Use of the guidewire technique
has been shown to be more successful than direct puncture alone (10). Using
this method, once the artery is located the guidewire is advanced into the
vessel (Fig. 1). The catheter is then advanced over the guidewire into the
artery. Upon removal of the guidewire and needle, pulsatile blood return
should be seen. The cannula is then connected to the transducer tubing and
securely sutured into the skin, and dressing is then applied.
spine and the symphasis pubis, lateral to the femoral vein, and medial to the
femoral nerve. The artery is cannulated using the Seldinger technique. Difculty
in cannulation is normally because of atherosclerosis, or prior vascular pro-
cedures. Specic complications to this site include retroperitoneal hemorrhage
and bowel perforation in large inguinal hernias.
E. Complications
patients) and position (sitting or upright), air embolism may travel in a retrograde
direction into the cerebral circulation via the vertebral arteries. Radial artery cath-
eters appear to have a higher incidence of embolization (17). The risk is reduced
by ensuring all air is removed from the tubing prior to ushing, and opening the
ushing valve for no more than three seconds.
Infection
Catheter-related infections are discussed in detail in the Central Venous
Catheters section. As when placing all intravenous access, thorough site prepa-
ration and sterile technique is paramount. The risk of infection increases with dur-
ation of cannulation (18,19). The incidence of arterial catheter colonization has
been reported to be from 5% to 10%, and it is similar in both radial and
femoral catheters (20 22). Routine changing of arterial lines has not been
shown to lower infection rates. Guidewire exchange is not recommended.
Systemic infections may also be caused by contaminated ush solutions or
equipment such as stopcocks or transducer domes. Staphylococcus epidermidis
remains the most common organism causing line infection (23). Gram-negative
organisms and Candida have also been reported, the latter more so in immuno-
compromised patients. Catheter infections should be treated with the appropriate
antibiotic for 7 to 14 days, and complicated cases may require prolonged therapy.
Catheters should be removed immediately in suspected or conrmed catheter
infections.
Most institutions do not replace or relocate arterial catheters routinely for
infection control. Transducers and continuous ush devices are replaced at 72
hours, along with intravenous tubing and ush solution.
output states. Overdamping may occur with catheter kinking or thrombus occlu-
sion, blood in the transducer dome or line, and air bubbles in the tubing or
stopcocks.
Improper zeroing is the most common source of error. The transducer can
be zeroed by opening the transducer stopcock to air and aligning it at the fourth
intercostals space using a level. If the transducer is positioned too high, it will
underestimate the true readings, giving falsely low pressures. Conversely, if it
is positioned too low, it will record falsely high pressures. This process should
be repeated each time the patients position is changed; when signicant
changes in blood pressure occur; and routinely every six to eight hours (29).
Central venous catheters are used extensively both in intensive care units (ICUs)
and general wards. Knowledge of indications for placement, the appropriate site
selection, and complications are essential for all physicians, not only intensive
care specialists.
It has been estimated that over ve million central venous and pulmonary
artery catheters are inserted each year in the U.S.A (30). Their uses include
providing secure access to the central circulation for infusion therapy, hemody-
namic monitoring, nutritional support, temporary transvenous cardiac pacing,
and hemodialysis. The complication rates of venous catheters have been reported
to be over 15% (31 33), and some have been associated with signicant morbid-
ity and economic consequences.
B. Indications
Common indications for central venous catheter placement are listed in Table 4.
The indication for placement and patient-related factors, such as hemodynamic
stability, pulmonary problems, and sites of infection, determine the most appro-
priate site for catheter placement. The most common sites used are the internal
jugular, the subclavian, and the femoral veins. Advantages and disadvantages of
each site are listed in Table 5.
C. General Technique
followed by the catheter itself. The wire is then removed, leaving the catheter
within the vessel. Triple-lumen catheters are generally used over single-lumen
catheters because of their multiple ports.
Sites other than the internal jugular vein, subclavian vein, and femoral
vein include the external jugular vein and brachiocephalic vein. Insertion tech-
niques at the three most frequently used sites are discussed in the following
sections.
Figure 3 (A) Catheterization of the right internal jugular vein. The apex of the triangle
formed by the two heads of the sternocliedomastoid and clavicle is identied. The internal
jugular vein lies lateral to the internal carotid artery. (B) Catheterization of the right sub-
clavian vein. The skin is punctured 2 3 cm caudal to the midpoint of the clavicle. The
needle is then advanced in the direction of the sternal notch, hugging the inferior
surface of the clavicle. Source: From Mc Gee DC, Gould MK. Pulmonary complications
of central venous catheterization. N Engl J Med 2003; 348:1123 1133.
Procedures in the ICU 229
nipple applying constant back pressure to the syringe. The vein is normally ident-
ied within 2 to 5 cm from the skin surface. If the vein is not located, the needle
should be withdrawn slowly, with continued gentle back pressure as the vein is
commonly located during withdrawal of the needle. Repeated attempts should
be just medial or lateral to the initial attempt. Once the vein is located with the
nder needle, the operator can either withdraw the nder needle and insert the
larger needle in the same plane or leave the nder needle in and insert the
larger needle alongside it into the vein. Finder needles may not be necessary if
the cannulation is performed with ultrasound guidance. The syringe is then
removed leaving the hub in place. Backow is checked, ensuring it is not pulsi-
tile, an indication of arterial blood. The J-tip guidewire is then passed through the
needle into the vein, where it should pass easily, encountering no resistance. The
needle is then removed while maintaining control of the guidewire. A small skin
nick contiguous with the guidewire is made using an upward facing scalpel blade.
The dilator is then advanced over the wire using a twisting motion; the operator
should hold on to the guidewire at all times. The dilator should then be with-
drawn, leaving the guidewire in place; gauze is placed at the puncture site to
control oozing. The catheter is threaded through the guidewire into the vein
15 17 cm. At no point should the operator let go of the actual wire. With
the catheter stabilized, the guidewire is removed. Each port of the catheter
should be aspirated and then ushed with normal saline. The catheter is securely
sutured or stapled into the skin and sterile dressing applied. A chest X-ray is then
obtained to check for complications and position of the tip of the catheter.
In the anterior approach, the needle is inserted along the medial edge of
the sternocleidomastoid muscle at the level of the inferior margin of the
thyroid cartilage. The needle is inserted 1 cm lateral to the pulsation of the
carotid artery.
For the posterior (lateral) approach, the needle is inserted at the posterio-
lateral margin of the sternocleidomastoid, approximately 5 cm above the sterno-
clavicular joint. The needle is directed toward the suprasternal notch at a 158
angle to the skin. Venopuncture normally occurs within 5 to 7 cm.
Subclavian Vein
The subclavian vein approach has several advantages. This approach has more
easily identiable bony landmarks, a lower incidence of infection, easier dres-
sing care and maintenance, and improved patient comfort. It is the preferred
route for long-term total parental nutrition and central access in hypotension
when placed by experienced operators. The subclavian vein is a 1 to 2 cm
diameter vessel, a continuation of the axillary vein beginning at the lateral
border of rst rib. It is usually xed by brous attachment directly beneath
the clavicle for 3 to 4 cm before becoming the brachiocephalic vein. The
brous attachments prevent the vein from collapsing even in severe
230 Vadgama, Au, and Kamangar
hypovolemia. The anterior scalene muscle separates the subclavian vein from
the subclavian artery.
As with the internal jugular approach, the patient is placed in
Trendelenbergs position. Additionally, a small rolled towel is placed between
the shoulder blades. There are two basic techniques for inserting the subclavian
catheter, the supraclavicular approach and the infraclavicular approach. Catheter
malposition and pneumothorax rates may be a little lower using the supraclavi-
cular approach. There is no difference in the success rates between both
approaches.
With the patients head turned towards the contralateral side, the area is
prepped with 2% chlorhexidine. Using sterile technique, the area is draped to
create a sterile eld. For the infraclavicular approach, the skin is punctured 2
to 3 cm caudal to the midpoint of the clavicle (Fig. 3B). Following inltration
with 1% lidocaine, an 18-gauge needle attached to a 10-mL syringe is advanced
in the direction of the sternal notch until the tip of the needle abuts the clavicle.
The needle is then walked down the clavicle until it slips under its inferior
edge. Keeping the bevel pointed cephalad, the needle is advanced towards the
suprasternal notch, hugging the inferior surface of the clavicle. If no blood
return is seen, the needle is gently withdrawn, maintaining gentle suction. If
venopuncture does not occur, the needle should be angled in a slightly more
cephalad direction for the next attempt. Once blood return is achieved, the
needle is rmly held, whereas the syringe is detached from the hub. A nger is
placed over the hub to prevent air embolism. The blood return should be non-pul-
sitile. The guidewire is then passed through the needle, turning the bevel as the
guidewire is advanced; the needle is then withdrawn. As with the internal
jugular approach, the catheter is passed over the wire following dilatation. The
catheter is sutured 16 to 17 cm if placed on the right side of the chest and 18
to 19 cm if placed on the left. A post-procedure chest X-ray is taken.
For the supraclavicular approach, the operator is positioned at the head of
the patient on the side to be cannulated. The venopuncture site is the claviculo-
sternocleidomastoid angle, found lateral to the insertion of the clavicular head of
the sternocleidomastoid muscle, above the clavicle. The needle is advanced
underneath the clavicle, at an angle of 158 toward the contralateral nipple. The
vein is usually found at a depth of 1 to 4 cm from the surface.
Femoral Vein
Cannulation of the femoral vein has several advantages. As the artery is
usually easy to palpate, the vein, which lies medially, is usually easy to
locate. If the femoral artery is accidentally punctured, it is easily compressible.
Patients can be cannulated without being placed in Trendelenbergs position,
important when severe orthopnea may be a problem. It also avoids the risk of
pneumothorax. Femoral access is useful during cardiopulmonary resuscitation
when chest compressions and airway management may make subclavian and
Procedures in the ICU 231
internal jugular vein cannulation difcult. Disadvantages of this site include the
higher infection rates and unreliable drug delivery to the heart in low ow states.
The femoral vein is a continuation of the popliteal vein, becoming the
external iliac vein at the inguinal ligament. The femoral vein lies within the
femoral sheath, medial to the femoral artery, which is usually easily palpable.
If possible, the patient is placed in the supine position. The femoral vein lays
approximately 1 to 1.5 cm medial to the femoral artery and 1 to 3 cm below the
inguinal ligament. Once again the site is meticulously sterilized and prepped as
described earlier. An 18-gauge needle is advanced at a 458 to 608 angle, pointing
the needle towards the head. In obese patients it may be necessary to advance the
needle up to the hub. Once venous return is conrmed, the syringe is removed.
Ensuring non-pulsatile blood ow, the guidewire is passed through the needle
and venous catheter placed as described previously.
Failure rates for insertion of central venous catheters have been found to be as
high as 12%, with a complication rate of 10% (34). Commonest causes of
failure or complication include previously difcult catheterization, limited
sites, difcult landmarks, and thrombosed vessels. Studies have clearly demon-
strated the superiority of using ultrasound guidance over blind landmark-
guided techniques in the time taken to place the catheter, the number of attempts,
and the complication rate (35 38). The ultrasound also identies thrombosed or
unusually small vessels, thus avoiding inaccessible sites. Sterile ultrasound probe
covers and needle guidance clips allow cannulation of the vessel as it is seen in
real-time on the monitor (Fig. 4A). Although helpful, the guidance clips can be
cumbersome, and some operators prefer cannulation under real-time conditions.
Real-time ultrasound uses high-frequency sound waves (2 10 MHz), generating
a two-dimensional grey-scale image of the vein and surrounding tissues (Fig.
4B). Fluid such as blood transmits sound completely and is seen as a dark
image. Veins are identied by their non-pulsatile appearance, compressibility,
and distension with the patient in Trendelenbergs position (Fig. 4C). The
artery is generally not compressible on application of gentle pressure. Ultra-
sound-guided catheter placement appears to be the most benecial for the internal
jugular site and should be used whenever available.
placement attempts. Accidental cannulation of the internal carotid artery has been
reported (40), leading to retrograde dissection of the subclavian, innominate
artery, and ascending aorta (41). Subclavian artery puncture is less common,
usually managed by applying pressure above and below the clavicle. Femoral
artery puncture occurs in 5% to 10% of adults (42 44); however, this is
mostly uncomplicated, controlled by local pressure.
Cannulation of the left internal jugular vein has its own complications.
These include injury to the left innominate vein and thoracic duct. Because of
its anatomy, cannulation of the left innominate vein has a greater risk for
perforation. Care should be taken when dilation over the guidewire is attemp-
ted, being careful not to force the dilator beyond the internal jugular vein.
Damage to the thoracic duct can result in chylothorax and rarely chylo-
pericardium (45).
Catheter malposition rates have been reported at 5.3% versus 9.3% for
internal jugular versus the subclavian site (39) and lower for the femoral vein.
When using the internal jugular and subclavian sites, care should be taken to
ensure the catheter tip is not placed too low into the heart. The ideal tip position
is 3 to 5 cm proximal to the caval atrial junction. Fatal tamponade can occur sec-
ondary to perforation of the right atrium (RA) or ventricle. Arrhythmias have also
been reported (46).
Figure 4 (A) An illustration of the ultrasound probe with needle guide attached. Site-
rite ultrasound system. Source: C.R. Bard, Inc., (B) Real-time ultrasound demonstrating
the internal jugular vein and common carotid artery, with the patient in Trendelenbergs
position. (C) Upon application of gentle downward pressure from the ultrasound probe,
the vein collapses but the artery remains clearly seen. (Continued next page.)
Procedures in the ICU 233
Figure 4 (Continued.)
Malposition of the subclavian catheters into the ipsilateral jugular vein has
been well documented (47). Catheters can enter the contralateral brachiocephalic
vein more commonly in left-sided insertions. An example of malposition is
shown in Figure 5A; the central venous catheter was inserted into the left
internal jugular vein and is shown looping back and entering the right internal
jugular vein.
234 Vadgama, Au, and Kamangar
is higher for the subclavian approach than internal jugular, where pneumothorax
only occurs if attempts at catheterization occur close to the clavicle. Treatment of
the pneumothorax depends on the patients clinical condition, severity of under-
lying disease, and the size of the pneumothorax. Although most pneumothoracies
are immediately apparent, they can also occur as a late complication, occurring
several days after subclavian vein catheterization (50).
Catheter and wire embolism occur most commonly when they are with-
drawn through the needle as a result of the shearing action over the barrel of
the needle. The fragments of the catheter can embolize, resulting in potentially
serious complications (51). Shearing of the wire is more common than of the
catheter. Air embolism occurs most commonly when catheters are accidentally
disconnected, becoming open to the atmosphere. Complications range from tran-
sient hypoxemia and chest pain, focal cerebral lesions with hemiparesis, or hemi-
anopia (52) to cardiovascular collapse and death (53). Placing the patient in
Trendelenbergs position during catheter insertion and removal is important in
the prevention of air embolism and its complications. Infrequently, the operator
may lose the guidewire during insertion as is shown in Figure 5B, where the
guidewire can be seen in the descending aorta.
Thrombosis
Infection
Intravascular catheters play an essential role in the management of critically and
chronically ill patients, with millions purchased by healthcare institutions.
Catheter-related infections are associated with increased morbidity, longer
duration of hospitalization, and substantial nancial costs. The incidence of
catheter-related infections varies according to the type of catheter used, the site
of placement, the conditions under which the catheter was placed, frequency of
manipulation, and patient-related factors, such as the severity of their underlying
disease (58). The diagnosis of catheter-related infection is often difcult because
symptoms of fever, chills, and signs of sepsis may occur from other foci of infec-
tion. Catheter-related infection is suggested by several factors. These include
inammation or frank pus at the catheter insertion site or exit site, dysfunction
of the catheter secondary to intraluminal clot, isolation of coagulase negative sta-
phylococci, corynebacterium species or fungi, and clinical improvement on
removal of the catheter.
There should also be an absence of infection from other sites, such as the
lung and urine. The presence of blood stream infection is conrmed by two
236 Vadgama, Au, and Kamangar
positive blood cultures drawn from a peripheral vein, not the catheter itself,
which may be colonized with contaminants. Catheter-related blood stream infec-
tions are unlikely if blood cultures drawn from the catheter are negative (59).
Positive cultures of the catheter tip should demonstrate more than 15 colony
forming units of the same organism (60). The evaluation and management of
catheter-associated infections is discussed in more depth in the chapter on
infection control in the ICU.
With infection of central venous catheters being such a troublesome cause
of nosocomial infections, prevention of infections is paramount. Aseptic tech-
nique and strict adherence to hand washing have been found to signicantly
reduce the rates of catheter-related infections (61 63). Whenever placing a cath-
eter, full barrier precautions are necessary, including sterile gloves, long-sleeved
gowns, surgical masks, and sterile drapes (64). Two-percent chlorhexidine has
been shown to be a more effective cutaneous disinfectant than povidone
iodine (65). Use of the subclavian vein has been shown to have a lower infection
rate compared with both internal jugular and femoral veins (31,66,67). Femoral
veins have the highest rate of infection and hence the use of this site is generally
discouraged. Chlorhexidine silver impregnated catheters have been shown to
have a lower rate of catheter colonization and catheter-related bacteremia (68),
as have minocycline rifampin bonded catheters (69). At present, antimicrobial
impregnated catheters are reserved for use when rates of catheter-related infec-
tions remain high despite appropriate prevention measures.
Catheters should be removed as soon as they are no longer needed. Infec-
tion rates increase after the fth to seventh day (70 72). Routine exchange of
catheters over guidewire or new puncture has not been shown to reduce the
rates of catheter-related infections (73,74) and carries the risk of mechanical
complications. The catheter site should be examined daily for evidence of puru-
lence or erythema, suggestive of infection requiring removal of catheter. The
absence of these signs does not exclude the possibility of a catheter-related
infection.
Antibiotic therapy should be directed by blood and catheter tip cultures, the
severity of the patients underlying disease, and the hospitals infection and
resistance patterns.
Future developments will be aimed at the use of more effective preventa-
tive strategies, including more effective antiseptics and catheters made of
new polymer-antibiotic systems that will inhibit the formation of a bacterial
biolm.
in his arm and into his heart. He took a chest radiograph to conrm the position of
the catheter and published the procedure as a brief report (75). In 1947, Dexter
et al. measured PA pressure by wedging a catheter in a distal branch of a pulmon-
ary artery. Subsequently, they found that the pressure in the wedge position was
the same as the lling pressure in the left ventricle (LV). By measuring the
pressure and oxygen content of blood in the RA, right ventricle (RV), and
pulmonary artery, they were able to diagnose congenital heart disease, valvular
heart disease, and left ventricular failure (76,77). Bedside catheterization of the
pulmonary artery was introduced by Swan et al. in 1970 (78). An inatable
balloon at the tip of a exible catheter enabled the catheter to be directed by
blood ow into the pulmonary artery, thus avoiding the need for uoroscopy-
guided placement. This ow-directed pulmonary artery catheter, also known as
the Swan Ganz catheter, was originally used to assess and guide therapy of
patients following acute myocardial infarction.
With subsequent improvements in PA catheter technology, it is now used in
both diagnosis and management of a wide range of conditions in critically ill
patients.
Despite its extensive use over the past 30 years, there has been increasing
concern over the use and interpretation of the data obtained by the PA catheter,
with some studies suggesting an increased mortality with its use. These contro-
versies will be discussed later.
Diagnosis
Differentiation and assessment of shock
Cardiogenic
Hypovolemic
Septic
Pulmonary embolus
Differentiation of pulmonary edema
Cardiogenic versus non-cardiogenic (e.g., ARDS)
Primary versus secondary pulmonary hypertension
Diagnosis of pericardial tamponade
Diagnosis of left to right intracardiac shunt
Therapeutic
Management of complicated myocardial infarction
Hypotension unresponsive to volume challenge
Hemodynamic instability (cadiogenic shock versus hypovolemia)
Ventricular septal rupture versus acute mitral regurgitation
Right ventricular infarction
Assessment of valvular heart disease
Guide to uid management
Gastrointestinal hemorrhage
Sepsis
Heart failure
Acute renal failure
Burns
Decompensated cirrhosis
Guide to pharmacological therapy
Afterload and preload reducing agents, inotropics, vasopressors, beta-blockers
Management of perioperative cardiac and noncardiac surgical patients with cardiac
instability
Post-operative management of open heart surgical patients
Management of severe pre-eclampsia
Ventilator management (assessment of best PEEP for oxygenation)
Abbreviations: ARDS, acute respiratory distress syndrome; PEEP, positive end-expiratory pressure.
Absolute
Right sided endocarditis
Mechanical tricuspid (or pulmonic) valve
Tumor or thrombus in the right atrium or right ventricle
Terminal illness where aggressive is considered futile
Latex allergy
Relative
Coagulopathy (or anticoagulation)
Recently placed permanent pacemakers or implantable debrillators
supine position. The transducer is xed at this height, the membrane exposed to
atmospheric pressure, and the monitor adjusted to zero. For most catheters, cali-
bration of monitors is not required.
It is important for the monitoring system to have an appropriate frequency
response. Flicking or gently shaking the catheter tip should elicit a brisk high-
frequency response in the waveform. Damped waveforms are caused by air
bubbles, long non-compliant tubing, vessel wall impingement, intra catheter
debris, and loose connections in the tubing. Following insertion, a rapid ush
test is performed similar to that described in the Arterial Catheters section.
D. Insertion
PA catheters can be inserted through a choice of sites. The risks and benets of
the most popular sites are listed in Table 8. Most physicians prefer the right
internal jugular or the left subclavian vein approach. The right internal jugular
is the shortest and straightest path to the heart. The curvature of the catheter uti-
lizes the course of the left subclavian vein into the SVC and right heart.
A wide sterile procedural eld is prepared and aseptic technique and full
barrier precautions are taken. The catheter is ushed in all ports and the balloon
inated to test for leaks. The balloon is then deated. Pressure transducer function
is checked by icking or waving the distal catheter tip gently. Using techniques
described in the central lines section, the appropriate site for insertion is identied,
and the vein is located and cannulated using the Seldinger technique. A dilator is
passed through the introducer, and together these are advanced over the guidewire
into the vein. The guidewire and dilator are then removed simultaneously, leaving
the introducer in the vessel. The catheter is introduced through the sterile sheath
adapter, which at the end of the procedure is pulled over the catheter. The distal
end of the sleeve adapter attaches to the introducer sheath limb. The catheter con-
tains markings spaced 10 cm apart from the distal tip. Using these markings, and
continuous pressure monitoring, the catheter is advanced into the RA. Table 8 illus-
trates the appropriate distances for the three most frequently used venous sites.
Table 8 Comparison of the Three Most Frequently Used PA Catheter Insertions Sites
The catheter is advanced until it is in the RA; the appropriate right atrial
waveform should be seen with characteristic a and v waves. Right atrial blood
samples are obtained, if needed, and the pressure recorded. The balloon is then
inated with the recommended amount of air (no more than 1.5 mL) until a
small amount of resistance is felt. If no resistance is felt, one should suspect
balloon rupture and re-examine the catheter immediately. If signicant resistance
is encountered, then one should suspect malposition. The catheter is withdrawn
and re-advanced to a new position. With the balloon inated, the catheter is
advanced into the RV. As it crosses over the tricuspid valve into the RV, the
pressure waveform changes and a rise in systolic pressure is seen (Fig. 6). The
pressures within the RV are recorded. The catheter is then advanced until the
diastolic pressure increases above that seen in the RV. At this point, the catheter
tip has moved across the pulmonary valve into the PA. In this position, the
characteristic dicrotic notch appears in the waveform, indicating closure of
the pulmonic valve. If the RV tracing is still seen after 40 cm of the catheter is
introduced, it is likely that the catheter is coiled within the RV. If this occurs,
the balloon is deated and catheter withdrawn until the RA waveform is seen.
The balloon can then be inated and advanced again as described. The inated
catheter is advanced into the PA until a fall in pressure and a change in waveform
is seen. This is the pulmonary capillary wedge pressure (PCWP). The PCWP is
recorded as a mean value, not systolic or diastolic, and is measured at the end of
expiration. Once this is recorded, the balloon should be deated, at which point
the PA tracing should reappear. The balloon ination volume needed to change
the PA tracing to the PCWP should be noted. If the volume needed is signicantly
lower than the recommended volume of 1.5 mL or subsequent PCWP readings
require smaller volumes, then the catheter tip has migrated too far peripherally
and should be pulled back immediately (with the balloon deated). Characteristic
waveforms in the right heart chambers, pulmonary artery, and PCWPs are shown
in Figure 7.
As a rule, the catheter tip should never be advanced until the balloon is
inated and must always be deated if the catheter is withdrawn. In patients
with an enlarged RA and RV, it may be difcult to advance the PA catheter
into PA and wedge position. The catheter softens as it is exposed to body
temperature, which makes it more difcult to pass in patients with pulmonary
hypertension. The catheter stiffness can be maintained by placing it in a
refrigerator or freezer prior to insertion. Alternatively, PA catheters that have
a guidewire within one of the lumens to maintain its form may be used. In dif-
cult cases, uoroscopy is useful in guiding the PA catheter to the appropriate
position.
Once the optimum position has been located, the catheter is secured by
suturing or taping it to the skin. The position of the catheter based on the centi-
meter markings at the site of insertion is noted. An appropriate dressing is then
applied. Finally, a chest X-ray is taken to conrm the catheter position and
exclude a pneumothorax. Daily chest X-rays are recommended.
242 Vadgama, Au, and Kamangar
(A)
(B)
(C)
(D)
The right atrial pressure waveform reects the venous return to the RV and,
in the absence of tricuspid valve pathology, is an indicator of right ventricular end
diastolic pressure. The normal atrial pressure waveform consists of ve com-
ponentsthree positive deections, the a, c, and v waves, and two negative
deections, the x and y descents (Table 9). The normal pressures recorded by
the PA catheter are listed in Table 10. Changes in the right atrial pressure wave-
forms and pressures within the heart in various pathological conditions are dis-
cussed later in this section. Using data obtained from the PA catheter, useful
hemodynamic parameters can be derived. Table 11 shows how these values
are derived and their normal values.
When the mitral valve is open, there is no barrier between the tip of the catheter
and the left heart. Ination and wedging of the balloon closes off ow from the
right heart, so the catheter only reects a backward pressure from the left heart.
Ideally, left ventricular end diastolic volume (LVEDV) is the desired value for
hemodynamic monitoring, but it unfortunately requires cardiac catheterization.
244 Vadgama, Au, and Kamangar
Abbreviations: BSA, body surface area; CO, cardiac output; CVP, central venous pressure; CaO2,
arterial oxygen content; CcO2, pulmonary capillary blood oxygen content; CvO2, venous oxygen
content; CvO2, mixed venous oxygen content; HR, heart rate; MAP, mean arterial pressure;
PAP, pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; VO2, oxygen
consumption.
Thermodilution CO
Cardiac Output
The PA catheter is equipped with a thermistor at the tip, which allows CO to be
calculated using the Ficks principle (91). In practice, 5 to 10 mL of cold D5W or
saline is injected into the RA via the proximal port of the PA catheter. The
average of three separate measurements should be obtained at one-minute inter-
vals. The solution is rapidly injected, preferably at end-expiration.
_ 2 (mL= min )
VO
CO mixed venous O2 content (Cv O2 )
arterial O2 content (Ca O2 )
or
_ 2
VO
CO Cv O
Ca O2 2
_ 2
VO
Cv O2 Ca O2
CO
The oxygen content is the sum of both the oxygen bound to hemoglobin and that
in solution, therefore
SaO2 and SvO2 represent arterial oxygen saturations and mixed venous oxygen
saturations, respectively. The amount of oxygen (in milliliters) that can bind to
1 g of hemoglobin (Hb) is 1.34. The solubility of oxygen in the blood is 0.003.
Under normal conditions this is negligible, therefore
Cv O2 Hb Sv O2 1:34
Ca O2 Hb Sa O2 1:34
250 Vadgama, Au, and Kamangar
Substituting these simplications with SvO2 and SaO2 into the Fick
equation gives
_ 2
VO
Sv O2 Sa O2
CO Hb 1:34
It becomes easy to see that the four factors that can inuence SvO2 are arterial
oxygen saturation, oxygen consumption, CO, and hemoglobin level. Therefore
changes in SvO2 may reect changes in any or all of these parameters.
The normal SvO2 is 75% (range: 60 80%). When less than 60%, this
suggests anaerobic metabolism (e.g., lactic acidosis), and values below 40%
are associated with severely insufcient tissue oxygenation due to shock.
When the hemoglobin concentration, oxygen consumption, and arterial oxygen
saturation are constant, changes in SvO2 are due to changes in CO. Continuous
mixed venous oxygen saturation is possible with PA catheters by the use of ber-
optic reectance spectrophotometry. The continuous mixed oxygen saturation
reading from the catheter is therefore a useful monitor of CO in hemodynamically
unstable patients. Increases in SvO2 would indicate improvement in CO or
tissue perfusion, and conversely, a decline would represent further deterioration
of hemodynamics and tissue compromise. These changes may be evident prior to
other changes in routinely monitored variables (104). In addition, there may be
metabolic derangements, such as lactic acidosis, that may be persist despite
normal hemodynamics that can be identied with SvO2 monitoring. Despite the
potential benet of continuous monitoring of SvO2, there remains considerable
uncertainty of its role in the management of critically ill patients and advantage
over other available monitoring devices (105). The mixed venous oxygen satur-
ation is affected in different clinical conditions as shown in Table 12.
Mechanism Example
Increased SvO2
Increased arterial O2 supply by increased Sepsis
cardiac output
Decreased tissue demand by decreased VO2 Anesthesia, coma, hypothermia,
hypothyroid, cirrhosis, left to right
cardiac shunt
Decreased SvO2
Decreased cardiac output Hypovolemic shock, acute myocardial
infarction
Decreased arterial oxygen (CaO2) Hypoxemia
Decreased Hemoglobin Anemia, hemorrhage
Increased tissue demand by increased VO2 Fever, seizures, burns, hyperthyroid,
pain, shivering (34 37)
Heart Failure
There are six conditions in patients with heart failure for which PA catheteriza-
tion is warranted:
1. To differentiate pulmonary edema caused by heart failure or by pul-
monary disease or both.
2. To distinguish and guide therapy between cardiogenic and non-
cardiogenic shock.
3. To guide therapy in patients with concomitant manifestations of
forward (hypotension, oliguria, and azotemia) and backward
(dyspnea and hypoxemia) heart failure.
4. To diagnose pericardial tamponade when clinical assessment is incon-
clusive and echocardiography is unequivocal.
5. To guide perioperative management in selected patients with decom-
pensated heart failure undergoing intermediate or high-risk non-
cardiac surgery.
252 Vadgama, Au, and Kamangar
Pulmonary Hypertension
There are four conditions for which PA catheterization is warranted in the man-
agement of pulmonary hypertension according to the ACC expert consensus:
1. To exclude postcapillary (elevated PAWP) causes of pulmonary
hypertension.
2. To assess the diagnosis and severity of precapillary (normal PAWP)
pulmonary hypertension.
3. To determine the safety and efcacy of vasodilator therapy based on
acute hemodynamic response.
4. To evaluate different hemodynamic variables before lung transplan-
tation (79).
Although there were no conditions in which right heart catheterization was
not warranted, there were conditions in which reasonable differences in opinions
between the ACC experts existed. These conditions are the evaluation of the
long-term efcacy of vasodilator therapy, in particular prostacycline, and the
exclusion of signicant left to right or right to left intracardiac shunts.
Figure 10 The tracing in a patient with mitral regurgitation. The characteristic tall v
waves are demonstrated.
the LA is distended and non-compliant due to left ventricular failure from any
cause, that is, dilated cardiomyopathy (108,109).
Tricuspid Regurgitation
Tricuspid regurgitation usually occurs in the setting of pulmonary hypertension
and right ventricular dilatation. It is associated with elevated right atrial and
right ventricular end-diastolic pressures. It usually represents chronic valvular
insufciency. A prominent early right atrial v wave with a steep y descent
is seen. In severe tricuspid regurgitation, the tracings of the right atrial and
right ventricular pressures may look similar (110).
Cardiac Tamponade
Constrictive Pericarditis
Chronic brous thickening of the pericardium can lead to constrictive pericardi-
tis. Causes of constrictive pericarditis include tuberculosis, malignancy, post-
chest radiotherapy, and idiopathic causes. Acute or subacute presentations are
seen after cardiac surgery. On examination, the jugular venous pressure (JVP)
is commonly raised and may rise or fail to fall with inspiration (Kussmals
sign), which distinguishes this condition from cardiac tamponade, in which
there is a decline in the right atrial pressure (JVP) with inspiration. The JVP
also shows a prominent y descent. At catheterization, low CO is seen with
prominent x and y descents on the atrial waveform, producing m or w
waveform appearance. The waveforms have a characteristic diastolic dip and
plateau pattern (or square root sign) in the ventricular waveform, reecting
abrupt termination of the ventricular lling due to the rigid pericardium. Equili-
bration of right ventricular and left ventricular end diastolic pressures
(,5 mmHg) is often seen, although this can occur in patients with heart
failure or acute volume overload. Unlike pericardial tamponade, the PCWP
may be as high as 20 to 25 mmHg and similar in appearance to the RA waveform.
Shock States
In clinical practice, specic conditions are associated with characteristic ndings
in the PAWP, CO, systemic, and pulmonary vascular resistance. This can be
Procedures in the ICU 255
or trauma. These patients usually will be followed by care in the ICU and with the
information provided by the PA catheter, intensivists are be able to provide
earlier goal-directed therapy. In low-risk patients undergoing low-risk surgery,
e.g., knee replacement, the PA catheter is not necessary. Hence, the use of the
PA catheter depends on clinician knowledge, patient classication, and the sever-
ity of the surgery.
In summary, there is no absolute requirement for the use of the PA catheter
in critically ill patients. Many studies over the last four decades have been done,
but none have provided conclusive support or against its routine use. The above
examples are suggestions and recommendations for the use of the catheter in
specic clinical situations. Clinician knowledge about the insertion and compli-
cations of the catheter, appropriate interpretation of the data, and appropriate
application of the data are crucial for the safe use of this device, and this expertise
may be lacking by some. Current training, credentialing, and quality improve-
ment issues need to be evaluated at each institution. Clinicians should continue
to carefully weigh the risks and benets of the PA catheterization in management
of individual patients before its insertion.
G. Complications of PA Catheters
Complications of PA Catheterization are listed in Table 14. These have been sep-
arated into complications related to placement of the catheter and complications
related to the catheter itself. Dysrhythmias, such as atrial brillation and utter,
premature ventricular tachycardia, or brillation, may occur during catheter
insertion or withdrawal but usually resolve spontaneously after the catheter is
advanced or withdrawn through the right heart chambers (121). Dysrhythmias
lower mortality in patients in septic shock with the rationale that better tissue per-
fusion was obtained (81,84). In a meta-analysis of studies published between
1970 and 1996, there was decreased morbidity noted in patients with PA catheters
compared with those without, calculating a relative risk ratio of 0.78 (0.65 0.94)
for the incidence of organ failure (151).
However, there are several investigations that are contrary to this experience.
Gore et al. reported that the use of the PA catheter was associated with a higher
fatality rate in patients with congestive heart failure, 44.8% compared with
25.3% (P , 0.001) (85). In patients with hypotension, the case fatality rate was
48.3% in those with PA catheterization, compared with 32.2% for those without
(P , 0.001). Length of hospital stay was signicantly longer in patients receiving
a PA catheter. A study from Israel reported a similar lack of benet in an observa-
tional cohort of 5841 patients with acute myocardial infarction (86). In 1996,
Connors et al. (152) performed a post hoc analysis on 5735 critically ill patients
in nine disease categories admitted to the ICUs in ve medical centers as part of
the SUPPORT study. This was an observational study, and as part of their retro-
spective analysis, they developed a propensity score to adjust for disease severity
and other demographic factors. They identied two groups of patients with similar
propensity scores. Patients who underwent PA catheterization had higher mortality
rates, longer length of stay in the hospital, and subsequently higher hospital costs
than those who did not (87). This suggested that the PA catheter was associated
with increased morbidity and mortality and spawned considerable debate about
the use of the device in the management of critically ill patients (153). It should
be noted that the device was never subject to a prospective evaluation of efcacy
prior to its introduction into clinical use.
Subsequent, prospective randomized trials have not identied an increase in
mortality associated with the use of the PA catheter. The randomized, controlled
trial of PA catheters in high-risk surgical patients by Sandham et al. (88) has been
previously outlined. There was no signicant difference in mortality rate between
the two groups, but there is a signicant increase in the incidence of pulmonary
embolism in the catheter group (8 events vs. 0 events). The authors found no
benet to routine insertion of the PA catheter perioperatively in high-risk surgical
patients and concluded it was not warranted (88). Richard et al. randomized 676
with shock, ARDS, or both to early use (within 24 hours) of a PA catheter in man-
agement. The placement of a PA catheter did not signicantly affect mortality or
morbidity as dened by organ system failure, need for vasoactive agents, duration
of mechanical ventilation, ICU, or hospital stay. On day 28, the mortality rate in
the PA catheter group was 59.4%, and the control group was 61.0%. They con-
cluded that there was no observed benet with the insertion of a PA catheter in
these patients (89).
sponsored PAC-Man trial (154). The latter trial has now been completed. It
enrolled 1041 patients from 65 United Kingdom intensive care units with a
broad range of diagnoses. The focus was on critically ill patients with a perceived
need for PA catheter placement by their physician. The study demonstrated no
difference in hospital mortality in critically ill patients managed with or
without pulmonary artery catheters. There were more complications related to
catheter insertion and changes in management based on results from the catheter
were modest. It is unclear whether PA catheter use improved outcome. Another
trial evaluated the utility of the PA catheter in the management of patients with
severe congestive heart failure (New York Heart Association classes III and IV)
(155). This is the Evaluation Study of Congestive Heart Failure and Pulmonary
Artery Catheterization Effectiveness (ESCAPE) trial, which has completed
enrollment and is currently undergoing analysis. The results from these three
major studies should further dene the role of the PA catheter in the management
of critically ill patients.
J. General Principles in PA Catheter Use
Until the results of these future studies are available, clinicians using the PA cath-
eters for hemodynamic monitoring should carefully assess the risk-to-benet
ratio on an individual patient basis. The indications, insertion techniques, equip-
ment, patients health status, and interpretation of data obtained from the catheter
should be taken into consideration before the PA catheter is used (156). PA cathe-
terization should only be used as a complementary device to diagnose and guide
therapy after clinical evaluation and treatment have been exercised. It is import-
ant to recognize that diagnostic tests and monitoring devices do not determine
clinical outcomes, but that outcomes are inuenced by therapy directed by infor-
mation obtained with the catheter. Minimizing the complications associated with
its placement and limiting the duration of placement should be the major guiding
principles in its use. The PA catheter should be removed as soon as information
derived from its placement is no longer clinically useful. In some cases, this may
occur within a few hours, and certainly placement should not exceed three to four
days in any case. Although there is great hope that pending research studies will
further dene the role of the PA catheter in patient management, it is important to
recognize that there will remain areas of uncertainty in its use. Even with these
trials, there may still be signicant debate surrounding the use of the PA catheter.
It is probably safe to conclude that there will always be a role for this device in the
management of critically ill patients. Therefore, it behooves all involved to be
cognizant of its potential and limitations.
Chest tube placement and management is important not only in the ICU but also
in Advanced Trauma and Life Support (157).
B. Anatomy
The parietal pleura lines the thoracic wall and covers the superior surface of the
diaphragm. It can be divided into four sections: the costal pleura, cervical
pleura, mediastinal pleura, and the diaphragmatic pleura. The parietal pleura is
supplied with blood from the intercostal arteries and branches of the internal thor-
acic artery. It is innervated by the intercostal nerves and the phrenic nerve. Pain
from these nerve bers may be referred to the thoracic and abdominal walls,
neck, and shoulder (phrenic nerve). At the hilum and pulmonary ligament, the
pleura becomes the visceral pleura as it is reected and adheres to the surface
of the lung. The visceral pleura is mainly perfused by the systemic circulation.
The pleural layers are in close opposition to each other, with the potential space
between them containing a thin layer of uid. This uid acts as a lubricant, allow-
ing the pleural membranes to slide easily against each other. Under normal
circumstances, there is a negative intrapleural pressure of 2 to 5 cm of water.
Disruption in visceral or parietal pleura results in pneumothorax or hemothorax.
Fluid can accumulate in the pleural space because of the changes in hydrostatic
or oncotic pressures due to impaired lymphatic drainage or inammatory
diseases.
C. Indications
Chest tubes are inserted to evacuate air or uid. The indications for chest tube
placement are listed in Table 15 and also discussed in the following sections.
Pneumothorax
Pneumothorax is the most common indication for chest tube placement. Patients
may present with symptoms of dyspnea, tachypnea, and chest pain. Clinical
examination reveals diminished breath sounds and hyper resonance to percus-
sion. The diagnosis is usually conrmed by chest X-ray. Inspiratory and expira-
tory lms are helpful in equivocal cases as is chest computed tomographic
scanning. Spontaneous or iatrogenic pneumothorax of less than 25% volume in
Hemothorax
Hemothorax refers to a collection of blood in the pleural cavity. The pleural uid
hematocrit is greater than or equal to 50% of the blood hematocrit. Etiology of
hemothorax can be spontaneous, iatrogenic, or traumatic in origin. Spontaneous
hemothorax may result from primary or metastatic malignancy in the lung or
pleura, pulmonary infarcts, arteriovenous malformations, and necrotizing pul-
monary infections. Common traumatic and iatrogenic causes include rib frac-
tures, attempted thoracentesis, or chest tube placement resulting in damage to
intercostal or internal mammary arteries. Large bore drainage tubes in this
setting help in the assessment of the degree of blood loss. Evidence of persistent
bleeding following trauma, with hemodynamic instability, is an indication for
open thoracotomy. Late-stage complications of unresolved hemothorax include
empyema and brothorax.
Pleural Effusion
Fluid can accumulate in the pleural space for a number of reasons. These include
changes in hydrostatic and oncotic pressures, increased permeability of the
microvasculature by inammatory mediators, and impaired lymphatic drainage.
Thoracentesis is usually done to determine whether the effusion is transudative or
exudative in origin. Transudates result from imbalances in hydrostatic and
oncotic pressures. Causes of transudative effusions include cirrhosis, congestive
cardiac failure, and nephrotic syndrome. Management of transudative effusions is
aimed at identifying and treating the underlying cause. Chest tubes are usually
not indicated in the management of transudates except in selected cases of
massive effusions. Exudates are formed by the disruption in the integrity of the
endothelial membranes of pleural capillaries and venules. Inammatory pro-
cesses result in an increased permeability of these membranes with subsequent
protein leak. Impaired lymphatic drainage decreases the removal of protein in
the pleural space. Fluid from the peritoneum can also cause exudates. Infection,
malignancy, and inammatory diseases are all causes of exudative pleural
effusions.
Chemical analysis of the pleural uid distinguishes transudative effusions
from exudates. Measurement of both serum and pleural uid protein and LDH
is the most practical method of distinguishing the two. Lights criteria (163),
shown in Table 16, are used to distinguish exudative pleural effusions. If one
of these three criteria is present, the uid is almost always an exudate. If none
of these three are present, it is virtually always a transudate. If paired testing is
not an option, any one of the three tests on a single pleural uid sample shown
in Table 16 will identify an exudate (164).
Procedures in the ICU 265
Decubitus lms help determine if the pleural effusion is free owing. Safe
thoracentesis requires at least 1 cm of visible uid on the decubitus lm. Ultra-
sound, as already mentioned, can help determine the most appropriate site for
thoracentesis or chest tube placement. In malignant effusions, chest tubes are
placed for both symptomatic improvement and, often, subsequent instillation
of sclerosing agents for pleurodesis. Sclerosing agents, such as talc, tetracycline
derivatives, and bleomycin, are used once the lung has fully re-expanded and
pleural uid drained.
Chylothorax
Chyle is lymphatic uid of intestinal origin carried in the thoracic duct across the
chest. It has a high content of triglycerides, typically over 110 mg/dL, and lym-
phocytes. The thoracic duct enters the mediastinum on the right side of the chest
and crosses over to the left at the level of the fth thoracic vertebrae in most
adults. Between 1.5 and 2.4 L of uid ow through the thoracic duct in a day.
The causes of chylothrorax are divided into non-traumatic and traumatic
(Table 17). Lymphoma is the most common non-traumatic cause; surgical pro-
cedures account for most traumatic cases. Treatment in non-traumatic cases is
aimed at the underlying cause. Failure of chemotherapy or radiotherapy or the
development of symptoms due to an increasing pleural effusion require chest
tube drainage. Symptomatic effusions of idiopathic origin are treated with
chest tube drainage and either decreased oral intake or parental nutrition with
or without medium chain triglycerides. Most cases resolve within two to three
weeks. Treatment options for resistant cases include pleurodesis, pleurectomy,
and ligation of the thoracic duct at the hiatus.
Prolonged and copious drainage from chest drains can result in weight loss,
progressive hypoproteinemia, and immunosuppression. In these cases, earlier
surgical intervention is warranted.
266 Vadgama, Au, and Kamangar
Causes
Non-traumatic
Malignant Lymphomatous, non-Lymphomatous (e.g., pulmonary and
mediastinal malignancies)
Non-malignant Idiopathic, others (e.g., cirrhosis, tuberculous, venous thrombosis)
Traumatic
Surgical Cardiovascular, aortic lobectomy, pneumonectomy etc.
Nonsurgical Penetrating, nonpenetrating trauma to the neck,
thorax, upper abdomen
D. Contraindications
E. Equipment
The equipment necessary for chest tube placement is listed in Table 18. This
includes the appropriate-size chest tube, a drainage device with or without suction
source, connecting hoses with connectors, and a tray with insertion instruments.
F. Drain Size
Modern chest tubes are plastic, pliable, fenestrated tubes made of minimally
thrombogenic material. Chest tube sizes range from number 6 to 40 French.
4F 0.053 1.35
8F 0.105 2.7
10F 0.131 3.3
16F 0.210 5.3
18F 0.236 6.0
22F 0.288 7.3
26F 0.341 8.7
28F 0.367 9.3
32F 0.419 10.7
36F 0.471 12.0
38F 0.498 12.7
40F 0.524 13.3
Table 19 lists the internal diameter correlate to the French grading system for a
range of chest tube sizes. The use of large bore tubes has previously been rec-
ommended (165 167) as it was felt they decreased the incidence of blockage,
particularly in thick malignant or infected uid. Most physicians now use
smaller tubes (10 14 French) as studies have shown these to be as effective as
the larger bore tubes (168,169), more comfortable, and better tolerated (170).
There remains intense debate over the optimal drain size. The 9 French tubes
have been used with success rates of up to 87% in pneumothoracies (171).
Ultrasound-guided insertion of small-bore tubes for treatment of malignant
pleural effusions for sclerotherapy has been well studied with good effect
(169,172,173). Small- and medium-bore chest tubes are usually placed using
the Seldinger technique.
Complicated parapneumonic effusions not amenable to a single large cath-
eter may require more than one small-bore tube placed under image guidance.
Larger bore tubes (32 40 French) are recommended for hemothorax, to allow
for successful drainage and assessment of continuing blood loss (174). The
larger bore tubes are inserted by blunt dissection, which allows the operator to
manually break down loculations if necessary.
G. Technique
Prior to inserting the chest tube, it is important to place the patient in the correct
position. The preferred position is on the bed, slightly rotated with the arm on the
side of the lesion behind the patients head, exposing the axillary area (166,175).
Alternatively, the patient can be sitting upright, leaning on a pillow over a table.
Aim for insertion in the triangle of safety (157), which lies between the
anterior and posterior axillary lines, through the fourth or fth intercostal
268 Vadgama, Au, and Kamangar
space, just above the rib, thus avoiding the intercostal neurovascular bundle.
Under sterile conditions, the area is prepped with pivodine iodine solution.
Local anesthesia is inltrated into the site of insertion. A small gauge needle is
used to raise a small bleb prior to inltration of the interocostal muscles and
pleural surface. Up to 30 40 mL of lidocaine may be needed to ensure adequate
anesthesia.
Thoracentesis is often initially performed to conrm the presence of air or
uid at the site of insertion. Chest tubes can be inserted by either blunt dissection
with or without the use of a trochar or the Seldinger technique. Whichever tech-
nique is used, the use of substantial force must be avoided to avoid penetration
and damage to vital intrathoracic structures. For blunt dissection, a 2-cm incision
is made parallel to the intercostal space, immediately above the rib. A Kelly
clamp is used for blunt dissection (Fig. 12), creating a subcutaneous tunnel to
the intercostal space. The closed clamp is gently inserted through the parietal
pleura over the superior aspect of the rib. Once through the parietal pleural,
the clamp is gently opened to spread the parietal pleura and intercostal
muscles. For a chest tube similar to the size of a nger, the tract should be
gently explored with a nger, performing a gentle sweep and taking care not
to disrupt rm organized adhesions. The proximal end of the tube, rmly held
Figure 12 Appropriate angle of penetration into the pleural space during the blunt dis-
section technique. Source: From Miller, KS, Sahn, SA. Chest tubes, indications, tech-
niques, management and complications. Chest 1987; 91(2):258 264.
Procedures in the ICU 269
by the Kelly clamp, is then inserted into the pleural space. If using a trochar, care
must be taken to ensure the sharp point of the trochar is retracted a few centi-
meters from the chest tube tip. Trochar insertion is relatively fast but runs the
risk of impaling the lung or other organs.
The position of the tip of the chest tube should ideally be aimed anteriorly
towards the apex of the lung in the treatment of a pneumothorax. For free-owing
effusions, chest tubes are aimed downwards and in the posterior direction, into
the costovertebral angle. For basal and lateral empyemas or loculated collections,
the chest tube is placed in the most dependant region. However, any tube position
can be effective at draining air or uid, and an effective drain should not be re-
positioned solely because of its radiographic position. The depth of insertion of
the chest tube ranges from 5 to 15 cm, ensuring all sideports are within the chest,
and the proximal port must be at least 2 cm beyond the rib margin. The centi-
meter markings on the side of the chest tube start from the most proximal port;
therefore, a tube at a 5-cm mark at the skin means it is 5 cm from the most prox-
imal port. The nal depth is dependant on the ease of aspiration of the pleural
uid and the patient size.
Correct insertion of the chest tube is conrmed by visualization of conden-
sation within the tube with respiration or the drainage of pleural uid. Two simple
horizontal mattress sutures are placed on either side of the insertion site to anchor
the tube. Complicated purse string sutures should be avoided as they convert a
linear incision into a circular wound, which is painful and can leave a prominent
scar (176). A sterile dressing is then applied over the site, which is covered and
secured with surgical tape. Occlusive petroleum dressings are generally not
recommended.
Small- and medium-bore catheters are usually inserted with the aid of guide
wire by the Seldinger technique (170,177,178) (Fig. 13). This technique is fast
and relatively safe, making a smaller incision and insertion, causing less dis-
comfort than the blunt dissection method. It is, however, difcult to control the
direction of the tube entry into the pleura, and advancement may be difcult if
adhesions are present. Using this technique, the introducer needle is advanced
over the superior border of the rib into the pleural space. Air or uid is aspirated
to conrm the correct position. The guide wire is then advanced through the
needle into the pleural space without any resistance. The needle is then
removed, leaving the guide wire in place. Ensuring adequate analgesia, a series
of dilators are passed over the guide wire into the pleural space. Introduction
into the pleural space is facilitated by rotating and advancing the dilators in
the same plane of the guidewire to prevent kinking. The chest tube assembly is
advanced into the pleural space over the guide wire. The guide wire and chest
tube inserter are removed, leaving the chest tube in place. All ports of the
chest tube should be in the pleural space.
Once the chest tube is inserted and secured, it is connected to the pleural
drainage system. All connections between the chest tube and its drainage
system should be tight and securely taped.
270 Vadgama, Au, and Kamangar
Figure 13 Percutaneous insertion of a chest tube using the Seldinger technique. Source:
Thal-Quick Chest Tube Instruction Manual. Cook Critical Care. Cook Incorporated, 1987.
H. Complications
Complications of chest tube placement can be divided into three categories:
placement, maintenance, and discontinuation (Table 20).
Complication rates from chest tube insertion in blunt trauma have been
shown to be between 9% and 21% (179,180). Insertion alone has a complication
rate of 1% to 2% (179,181). Perforation of the RA, RV, and abdominal organs has
also been reported (182 185), as has chylothorax from thoracic duct injury
(186). Perforation of the diaphragm is avoided by inserting the chest tube no
lower than the fth intercostal space. Ipsilateral hemidiaphragmatic palsy from
isolated injury to the intrathoracic phrenic nerve has also been reported (187).
Subcutaneous emphysema from the thoracostomy site involving the chest wall
and extending into the neck can also occur, but this is usually a cosmetic problem.
Rapid re-expansion of previously collapsed lung may result in unilateral
pulmonary edema (188), which may be fatal (189). It can occur in young patients
with pneumothoracies (190) and also following removal of large volumes of uid
or after removal of an obstructive tumor (191 193). The development of
re-expansion pulmonary edema probably correlates with the amount of negative
intrathoracic pressure, which in turn is related to the rate of uid removal. The
onset of cough or chest tightness is an important warning sign to end the
procedure. Most experts recommend the removal of 1 to 1.5 L at any one time.
Supportive treatment is usually sufcient.
Secondary infection of the pleural space following chest tube insertion is a
rare complication that occurs most frequently following treatment of traumatic
hemothorax.
Once secured, the chest tube is attached to a drainage device. Although there are
many commercially available devices, they consist of one or more of the follow-
ing: (1) water seal, (2) drainage tap, (3) pressure control change, and (4) connect-
ing hoses. The water seal prevents air from entering the pleural cavity whilst air
or uid is being drained. Most institutions use a three-chamber system (Fig. 14).
The closed water seal is placed under water at 3 cm. This allows the operator to
see air bubble out as the lung re-expands in pneumothorax or the uid evacuation
rate in effusions or empyemas. If suction is required, it is performed via the water
seal at a level of 15 to 20 cm of water. This water level should be checked daily to
ensure an adequate water seal and suction regulator is maintained. If suction is
applied, bubbling is seen in the suction control chamber. Periodic milking or
stripping of the connecting hoses is generally discouraged because as it has not
been shown to increase tube patency (194). If continuous bubbling through the
water seal is seen, intermittent clamping of the connecting hoses, starting at
the drainage device and moving proximally to the chest wall, should reveal the
site of leakage. Connecting sites are common sources of leaks. Persistent air
leaks may be due to bronchopleural stulas. If there is no variation of the
water seal with respiration, then occlusion of the chest tube should be suspected.
Saline irrigation should initially be attempted. Prolonged clamping of the chest
tube should be avoided as this may result in tension pneumothorax. Serial
chest X-rays should be obtained to assess tube position, exclude complications
272 Vadgama, Au, and Kamangar
Figure 14 Example of a three-chamber water seal chest drain. Source: Atrium Ocean
Water Seal Chest Drain, Atrium Medical Corporation, 2004.
References
22. Thomas F, Burke JP, Parker J, et al. The risk of infection related to radial vs femoral
sites for arterial catheterization. Crit Care Med 1983; 11(10):807 812.
23. Maki DG. Nosocomial bacteremia. An epidemiologic overview. Am J Med 1981;
70(3):719 732.
24. Peruzzi WT, Parker MA, Lichtenthal PR, et al. A clinical evaluation of a blood con-
servation device in medical intensive care unit patients. Crit Care Med 1993;
21(4):501 506.
25. Smoller BR, Kruskall MS. Phlebotomy for diagnostic laboratory tests in adults.
Pattern of use and effect on transfusion requirements. N Engl J Med 1986;
314(19):1233 1235.
26. Silver MJ, Jubran H, Stein S, McSweeney T, Jubran F. Evaluation of a new blood
conserving arterial line system for patients in intensive care units. Crit Care Med
1993; 21(4):507 511.
27. Shapiro BA, Mahutte CK, Cane RD, Gilmour IJ. Clinical performance of a blood
gas monitor: a prospective, multicenter trial. Crit Care Med 1993; 21(4):487 494.
28. Veremakis C, Holloran TH. The technique of monitoring arterial blood pressure.
J Crit Illn 1989; 4:82.
29. Gardner RM, Hollingsworth KW. Optimizing the electrocardiogram and pressure
monitoring. Crit Care Med 1986: 14(7):651 658.
30. McGee DC, Gould MK. Preventing complications of central venous catheteriza-
tion. N Engl J Med 2003; 348(12):1123 1133.
31. Merrer J, De Jonghe B, Golliot F, et al. Complications of femoral and subclavian
venous catheterization in critically ill patients: a randomized controlled trial. J
Am Med Assoc 2001; 286(6):700 707.
32. Sznajder JI, Zveibil FR, Bitterman H, Weiner P, Bursztein S. Central vein catheter-
ization. Failure and complication rates by three percutaneous approaches. Arch
Intern Med 1986; 146(2):259 261.
33. Veenstra DL, Saint S, Saha S, Lumley T, Sullivan SD. Efcacy of antiseptic-
impregnated central venous catheters in preventing catheter-related bloodstream
infection: a meta-analysis. J Am Med Assoc 1999; 281(3):261 267.
34. Manseld PF, Hohn DC, Fornage BD, Gregurich MA, Ota DM. Compli-
cations and failures of subclavian-vein catheterization. N Engl J Med 1994;
331(26):1735 1738.
35. Hind D, Calvert N, McWilliams R, Davidson A, Paisley S, Beverley C, Thomas
S. Ultrasonic locating devices for central venous cannulation: meta-analysis.
BMJ 2003; 327(7411):361.
36. Randolph AG, Cook DJ, Gonzales CA, Pribble CG. Ultrasound guidance for place-
ment of central venous catheters: a meta-analysis of the literature. Crit Care Med
1996; 24(12):2053 2058.
37. Armstrong PJ, Cullen M, Scott DH. The SiteRite ultrasound machinean aid to
internal jugular vein cannulation. Anaesthesia 1993; 48(4):319 323.
38. Hateld A, Bodenham A. Portable ultrasound for difcult central venous access. Br
J Anaesth 1999; 82(6):822 826.
39. Ruesch S, Walder B, Tramer MR. Complications of central venous catheters:
internal jugular versus subclavian accessa systematic review. Crit Care Med
2002; 30(2):454 460.
Procedures in the ICU 275
40. Golden LR. Incidence and management of large-bore introducer sheath puncture of
the carotid artery. J Cardiothorac Vasc Anesth 1995; 9(4):425 428.
41. Applebaum RM, Adelman MA, Kanschuger MS, Jacobowitz G, Kronzon
I. Transesophageal echocardiographic identication of a retrograde dissection of
the ascending aorta caused by inadvertent cannulation of the common carotid
artery. J Am Soc Echocardiogr 1997; 10(7):749 751.
42. Getzen LC, Pollak EW. Short-term femoral vein catheterization. A safe alternative
venous access? Am J Surg 1979; 138(6):875 878.
43. Williams JF, Seneff MG, Friedman BC, McGrath BJ, Gregg R, Sunner J,
Zimmerman JE. Use of femoral venous catheters in critically ill adults: prospective
study. Crit Care Med 1991; 19(4):550 553.
44. Dailey RH. Femoral vein cannulation: a review. J Emerg Med 1985; 2(5):367
372.
45. Kurekci E, Kaye R, Koehler M. Chylothorax and chylopericardium: a complication
of a central venous catheter. J Pediatr 1998; 132(6):1064 1066.
46. Unnikrishnan D, Idris N, Varshneya N. Complete heart block during central venous
catheter placement in a patient with pre-existing left bundle branch block. Br J
Anaesth 2003; 91(5):747 749.
47. Wechsler RJ, Steiner RM, Kinori I. Monitoring the monitors: the radiology of thor-
acic catheters, wires, and tubes. Semin Roentgenol 1988; 23(1):61 84.
48. Rosen M, Latto IP, Ng WS. Handbook of percutaneous central venous catheteri-
zation. London: WB Sanders, 1981.
49. Lefrant JY, Muller L, De La Coussaye JE, Prudhomme M, Ripart J, Gouzes C,
Peray P, Saissi G, Eledjam JJ. Risk factors of failure and immediate complication
of subclavian vein catheterization in critically ill patients. Intensive Care Med 2002;
28(8):1036 1041.
50. Sivak SL. Late appearance of pneumothorax after subclavian venipuncture. Am J
Med 1986; 80(2):323 324.
51. Aronchick JM, Miller WT Jr. Tubes and lines in the intensive care setting. Semin
Roentgenol 1997; 32(2):102 116.
52. Heckmann JG, Lang CJ, Kindler K, Huk W, Erbguth FJ, Neundorfer B. Neurologic
manifestations of cerebral air embolism as a complication of central venous cathe-
terization. Crit Care Med 2000; 28(5):1621 1625.
53. Ely EW, Hite RD, Baker AM, Johnson MM, Bowton DL, Haponik EF. Venous air
embolism from central venous catheterization: a need for increased physician
awareness. Crit Care Med 1999; 27(10):2113 2117.
54. Trottier SJ, Veremakis C, OBrien J, Auer AI. Femoral deep vein thrombosis
associated with central venous catheterization: results from a prospective, random-
ized trial. Crit Care Med 1995; 23(1):52 59.
55. Joynt GM, Kew J, Gomersall CD, Leung VY, Liu EK. Deep venous thrombosis
caused by femoral venous catheters in critically ill adult patients. Chest 2000;
177:178 183.
56. Mian NZ, Bayly R, Schreck DM, Besserman EB, Richmand D. Incidence of deep
venous thrombosis associated with femoral venous catheterization. Acad Emerg
Med 1997; 4(12):1118 1121.
276 Vadgama, Au, and Kamangar
120. Pulmonary Artery Catheter Consensus Conference. Consensus Statement. Crit Care
Med 1997; 25(6):910 925.
121. Shaw TJI. The Swan Ganz pulmonary artery catheter: incidence of complications,
with particular reference to ventricular dysrhythmias, and their prevention.
Anesthesia 1979; 34:651 656.
122. Thomson IR, Dalton BC, Lappas DG, Lowenstein E. Right bundle-branch block
and complete heart block caused by the Swan Ganz catheter. Anesthesiology
1979; 51:359 362.
123. Sprung CL, Jacobs LJ, Caralis PV, Karpf M. Ventricular arrhythmias during
Swan-Ganz catheterization of the critically ill. Chest 1981; 79:413 415.
124. Sprung CL, Elser B, Schein MH, Marcial EH, Shrager BR. Risk of right bundle-
branch block and complete heart block during pulmonary artery catheterization.
Crit Care Med 1989; 17:1 3.
125. Baldwin IC, Heland M. Incidence of cardiac dysrhythmias in patients during
pulmonary artery catheter removal after cardiac surgery. Heart Lung 2000;
29:155 160.
126. Iberti TJ, Benjamin E, Gruppi I, Raskin JM. Ventricular arrhythmias during
pulmonary artery catheterization in the intensive care unit: prospective study.
Am J Med 1985; 78:451 454.
127. Morris D, Mulvhill D, Lew WYW. Risk of developing complete heart block during
bedside pulmonary artery catheterization in patients with left bundle-branch block.
Arch Intern Med 1987; 147:2005 2010.
128. Lipp H, ODonoghue K, Resnekov L. Intracardiac knotting of a ow-directed
balloon catheter. N Engl J Med 1971; 284:220.
129. Foote GA, Schabel SI, Hodges M. Pulmonary complications of the ow-directed
balloon-tipped catheter. N Engl J Med 1974; 290:927 931.
130. Boyd KD, Thomas SJ, Gold J, Boyd AD. A prospective study of complications
of pulmonary artery catheterizations in 500 consecutive patients. Chest 1983;
84:245 249.
131. Carlson TA, Goldenberg IF, Murray PD, Tadavarthy SM, Walker M, Gobel FL.
Catheter-induced delayed recurrent pulmonary artery hemorrhage; J Am Med
Assoc 1989; 261:1943 1945.
132. Barash PG, Nardi D, Hammond G, Walker-Smith G, Capuano D, Laks H,
Kopriva CJ, Baue AE, Geha AS. Catheter-induced pulmonary artery perforation:
mechanisms, management and modications. J Thorac Cardiovasc Surg 1981;
82:5 12.
133. Sprung CL, Marcial EH, Garcia AA, Sequeira RF, Pozen RG. Prophylactic
use of lidocaine to prevent advanced ventricular arrhythmias during pulmonary
artery catheterization: prospective double blind study. Am J Med 1983;
75:906 910.
134. Kearney TJ, Shabot MM. Pulmonary artery rupture associated with the Swan Ganz
catheter. Chest 1995; 108:1349 1352.
135. Slaughter TF, Greenberg CS. Heparin-associated thrombocytopenia and throm-
bosis: Implications for perioperative management. Anesthesiology 1997;
87:667 675.
136. Thomas R., Siproudhis L, Laurent JF, Bouget J, Bousser J, Camus C, Michelet C.
Massive hemoptysis from iatrogenic balloon catheter rupture of pulmonary
Procedures in the ICU 281
artery: successful early management by balloon tamponade. Crit Care Med 1987;
15:272 273.
137. Kelly TF Jr, Morris GC, Crawford ES, Espada R, Howell JF. Perforation of the pul-
monary artery with Swan Ganz catheters: diagnosis and surgical management.
Ann Surg 1981; 193:686 692.
138. Slacken A. Complications of invasive hemodynamic monitoring in the intensive
care unit. Curr Probl Surg 1988; 25:69 145.
139. Michel L, Marsh M, McMichan JC, Southorn PA, Brewer NS. Infection of pulmon-
ary artery catheters in critically ill patients. J Am Med Assoc 1981; 245:
1032 1036.
140. Raad I, Umphrey J, Khan A, Truett LJ, Bodey GP. The duration of placement as a
predictor of peripheral and pulmonary arterial catheter infections. J Hosp Infect
1993; 23:17 26.
141. Rowley KM, Clubb KS, Smith GJW, Cabin HS. Right-sided infective endo-
carditis as a consequence of ow-directed pulmonary artery catheterization: A
clinicopathological study of 55 autopsied patients. N Engl J Med 1984;
311:1152 1156.
142. Mermel LA, McCormick RD, Springman SR, Maki DG. The pathogenesis and
epidemiology of catheter-related infection with pulmonary artery Swan Ganz
catheters: A prospective study utilizing molecular subtyping. Am J Med 1991;
91(3B):197S 205S.
143. Myers ML, Austin TW, Sibbald WJ. Pulmonary artery catheter infections: a
prospective study. Ann Surg 1985; 201:237 241.
144. Singh SJ, Puri VK. Prevention of bacterial colonization of pulmonary artery cath-
eters. Infect Surg 1984; 853 856.
145. Kopman EA, Sandza JG Jr. Manipulation of the pulmonary artery catheter after
placement: maintenance of sterility. Anesthesiology 1978; 48:373 374.
146. Heard SO, Davis RF, Sherertz RJ, Mikhail MS, Gallagher RC, Layon A, Gallagher
TJ. Inuence of sterile protective sleeves on the sterility of pulmonary artery cath-
eters. Crit Care Med 1987; 15:499 502.
147. Iberti T, Fischer E, Leibowitz AB, Panacek EA, Silverstein JH, Albertson TE, and
the Pulmonary Artery Catheter Study Group. A multicenter study of physicians
knowledge of the pulmonary artery catheter. J Am Med Assoc 1990; 264:2928
2932.
148. Bone RC. High-tech predicament: pulmonary artery catheter. J Am Med Assoc
1990; 264:2933.
149. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of
supranormal values of survivors as therapeutic goals in high-risk surgical patients.
Chest, 1988; 94:1176 1186.
150. Shoemaker WC, Appel PC, Kram HB. Oxygen transport measurements to evaluate
tissue perfusion and titrate therapy: dobutamine and dopamine effects. Crit Care
Med 1991; 19:672 688.
151. Ivanov R, Allen J, Calvin JE. The incidence of major morbidity in critically in
patients managed with pulmonary artery catheters: a meta-analysis. Crit Care
Med 2000 28:615 619.
152. Connors AF Jr, Speroff T, Dawson NV, Thomas C, Harrell FE Jr, Wagner D,
Desbiens, N, Goldman L, Wu AW, Califf RM, et al., for the SUPPORT
282 Vadgama, Au, and Kamangar
170. Patz EF Jr, Goodman PC, Erasmus JJ. Percutaneous drainage of pleural collections.
J Thorac Imaging 1998; 13(2):83 92.
171. Conces DJ Jr, Tarver RD, Gray WC, Pearcy EA. Treatment of pneumothoraces
utilizing small caliber chest tubes. Chest 1988; 94(1):55 57.
172. Goff BA, Mueller PR, Muntz HG, Rice LW. Small chest-tube drainage followed by
bleomycin sclerosis for malignant pleural effusions. Obstet Gynecol. 1993;
81(6):993 996.
173. Thompson RL, Yau JC, Donnelly RF, Gowan DJ, Matzinger FR. Pleurodesis with
iodized talc for malignant effusions using pigtail catheters. Ann Pharmacother
1998; 32(7 8):739 742.
174. Parry GW, Morgan WE, Salama FD. Management of haemothorax. Ann R Coll
Surg Engl 1996; 78(4):325 326.
175. Iberti TJ, Stern PM. Chest tube thoracostomy. Crit Care Clin 1992; 8(4):879 895.
176. Tomlinson MA, Treasure T. Insertion of a chest drain: how to do it. Br J Hosp Med
1997; 58(6):248 252.
177. Thal AP, Quick KL. A guided chest tube for safe thoracostomy. Surg Gynecol
Obstet 1988; 167(6):517.
178. Guyton SW, Paull DL, Anderson RP. Introducer insertion of mini-thoracostomy
tubes. Am J Surg 1988; 155(5):693 696.
179. Daly RC, Mucha P, Pairolero PC, Farnell MB. The risk of percutaneous chest tube
thoracostomy for blunt thoracic trauma. Ann Emerg Med 1985; 14(9):865 870.
180. Etoch SW, Bar-Natan MF, Miller FB, Richardson JD. Tube thoracostomy. Factors
related to complications. Arch Surg 1995; 130(5):521 525.
181. Millikan JS, Moore EE, Steiner E, Aragon GE, Van Way CW III. Complications of
tube thoracostomy for acute trauma. Am J Surg 1980; 140(6):738 741.
182. Chest drainage tubes. Surg Clin North Am 1989; 69(1):41 46.
183. Dalbec DL, Krome RL. Thoracostomy. Emerg Med Clin North Am 1986;
4(3):441 457.
184. Kopec SE, Conlan AA, Irwin RS. Perforation of the right ventricle: a complication
of blind placement of a chest tube into the postpneumonectomy space. Chest 1998;
114(4):1213 1215.
185. Meisel S, Ram Z, Priel I, Nass D, Lieberman P. Another complication of thora-
costomyperforation of the right atrium. Chest 1990; 98(3):772 773.
186. Kumar SP, Belik J. Chylothoraxa complication of chest tube placement in a
neonate. Crit Care Med 1984; 12(4):411 412.
187. Palomeque A, Canadell D, Pastor X. Acute diaphragmatic paralysis after chest tube
placement. Intensive Care Med 1990; 16(2):138.
188. Murphy K, Tomlanovich MC. Unilateral pulmonary edema after drainage of a
spontaneous pneumothorax: case report and review of the world literature.
J Emerg Med 1983; 1(1):29 36.
189. Gascoigne A, Appleton A, Taylor R, Batchelor A, Cook S. Catastrophic circulatory
collapse following re-expansion pulmonary oedema. Resuscitation 1996;
31(3):265 269.
190. Fitzpatrick S, Acheson J, Curran P. Re-expansion pulmonary edema and circulatory
shock in a 20-year-old man Eur J Emerg Med 2003; 10(2):146 148.
191. Mahfood S, Hix WR, Aaron BL, Blaes P, Watson DC. Reexpansion pulmonary
edema. Ann Thorac Surg 1988; 45(3):340 345.
284 Vadgama, Au, and Kamangar
192. Timby J, Reed C, Zeilender S, Glauser FL. Mechanical causes of pulmonary edema.
Chest 1990; 98(4):973 979.
193. Matsumiya N, Dohi S, Kimura T, Naito H. Reexpansion pulmonary edema after
mediastinal tumor removal. Anesth Analg 1991; 73(5):646 648.
194. Lim-Levy F, Babler SA, De Groot-Kosolcharoen J, Kosolcharoen P, Kroncke GM.
Is milking and stripping chest tubes really necessary? Ann Thorac Surg 1986;
42(1):77 80.
8
Bronchoscopy
SILVERIO SANTIAGO
Pulmonary and Critical
Care Section
West Los Angeles Healthcare Center
VA Greater Los Angeles Healthcare
System and Geffen School
of Medicine at UCLA
Los Angeles, California, U.S.A.
I. Introduction
285
286 Shah, Susanto, and Santiago
beroptics enable the bronchoscope to bend, and this allows for easy navigation
throughout the tracheo-bronchial tree. Flexible bronchoscopy currently plays an
important role for diagnosis and treatment in both outpatient and inpatient settings.
Rigid bronchoscopes are straight metal hollow tubes that allow for diagnostic
procedures and therapeutic interventions. They have ports that allow for venti-
lation and administration of anesthetic gases and suction ports for removal of
secretions and blood. The hollow tube allows for the introduction of optical tele-
scopes or a exible beroptic bronchoscope. Other types of accessory equipment
are available, including biopsy forceps, baskets for foreign body removal, intro-
ducers for stent placement, and laser bers for treatment of endobronchial
lesions. Rigid bronchoscopy is typically performed under general anesthesia.
The indications for rigid bronchoscopy are listed in Table 1.
The exible beroptic bronchoscope has continued to evolve since its intro-
duction in 1967. The average adult bronchoscope has an external diameter of 4 to
6 mm and a typical viewing radius of 1208. The working channel for introduction of
accessory instruments or suction is typically 1 to 3 mm in diameter. Ultrathin
bronchoscopes for the evaluation of small distal airways are now available and
have an external diameter of 2.7 mm. Another relatively new development is the
videoscope that contains a video chip at the distal tip of the bronchoscope. These
chips have a lower propensity for damage when compared with ber bundles and
also allow for digital processing of images. A wide variety of ancillary equipment
may be used with the exible beroptic bronchoscope, including biopsy forceps,
protected and non-protected brushes for cytologic and microbiologic studies, and
devices for transbronchial needle aspiration. Flexible bronchoscopy is usually per-
formed with local anesthesia and may or may not require conscious sedation. The
indications for exible bronchoscopy are listed in Table 2 (5,6).
There are relatively few absolute contraindications to both rigid and ex-
ible bronchoscopy. These include lack of consent for the procedure and the pre-
sence of malignant cardiac arrhythmias and refractory hypoxemia. Additional
absolute contraindications for rigid bronchoscopy include the presence of cervi-
cal injury, an unstable neck, and microstomia.
Dilatation of strictures/stenosis
Placement of airway prostheses (stents)
Retrieval of foreign body
Massive hemoptysis
Resection of bulky tumors
Laser therapy
Cryotherapy
Photodynamic therapy
Bronchoscopy 287
Diagnostic indications
Abnormal chest roentgenogram
Unexplained cough, hemoptysis, wheeze, or stridor
Suspected pulmonary infections
Refractory lung abscess
Determination of the presence and extent of thermal burns
Evaluation of the airways for suspected injury after chest trauma
Unexplained vocal cord paralysis and diaphragmatic paralysis
Abnormal sputum cytology
Suspected tracheo-esophageal or broncho-pleural stulas
Carcinoma of the lung
Evaluation of problems associated with endotracheal tubes
To obtain diagnostic material from patients with non-infectious diffuse lung disease
Foreign bodies
Assessment of airway patency
Evaluation of lung transplant rejection
To investigate the etiology of SVC syndrome, chylothorax, or unexplained pleural
effusion
Therapeutic indications
Atelectasis secondary to retained secretions, mucous plugs, or clots not mobilized by
non-invasive techniques
Debulking of malignant neoplasm
Retrieval of foreign bodies
Hemoptysis
Treatment of Bronchopleural stula
Drainage of bronchogenic cysts
Endotracheal tube placement
Percutaneous dilatational tracheostomy
Electrocautery
Cryotherapy
Photodynamic therapy
Laser therapy
Brachytherapy
Placement of metal stents
Dilatation of strictures and stenosis
Therapeutic BAL for Pulmonary Alveolar Proteinosis
Conditions involving increased risk for both rigid and exible broncho-
scopy include lack of patient cooperation, recent myocardial infarction, unstable
angina, and active bronchospasm. Increased risk of bleeding after transbronchial
biopsy is seen in patients with pulmonary hypertension. Similarly, increased risk
of bleeding after biopsy is seen in patients with uremia, thrombocytopenia, and
coagulopathy. Bleeding and laryngeal edema may also occur in patients with
superior vena cava syndrome.
288 Shah, Susanto, and Santiago
The rigid bronchoscope can be inserted through the mouth or through a tracheost-
omy stoma. On the other hand, the exible bronchoscope can be introduced
through the nose, mouth, tracheostomy, or endotracheal tube. The oral route is
preferred for exible bronchoscopy if frequent removal and reinsertion of the
bronchoscope is anticipated or if endotracheal intubation is to be performed.
The decision to intubate the patient is dependent on the clinical status of the
patient. Patients are frequently intubated over the bronchoscope if their respirat-
ory status is tenuous or if they are at increased risk for bleeding after biopsies.
The nasal route avoids the possible complication of damage to the bronchoscope
if the patient bites down in the absence of a bite block. In the intensive care unit
(ICU), a patient requiring bronchoscopy may already have an endotracheal tube
in place. This can pose problems if the tube is too small to accommodate an
average bronchoscope. A size 7.5 (mm, inner diameter) endotracheal tube will
allow passage of the bronchoscope but can limit adequate ventilation. A size
8.0 or greater endotracheal tube is recommended for ease of passage of the
bronchoscope and for adequate ventilation. In patients with signicant lung
disease, the bronchoscope may need to be withdrawn multiple times throughout
the procedure to allow the patient to oxygenate and ventilate adequately. Smaller
pediatric bronchoscopes can also be used. When compared with larger diameter
adult bronchoscopes, use of pediatric bronchoscopes offers advantages in the
mechanically ventilated patient, with less signicant cardiovascular and respirat-
ory side effects while providing comparable cell and microbiological yields (12).
When performing bronchoscopy on patients on mechanical ventilation, the FIO2
should be increased to 1.0 and maintained at that level throughout the procedure.
V. Diagnostic Techniques
container, with care being taken not to apply excessive suction to avoid collapse
of the airway or suction trauma. It is estimated that this allows for the sampling of
approximately one million alveoli. The uid can then be sent for a variety of ana-
lyses, including differential cell counts, cytologic examination, and microbiolo-
gic studies (13,14). Complications of BAL are uncommon but include fever,
pneumonitis, bleeding, and bronchospasm.
Endobronchial brushing is another frequently used technique for obtaining
samples for cytology and microbiology. Brushes may be protected (sheathed) or
unprotected. When attempting to obtain an uncontaminated material for culture,
the protected brush is typically favored. Discrete peripheral lesions suspicious for
malignancy can also be brushed to obtain specimens for cytology and is performed
under uoroscopy to ensure that the brush is in the vicinity of the lesion. The most
frequent complication seen with brushing is bleeding, which is usually self-limited.
Bronchoscopic biopsies can be obtained from both visible endobronchial
lesions and distal lesions. Different biopsy forceps are available, including
toothed, non-toothed, and spiked forceps. Endobronchial lesions can be biopsied
under direct visualization, thus minimizing risk of complications. Spiked forceps
with an impaler needle can facilitate biopsies of endobronchial lesions that are dif-
cult to grab. Four to six biopsies are generally recommended to optimize diagnostic
yield in bronchogenic carcinoma, although as many as 10 biopsies may be necessary
to maximize yield for peripheral carcinomas (15). For more peripheral lung lesions
or diffuse lung disease, samples are obtained using the transbronchial approach.
Transbronchial biopsy in diffuse lung disease can be performed with or without
uoroscopic guidance. Whether use of uoroscopy decreases the incidence of
pneumothorax after transbronchial biopsy is debated (10,16,17). The other major
complication that occurs with bronchoscopic biopsies is bleeding. Bleeding is
usually self-limited but may be severe. Thus, it is important to ensure that the
patient does not have coagulopathy, thrombocytopenia, signicant renal failure,
or severe pulmonary hypertension prior to proceeding with biopsies. Topical appli-
cation of 5 mL of 1:20,000 epinephrine prior to biopsy may minimize bleeding (18).
Transbronchial needle aspiration is a useful procedure that allows sampling
of mediastinal lymph nodes for diagnosis and staging of lung cancer. The bronch-
oscopist must be familiar with mediastinal anatomy prior to performance of
transbronchial needle aspiration. In addition, a CT scan of the chest is required
to determine the exact location of adenopathy or tumor. Transbronchial
needles come in different sizes, with smaller needles providing cytology speci-
mens and larger ones providing biopsy specimens. Complications from this pro-
cedure are rare and include minor bleeding (19).
Endobronchial ultrasound (EBUS) is a recent addition in the armamentar-
ium of the interventional pulmonologist. EBUS extends the imaging capability of
the bronchoscopist beyond the airway lumen in assessing bronchial wall tumor
invasion, locating lymph nodes or tumor for needle biopsy, and locating vascular
structures adjacent to the airway. EBUS requires transient occlusion of the airway
for 20 to 30 seconds, and it provides a 3608 view of the soft tissues surrounding
the airway. It can be used to identify the cause of an extrinsic compression and
Bronchoscopy 291
Prior to discharge, patients should be monitored until sedation has worn off.
Post-procedure chest radiographs are routinely recommended after rigid
bronchoscopy or after bronchoscopy in patients on mechanical ventilation.
This may not be necessary after transbronchial biopsies, particularly if biopsies
are performed under uoroscopic guidance and the patient does not complain
of shortness of breath or chest pain. In a large survey, however, the majority
of pulmonologists replied that they routinely obtain chest radiographs after
transbronchial biopsy (8).
VII. Complications
(26). In addition, higher rates of complications related to anesthesia were seen for
both procedures.
Although the etiologic agent of most infections can be determined based on clini-
cal history and non-invasive testing, bronchoscopy often serves as a useful
adjunct when therapy is ineffective or when prompt diagnosis is necessary. In
addition to providing specimens for microbiologic studies, bronchoscopy pro-
vides information regarding the severity of airway inammation and the quantity
and nature of secretions. BAL and protected brushings are particularly useful for
obtaining specimens for microbiologic studies. It is important to keep in mind
that although the isolation of certain organisms is diagnostic for the etiology of
disease (e.g., Nocardia), other organisms may be non-pathogenic colonizers
(14). All BAL culture results must, therefore, be interpreted in the proper clinical
context. Transbronchial biopsy is an additional technique that may add to the
diagnostic yield, particularly if invasive fungal or viral disease is suspected
(29). Transbronchial biopsies may also help establish the diagnosis of non-
infectious processes, such as Kaposi sarcoma or lymphoma, especially in immu-
nocompromised hosts.
Although the effect may be negligible, the inhibitory properties of lidocaine
on bacterial growth in vitro should be kept in mind when using liberal amounts of
lidocaine during diagnostic bronchoscopy for infection (9,30). On the other hand,
BAL or brush specimens may be falsely positive if withdrawn through a working
channel contaminated with upper airway secretions. Therefore, care must be
taken to obtain non-contaminated specimens. In addition, bronchoscopes must
Bronchoscopy 293
C. Specic Pathogens
Pneumocystis jiroveci (previously Pneumocystis carinii) is a common pulmonary
pathogen in patients with AIDS and other immunocompromised hosts. The diag-
nostic yield of BAL for Pneumocystis is as high as 86% to 97% in patients with
294 Shah, Susanto, and Santiago
AIDS (44 46). Transbronchial biopsy, in addition, increases the diagnostic yield
with a reported sensitivity of 100% when combined with BAL (46). The yield
tends to be lower, however, in patients receiving aerosolized pentamidine (47)
or in patients with disease processes other than HIV. Site-directed multiple
lobe BAL combined with the use of indirect uorescent antibodies have also
been reported to increase the diagnostic yield for Pneumocystis (48). Bronchial
brushings, on the other hand, do not provide additional value to BAL and trans-
bronchial biopsy (49).
Mycobacterial diseases, including those caused by Mycobacterium tuber-
culosis as well as the many atypical mycobacteria, continue to be a major
health concern both in the United States and throughout the world (50). Diagnosis
of tuberculosis is typically made by obtaining sputum samples. In patients,
however, who cannot produce sputum even with induction, bronchoscopy with
BAL may provide the diagnosis. The diagnostic yield from BAL has been
reported to be comparable to that from sputum (51 53). Additional diagnostic
yield has been reported with transbronchial biopsy in conjunction with bronchial
brushings (54). Another report, however, did not nd transbronchial biopsies
helpful in establishing diagnosis (55). Following bronchoscopy, patients fre-
quently continue to produce sputum for one to two days. These samples should
be collected and sent to the laboratory for analysis as they will often be positive
for mycobacteria (56). It should be kept in mind that bronchoscopy in patients
with tuberculosis increases the risk of transmission of disease to medical person-
nel, and appropriate precautions should be taken.
Bronchoscopy is also useful in the diagnosis of infection caused by
non-tuberculous mycobacteria (NTM). The most frequent pulmonary pathogens
include Mycobacterium avium complex, Mycobacterium kansasii, Mycobacterium
abscessus, Mycobacterium fortuitum, Mycobacterium xenopi, and Mycobacterium
malmoense (57). NTM tend to cause disease in elderly patients, patients with
chronic diseases, and immunocompromised hosts. The presence of atypical
mycobacteria in bronchoscopic samples is not diagnostic of disease, however,
in the absence of radiographic and clinical criteria because of their ubiquitous
nature and the high prevalence of colonization (58,59).
Fungal infections have also been diagnosed by bronchoscopy. These
include histoplasmosis, coccidioidomycosis, blastomycosis, and aspergillosis.
In general, the diagnosis of pulmonary histoplasmosis requires a tissue sample
obtained via thoracoscopy or open lung biopsy. This is especially true if
disease is limited to a solitary pulmonary nodule. However, when utilized in
patients with other radiographic features, such as inltrates and cavitary
lesions, bronchoscopy samples are associated with a higher diagnostic yield com-
pared with sputum (60).
Similarly, bronchoscopy samples are more efcacious in the diagnosis of
coccidioidomycosis and blastomycosis compared with sputum. Fungal stains of
BAL samples for Coccidioidomycosis immitis tend to have a low yield, but cul-
tures and transbronchial biopsies have high yields. In addition, serologic tests for
Bronchoscopy 295
F. Chest Trauma
Blunt chest trauma occasionally results in injury to the tracheobronchial tree.
Patients typically complain of dyspnea and hemoptysis and may develop subcu-
taneous emphysema. Bronchoscopy, in conjunction with radiographic imaging, is
the procedure of choice in establishing the presence and extent of traumatic
airway injury.
G. Thermal Injuries
Thermal injuries are typically the result of exposure to hot smoke from burning
material. Edema, inammation, necrosis, and ulceration of the airways lead to
upper airway obstruction that can develop rapidly, necessitating emergent intuba-
tion. As such, serial bronchoscopy should be considered as edema can be pro-
gressive during the rst 24 hours. However, initial bronchoscopic assessment
does not predict the duration or degree of ventilatory support necessary during
hospitalization (73,74). In addition to damage caused by direct thermal burns,
the lung parenchyma can also be damaged by other products of combustion,
leading to pulmonary edema and contributing further to the mortality of patients
with inhalation injury.
B. Bronchopleural Fistula
Bronchopleural stula is a common complication seen in patients requiring
mechanical ventilation or following lung resection surgery. In patients with per-
sistent air leak, the bronchoscope can be used for localization of the site of leak
and for attempting therapeutic maneuvers. Localization is accomplished by
sequential ination and deation of a Fogarty balloon catheter in segmental
bronchi and watching for cessation of air leak in the water-seal apparatus. Alter-
natively, a bronchogram can be performed by injecting contrast material through
a catheter in the working channel of the bronchoscope. Various agents have been
utilized to seal the airways leading to the stula, including oxidized regenerated
cellulose (Surgicel), brin clot, silver nitrate, methyl-2-cyanoacrylate, tetra-
cycline, and doxycycline. In addition, laser therapy has been used to coagulate
the site of leak (77 80).
C. Foreign Bodies
Foreign body aspiration is a common problem seen in pulmonary medicine.
Aspiration most often occurs in children with equal prevalence of right and
298 Shah, Susanto, and Santiago
left lung aspiration. In adults, the right side is more often involved than the left
because of the branch angle of the left main stem bronchus. Patients with foreign
bodies in the upper airway may be profoundly symptomatic, whereas those with
passage of the foreign body to more distal airways may be asymptomatic. Follow-
ing imaging studies of the chest, the next diagnostic and treatment modality is
bronchoscopy. In the pediatric population, rigid bronchoscopy is preferred. In
adults, exible bronchoscopy is usually attempted rst and is often successful
in removal of the foreign body with use of ancillary equipment, such as
forceps and wire baskets. In difcult cases, rigid bronchoscopy may be necessary
for extraction of the foreign body and control of the airway (81).
A. Atelectasis
Atelectasis and retained secretions occur commonly among ICU patients.
Approximately 50 to 60% of bronchoscopies performed in the ICU are for
treatment of atelectasis. The efcacy of bronchoscopy for atelectasis varies in
different reports. A recent review found that patients with lobar or segmental
atelectasis benet more from bronchoscopy compared to patients with subseg-
mental atelectasis or air bronchograms on chest roentgenograms. There are,
however, insufcient studies at this time comparing the efcacy of bronchoscopy
to non-invasive treatments, such as chest physiotherapy, mucolytic agents, and
kinetic beds. The decision to proceed with bronchoscopy for atelectasis must
be made on an individual basis (82).
Bronchoscopy 299
C. Laser Therapy
Laser bronchoscopy can be performed using either a exible or rigid broncho-
scope as described above. For most airway interventions below the vocal
cords, the Nd:YAG laser is generally the laser of choice. It is important to
300 Shah, Susanto, and Santiago
Figure 1 A 52-year-old man with known metastatic colon carcinoma who presented
with worsening dyspnea and on chest roentgenography was found to have complete
right lung collapse. (A) Bronchoscopy shows complete obstruction of the right main
stem bronchus at the level of the major carina. This tumor was known to arise out of
the right upper lobe encroaching into the right main stem bronchus. (B) The chest roent-
genogram before tumor resection. (C) The chest roentgenogram immediately following
rigid bronchoscopic resection of tumor with opening of the right middle and lower
lobes and post obstructive pneumonia. (D) The size of tumor fragments resected from
the right main stem bronchus. (Continued.)
Bronchoscopy 301
Figure 1 (Continued.)
D. Cryotherapy
Endobronchial cryotherapy works by causing tissue destruction through rapid
freezing followed by slow thawing. Nitrous oxide (N20) gas is used to lower
the temperature to approximately 2898C. Rapid freezing results in the formation
of ice crystals within the cytoplasm that grind and abrade cellular organelles
during thawing, resulting in cell death. Cryo-sensitive tissues include skin,
mucous membranes, granulation tissues, tumors, and other tissues with a signi-
cant amount of water content. Cryo-resistant tissues include fat, cartilage, bone,
Bronchoscopy 303
and brous connective tissue with little water content. The exible cryotherapy
probe can be delivered through a exible or a rigid bronchoscope. The advan-
tages of endobronchial cryotherapy as opposed to Nd:YAG laser therapy or elec-
trocautery include its safety, low cost, ease of use, and the ability to remove
certain foreign bodies and blood clots. The disadvantage is that it is slow in
removing tumor or granulation tissue from the airways, with repeat bronchoscopy
being generally required within a week. Some success has also been reported with
the use of endobronchial cryotherapy with curative intent in the management of
airway carcinoma in situ (87).
E. Electrocautery
F. Photodynamic Therapy
Photodynamic therapy (PDT) is an effective technique for the management of
small endobronchial tumors. It is used primarily in non-surgical candidates
with small lesions and early stages of lung cancer, with a reported cure rate of
approximately 75% and a recurrence rate of approximately 30% (89). PDT
involves intravenous injection of a tumor photosensitizing agent derived from
hematoporphyrin, which is retained in tumor cells and cleared from most
healthy tissues in six hours except for the lung, reticuloendothelial tissues, and
the skin. Airway tumor cells retain the photosensitizer 48 to 72 hours after its
administration and undergo selective destruction by photo-activation upon
exposure to 630 nm wavelength of laser light introduced in the airway using a
exible bronchoscope. Tumor necrosis occurs as a result of cellular destruction
from superoxide and hydroxyl radicals and vascular occlusions related to throm-
boxane-A2 release (90). A clean-up bronchoscopy two to three days later is
required to remove necrotic tumor debris from the airway. Photodynamic
therapy makes the skin and eyes sensitive to light for six weeks or more after
treatment, and patients are advised to avoid direct sunlight and bright indoor
light during this time period.
304 Shah, Susanto, and Santiago
G. Brachytherapy
Endobronchial brachytherapy involves delivery of radiation therapy from a
radioactive source placed within or very near the tumor mass. Direct placement
of the radioactive source is accomplished through an airway catheter placed
bronchoscopically. When used in early stage endobronchial cancer, brachyther-
apy produces results comparable to that of photodynamic therapy, with reported
cure rates of 75% to 85% (89). In addition, the use of brachytherapy in conjunc-
tion with Nd:YAG laser treatment increases survival compared with laser therapy
alone (91).
through the wire mesh. Mucostasis and mucus plug formation tend to occur with
silicone stents and stents covered with a silicone membrane. The edges of the
stents may also elicit formation of granulation tissue, which may potentially
cause future obstruction.
References
1. Patterson EJ. History of bronchoscopy and esophagoscopy for foreign body. The
Laryngoscope 1926; 36:157 175.
2. Jackson C. Bronchoscopy: past, present and future. N Engl J Med 1928; 199:758763.
3. Andersen HA, Fontana RS, Harrison EG, Jr. Transbronchoscopic lung biopsy in
diffuse pulmonary disease. Dis Chest 1965; 48:187 192.
4. Ikeda, S, Yanai N, Ishikawa S. Flexible bronchoberscope. Keio J Med 1968; 17:116.
5. Sokolowski JW Jr, Burgher LW, Jones FL Jr, et al. Guidelines for beroptic broncho-
scopy in adults. Position paper of the American Thoracic Society Board of Directors.
Am Rev Resp Dis 1987; 136:1066.
6. Prakash UBS. Bronchoscopy in the Critical Care Unit. Semin Respir Crit Care Med
1997; 6:583 591.
7. Dajani AS, Taubert KA, Wilson W, Bolger AF, Bayer A, Ferrieri P, Gewitz MH,
Shulman ST, Nouri S, Newburger JW, Hutto C, Pallasch TJ, Gage TW, Levinson
ME, Peter G, Zuccaro G. Prevention of Bacterial Endocarditis. J Am Med Assoc
1997; 277:1794 1801.
8. Prakash UBS, Offord KP, Stubbs SE. Bronchoscopy in North America: The ACCP
Survey. Chest 1991; 100:1668 1675.
9. Kirkpatrick MB. Lidocaine topical anesthesia for exible bronchoscopy. Chest 1989;
96:965 967.
10. British Thoracic Society Guidelines on diagnostic exible bronchoscopy. Thorax
2001; 56(suppl 1):i1 i21.
11. Karim A, Ahmed S, Siddiqui R, Mattana J. Methemoglobinemia complicating topical
lidocaine used during endoscopic procedures. Am J Med 2001; 111:150153.
12. Ricou B, Grandin S, Nicod L, Thorens J, Suter PM. Adult and paediatric size
bronchoscopes for bronchoalveolar lavage in mechanically ventilated patients:
yield and side effects. Thorax 1995; 50(3):290 293.
13. Reynolds, HY. Bronchoalveolar lavage. Am Rev Respir Dis 1987; 135:250 263.
14. American Thoracic Society. Clinical role of bronchoalveolar lavage in adults with
pulmonary disease. Am Rev Resp Dis 1990; 142:481 486.
15. Popovich J Jr, Kvale PA, Eichenhorn MS, Radke JR, Ohorodnik JM, Fine G.
Diagnostic accuracy of multiple biopsies from exible beroptic bronchoscopy.
Am Rev Resp Dis 1982; 125:521 523.
16. de Fenoyl O, Capron F, Lebeau B, Rochemaure J. Transbronchial biopsy without
uoroscopy: A ve year experience in outpatients. Thorax 1989; 44:956 959.
Bronchoscopy 307
17. Zavala DC. Transbronchial biopsy in diffuse lung disease. Chest 1978; 73(suppl):
727 733.
18. Zavala DC. Flexible Fiberoptic Bronchoscopy: A Training Handbook. Iowa City:
University of Iowa 1978.
19. Wang KP, Terry PB. Transbronchial needle aspiration in the diagnosis and staging of
bronchogenic carcinoma. Am Rev Respir Dis 1983; 127:344 347.
20. Herth F, Becker HD, Ernst A. Conventional vs endobronchial untrasound-guided
transbronchial needle aspiration: a randomized trial. Chest 2004; 125:322 325.
21. Lam S, Kennedy T, Unger M, Miller YE, Gelmont D, Rusch V, Gipe B, Howard D,
LeRiche JC, Coldman A, Gazdar AF. Localization of bronchial intraepithelial
neoplastic lesions by uorescence bronchoscopy. Chest 1998; 113:696 702.
22. Suratt PM, Smiddy JF, Gruber B. Deaths and complications associated with
beroptic bronchoscopy. Chest 1976; 69:747 751.
23. Credle WF, Smiddy JF, Elliott RC. Complications of beroptic bronchoscopy. Am
Rev Respir Dis 1974; 109:67 72.
24. Simpson FG, Arnold AG, Purvis A, et al. Postal survey of bronchoscopic practice by
physicians in the United Kingdom. Thorax 1986; 41:311 317.
25. Pereira W Jr, Kovnat DM, Snider GL. A prospective cooperative study of
complications following exible beroptic bronchoscopy. Chest 1978; 73:813 816.
26. Lukomsky GI, Ovchinnikov AA, Bilal A. Complications of bronchoscopy: com-
parison of rigid bronchoscopy under general anesthesia and exible beroptic
bronchoscopy under topical anesthesia. Chest 1981; 79:316 321.
27. Gellert AR, Rudd RM, Sinha G. Fiberoptic bronchoscopy: effect of multiple biopsies
on diagnostic yield in bronchial carcinoma. Thorax 1982; 37:684 687.
28. Zavala D. Diagnostic beroptic bronchoscopy: techniques and results of biopsy in
600 patients. Chest 1975; 68:12 19.
29. Salzman SH. Bronchoscopic techniques for the diagnosis of pulmonary compli-
cations of HIV infection. Semin Respir Infect 1999; 14:318 326.
30. Wimberley N, Willey S, Sullivan N, Bartlett JG. Antibacterial properties of
lidocaine. Chest 1979; 76:37 40.
31. Ramsey AH, Oemig TV, Davis JP, Massey JP, Torok TJ. An outbreak of broncho-
scopy-related Mycobacterium tuberculosis infections due to lack of bronchoscope
leak testing. Chest 2002; 121:976 981.
32. Pseudomonas aeruginosa infections associated with defective bronchoscopes.
MMWR 2002; 51:190.
33. Dakin J, Grifths M. Pulmonary investigations for acute respiratory failure. Thorax
2002; 57:79 85.
34. Chastre J, Fagon JY. Ventilator-associated pneumonia. Am J Respir Crit Care Med
2002; 165:867 903.
35. Craven DE, De Rosa FG, Thornton D. Nosocomial pneumonia: emerging
concepts in diagnosis, management, and prophylaxis. Curr Opin Crit Care 2002:8:
421 429.
36. American Thoracic Society. Hospital-acquired pneumonia in adults. Diagnosis,
assessment of severity, initial antimicrobial therapy, and preventive strategies. A
consensus statement. Am J Respir Crit Care Med 1995; 153:1711 1725.
37. Fagon JY, Chastre J, Hance AJ, Guiguet M, Trouillet JL, Domart Y, Pierre J, Gibert
C. Detection of nosocomial lung infection in ventilated patients: use of a protected
308 Shah, Susanto, and Santiago
specimen brush and quantitative culture techniques in 147 patients. Am Rev Respir
Dis 1988; 138:110 116.
38. Meduri GU. Diagnosis and differential diagnosis of ventilator-associated pneumo-
nia. Clin Chest Med 1995; 16:61 93.
39. Meduri GU, Chastre J. The standardization of bronchoscopic techniques for
ventilator-associated pneumonia. Chest 1992; 102(suppl):557 564.
40. Grifn JJ, Meduri GU. New approaches in the diagnosis of nosocomial pneumonia.
Med Clin North Am 1994; 78:1091 1122.
41. Baselski VS, El-Torky M, Coalson JJ, Grifn JP. The standardization of criteria for
processing and interpreting laboratory specimens in patients with suspected venti-
lator-associated pneumonia. Chest 1992; 102(suppl):571 579.
42. Fagon JY, Chastre J, Wolff M, Gervais C, Parer-Aubas S, Stephan F, Similowski T,
Mercat A, Diehl JL, Sollet JP, Tenaillon A. Invasive and noninvasive strategies for
management of suspected ventilator-associated pneumonia: a randomized trial. Ann
Intern Med 2000; 132:621 630.
43. Ruiz M, Tores A, Ewig S, Marcos MA, Alcon A, Lledo R, Asenjo MA, Maldonaldo
A. Noninvasive versus invasive microbial investigation in ventilator-associated
pneumonia: evaluation of outcome. Am J Respir Crit Care Med 2000; 162:119 125.
44. Gal AA, Klatt EC, Koss MN, Strigle SM, Boylen CT. The effectiveness of broncho-
scopy in the diagnosis of Pneumocystis carinii and Cytomegalovirus pulmonary
infections in acquired immunodeciency syndrome. Arch Pathol Lab Med 1987;
111:238 241.
45. Golden JA, Hollander H, Stulbarg MS, Gamsu G. Bronchoalveolar lavage as the
exclusive diagnostic modality for Pneumocystis carinii pneumonia. A prospective
study among patients with acquired immunodeciency syndrome. Chest 1986;
90:18 22.
46. Broaddus C, Dake MD, Stulbarg MS, Blumenfeld W, Hadley WK, Golden JA,
Hopewell PC. Bronchoalveolar lavage and transbronchial biopsy for the diagnosis
of pulmonary infections in the acquired immunodeciency syndrome. Ann Intern
Med 1985; 102:747 752.
47. Jules-Elysee KM, Stover DE, Zaman MB, Bernard EM, White DA. Aerosolized pen-
tamidine: effect on diagnosis and presentation of Pneumocystis carinii pneumonia.
Ann Intern Med 1990; 112:750 757.
48. Levine SJ, Kennedy D, Shelhamer JH, Kovacs A, Feuerstein IM, Gill VJ, Stock F,
Solomon D, Boylen CT, Masur H, Ognibene FP. Diagnosis of Pneumocystis
carinii pneumonia by multiple lobe, site-directed bronchoalveolar lavage with
immunouorescent monoclonal antibody staining in human immunodeciency
virus-infected patients receiving aerosolized pentamidine chemoprophylaxis. Am
Rev Respir Dis 1992; 146:838 843.
49. Metersky ML, Harrell JH, Moser KM. Lack of utility of bronchial brush biopsy
in patients infected with the human immunodeciency virus. Chest 1992;
101:680 683.
50. American Thoracic Society. Diagnostic standards and classication of tuberculosis
in adults and children. Am J Respir Crit Care Med 2000; 161:1376 1395.
51. Anderson C, Inhaber N, Menzies D. Comparison of sputum induction with ber-
optic bronchoscopy in the diagnosis of tuberculosis. Am J Respir Crit Care Med
1995; 152:1570 1574.
Bronchoscopy 309
52. Conde MB, Soares SL, Mello FC, Rezende VM, Almeida LL, Reingold AL, Daley
CL, Kritski AL. Comparison of sputum induction with beroptic bronchoscopy in
the diagnosis of tuberculosis. Am J Respir Crit Care Med 2000; 162:2238 2240.
53. McWilliams T, Wells AU, Harrison AC, Lindstrom S, Cameron RJ, Foskin E.
Induced sputum and bronchoscopy in the diagnosis of pulmonary tuberculosis.
Thorax 2002; 57:1010 1014.
54. Palenque E, Amor E, Bernaldo de-Quiros JC. Comparison of bronchial washing,
brushing and biopsy for diagnosis of pulmonary tuberculosis. Eur J Clin Microbiol
1987; 6:191 192.
55. Stenson W, Aranda C, Bevelaque FA. Transbronchial biopsy culture in pulmonary
tuberculosis. Chest 1983; 83:883 884.
56. Sarkar SK, Sharma GS, Gupta PR, Sharma RK. Fiberoptic bronchoscopy in the diag-
nosis of pulmonary tuberculosis. Tubercle 1980; 61:97 99.
57. American Thoracic Society. Diagnosis and treatment of disease caused by non-TB
mycobacteria. Am J Respir Crit Care Med 1997; 156(suppl):1 25.
58. Good RC, Snider DE. Isolation of nontuberculous mycobacteria in the United States,
1980. J Infect Dis 1982; 146:829 833.
59. OBrien RJ, Geiter LJ, Snider DE. The epidemiology of nontuberculous mycobacter-
ial diseases in the United States: results from a national survey. Am Rev Respir Dis
1987; 135:1007 1014.
60. Prechter GC, Prakash UBS. Bronchoscopy in the diagnosis of pulmonary histopla-
mosis. Chest 1989; 95:1033 1036.
61. Sobonya RE, Barbee RA, Wiens J, Trego D. Detection of fungi and other pathogens
in immunocompromised patients by bronchoalveolar lavage in an area endemic for
coccidioidomycosis. Chest 1990; 97:1349 1355.
62. DiTomasso JP, Ampel NM, Sobonya RE, Bloom JW. Bronchoscopic diagnosis of
pulmonary coccidioidomycosis. Comparison of cytology, culture, and transbronchial
biopsy. Diagn Microbiol Infect Dis 1994; 18:83 87.
63. Martynowicz M, Prakash UBS. Pulmonary blastomycosis: an appraisal of diagnostic
techniques. Chest 2002; 121:768 773.
64. Malabonga VM, Basti J, Kamholz SL. Utility of bronchoscopic sampling techniques
for cryptococcal disease in AIDS. Chest 1991; 99:370 372.
65. Kahn FW, Jones JM, England DM. The role of bronchoalveolar lavage in the diag-
nosis of invasive pulmonary aspergillosis. Am J Clin Pathol 1986; 86:518 523.
66. Albelda SM. Talbot GH, Gerson SL, Miller WT, Cassileth PA. Role of beroptic
bronchoscopy in the diagnosis of invasive pulmonary aspergillosis in patients with
acute leukemia. Am J Med 1984; 76:1027 1034.
67. Lortholary O, Meyoha MC, Dupont B, Cadranel J, Salmon-Ceron D, Peyramond D,
Simonin D. Invasive aspergillosis in patients with acquired immunodeciency syn-
drome: Report of 33 cases. Am J Med 1993; 95:171 187.
68. Sanguinetti M, Posteraro B, Pagano L, Pagliari G, Fianchi L, Mele L, La Sorda M,
Franco A, Fadda G. Comparison of real-time PCR, conventional PCR, and galacto-
mannan antigen detection by enzyme-linked immunosorbent assay using bronchoal-
veolar lavage uid samples from hematology patients for diagnosis of invasive
pulmonary aspergillosis. J Clin Micro 2003; 41:3922 3925.
69. Graham BS, Snell JD Jr. Herpes simplex virus infection of the adult lower respiratory
tract. Medicine 1983; 62(6):384 393.
310 Shah, Susanto, and Santiago
89. Mathur PN, Edell E, Sutedja T, Vergnon JM. Treatment of early stage non-small cell
lung cancer. Chest 2003; 123:176s 180s.
90. Edell ES, Cortese DA. Photodynamic therapy: its use in the management of broncho-
genic carcinoma. Clin Chest Med 1995; 16:455 463.
91. Shea JM, Allen RP, Tharratt RS, Chan AL, Siefkin AD. Survival of patients
undergoing Nd:YAG laser therapy compared with Nd:YAG laser therapy and
brachytherapy for malignant airway disease Chest 1993; 103:1028 1031.
92. Colt HG, Dumon JF. Airway stents: present and future. Clin Chest Med 1995;
16:465 478.
93. Ingenito EP, Reilly JJ, Mentzer SJ, Swanson SJ, Vin R, Keuhn H, Berger RL,
Hoffman A. Bronchoscopic volume reduction: a safe and effective alternative to
surgical therapy for emphysema. Am J Respir Crit Care Med 2001; 164:295 301.
94. Ingenito EP, Berger RL, Henderson AC, Reilly JJ, Tsai L, Hoffman A. Broncho-
scopic lung volume reduction using tissue engineering principles. Am J Respir
Crit Care Med 2003; 167:771 778.
95. Maxeld RA. New and emerging minimally invasive techniques for lung volume
reduction. Chest 2004; 125:777 783.
96. Snell GI, Holsworth L, Borrill ZL, Thomson KR, Kalff V, Smith JA, Williams TJ.
The potential for bronchoscopic lung volume reduction using bronchial prostheses.
Chest 2003; 124:1073 1080.
97. Rendina EA, De Giacomo T, Venuta F, Coloni GF, Meyers BF, Patterson GA,
Cooper JD. Feasibility and safety of the airway bypass procedure for patients with
emphysema. J Thorac Cardiovasc Surg 2003; 125:1294 1299.
98. Lausberg HF, Chino K, Patterson GA, Meyers BF, Toeniskoetter PD, Cooper JD.
Bronchial fenestration improves expiratory ow in emphysematous human lungs.
Ann Thorac Surg 2003; 75:393 397.
This page intentionally left blank
9
Percutaneous Tracheostomy
I. Introduction
313
314 Sakkour and Susanto
users of the Ciaglia PDT technique have largely adopted this Blue Rhino
technique over the serial dilator technique.
Laryngeal injury
Vocal cord paralysis
Glottic and subglottic stenosis
Infectious complications: sinusitis, otitis, pneumonia
Tracheal injury (tracheomalacia, tracheal dilation, and
tracheal stenosis)
the contraindications at any particular time in the clinical course of the critically
ill patient. Some evidence suggests that the risk of long-term airway compli-
cations signicantly increase beyond the 10th day of translaryngeal intubation
(23). Many of the studies that compared outcomes of patients converted at
various times in their clinical course to tracheostomy vary in design and sampling
and suffer from multiple design problems (24). Armstrong et al. (25) found that
early use of tracheostomy in blunt trauma patients with ventilator dependence
reduces length of stay in ICU and in the hospital with no adverse effect on mor-
bidity and mortality rates. However, Sugerman et al. (26) failed to show any sig-
nicant difference with early tracheostomy. Brook et al. (27) concluded in their
study that early tracheostomy was associated with shorter hospital stay, lower
hospital cost, shorter length of stay in the ICU, and shorter duration of mechan-
ical ventilation. Kluger et al. (28) found that tracheostomy performed between
day 0 and day 3 of mechanical ventilation may decrease the incidence of pneu-
monia in critically ill trauma patients. A recent prospective randomized study of
120 critically ill medical patients by Rumbak et al. (29) demonstrated that early
(within 48 hr) percutaneous tracheostomy was associated with signicantly lower
rates of mortality, pneumonia, and accidental extubations compared with pro-
longed translaryngeal intubation, in which tracheostomy was performed on day
14 to day 16.
The 1989 ACCP consensus conference on articial airway in mechanically
ventilated patients recommended tracheostomy in patients whose anticipated
need for articial airway is more than 21 days. It also recommended that once
the decision for tracheostomy is made, the procedure should be performed as
early as possible to minimize the duration of translaryngeal intubation, based
on individual patient factors (30).
III. Contraindications
absolute, and what is feasible and what is not, becomes less clear with the increas-
ing experience and in the hands of highly skilled operators. Beiderlinden et al.
(31) found that bronchoscopically guided PDT in patients on high PEEP did
not jeopardize oxygenation one hour and 24 hours following PDT (31). Meyer
et al. (32) found that trained physicians can safely perform bedside percutaneous
tracheostomy in patients who had undergone previous tracheostomy. Although
not considered a standard practice, several studies have reported that emergency
percutaneous tracheostomy is feasible and safe (33,34). Morbid obesity has
traditionally been a contraindication for percutaneous tracheostomy because of
the presumed difculty in identifying landmarks. This assumption was never sup-
ported by any trials. In experience, percutaneous tracheostomy in morbidly obese
patients can be performed safely as long as proper precautions and adjustments
are made, including using a longer tracheostomy tube to accommodate the
additional soft tissue of the neck. Mansharamani et al. (35) reported 13
consecutive cases of PDT in patients with a body mass index of 28 to 62.
They reported no complications, technical difculty, or failure to place the
tracheostomy. Laryngeal mask airway has been shown to be as an effective
and reliable alternative to endotracheal intubation as a temporary articial
airway during percutaneous tracheostomy (36,37).
Figure 1 A large bore needle with a catheter sheath over it is introduced into the tra-
cheal lumen at the level of the rst tracheal interspace. Source: Courtesy of Cook Critical
Care, Inc., Bloomington, Indiana, U.S.A.
in individual ways. We will focus on describing the Ciaglia PDT the way we
do it.
The preparatory steps before PDT are just as important as the tracheostomy
itself. Proper patient preparation cuts down on the potential complications that
may arise during the procedure. The mechanical ventilator settings are adjusted
Figure 2 After the needle is removed, a guidewire is inserted through the catheter
sheath into the tracheal lumen. Once the guidewire is in place, the catheter is removed.
Source: Courtesy of Cook Critical Care, Inc., Bloomington, Indiana, U.S.A.
Percutaneous Tracheostomy 319
Figure 3 Following dilation of the anterior tracheal wall with an 11 French dilator
(not shown), a guiding catheter is introduced over the guidewire. Source: Courtesy of
Cook Critical Care, Inc., Bloomington, Indiana, U.S.A.
Figure 4 A dilator is passed over the guiding catheter and guidewire to dilate the
anterior tracheal wall. Dilation can be achieved serially using a progressively larger
serial dilators or a single tapered dilator (Blue Rhinow). Source: Courtesy of Cook
Critical Care, Inc., Bloomington, Indiana, U.S.A.
320 Sakkour and Susanto
to an FIO2 of 1.0, with appropriate alarm settings, and placed on volume control
with appropriate rate and tidal volume to maintain adequate basal minute venti-
lation when a paralytic agent is used. The high-pressure limit is increased to
accommodate the increase in peak airway pressure due to the presence of the
bronchoscope in the endotracheal tube. Rolled towels are placed behind the
patients back in a supine position to extend the head and neck. The blood
pressure, heart rate, and pulse oximetry are continuously monitored throughout
the procedure. The surface markers of the trachea are carefully palpated and
marked. The rst or second tracheal interspace is the preferred tracheostomy
site. The subcricoid space may be used in patients with short neck or severe
kyphosis. We avoid the space below the third tracheal ring to decrease the risk
of accidental erosion into the innominate artery and minimize the potential
trauma to the thyroid isthmus. The skin is prepped and draped in a sterile fashion.
The patient is premedicated with a combination of intravenous benzo-
diazepine and opiate analgesics, for example, meperidine, morphine, or fentanyl.
Alternatively, the patient could be maintained on intravenous proprofol. A non-
depolarizing neuromuscular blocking agent is also used. A 2-cm skin incision
is made vertically or transversely at midline over the selected site, at least one
ngerbreadth above the suprasternal notch. A blunt dissection is then done
using a curved clamp until the pretracheal fascia is felt. The left middle nger
and thumb are used to secure the lateral edges of the trachea, while the index
nger is used to palpate the anterior tracheal wall to select the appropriate
tracheal interspace for insertion site. A exible bronchoscope is introduced
into the endotracheal tube. The tip of the bronchoscope is kept inside the tube
to prevent damage by the needle. The cuff of the endotracheal tube is then
deated, and the tube is slowly withdrawn to the subglottic space, guided by
Percutaneous Tracheostomy 321
V. Complications
Bleeding
Infection
Accidental extubation
Paratracheal insertion
Esophageal perforation
Subcutaneous emphysema
Pneumothorax
Tracheal ring fracture
Tracheal stenosis
Tracheoamalacia
Percutaneous Tracheostomy 323
The trend toward minimally invasive surgery and the development of interven-
tional services within the non-surgical specialties has spurred considerable
interest in bedside PDT in the ICU. The economic issue was one of the important
aspects of PDT that led to its initial popularity. OST has traditionally been
performed in the operating room. In many institutions it remains so. However,
performing tracheostomy in the ICU avoids the potential hazards of transporting
critically ill patients to and from the operating room. When rst introduced,
bedside PDT was touted not only for its ease of performance with safety
prole comparable to OST, but also for a signicant decrease in hospital
charges and a more efcient utilization of ICU resources. The latter reects the
absence of operating room charges and anesthesia fees associated predominantly
with OST performed in the operating room. More recently, there are reports of a
comparable safety prole in performing bedside OST (61 63). Bedside OST is
usually performed with a standard reusable tracheostomy tray and electrocautery,
without operating room charges and anesthesia fees. The professional fee for
tracheostomy is the same, regardless of where or how tracheostomy is performed.
The cost of using the surgical instruments at bedside is minimal. PDT is usually
performed using disposable kits under bronchoscopic guidance; therefore, there
are additional costs and charges for these items, rendering PDT somewhat
more costly than OST when both are performed at bedside. Although most
studies have reported a shorter operative time for PDT when compared with
OST (39,47,62,64,65), this is only an issue when the procedure is performed in
the operating room, in which longer operating time entails higher charges. The
time from decision to performance of the tracheostomy is shorter in PDT
compared with OST (47,66). Whether this faster decision to performance time
will reect in overall decreased hospital and ICU days and the overall cost
remains to be seen. Some studies also report less hemorrhage and wound
infection with PDT, while others have shown no difference (47,65,67 69).
Most of the studies looking at the safety and outcome prole of the two
techniques lack the rigorous design for this comparison to be meaningful.
Many of the OST in the studies were performed in the operating room, not at
bedside. Even the meta-analyses comparing percutaneous tracheostomy and sur-
gical tracheostomy are plagued with inconsistencies, including the heterogeneity
of the techniques included under percutaneous tracheostomy and the wide
ranging quality of studies included (39,40). In general, the superiority of one
procedure over the other has not been established. What has been established
in multiple studies is that PDT has a comparable safety prole to OST (70 74).
In the early post-tracheostomy period, care should focus on keeping the wound
clean. Dressing changes are indicated twice daily or whenever dressing is
Percutaneous Tracheostomy 325
soiled. The site can be cleaned with a mixture of hydrogen peroxide and saline
solution. If the tracheostomy tube is secured with sutures, then sutures can be
removed in seven to ten days. The tracheostomy tube can be secured comfortably
with a foam-pad Velcrow neck strap. The inner cannula can be changed daily in
the early post-operative period. We do not routinely change the tracheostomy
tube after placement, unless a different tube is needed or there is a malfunction
of the existing tube. Dislodgement of the tracheostomy tube in the rst seven
to ten days before the tract has matured can be problematic. Endotracheal
intubation should be performed if the tracheostomy tube cannot be reinserted
safely. Bronchoscopic guidance and conrmation of placement is strongly
recommended.
Routine care for an established tracheostomy generally centers on suction-
ing, appropriate humidication, weekly cleaning of the tube, and daily change
of the inner cannula. Cuffed tubes are used when patients are still mechanically
ventilated, or when risks of aspiration are high. Cufess tubes promote better
clearance of secretions, and less likelihood of infection.
The intervals to downsizing of tracheostomy tubes and decannulation
should be individualized. Factors to consider include the initial indication for tra-
cheostomy and the current respiratory status, the amount of secretions and the
patients ability to clear the secretions, and the mental status. Once decannulated,
the stoma closes rapidly within days, especially if the stoma has been downsized.
VIII. Future
Many percutaneous tracheostomy methods have been introduced over the past
half-century. Almost two decades since PDT was rst introduced, a large body
of literature has conrmed that in skilled hands, the safety prole of the technique
is comparable to OST. PDT comes in the era of cost containment, increasing role
of minimally invasive surgery, and the growing intensivist movement, all of
which have contributed to the momentum towards its acceptance as a safe and
viable bedside technique in the ICU. PDT should not be viewed as a replacement
for OST because OST still has its vital role and place. In the future, we will likely
see more renements of percutaneous tracheostomy techniques beyond PDT.
In our opinion, PDT will likely be the benchmark to which future percutaneous
tracheostomy techniques will be compared.
References
1. Frost EA. Tracing the tracheostomy. Ann Otol Rhinol Laryngol 1976; 85:618 624.
2. Sheldon CH, Pudenz RH, Feshwater DB, Crue BL. A new method for tracheotomy.
J Neurosurg 1957; 12:428 431.
3. Toye FJ, Weinstein JD. A percutaneous tracheostomy device. Surgery 1969;
65:384 389.
326 Sakkour and Susanto
25. Armstrong PA, Mc Carthy MC, Peoples JB. R. Reduced use of resources by early
tracheostomy in ventilator dependant patients with blunt trauma. Surgery 1998;
124:763 767.
26. Sugerman HJ, Wolfe L, Pasquale MD, Rogers FB, OMalley KF, Knudson M,
DiNardo L, Gordon M, Schaffer S. Multicenter, randomized, prospective trial of
early tracheostomy. J trauma 1997; 43:741 747.
27. Brook A, Sherman G, Malen J, Kollef MH. Early vs late tracheostomy in
patients who require prolonged mechanical ventilation. Am J Crit Care 2000;
9:352 359.
28. Kluger Y, Paul DB, Lucke J, Cox P, Colella JJ, Townsend RN, Raves JJ, Diamond
DL. Early tracheostomy in trauma patients. Eur J Emerg Med 1996; 3:95 101.
29. Rumbak KJ, Newton M, Truncale T, Schwartz SW, Adams JW, Hazard PB. A pro-
spective, randomized study comparing early percutaneous dilational tracheostomy to
prolonged translaryngeal intubation (delayed tracheotomy) in critically ill medical
patients. Critical Care Med 2004; 32(8):1689 1694.
30. Plummer AL, Gracey DR. Consensus conference on articial airways in patients
receiving mechanical ventilation. Chest 1989; 96:178 180.
31. Beiderlinden M, Groeben H, Peters J. Safety of percutaneous dilatational tracheost-
omy in patients ventilated with high positive end-expiratory pressure (PEEP).
Intensive Care Med 2003; 29(7):1190.
32. Meyer M, Critchlow J, Manasharamani N, Angel LF, Garland R, Ernst A. Repeat
bedside percutaneous dilatational tracheostomy is a safe procedure. Crit Care Med
2002; 30(5):986 988.
33. Ben-Nun A, Altman E, Best LA. Emergency percutaneous tracheostomy in trauma
patients: an early experience. Ann Thorac Surg 2004; 77(3):1045 1047.
34. Ault MJ, Ault B, Ng PK. Percutaneous dilatational tracheostomy for emergent
airway access. J Intensive Care Med 2003; 18(4):222 226.
35. Mansharamani NG, Koziel H, Garkland R, LoCicero J III, Critchlow J, Ernst
A. Safety of bedside percutaneous tracheostomy in obese patients in the ICU.
Chest 2000; 118:1412 1418.
36. Dosemeci L, Yilmaz M, Gurpinar F, Ramazanoglu A. The use of the laryngeal mask
airway as an alternative to the endotracheal tube during percutaneous dilatational
tracheostomy. Intensive Care Med 2002; 28(9):1368.
37. Craven RM, Laver SR, Cook TM, Nolan JP. Use of pro-seal LMA facilitates percu-
taneous dilational tracheostomy. Can J Anaesth 2003; 50(7):718 720.
38. Datta D, Onyirimba F, McNamee MJ. The Utility of Chest Radiographs following
percutaneous dilatational tracheostomy. Chest 2004; 123(5):1603 1606.
39. Freeman BD, Isabella K, Lin N, Buchman TG. A meta-analysis of prospective trials
comparing percutaneous and surgical tracheostomy in critically ill patients. Chest
2000; 118:1412 1418.
40. Dulguerov P, Gysin C, Perneger TV, Chevrolet JC. Percutaneous or surgical tra-
cheostomy: a meta-analysis. Crit Care Med 1999; 27:1617 1625.
41. Goldenberg D, Ari EG, Golz A, Danino J, Nezer A, Joachims HZ. Tracheotomy
complications: a retrospective study of 1130 cases. Otolaryngol Head Neck Surg
2000; 123:495 500.
42. Polderman KH, Spijkstra JJ, de Bree R, et al. Percutaneous dilatational tracheostomy
in the ICU. Chest 2003; 123(5):1336 1338.
328 Sakkour and Susanto
64. Freeman BD, Isabella K, Cobb JP, Boyle WA III, Schmieg RE Jr, Kolleff MH, Lin N,
Saak T, Thompson EC, Buchman TG. A prospective, randomized study comparing
percutaneous with surgical tracheostomy in critically ill patients. Crit Care Med
2001; 29(5):926 930.
65. Hazard P, Jones C, Benitone J. Comparative clinical trial of standard operative tra-
cheostomy with percutaneous tracheostomy. Crit Care Med 1991; 19(8):1018 1024.
66. Friedman Y, Mizock BA. Percutaneous vs surgical tracheostomy: procedure of
choice or choice of procedure. Crit Care Med 1999; 27(8):1684 1685.
67. Griggs WM, Myburg JA, Worthley LI. A prospective comparison of a percutaneous
tracheostomy technique with standard surgical tracheostomy. Intensive Care Med
1991; 17(5):261 263.
68. Cobean R, Beals M, Moss C, Brendenberg CE. Percutaneous dilatational tracheost-
omy. A safe, cost-effective bedside procedure. Arch Surg 1996; 131(3):265 271.
69. Holdgaard HO, Pedersen J, Jensen RH, Outzen KE, Midtgaard T, Johansen LV,
Moller J, Paaske PB. Percutaneous dilatational tracheostomy versus conventional
surgical tracheostomy. A clinical randomized study. Acta Anaesthesiol Scand
1998; 42(5):545 550.
70. Sustic A, Krstulovic B, Eskinja N, Zelic M, Ledic D, Turina D. Surgical tracheost-
omy vs percutaneous dilational tracheostomy in patients with anterior cervical spine
xation: preliminary report. Spine 2002; 27(17):1942 1945.
71. Khalili TM, Koss W, Margulies DR, Morriaon E, Shabot MM. Percuateous dilational
tracheostomy is as safe as open tracheostomy. Am Surg 2002; 68(1):92 94.
72. Lukas J, Stritesky M. Tracheostomy in critically ill patients. Bratisl Lek Listy 2003;
104(7 8):239 242.
73. Kaylie DM, Andersen PE, Wax MK. An analysis of time and staff utilization for
open versus percutaneous tracheostomies. Otolaryngol Head Neck Surg 2003;
128(1):109 114.
74. Wu JJ, Huang MS, Tang GJ, Shih SC, Yang CC, Kao WF, Huang MH, Lee CH.
Percutaneous dilatational tracheostomy vs open tracheostomya prospective,
randomized controlled trial. J Chin Med Assoc 2003; 66(8):467 473.
This page intentionally left blank
10
Radiology in the Intensive Care Unit
I. Introduction
The care of the intensive care unit (ICU) patient requires frequent radiological
consultation. Daily review of radiographs is an integral part of critical care
rounds. This chapter will address the role of thoracic radiology in the ICU patient.
The indications, technique, interpretation, and limitations of the portable
chest radiograph will be discussed. The proper positioning and complications
of devices used in the ICU and the radiologic ndings of common thoracic
abnormalities encountered will also be described. Lastly, the role of multi-
detector computed tomography (MDCT) imaging and the role of interventional
radiology will be summarized.
331
332 Barack et al.
B. Technique
Portable Antero-Posterior View
Portable chest radiography accounts for 50% of all chest radiographs performed
in hospitals (9). The key to obtaining optimal repeat portable chest radiographs
on the same ICU patient is to minimize variation in the technical parameters
of successive radiographs (10 12). At our institution, all portable ICU chest
radiographs are performed at a 50 in. source image distance (SID), with the
Radiology in the ICU 333
patient in the supine position, during peak inspiration with 85 kVp. The exposure
is timed with respect to peak inspiratory pressure if the patient is on mechanical
ventilation. The technique is recorded on an adhesive label, which is placed at
the patients bedside for use on all subsequent examinations. It has been found
that the supine position minimizes position errors secondary to rotation,
distortion errors secondary to kyphosis, and unwanted body artifacts secondary
to head or extremity obscuring the thoracic anatomy.
Portable chest radiographs result in the patient receiving two to four times
the radiation dose of a radiograph obtained with stationary equipment, primarily
because of increased scatter radiation (12). Scatter radiation at ICU nursing
stations has been demonstrated to be signicantly below the maximal permissible
dose for non-occupational workers (13,14).
Special Views
Special portable views that may be useful in the ICU patient include lateral,
transthoracic lateral, and lateral decubitus views of the chest, left lateral decubi-
tus view of the abdomen, and an antero-posterior (AP) view of the lower chest
and upper abdomen.
In one series, the portable lateral chest radiograph yielded an 11% inci-
dence of either unexpected conditions, such as catheter or tube malpositions,
or improved interpretation on the AP radiograph (15). The transthoracic lateral
view may conclusively demonstrate a pneumothorax. The lateral decubitus
view also demonstrates a pneumothorax and differentiates free from loculated
pleural uid and pneumothorax from pneumomediastinum. The left lateral decu-
bitus view of the abdomen is frequently used to demonstrate free intraperitoneal
air, which collects superiorly to the liver. An AP view of the lower chest and
upper abdomen is useful in localizing the tip of a nasoenteric tube.
Oral Contrast
(Fig. 2). Because of the compromised condition of the ICU patient, isoosmolar
water soluble contrast is used to avoid possible uid compartmental shifts.
quality over a wide range of exposures, the ability to postprocess the image, and
advanced imaging processing algorithms that minimize scatter and reduce radia-
tion dose. The major disadvantages of slightly higher image noise and lower
spatial resolution are not signicant problems with portable chest radiographs
(17). Although the classic technical errors of patient motion and malpositioning,
incorrect patient identication, incorrect examination, and double exposure occur
with the same frequency as with conventional lm-screen systems (18), repeat
rates of less than 1% are reported with CR systems (19). This is primarily attri-
buted to fewer repeat examinations because of under and overexposure.
Digital radiography is usually an integral part of an information system
called picture archiving and communications system (PACS). At present, the
majority of radiographic images are acquired in a digital format, and it is esti-
mated that by the end of this decade, over half of all radiology departments in
medical centers will be completely or almost completely digital (20).
Major advantages of PACS include elimination of lost studies, marked
reduction in lm costs, fewer repeat examinations, improved image quality,
rapid availability of images, rapid viewing of large studies such as MDCT or
magnetic resonance imaging (MRI), and improved efciency of radiology
personnel and clinicians.
336 Barack et al.
C. Interpretation
Accurate interpretation of the portable chest radiograph demands a careful, sys-
tematic evaluation, which requires discipline and practice. The following
approach is intended as a guide, although individual approaches may vary.
The initial observation should include the technical quality of the radiograph
for exposure factors and patient positioning, and insure that the lung bases and
apices have been included on the lm. The radiograph should then be inspected
for any tubes, lines, or catheters, and any malposition, complication, or change
from the prior radiograph noted. Examination of the osseous structures, the
mediastinum (including the vascular pedicle), the heart, the pulmonary vessels,
and the lungs should follow. An attempt should be made to form a preliminary
impression and a relevant differential diagnosis.
There are ve additional factors to consider when interpreting a portable
chest radiograph. First, patients are expected to inspire to the same lung
volumes on successive radiographs in the absence of intra-abdominal changes,
cardio-respiratory problems, or cardiac decompensation (10). Then, lower lung
volumes are only because of a decrease in lung, chest wall, or abdominal
compliance.
Second, the heart size does not change when the patient goes from the erect
to the supine position. The apparent increase in heart size on a supine lm is
because of magnication. As the heart is an anterior structure within the chest,
it will be magnied on an AP view when the cassette is placed behind the
patient, as opposed to a postero-anterior (PA) view when the cassette is closer
to the heart. The upper limit of normal for the cardiothoracic ratio in supine
patients with a normal heart is 0.57 to 0.58 (10,21).
Third, the vascular pedicle extends from thoracic inlet to the superior
aspect of the heart. The right border is formed by the right brachiocephalic
vein superiorly and the superior vena cava inferiorly. The left border is usually
formed by the left subclavian artery above the aortic arch. Therefore, the right
side of the vascular pedicle is venous and the left side arterial. As veins are
more compliant than arteries, changes in intravascular volume will be reected
by a larger change in the right side of the vascular pedicle than the left side.
The width of the vascular pedicle is measured horizontally from where the
right mainstem bronchus crosses the superior vena cava to a perpendicular
line drawn from where the left subclavian artery arises from the aorta (22).
Fourth, the size of the azygous vein changes linearly with the width of the
vascular pedicle, and both closely correlate with changes in intravascular systemic
blood volume (10). Assuming the supine position, rotation to the right, uid over-
load, acute right heart failure, and pericardial tamponade (Fig. 3) will increase the
width of the vascular pedicle, whereas increasing ventilatory pressure, decreasing
intravascular volume, and rotation to the left will decrease the width of the vas-
cular pedicle (10, 22). The vascular pedicle width (VPW) varies according
to the patients body habitus. The normal VPW is 48 + 5.0 mm on a 72-in.
Radiology in the ICU 337
Tubes
Endotracheal Tube
The tip of the endotracheal tube should ideally be located 3 7 cm above the
carina with the head in neutral position (23,24). This position allows for
maximal excursion of the endotracheal tube within the airway, with changing
head position preventing the possibility of subsequent malposition. Neutral
head position is best insured with the patient supine, and this is the recommended
position for the immediate post-intubation radiograph.
When the carina is not identied on the immediate post-intubation radio-
graph, a repeat radiograph is recommended. The method of estimating the
location of the carina from the projection of the tip of the endotracheal tube in
gastric distention, esophageal air, and deviation of the trachea by the balloon cuff.
Examination of the patient in a 258 right posterior oblique projection accurately
demonstrates the endotracheal tube position in 92% of radiographs (34).
Tracheal rupture is a serious, less common complication of endotracheal
tube intubation. Early radiographic ndings are orientation of the distal portion
of the tube to the right of the tracheal air column with overdistention of
the tube cuff towards the endotracheal tube tip and subsequent development of
pneumomediastinum, subcutaneous air, and pneumothorax (35).
Aspiration pneumonitis has been reported to occur in 8% of endotracheal
tube intubations (29,36). It is usually easily identied on the post-intubation
radiograph and frequently becomes more apparent on subsequent examinations.
Fillings, teeth, and dentures may also be aspirated or swallowed secondary to
dislodgment during intubation (37).
Tracheostomy Tube
Tracheostomy is performed on patients with upper airway obstruction or on
patients who require long-term ventilator support (23). The tracheostomy tube
is inserted between the rst and second tracheal ring or second and third tracheal
ring. The tip of the tracheostomy tube should be one-half to two-thirds of the
distance between the tracheal stoma and the carina at the level of the third thor-
acic vertebral body. Neither head exion nor extension affects the location of
the tip of the tracheostomy tube (12). The lumen of the tube should be one-
half to two-thirds the diameter of the trachea, and the cuff should not distend
the tracheal wall. A post-tracheostomy lm showing the internal and external
ends of the tube in the same plane may indicate that the tracheostomy tube has
not distended properly into the trachea, which can be conrmed on a portable
lateral radiograph (38).
A small amount of subcutaneous air in the neck and a small pneumo-
mediastinum are frequent ndings on the post-tracheostomy radiograph (39).
However, fascial plane disruption may result in a large amount of air in the
neck, with a larger pneumomediastinum and a pneumothorax frequently seen
following injuries to the lung apex. Pneumothorax may also be seen in tracheal
perforation. Late complications include tracheo-innominate artery stulas or
tracheoesophageal stula (40 42). The stulas are usually secondary to pro-
longed hyperination of the cuff and occur at the level of the cuff with erosion
anteriorly (tracheo-innominate artery) or posteriorly (tracheoesophageal). In tra-
cheoesophageal stulas above the cuff, gastric contents may accumulate in the
upper trachea. With stulas below the cuff, aspiration of gastric contents into
the lungs is a routine occurrence. Other late complications include stricture, tra-
cheomalacia, and tracheal stenosis.
Nasoenteric Tubes
Nasoenteric tubes are commonly used in ICU patients for gastric decompression
or feeding. Sideholes are present along the distal 10 cm of nasogastric tubes.
Radiology in the ICU 341
Therefore, the tip of the tube should extend at least 10 cm beyond the-
gastroesophageal junction. A more proximal position may result in obstruction
of the distal esophagus with applied suction or may allow contents entering the
distal portion of the tube to exit into the distal esophagus and predispose to
aspiration (23). Incorrect placement was seen in 4.4% of nasogastric tube
insertions in one series of 340 patients (43). Similarly, the most proximal sidehole
of the tube should be placed inferior to the anastomotic site in postoperative
esophagectomy patients.
Feeding tubes are used for nutritional support as an alternative to intrave-
nous feeding. Ideally, they should be positioned distal to the gastric pylorus to
reduce gastroesophageal reux. A post-placement radiograph should always be
obtained, as malpositions are common following insertion (44,45) (Fig. 5).
Malpositioned tubes may be coiled in the pharynx, esophagus, or stomach. In
intubated or obtunded patients, or patients lacking a gag reex, tubes may pass
into the trachea and preferentially enter the right mainstem bronchus.
Immediate complications of nasoenteric tube insertion include traumatic
esophageal perforation with pneumomediastinum or, more rarely, bronchial
perforation and pneumothorax (23). Parietal pleural perforation may also result
from malposition of the tube within the esophagus and should be suspected
when a pleural effusion, pneumomediastinum, mediastinal widening, or
mediastinal air uid levels appear rapidly after beginning tube feedings (36,46).
Figure 5 Feeding tube in the left lower lobe bronchus. Antero-posterior post-insertion
radiograph demonstrates the tip of the feeding tube in the left lower lobe bronchus (arrow-
head). The tube was subsequently removed and reinserted without complication.
342 Barack et al.
Thoracostomy Tube
Thoracostomy tubes are commonly used in ICU patients to drain pleural uid col-
lections and pneumothoraces. In the supine patient, mobile pleural uid collects
posteriorly and pleural air anteriorly. A pleural drainage tube for mobile uid col-
lection should therefore be positioned postero-inferiorly through the sixth to
eighth intercostal space at the level of the mid axillary line, preferably guided
by ultrasonic localization (52). If there is uncertainty about the potential for
injury to abdominal structures, placement through the fth intercostal space pro-
vides a margin of safety. A thoracostomy tube for pneumothorax treatment
should be positioned near the lung apex at the level of the anterior axillary line
and be directed antero-superiorly. Loculated pleural uid collections, as seen
in empyemas, require precise localization with ultrasonography or computerized
axial tomography (23). Rapid improvement is expected on post-drainage radio-
graphs if the tube is properly positioned. Conversely, a persistent pneumothorax
or undrained pleural uid collection suggests a malpositioned or malfunctioning
tube. A malfunctioning tube may result from kinking or plugging by debris (12).
Improper tube position may be appreciated on the AP radiograph.
However, a lateral radiograph or computerized axial tomography may be
required to determine the exact tube position (15,53). A malpositioned tube
may be in the subcutaneous or extrapleural soft tissues, within an interlobar
ssure (Fig. 6) or within the lung parenchyma (Fig. 7).
The proximal side hole of the chest tube should be medial to the inner
margin of the ribs and can be identied along the radio-opaque edge of the
Radiology in the ICU 343
Figure 6 Tip of right chest tube in the right major ssure. Sagittal computed tomogra-
phy reformation (mediastinal window) in a patient with a non-draining chest tube demon-
strates the tip of the right chest tube within the right major ssure (arrow). A large
empyema (E) is present in the right posterior hemithorax, causing anterior displacement
and compression of the right lower lobe (arrowheads).
tube (54). Malposition of the tube in the subcutaneous tissue may result in
subcutaneous emphysema and occurs more frequently in patients with excessive
subcutaneous fat (44). Malpositioned tubes within an interlobar ssure are associ-
ated with a 29% rate of unsatisfactory drainage (55). Malposition within the lung
parenchyma may be associated with bronchopleural stula, pulmonary lacera-
tion, and hematoma (36). Pulmonary laceration and hematoma are particularly
seen in patients with pleural adhesions or decreased lung compliance.
Complications of drainage tubes because of tube insertion include bleeding
secondary to laceration of an intercostal artery, laceration of the diaphragm with
associated splenic, hepatic, and gastric injuries, and mediastinal and parenchymal
lung injury (56). A delayed complication of thoracostomy tubes is unilateral
pulmonary edema on the side of tube placement due to rapid pulmonary
344 Barack et al.
Figure 7 Tip of the right chest tube in the right middle lobe. Axial computed tomogra-
phy image (mediastinal window) in a patient with a right empyema and a non-functioning
right chest tube demonstrates the tip of the right chest tube within the lateral segment of
the right middle lobe (arrow). The right major ssure is clearly visualized (arrowhead).
The tube was removed without complications.
re-expansion after removal of a large amount of uid or air from the pleural space
(36). A less common delayed complication is pulmonary infarction due to suc-
tioning of lung tissue by the chest tube even at low suction pressures (57).
After a thoracostomy tube has been removed, a tubular structure containing air
or uid may persist in the position previously occupied by the thoracostomy
tube. Because of a local inammatory response and associated pleural thickening,
free communication between this tubular tract and the rest of the pleural space
may not exist and the isolated tube tract may simulate a localized pneumothorax
or abscess on the chest radiograph (52). Such tracts usually decrease and
disappear within a few days and rarely become infected.
Catheters
Central Venous Catheters
Approximately 38% of patients in the ICU have central venous catheters (58).
The most proximal port is 5 cm from the tip. The ideal location of the tip of
the central venous catheter is the superior vena cava at or slightly above the
entrance of the azygous vein. This location assures that all ports of the central
venous catheter are beyond the most proximal venous valves. Below the
azygous vein, the superior vena cava becomes intrapericardial, and caval perfor-
ation is associated with a pericardial effusion and eventual tamponade (11).
Radiology in the ICU 345
If the catheter is to be used to monitor central venous pressure, the tip of the
pressure monitoring port must be proximal to any competent valve. The last valve
in the subclavian vein occurs 2 cm distal to the junction of the subclavian and
internal jugular veins at about the level of the rst anterior rib (59). Therefore,
the tip of the pressure monitoring port should be visualized medial to the anterior
portion of the rst rib at the level of the rst anterior intercostal space. The
right and left brachiocephalic veins join to form the superior vena cava at this
level (23). Neither the brachiocephalic veins nor the superior vena cava
contains valves.
Malposition of central venous lines occurs in approximately one-third of
catheter insertions, and most malpositions are not clinically suspected (60).
The subclavian vein is the most often used site for central venous placement.
A retrospective study of 500 subclavian vein catheterizations revealed only
68% to be properly positioned, with 21.4% to be positioned in the right atrium
and 0.4% in the right ventricle. Cardiac malposition was more common with a
right-sided approach (30.1%) than a left-sided approach (61). The average
safe insertion length of the central venous catheter from either the right or left
subclavian or internal jugular vein is 16.5 cm in most adults, and no central
venous catheter should be inserted greater than 20 cm from these access
sites (62).
The most common venous malposition of the subclavian venous catheter is
the ipsilateral internal jugular vein (Fig. 8), which is reported to occur in 15% of
such central venous insertions (36). A malpositioned catheter in this location
may produce abnormal sensations in the ear or severe headache (63). Retrograde
infusion of intravenous uids into the internal jugular vein in this malposition
may result in thrombosis, erosion of the vein, and pooling of infused solutions
near the brain in the supine patient (64).
Another common venous malposition occurs when the catheter enters
the contralateral brachiocephalic vein (Fig. 9). This is more common when the
junction of the brachiocephalic veins is higher and at a less acute angle and
more common with left-sided insertions. Malposition of the catheter in the
azygous vein occurs occasionally. Rarely, a catheter may be positioned in the
right superior intercostal vein, the internal mammary vein, the pericardiophrenic
vein, the left superior intercostal vein, and an inferior thyroid vein.
The coronary sinus, which drains the cardiac veins and empties into the
posterior aspect of the right atrium near the entry of the inferior vena cava,
may be entered by a malpositioned catheter in the right atrium, as may the
hepatic veins and inferior vena cava (23).
Catheter malposition within the superior vena cava may occur in left-sided
catheter insertions when the catheter tip abuts perpendicularly upon the right
lateral wall of the superior vena cava. This position causes the tip to repeatedly
hit the lateral caval wall with inspiratory movements or with head and neck
movements with left internal jugular catheters. This can eventually damage the
caval epithelium and cause caval rupture hours or days after insertion (65).
346 Barack et al.
Figure 8 Tip of right subclavian venous catheter in the right internal jugular vein.
Antero-posterior radiograph obtained after insertion demonstrates right subclavian
venous catheter extending into the right internal jugular vein (arrow).
A gentle curve of the tip of the catheter may be seen when the catheter tip is in
contact with the lateral caval wall, and the catheter should promptly be
repositioned (66).
A simulated catheter malposition occurs in the presence of a persistent
left superior vena cava, which is the most common anomaly of systemic
venous drainage (Fig. 10). This anomaly occurs in 0.3% of the normal population
and 4.3% of patients with congenital heart lesions when the left anterior and
common cardinal veins fail to regress normally (67,68). This vein courses
inferiorly along the left side of the mediastinum and drains into the coronary
sinus. Most patients with this anomaly also have a right-sided superior vena
cava, which may be smaller than normal (23).
The most common complication of central venous line placement is
pneumothorax, which has been reported in 5.6% of patients with any central
line placement (3) and 6% of subclavian venous insertions (69). The latter is
explained by the 1 cm or less proximity of the apical pleura to the subclavian
vein. Therefore, a radiograph is recommended following any unsuccessful
subclavian venous insertion to exclude a pneumothorax before contralateral
Radiology in the ICU 347
Figure 9 Tip of right subclavian catheter in the left brachiocephalic vein. Antero-
posterior radiograph demonstrates the right subclavian catheter extending into the left
brachiocephalic vein (arrow).
Figure 10 Pulmonary artery catheter in the left superior vena cava. Antero-posterior
radiograph demonstrates the tip of the pulmonary artery catheter in the right interlobar pul-
monary artery. The catheter was inserted into the right subclavian vein and coursed
through the right brachiocephalic vein, a left superior vena cava (arrow), the coronary
sinus, right atrium, right ventricle, and main pulmonary artery to reach the nal location.
more acute and curvaceous course than the right before joining to form the
superior vena cava, closer scrutiny for malpositions and complications is required
when left-sided insertions are reported. On the right side, the internal jugular
insertion is safer than the subclavian insertion because of the higher rate of
pneumothorax with the latter approach.
Dialysis Catheters
The types of dialysis catheters used in the ICU are acute dialysis catheters (83),
cuffed tunneled catheters (84), and subcutaneous vascular port catheters.
Acute dialysis catheters are primarily used as short-term access in patients
with malfunction of permanent access or in bed-bound patients with acute
renal failure.
Most acute dialysis catheters are non-cuffed, non-tunneled, dual lumen
catheters composed of polyurethane. Polyurethane is quite sturdy, permitting a
larger internal lumen for a given outer diameter, and is less thrombogenic than
silicone (83).
Bedside ultrasound to localize the vein reduces the number of needle passes,
failed placements, and insertion-related complications and increases the success
rate of inexperienced operators to 95% (85). In addition, 28% to 35% of dialysis
patients have demonstrated signicant vein abnormalities, such as total occlusion,
non-occlusive thrombus, stenosis, and anatomic variation (86,87).
The tip of acute dialysis catheters placed in the internal jugular vein should
lie in the superior vena cava. Catheters placed in the femoral vein should be long
enough to insure that their tip lies in the inferior vena cava to prevent excessive
recirculation. Right internal jugular placement provides superior blood ow and
is less likely to malfunction than left internal jugular or femoral placement. High
complications rates temper subclavian vein placement of these catheters (83).
Complications of acute dialysis catheters are both immediate insertion
related and delayed (83) and similar to those associated with central venous
catheter placement. Because acute dialysis catheters are relatively stiff, they
are associated with a higher incidence of vessel and right atrial perforation
than cuffed, tunneled catheters. Late complications include central vein steno-
sis/thrombosis, vessel perforation, cardiac tamponade, hemomediastinum, and
infection. Stenosis/thrombosis has been reported in up to 28% of subclavian
dialysis catheters with infection, increasing the risk in such insertions (88).
Because of the high rate of complication, subclavian insertions should be
avoided and are contraindicated in patients expected to receive permanent
access on the same side. In contrast, internal jugular insertions are associated
with only a 2% to 3% thrombosis/stenosis rate (89 92).
Because cuffed, tunneled catheters are softer and less rigid than acute dialy-
sis catheters, vessel and atrial perforation is not a major complication. The
optional location of the tip of cuffed, tunneled catheters is the mid right atrium
(84), as opposed to the superior vena cava. The right internal jugular site is the
preferred site for catheter placement because it has a lower rate of complication
during and after insertion than other locations (93).
Radiology in the ICU 351
The value of the post-insertion chest radiograph has been questioned when
a cuffed tunneled catheter is inserted under uoroscopy (94). Only seven pro-
cedural complications were identied in 937 consecutive central venous access
procedures, and all were identied at uoroscopy. No procedural complication
was detected on post-procedure chest radiography that was not detected on
uoroscopy, and only one malpositioned catheter was detected by chest radio-
graphy. The most common acute complications are inadvertent carotid artery
puncture, air embolism, hemothorax, and pneumothorax (95 97). Late malfunc-
tions of cuffed, tunneled catheters are primarily associated with thrombosis,
which may be either extrinsic or intrinsic (96).
Other Devices
Intra-Aortic Balloon Pump
The intra-aortic counterpulsation balloon pump is used in patients in cardiogenic
shock or with severe ventricular dysfunction and patients who undergo high-risk
cardiac surgical procedures (53). The pump is composed of a catheter surrounded
by a long balloon. The catheter is placed in the descending thoracic aorta with its
tip immediately distal to the origin of the left subclavian artery (23). The balloon
is inated during systole, increasing coronary artery perfusion, and forcibly
deated during diastole, facilitating aortic blood ow and decreasing ventricular
afterload (98). On the portable chest X-ray, the radio-opaque tip is identied
within the aortic knob (Fig. 12). The inated balloon is a long radiolucent
tubular structure extending the length of the descending thoracic aorta (11), and
the deated balloon cannot be visualized on the radiograph.
The location of the tip should be noted on daily radiographs to avoid
complications. Advancement of the catheter too far may result in cerebral embo-
lization or occlusion or dissection of the left subclavian or vertebral artery. If the
balloon tip is positioned more distally in the descending thoracic aorta, counter-
pulsation is less effective, and potential occlusion of the major branches of the
abdominal aorta, including the renal arteries, may occur (99). Dissection of the
aorta at the time of insertion is a rarely reported complication (100). This is
accompanied by clinical symptoms and may be suggested by loss of denition
of the descending aorta or widening of the para-aortic line on the portable radio-
graph. The diagnosis may be conrmed by either contrast enhanced computer-
ized tomography (CT) or magnetic resonance angiography (MRA).
have been divided into three categories: (1) physiologic effects on the heart and
pulmonary vasculature, (2) direct lung injury, and (3) air leak phenomena (105).
Positive pressure ventilation (PPV) may affect cardiopulmonary function,
particularly when used with positive end-expiratory pressure (PEEP). Cardiac
output may be diminished by PPV with PEEP in patients with suboptimal
volume status, resulting in further reduction in cardiac blood volume. The dimin-
ished cardiac output and reduction in cardiac blood volume may be manifest
radiographically as a decrease in heart size.
However, the decrease in heart size does not necessarily indicate an
improvement in the cardiovascular status of the patient. Conversely, the increase
in heart size, which may occur when PPV and PEEP are discontinued or
decreased to lower levels, does not necessarily indicate cardiac disease or uid
overload. Therefore, changes in heart size in patients receiving mechanical ven-
tilation should be carefully interpreted along with information on current venti-
lator settings and changes in settings from the time of the previous radiograph.
The effect of PPV and PEEP on pulmonary vasculature is complex, and
large radiographically visible pulmonary vessels can either increase or decrease
in size depending on lung volume, ination pressures, and pre-existing pulmonary
vascular tone (105). Because the air to tissue ratio increases when the patient
is placed on PPV or PEEP, the post-PPV or -PEEP chest radiograph may
show signicant apparent improvement despite the presence of severe residual
underlying disease (106 110).
Additional lung injury, manifest as non-cardiogenic pulmonary edema or
diffuse alveolar damage, may occur with PPV and PEEP. Experimentally, the
threshold peak inspiratory pressures (PIP) that cause lung injury appear to be
approximately 25 to 30 cmH2O (111), that is, the same PIP at which perivascular
interstitial emphysema (PIE) occurs (110). Further increases in PIP produce more
severe lung injury (112). On the basis of the experimental studies, it has been rec-
ommended that maximum transalveolar pressure should not exceed 30 to
35 cmH2O, which usually corresponds to 45 cmH2O end-inspiratory pressure
(110).
The other major complication of mechanical ventilation is air-leak
phenomena, PIE and its sequellae, commonly recognized manifestations of baro-
trauma and oxygen toxicity. The incidence of barotrauma in the general
population of ventilator dependent patients varies from 4% to 15% (113,114)
and rises to as high as 60% when underlying lung disease, such as pneumonia
or acute respiratory distress syndrom (ARDS), is present (115).
Although PIE may be seen in any age, the majority of severe cases have
been reported in infants, children, and adults under 40, suggesting that peri-
vascular and connective tissues in older adults are less easily dissected than in
younger patients (105). Ventilator-related risk factors for development of air
leak include large tidal volumes, PIP greater than 40 cmH2O, plateau pressures
greater than 35 cmH2O, high levels and long-term duration of PEEP, and
higher minute ventilation. A PIP of 40 cmH2O or greater is usually required to
356 Barack et al.
vascular resistance may be due to causes other than barotrauma (124). The early
stage of PIE prior to formation of large air cysts is easier to recognize in the pre-
sence of signicant concomitant alveolar consolidation. A mottled increase in
radiolucency of the lung anteriorly and medially around the heart and surface
of the diaphragm is the rst chest radiograph abnormality detected in adults
(105) (Fig. 15). This corresponds to the hyperinated pulmonary lobules and
interstitial and subpleural linear and cystic air collections demonstrated on CT
scans (125,126). These lucencies are extremely difcult to recognize in patients
with underlying emphysema or subcutaneous emphysema in the absence of
consolidated lung. However, without CT correlation, the plain lm diagnosis
of PIE is difcult at this stage because of similar ndings in microabscesses,
hyperinated acini, honeycombing, and emphysema.
Streaky, non-branching, xed caliber radiolucencies radiating from the hila
to the lung periphery corresponding to air in the perivascular connective tissue
also may be identied as an early radiographic sign of PIE and are more
suggestive of its diagnosis. The non-branching, xed caliber appearance of
these lucencies distinguishes them from the branching, peripherally tapering
appearance of air bronchograms. Rarely, perivascular halos, which are pathogno-
monic for PIE, may be seen (127). Other, more disorganized streaky
358 Barack et al.
radiolucencies probably correspond to air within the interlobular septa (105). The
subpleural cystic air collections that can be seen on CT scans in patients with
ARDS may originate in the interlobular septa.
The presence of subpleural air cysts is the most frequent plain lm radio-
graphic nding in PIE and often precedes a pneumothorax (124,128). Subpleural
air cysts may become large and may be effectively treated by percutaneous cath-
eter drainage (129). They also may become secondarily infected, demonstrating
progressive thickening of the cyst wall or an air uid level, with subsequent
rupture producing either a pneumothorax or tension pneumothorax (105,128).
Pneumomediastinum has been found in 37% of patients with PIE (117) and
has been reported to precede the appearance of pneumothorax in 50% of patients
with ARDS (115). On a supine radiograph, pneumomediastinum may be difcult
to distinguish from pneumopericardium, anteromedial pneumothorax, and an
optical illusion, the Mach effect, which is an apparent line of contrasting density
bordering a soft tissue shadow (130). All of these entities may produce a sharp
outline of the heart on the supine radiograph. The radiographic differentiation of
these four entities is mandatory given the potentially life threatening complication
of evolution to a tension pneumothorax in mechanically ventilated patients.
Radiology in the ICU 359
Figure 16 Pneumomediastinumthe ring around the artery sign. (A) Close-up postero-
anterior view of the left hilum demonstrates the left main pulmonary artery outlined by air
(arrows), the ring around the artery sign, and a small amount of air inferiorly within the
mediastinum (arrowhead). (B) Lateral radiograph demonstrates air surrounding the left
main pulmonary artery (arrows).
Radiology in the ICU 361
by air, and the medial surface of the lower lobe may be displaced laterally
(Fig. 19). Posteromedial pneumothorax is associated with lower lobe volume
loss and parenchymal disease, both of which are common in the ICU patient.
As air preferentially surrounds the surface of abnormal lung in lobar collapse,
the air in a posteromedial pneumothorax continues to surround the abnormal
collapsed lower lobe and fails to rise to less dependent pleural surfaces, even
when it is not loculated.
When the pressure in the pleural space exceeds atmospheric pressure, a
pneumothorax is considered to be under tension. Both a simple and a tension
pneumothorax may result in lung collapse and mediastinal shift (149).
However, both lung collapse and mediastinal shift may be absent in a tension
pneumothorax because of pleural adhesions and because of the stiffness of the
lungs in ARDS (150). Hence, a residual small loculated collection of air can
cause a tension pneumothorax even when the ipsilateral hemithorax is drained
by a thoracostomy tube.
Radiology in the ICU 363
develop during the course of the ICU stay. The etiology of any abnormality
should be interpreted in conjunction with the clinical history and the prior radio-
graphs (11).
A. Atelectasis
Atelectasis by denition means incomplete expansion, and it is used to describe
any condition where lung volume loss occurs. Atelectasis is a common radio-
graphic nding in ICU patients for a multitude of reasons including, but not
limited to, an impaired cough reex, supine position, and hypoventilation.
Most frequently, atelectasis is observed in the left lower lobe where the heart
causes mass effect on the left lower lobe bronchus in the supine position (155).
The radiographic appearance of atelectasis varies from undetectable in the
mildest cases to complete opacication of the affected hemithorax in cases of
total lung collapse. The latter represents the most extreme form of atelectasis.
Atelectasis can be described as subsegmental, segmental, or lobar, each of
which has characteristic radiographic ndings.
Subsegmental atelectasis manifests as linear, plate-like opacities, predomi-
nantly at the lung bases or lower lung zones. Segmental atelectasis has a
366 Barack et al.
B. Pneumonia
Types
Pneumonia is a frequent cause of ICU admission, as well as a complication of
hospitalization (nosocomial). The radiological ndings in pneumonia are
varied. The unilateral and focal airspace opacication of pneumonia may be
difcult to distinguish from atelectasis, pulmonary edema, or hemorrhage.
These entities comprise the differential of airspace disease. In one study, the
accuracy of bedside radiography in diagnosing pneumonia in ventilated patients
was only 50% (160). Generally, pulmonary opacities due to infection appear later
and resolve slower than do those seen with aspiration or atelectasis (161).
Radiology in the ICU 367
Aspiration
Predisposing factors for aspiration pneumonitis include vomiting, gastroesopha-
geal reux, altered mental status, intubation, and tracheostomy. Radiographic nd-
ings develop within 24 hours and commonly occur in the dependent portions of the
lung with a basilar distribution. However, in the supine patient, airspace opacities
may be seen in the upper lung zones due to aspiration into the posterior segments
of the upper lobes or the superior segments of the lower lobes. The radiographic
ndings of simple toxic aspiration begin to clear within 48 hours (169). Persist-
ence of opacities beyond this period suggests a complicating infection.
Complications of Pneumonia
Two of the more serious complications of pneumonia include lung abscess and
empyema. These entities may be difcult to diagnose and to differentiate by
plain chest radiography.
Lung Abscess
Lung abscess most frequently develops in pneumonias associated with infectious
aspiration, bronchial obstruction, and an immunocompromised state. A lung
abscess may also be a complication of septic emboli. Lung abscesses typically
develop one to two weeks after pneumonia. Approximately 20% are not apparent
on plain chest radiographs because a cavity is not evident (170).
The plain lm radiographic ndings of lung abscess are single or multiple
cavities, isolated or within areas of consolidation. In one study, 88% of cavities
demonstrated a smooth internal margin, and 72% of cavities contained air uid
levels (170). The wall thickness varies from 4 to 15 mm (171). Abscesses demon-
strating irregular margins and thick walls may simulate cavitary tumors radiogra-
phically and cannot be differentiated by MDCT. Lung abscesses may result in
formation of pneumatoceles, which can either resolve or persist indenitely
368 Barack et al.
(170). A rare and more serious complication of lung abscess is the development
of a bronchopleural stula (Fig. 21).
Empyema
Empyema is a collection of pus in the pleural space. Empyemas often progress
from parapneumonic effusions or may occur as a consequence of hematogenous
spread of infection, penetrating trauma, or any procedure disrupting the pleural
space.
Neither the plain lm nor the CT ndings of empyema are specic.
Thickening and enhancement of the visceral and parietal pleura, the split
pleura sign, may be seen in empyema and parapneumonic and malignant
pleural effusion (170). An air uid level may be present in an infection resulting
from a gas-forming organism. CT may be necessary to demonstrate the location
of the air uid level. In an empyema or a bronchopleural stula, the air uid
level will be in the pleural space. In contrast, the air uid level will be in the
lung parenchyma in a lung abscess. CT may also be useful in delineating
pleural uid loculations prior to drainage. Rare and more serious complications
of empyema include the spontaneous drainage through the chest wall (empyema
necessitans) or the development of a bronchopleural stula, which occurs
secondary to erosion into the tracheobronchial tree.
C. Pulmonary Edema
Introduction
Pulmonary edema represents an abnormal increase in extravascular water in the
lung and is commonly classied into two major categories, hydrostatic edema
and increased permeability edema (172). However, it is not possible to group
all patients into these two categories because many patients have pulmonary
edema resulting from both hydrostatic and increased permeability mechanisms.
Therefore, a third category, mixed pulmonary edema, has been created (173).
The three major categories of pulmonary edema and their subcategories
with disease groups are shown in Table 4.
Hydrostatic pulmonary edema can be further divided into cardiogenic
edema, decreased capillary osmotic pressure edema, and edema associated
with disease of the pulmonary veins. The subcategory of cardiogenic edema
includes patients whose edema results from left ventricular failure and mitral
valve disease. The subcategory of decreased capillary osmotic pressure edema
includes patients whose edema results from renal failure, uid overload, or
hypoproteinemia (cirrhosis and nephrotic syndrome). A decrease in serum
osmotic pressure can contribute to hydrostatic edema. Edema secondary to
disease of the pulmonary veins occurs with pulmonary veno-occlusive disease
and brosing mediastinitis. This form of edema is primarily hydrostatic but is
incompletely understood. Although this subclassication is useful for further
grouping patients according to the pathophysiology of the edema, these three
subcategories of hydrostatic pulmonary edema cannot be differentiated
radiographically.
Similarly, the category of increased permeability edema may be subdivided
into two subcategories, increased permeability edema with diffuse alveolar
damage (DAD) and increased permeability edema without DAD. The subcate-
gory of increased permeability edema with DAD includes ARDS. The subcate-
gory of increased permeability edema without DAD includes the edema
associated with cytokine administration, high altitude, and heroin overdose.
These two subcategories of increased permeability edema also cannot be
radiographically differentiated from each other.
Acute lung injury (ALI) refers to the spectrum of acute diffuse pulmonary
injury characterized by acute onset, gas exchange abnormalities (PaO2/
FIO2 , 300 mmHg regardless of PEEP level), and diffuse bilateral opacities in
the absence of an elevated pulmonary wedge pressure (172). For practical
purposes and discussion, increased permeability edema and the edema associated
with ALI are synonymous with non-cardiogenic pulmonary edema.
Left ventricular Disease of Hypervolemia Acute respiratory Cytokines Acute and chronic
failure pulmonary (uid overload) distress syndrome pulmonary emboli
veins
Mitral valve disease Renal failure High altitude Near drowning
Asymmetric Hypoproteinemia Heroin overdose Neurogenic
(cirrhosis, nephrotic
syndrome)
COPD Reperfusion
Patient position Lung transplantation
Mitral regurgitation Reexpansion
Post pneumonectomy
or lung reduction
Air embolism
Barack et al.
Abbreviation: COPD, chronic obstructive pulmonary disease.
Radiology in the ICU 371
Non-cardiogenic Cardiogenic
Finding edema edema
D. Pleural Effusion
Initially, free pleural uid accumulates in the most dependent aspect of the
thorax, which is located postero-inferiorly in the supine patient. With increasing
volume, pleural uid tracks into the subpulmonic space and along the convexity
of the lung postero-laterally, resulting in silhouetting of the diaphragm, obliteration
of the lateral costophrenic angle, and formation of a lateral meniscus sign. Fluid
tracks into the posterior superior aspect of the major ssures and the lateral
aspect of the minor ssure as it accumulates. With a large pleural effusion, an
apical cap develops. In one study, the reported sensitivity and specicity for detect-
ing pleural effusions on supine radiographs were 67% and 70%, respectively (199).
In a prospective study, the minimum amount of pleural uid necessary
for radiographic visualization on the supine radiograph was 175 mL (200).
Radiographic signs associated with the presence of small pleural effusions
(175 525 mL) included increased homogeneous density in the lower lung
zone in 91% of cases and silhouetting of the diaphragm in 45% of cases. Moder-
ate pleural effusions (.525 mL) demonstrated increased homogenous density in
the lower lung zones in all cases, silhouetting of the diaphragm in 71% of cases,
and blunting of the lateral costophrenic angle in 25% of cases. Large pleural
effusions showed increased homogenous density in the lower lung zones in all
cases, silhouetting of the diaphragm in 68% of cases, blunting of the lateral cost-
ophrenic angle in 41% of cases, and apical capping in 54% of cases. Thickening
of the minor ssure was present in 48% of cases (200).
Other radiographic ndings of free pleural uid on supine radiographs
include apparent elevation of the hemidiaphragm, which is caused by subpulmo-
nic uid accumulation, and decreased visualization of the pulmonary vessels
below the apparent dome of the hemidiaphragm because of the added density
of subpulmonic uid. However, these ndings are difcult to appreciate due to
the frequent presence of bilateral effusions silhouetting both hemidiaphragms
and simulating low lung volumes and insufcient penetration of the upper
abdominal structures on the supine radiograph (200). A subpulmonic pleural
effusion is characterized by apparent lateral displacement of the dome of the
hemidiaphragm and a more acute lateral costophrenic angle.
On the supine radiograph, pleural effusion may be difcult to recognize
because of overlying breast tissue, an enlarged heart, or a prominent epicardial
fat pad. Pleural effusion may be simulated by any condition resulting in an
increase or decrease in the density of a hemithorax. Conditions resulting in
increased density, such as chest wall or pleural masses, may simulate an ipsilat-
eral pleural effusion. Conditions causing decreased density, such as prior mas-
tectomy, unilateral emphysema, or bullous disease, and anterior pneumothorax
or atrophy of the pectoralis muscle may simulate a contralateral effusion (201).
The lateral decubitus lm is the most sensitive technique for detecting free
pleural uid, with a sensitivity threshold of 5 mL (202). It may be performed to
identify the presence of pleural uid and to distinguish free from loculated pleural
effusions. Bilateral decubitus views are preferred as the contralateral decubitus
view often yields additional important information.
Radiology in the ICU 377
E. Embolic Disease
Pulmonary Embolism
Pulmonary embolism results from occlusion of a pulmonary arterial branch by an
embolus. The most frequent source of emboli is the deep systemic veins. Predis-
posing factors for the development of pulmonary embolism include prolonged
immobilization, surgery, trauma, and hypercoagulable states.
Figure 25 Pseudotumor in the minor and upper portion of the major ssures. (A)
Antero-posterior radiograph demonstrates a sharply circumscribed, homogeneous, hori-
zontally oriented opacity in the right mid lung zone lateral to the right hilum (arrowheads)
and a second obliquely oriented opacity extending medial and superior to the right hilum
(arrows). Subpulmonic uid spills into the lateral costophrenic sulcus and extends upward
along the lateral chest wall. (B) Lateral view of chest demonstrates the horizontally
oriented opacity to lie in the minor ssure (arrowheads) and the obliquely oriented
opacity to lie in the superior aspect of the right major ssure (arrows). Subpulmonic
uid spills into the anterior and posterior costophrenic sulci and extends upward along
the anterior and posterior chest wall.
Radiology in the ICU 379
The chest radiograph is reported to be 33% sensitive and 59% specic for
diagnosing pulmonary embolism (205). In a large prospective study, the most
common radiographic abnormalities associated with pulmonary embolism were
cardiomegaly (27%), pleural effusion (23%), elevated hemidiaphragm
(20%), pulmonary artery enlargement (19%), atelectasis (18%), and parenchymal
opacity (17%). The chest radiograph was normal in 24% of the cases (206).
Radiographic ndings classically associated with pulmonary embolism are
hyperlucency of the lung distal to the occluded arterial segment because of
regional oligemia, the Westermarks sign (207), a prominent central pulmonary
artery, the Fleischner sign, and an abrupt cutoff of the interlobar pulmonary
artery, the knuckle sign (208) (Fig. 26). These ndings are highly specic but
insensitive for the diagnosis of pulmonary embolism (209). Pulmonary embolism
can also cause segmental and lobar consolidation secondary to edema and
hemorrhage, mimicking the radiologic appearance of pneumonia (210).
Pulmonary infarction is uncommon and is present in only 15% of pulmo-
nary embolisms because of the dual pulmonary arterial and bronchial blood
supply of the lung. Pulmonary infarction is more likely to occur in patients
with heart failure or severe hypotension. The characteristic radiographic appear-
ance of infarction is a homogeneous pleural-based opacity with convex medial
contour towards the hilum, the Hamptons hump (211).
At institutions where MDCT is available, CT pulmonary angiography is
the study of choice for suspected pulmonary embolism. Studies have shown
that MDCT pulmonary angiography is highly sensitive and specic for the
diagnosis of pulmonary embolism. Discrepancies with conventional angiogra-
phy are mainly at the subsegmental level, where even angiographers have
poor inter-observer variability (212 214). The newer scanners have thinner
slice acquisition, higher resolution, faster scanning times, and less motion arti-
fact, resulting in superior visualization of the peripheral pulmonary arteries. All
these factors improve detection of pulmonary emboli. In a prospective study
comparing the effect of different slice thickness on study interpretation,
MDCT at 1.25 mm collimation has been found to signicantly improve visu-
alization of the segmental and subsegmental arteries. In this study, 96% of
lobar, 90% of segmental, and 75% of subsegmental arteries were well visual-
ized using 1.25 mm slice thickness. Inter-observer agreement for detection of
pulmonary embolism at the segmental and subsegmental level was also
improved when these parameters were used (215). Animal studies have demon-
strated no difference in the sensitivity of MDCT at 1.25 mm collimation and
pulmonary angiography in the detection of subsegmental emboli. Specicity
was reported to be slightly higher for angiography, 88% versus 81% for
MDCT (216).
Clinical studies have shown that a negative MDCT pulmonary angiogram
has high negative predictive value for subsequent development of a pulmonary
embolism. The risk of pulmonary embolism in patients with a negative MDCT
pulmonary angiography is less than 1% at six months (217) and nine months
380 Barack et al.
after the study (218), and less than 2% at 12-month follow-up (219). Addition-
ally, a large prospective study has shown that MDCT reveals additional poten-
tially signicant ndings in up to 76% of patients, with 47% of these ndings
not suspected on chest radiography (218).
Radiology in the ICU 381
Septic Emboli
Septic emboli result from the hematogenous dissemination of infection to the lung
parenchyma. Septic emboli occur more often in people under the age of 40 (229).
The most common sources of emboli are the heart in association with tricuspid
endocarditis, or a ventricular septal defect, or the peripheral veins secondary to
septic thrombophlebitis or an indwelling vascular catheter (230,231). Common
predisposing factors associated with septic emboli include drug addiction, alcohol-
ism, infections in immunocompromised patients, congenital heart disease, and
dermal infections (232). The most common causative organisms are Staphy-
lococcus aureus, Streptococcus viridans, and Streptococcus pneumoniae (229).
On chest radiography, the diagnosis of septic emboli should be considered
when multifocal, peripheral, ill-dened, round or wedge-shaped opacities are
seen in the setting of a febrile patient (233). The opacities have a basilar predilec-
tion reecting greater blood ow. They may be uniform or variable in size,
reecting recurrent showers of emboli. Cavitation is frequent and early. The cav-
ities are usually thin-walled and occasionally contain a central loose body, the
target sign (234), which may simulate a fungus ball (234,235). The central
loose body represents a piece of necrotic lung. Air uid levels are conspicuously
absent in many of the cavities. Invasive aspergillosis is the major radiologic
differential diagnosis in the immunocompromised patient (233).
382 Barack et al.
F. Pulmonary Hemorrhage
Pulmonary hemorrhage can result from trauma, bleeding diatheses, infections,
pulmonary embolism, penicillamine therapy, and autoimmune diseases.
The most common autoimmune disease causing pulmonary hemorrhage is Good-
pastures syndrome (238).
Radiographically, pulmonary hemorrhage manifests as coalescent airspace
opacication. Unless there is recurrent bleeding, the airspace opacication clears
rapidly within two to three days. The chest radiograph ndings are nonspecic
and indistinguishable from pulmonary edema and pneumonia, and denitive
diagnosis is based on clinical, laboratory, and bronchoscopic ndings, if
necessary.
H. Traumatic Abnormalities
Aortic Injury
Rupture of the thoracic aorta is a common cause of death following blunt chest
trauma. Death occurs at the accident site in more than 80% of aortic ruptures.
Patients (60 70%) who reach the hospital survive if prompt treatment is insti-
tuted. Ninety percent of aortic injuries occur at the level of the ligamentum arter-
iosum. Less common sites of involvement are the aortic root and the
diaphragmatic hiatus (249).
384 Barack et al.
Esophageal Rupture
Esophageal rupture can be secondary to Boerhaaves syndrome, iatrogenic
causes, or blunt trauma. Boerhaaves syndrome is spontaneous perforation
of the thoracic esophagus due to sudden increase in intraluminal pressure.
Other
Any sequellae of trauma can result in admission to the ICU if severe enough to
require mechanical ventilation and close hemodynamic monitoring. Such
conditions include severe mediastinal and parenchymal hematoma, cardiac
contusion, osseous injuries (Fig. 29), and tracheobronchial rupture. The latter
should be suspected in a trauma patient following chest tube placement for a
pneumothorax with a persistent air leak.
V. Computerized Tomography
A. Indications
B. Technique
The VA Greater Los Angeles Health Care System Hospital at West Los Angeles
performs all thoracic imaging on a 16-row multidetector Toshiba scanner.
Routinely, axial images are acquired with 1-mm collimation. The 1-mm axial
386 Barack et al.
images with 3-mm axial reformations and 1.5-mm coronal and sagittal reforma-
tions are then sent to PACS for interpretation.
Nonionic contrast is administered when indicated and when renal function
permits (creatinine within 0 45 days of exam 1.5). When serum creatinine is
between 1.5 and 2.0 and in patients susceptible to developing nephrotoxicity,
contrast is administered depending upon the clinical situation and only after
close clinical consultation. Such patients are adequately hydrated and given
oral N-acetylcysteine 600 mg twice daily the day before and the day of the
exam. Isoosmolar nonionic contrast is used exclusively in this subset of patients,
and the volume of contrast administered is carefully tailored to the clinical
situation. The creatinine is closely monitored after the examination. Intravenous
contrast is not administered to patients whose serum creatinine is 2.0 unless
the patient is on dialysis. In these patients, contrast is administered on the day
of dialysis prior to the procedure. If the patient is on glucophage, the drug is
withheld on the day of the examination and for two days thereafter.
Radiation exposure is always a concern in these patients. Assuming a 2-mm
helical mode continuous axial acquisition is obtained on an average patient over a
30-cm length (lung apex to lung base) with a 16-row multidetector CT using the
following parameters (120 kVp, 360 mA, 0.5 sec gantry rotation, 1.25 effective
Radiology in the ICU 387
pitch, 16:1 acquisition, and 40% automatic dose reduction), the effective dose
equivalent given to the patient roughly corresponds to an effective dose of 120
PA chest radiographs obtained in a 70-kg patient using a technique of 110 kVp
and 3 mAs. It should be noted that variation among CT scanners, scanning
parameters, and actual patient size may result in considerable variation in this
number (Christopher H. Cagnon, Ph.D., Radiology Physicist, Department of
Radiological Science, David Geffen School of Medicine at U.C.L.A.).
Venous Evaluation
Before catheter placement, it is important to delineate the anatomy and patency
of the venous structures involved. Venography used to be the gold standard to
determine the patency of a venous segment. As ultrasound and MR techniques
become more rened, they largely replaced the need for venography, which
can cause patient discomfort because of needle puncture and contrast adminis-
tration. MRA (most frequently using contrast-enhanced and time of ight
techniques) can reliably delineate the central venous anatomy. Ultrasound com-
plemented by spectral and color Doppler analysis allows convenient assessment
of the venous system. Although ultrasound cannot provide direct visualization of
388 Barack et al.
the central veins [i.e., central subclavian and innominate veins], patency can be
inferred by evaluation of the spectral waveform for the presence of respiratory
phasicity and cardiac periodicity.
Malpositioned Catheters
Occluded Catheters
Catheter occlusions secondary to brin sheath formation or improper ushing
resulting in intraluminal clots are often treated with instillation of thrombolytic
agents (i.e., TPA or urokinase). Fibrin sheaths and clots at the tips of the catheters
can be treated by disruption with a curled heavy-duty wire, or even with a large
angioplasty balloon. However, resistant pericatheter brin sheaths require a
different approach. A non-cuffed catheter can be easily replaced over a guide-
wire. For cuffed or implanted catheters, exchange over a wire is usually not
feasible. A technique of catheter brin sheath stripping has been described by
Mewissen et al. (255) with an Amplatz snare from the femoral venous approach
for failing hemodialysis catheters.
Fibrin sheath typically acts as a ball valve, allowing infusion but occluding
side holes upon negative pressure application, therefore hindering catheter
aspiration. Its diagnosis is conrmed by venography. Once the brin sheath is
demonstrated, a guidewire is passed through the distal port of the central line
and manipulated through an open Amplatz snare (15, 25, or 35 mm) deployed
from the femoral approach. The loop is then guided around the catheter and
rmly snugged against it. As the snare is withdrawn, the brin sheath is
removed. Thirty-day patency rates are on the order of 30% to 50%. Repeated
stripping is necessary in maintaining long-term patency.
390 Barack et al.
Pleural Collections
The common etiologies for pleural collections that require drainage include com-
plicated parapneumonic effusions and post-traumatic or post-surgical collections.
It is important to identify the collections that would benet from drainage early in
the course, as delays in drainage allow the collection to organize and ultimately
diminish the speed and efcacy of percutaneous drainage.
Three stages have been described in the natural history of parapneumonic
effusions. Stage I (exudative) is characterized by a sterile exudative effusion with
a pH .7.30, glucose .60 mg/dL, and LDH ,500 U/L. In stage II (brinopuru-
lent), bacteria enter the pleural space, polymorphonuclear cells increase, and
glucose decreases to ,40 mg/dL, pH ,7.10, and LDH . 1000 U/L. As uid
protein increases, intrapleural brinolytic activity diminishes, and coagulation
of the uid can occur. Fibrin deposition and broblastic activity are initiated.
Stage III (organizational) is characterized by formation of a thick, brotic
pleural peel (256,257).
Radiologically placed tubes (8 14 French) are relatively small compared
with the traditional, surgically placed tubes (24 38 French). Percutaneous drai-
nage catheters can be inserted under ultrasound and uoroscopy guidance most
of the time. However, if the uid collection is loculated, poorly accessible
(e.g., overlying scapula, brachial plexus, or subclavian vessels) or close to vital
structures, CT guidance may be necessary for more precise localization. In
severely complex collections, multiple drainage catheters may be necessary.
The combination of a glide catheter and wires can be used to disturb septations
Radiology in the ICU 391
D. Conclusion
References
1. Marik PE, Janower ML. Impact of routine chest radiography on ICU management
decisions: an observational study. Am J Crit Care 1997; 6(2):95 98.
2. Strain DS, KinaseWitz GT, Vereen LE, George, RB. Value of routine daily chest
x-rays in the medical intensive care unit. Crit Care Med 1985; 13(7):534 536.
3. Bekemeyer WB, Crapo RO, Calhoon S, Cannon CY, Clayton PD. Efcacy of chest
radiography in a respiratory intensive care unit. A prospective study. Chest 1985;
88(5):691 696.
4. Hall JB, White SR, Karrison T. Efcacy of daily routine chest radiographs in intu-
bated, mechanically ventilated patients. Crit Care Med 1991; 19(5):689 693.
5. Henshke CI, Pasternack GS, Schroeder S, Hart KK, Herman PG. Bedside chest
radiography: diagnostic efcacy. Radiology 1983; 149(1):23 26.
6. Silverstein DS, Levingston DH, Elcavage J, Kovar L, Kelly KM. The utility of
routine daily chest radiography in the surgical intensive care unit. J Trauma
1993; 35(4):643 646.
Radiology in the ICU 393
7. Fong Y, Whalen G, Hariri RJ, Barie PS. Utility of routine chest radiographs in the
surgical intensive care unit. Arch Surg 1995; 130(7):764 768.
8. American College of Radiology ACR Appropriateness Criteria. Expert Panel on
Thoracic Imaging. Daily portable chest X-ray. Radiology 2000; 215(S):621 626.
9. Wandtke J. Bedside chest radiography. Radiology 1994; 190(1):1 10.
10. Milne EN. A physiological approach to reading critical care unit lms. J Thorac
Imaging 1986; 1(3):60 90.
11. Tocino I. Chest imaging in the intensive care unit. Eur J Radiol 1996; 23(1):
46 57.
12. Henschke CI, Yankelevitz DF, Wand A, Davis SD, Shiau M. Chest radiography in
the ICU. Clin Imaging 1997; 21(2):90 103.
13. Herman MW, Patrick J, Tabrisky J. A comparative study of scattered radiation
levels from 80-kVp and 240-kVp X rays in the surgical intensive care unit.
Radiology 1980; 137(2):552 553.
14. Boles JM, Boussert F, Manens JP, Le Cam B, Bellet M, Garre M. Measurement of
irradiation doses secondary to bedside radiographs in a medical intensive care unit.
Intensive Care Med 1987; 13(1):60 63.
15. Brandstetter RD, Garcia JC, Majumder M, Chisolm A. The benet of lateral
radiographs in an intensive care unit. Chest 1994; 105(2):560 562.
16. Schaefer-Prokop C, Uffmann M, Eisenhuber E, Prokop M. Digital radiography of
the chest: detector techniques and performance parameters. J Thorac Imaging
2003; 18(3):124 137.
17. Willis CE, Thompson SK, Sheppard J. Artifacts and misadventures in digital-
radiography. Applied Radiology 2004; 33(1):11 20.
18. Willis CE, Mercier J, Patel M. Modication of conventional quality assurance
procedures to accomodate computer radiography. Proceedings of the 13th Inter-
national Conference on Computer Applications in Radiology, Denver, Colarado,
June 7, 1996: 275 281.
19. Sagel SS, Jost RG, Glazer HS, Molina PL, Anderson DJ, Solomon SL, Schwarberg J.
Digital mobile radiography. J Thorac Imaging 1990; 5(1):3648.
20. Gamsu G, Perez E. Picture archiving and communication systems (PACS). J Thorac
Imaging 2003; 18(3):165 168.
21. van der Jagt EJ, Smits HJ. Cardiac size in the supine chestlm. Eur J Radiol 1992;
14(3):173 177.
22. Milne EN, Pistolesi M, Miniati M, Giuntini C. The vascular pedicle of the heart and
the vena azygos. Part I: The normal subject. Radiology 1984; 152(1):1 8.
23. Aronchick JM, Miller WT Jr. Tubes and lines in the intensive care setting. Semin
Roentgenol 1997; 32(2):102 116.
24. Goodman LR, Conrardy PA, Laing F, Singer MN. Radiographic evaluation of
endotracheal tube position. Am J Roentgenol 1976; 127(3):433 434.
25. Yap SJ, Morris RW, Pybus DA. Alterations in endotracheal tube position during
general anesthesia. Anaesth Intensive Care 1994; 22(5):586 588.
26. Conrardy PA, Goodman LR, Lainge F, Singer MM. Alteration of endotracheal tube
position. Flexion and extension of the neck. Crit Care Med 1976; 4(1):7 12.
27. Valentino J, Myers RK, Baker MD, Woodring JH. Utility of portable chest
radiographs as a predictor of endotracheal tube cuff pressure. Otolaryngol Head
Neck Surg 1999; 120(1):51 56.
394 Barack et al.
28. Khan F, Reddy NC, Khan A. Cuff/trachea ratio as an indicator of tracheal damage
[abstr]. Chest 1976; 70:431.
29. Stauffer JL, Olsen DE, Petty, TL. Complications and consequences of endotracheal
intubation and tracheotomy. A prospective study of 150 critically ill adult patients.
Am J Med 1981; 70(1):65 76.
30. Zwillich CW, Pierson DJ, Creagh CE, Sutton FD, Schatz E, Petty TL. Compli-
cations of assisted ventilation. A prospective study of 354 consecutive episodes.
Am J Med 1974; 57(2):161 170.
31. Lotano R, Gerber D, Aseron C, Santarelli R, Pratter M. Utility of postintubation
chest radiographs in the intensive care unit. Crit Care 2000; 4(1):50 53.
32. Brunel W, Coleman DL, Schwartz DE, Peper E, Cohen NH. Assessment of routine
chest roentgenograms and the physical examination to conrm endotracheal tube
position. Chest 1989; 96(5):1043 1045.
33. Schwartz DE, Lieberman JA, Cohen NH. Women are at greater risk than men for
malpositioning of the endotracheal tube after emergent intubation. Crit Care Med
1994; 22(7):1127 1131.
34. Smith GM, Reed JC, Choplin RH. Radiographic detection of esophageal malposi-
tioning of endotracheal tubes. Am J Roentgenol 1990; 154(1):23 26.
35. Rollins RJ, Tocino I. Early radiographic signs of tracheal rupture. Am J Roentgenol
1987; 148(4):695 698.
36. Wechsler RJ, Steiner RM, Kinori I. Monitoring the monitors: the radiology of
thoracic catheters, wires, and tubes. Semin Roentgenol 1988; 23(1):61 84.
37. Leeds WM, Morley TF, Zappasodi SJ, Giudice JC. Computed tomo-
graphy for diagnosis of tracheoesophageal stula. Crit Care Med 1986; 14(6):
591 592.
38. Goodman LR. Pulmonary support and monitoring apparatus. In: Goodman LR,
Putman CE, eds. Critical Care Imaging. Philadelphia: Saunders, 1992:35 39.
39. Pasternack G, OCain C. Thoracic complications of respiratory intensive care.
In: Herman P, ed. Iatrogenic Thoracic Complication. New York: Springer,
1983:59 77.
40. Heffner JE, Miller KS, Sahn SA. Tracheostomy in the intensive care unit. Part 1:
Indications, technique, management. Chest 1986; 90(2):269 274.
41. Heffner JE, Miller KS, Sahn SA. Tracheostomy in the intensive care unit. Part 2:
Complications. Chest 1986; 90(3):430 436.
42. Wechsler RJ, Steiner RM, Goodman LR, Teplick SK, Mapp E, Laufer I. Iatrogenic
esophageal-pleural stula: subtlety of diagnosis in the absence of mediastinitis.
Radiology 1982; 144(2):239 243.
43. Ghahremani GG, Gould RJ. Nasoenteric feeding tubes. Radiographic detection of
complications. Dig Dis Sci 1986; 31(6):574 585.
44. Maffessanti M, Berlot G, Bortolotto P. Chest roentgenology in the intensive care
unit: an overview. Eur Radiol 1998; 8(1):69 78.
45. Rassias AJ, Ball PA, Corwin HL. A prospective study of tracheopulmonary compli-
cations associated with the placement of narrow-bore enteral feeding tubes.
Crit Care Epub 1998; 2(1):25 28.
46. Miller KS, Tomlinson JR, Sahn SA. Pleuropulmonary complications of enteral tube
feedings. Two reports, review of the literature, and recommendations. Chest 1985;
88(2):230 233.
Radiology in the ICU 395
47. Aronchick JM, Epstein DM, Gefter WB, Miller WT. Pneumothorax as
a complication of placement of a nasoenteric tube. J Am Med Assoc 1984;
252(23):3287 3288.
48. Balogh GJ, Adler SJ, VanderWoude J, Glazer HS, Roper C, Weyman PJ. Pneu-
mothorax as a complication of feeding tube placement. Am J Roentgenol 1983;
141(6):1275 1277.
49. Stark P, Thordarson S, McKinney M. Manifestations of esophageal disease on plain
chest radiographs. Am J Roentgenol 1990; 155(4):729 734.
50. Teplick SK, Brandon JC, Shah SH. Acute gastrointestinal disorders. Goodman LR,
Putnam CE, eds. Critical Care Imaging. 3rd ed. Philadelphia: WB Saunders,
1992:314 318.
51. Wand A, Davis S, Yankelevitz DF. Esophageal perforation caused by Sengstaken
Blakemore tube. Complications Surg 1995; 14:79 80.
52. Zarshenas Z, Sparschu RA. Catheter placement and misplacement. Crit Care Clin
1994; 10(2):417 436.
53. Swensen SJ, Peters SG, LeRoy AJ, Gay PC, Sykes MW, Trastek VF. Radiology in
the intensive-care unit. Mayo Clin Proc 1991; 66(4):396 410.
54. Webb WR, Godwin JD. The obscured outer edge: a sign of improperly placed
pleural drainage tubes. Am J Roentgenol 1980; 134(5):1062 1064.
55. Maurer JR, Friedman PJ, Wing VW. Thoracostomy tube in an interlobar ssure:
radiologic recognition of a potential problem. Am J Roentgenol 1982;
139(6):1155 1161.
56. Miller KS, Sahn SA. Chest tubes. Indications, technique, management and compli-
cations. Chest 1987; 91(2):258 264.
57. Stahly TL, Tench WD. Lung entrapment and infarction by chest tube suction.
Radiology 1977; 122(2):307 309.
58. Wagner DP, Knaus WA, Draper EA. Identication of low-risk monitor admissions
to medical-surgical ICUs. Chest 1987; 92(3):423 428.
59. Henschke CI, Yankelevitz DF, Wand A, Davis SD, Shiau M. Accuracy and efcacy
of chest radiography in the intensive care unit. Radiol Clin North Am 1996;
34(1):21 31.
60. Langston CS. The aberrant central venous catheter and its complications.
Radiology 1971; 100(1):55 59.
61. Conces DJ Jr, Holden RW. Aberrant locations and complications in initial place-
ment of subclavian vein catheters. Arch Surg 1984; 119(3):293 295.
62. McGee WT, Ackerman BL, Rouben LR, Prasad VM, Bandi V, Mallory DL. Accu-
rate placement of central venous catheters: a prospective, randomized, multicenter
trial. Crit Care Med 1993; 21(8):1118 1123.
63. Gilner LI. The ear-gurgling sign. N Engl J Med 1977; 296(22):1301.
64. Langston CS. The aberrant central venous catheter and its complications.
Radiology 1971; 100(1):55 59.
65. Lumb PD. Complications of central venous catheters. Crit Care Med 1993;
21(8):1105 1106.
66. Tocino IM, Watanabe A. Impending catheter perforation of superior vena cava:
radiographic recognition. Am J Roentgenol 1986; 146(3):487 490.
67. Campbell M, Deurachar DC. The left-sided superior vena cava. Br Heart J
1954;16(4):423 439.
396 Barack et al.
68. Cha EM, Khoury GH. Persistent left superior vena cava. Radiologic and clinical
signicance. Radiology 1972; 103(2):375 381.
69. Gibson RN, Hennessy OF, Collier N, Hemingway AP. Major complications of
central venous catheterisation: a report of ve cases and a brief review of the litera-
ture. Clin Radiol 1985; 36(2):205 208.
70. Plaus WJ. Delayed pneumothorax after subclavian vein catheterization. J Parenter
Enteral Nutr 1990; 14(4):414 415.
71. Sivak SL. Late appearance of pneumothorax after subclavian venipuncture. Am J
Med 1986; 80(2):323 324.
72. Silverman SM, Olson KW. Avoiding unintentional arterial cannulation. Can J
Anaesth 1989; 36(6):734.
73. Wechsler RJ, Spirn PW, Conant EF, Steiner RM, Needleman L. Thrombosis and
infection caused by thoracic venous catheters: pathogenesis and imaging ndings.
Am J Roentgenol 1993; 160(3):467 471.
74. Alhimyary A, Fernandez C, Picard M, Tierno K, Pignatone N, Chan HS, Malt R,
Souba W. Safety and efcacy of total parenteral nutrition delivered via a peripher-
ally inserted central venous catheter. Nutr Clin Pract 1996; 11(5):199 203.
75. Cowl CT, Weinstock JV, Al-Jurf A, Ephgrave K, Murray JA, Dillon K. Compli-
cations and cost associated with parenteral nutrition delivered to hospitalized
patients through either subclavian or peripherally-inserted central catheters. Clin
Nutr 2000; 19(4):237 243.
76. Hadaway LC. An overview of vascular access devices inserted via the antecubital
area. J Intraven Nurs 1990; 13(5):297 306. Erratum in: J Intraven Nurs 1991;
14(2):135.
77. James L, Bledsoe L, Hadaway LC. A retrospective look at tip location and compli-
cations of peripherally inserted central catheter lines. J Intraven Nurs 1993;
16(2):104 109.
78. Cardella JF, Cardella K, Bacci N, Fox PS, Post JH. Cumulative experience with
1,273 peripherally inserted central catheters at a single institution. J Vasc Interv
Radiol 1996; 7(1):5 13.
79. Barber JM, Booth DM, King JA, Chakraverty S. A nurse led peripherally inserted
central catheter line insertion service is effective with radiological support.
Clin Radiol 2002; 57(5):352 354.
80. Wiedemann HP, Matthay MA, Matthay RA. Cardiovascular pulmonary monitor-
ing in the intensive care unit (Part 1). Chest 1984; 85(4):537 549.
81. Wiedemann HP, Matthay MA, Matthay RA. Cardiovascular pulmonary monitor-
ing in the intensive care unit (Part 2). Chest 1984; 85(5):656 668.
82. Sise MJ, Hollingsworth P, Brimm JE, Peters RM, Virgilio RW, Shackford SR.
Complications of the ow-directed pulmonary artery catheter: A prospective
analysis in 219 patients. Crit Care Med 1981; 9(4):315 318.
83. Oliver MJ. Acute dialysis catheters. Semin Dial 2001; 14(6):432 435.
84. Work J. Hemodialysis catheters and ports. Semin Nephrol 2002; 22(3):211 220.
85. Randolph AG, Cook DJ, Gonzales CA, Pribble CG. Ultrasound guidance for place-
ment of central venous catheters: a meta-analysis of the literature. Crit Care Med
1996; 24(12):2053 2058.
86. Forauer AR, Glockner JF. Importance of US ndings in access planning during
jugular vein hemodialysis catheter placements. J Vasc Interv Radiol 2000; 11(2
Pt 1):233 238.
Radiology in the ICU 397
87. Lin BS, Kong CW, Tarng DC, Huang TP, Tang GJ. Anatomical variation of the
internal jugular vein and its impact on temporary hemodialysis vascular access:
an ultrasonographic survey in uraemic patients. Nephrol Dial Transplant 1998;
13(1):134 138.
88. Hernandez D, Diaz F, Rufno M, Lorenzo V, Perez T, Rodriguez A, De Bonis E,
Losada M, Gonzalez-Posada JM, Torres A. Subclavian vascular access stenosis in
dialysis patients: natural history and risk factors. J Am Soc Nephrol 1998;
9(8):1507 1510.
89. Agraharkar M, Isaacson S, Mendelssohn D, Muralidharan J, Mustata S, Zevallos G,
Besley M, Uldall R. Percutaneously inserted silastic jugular hemodialysis catheters
seldom cause jugular vein thrombosis. ASAIO J 1995; 41(2):169 172.
90. Trerotola SO, Kuhn-Fulton J, Johnson MS, Shah H, Ambrosius WT, Kneebone PH.
Tunneled infusion catheters: increased incidence of symptomatic venous thrombo-
sis after subclavian versus internal jugular venous access. Radiology 2000;
217(1):89 93.
91. Vanherweghem JL, Cabolet P, Dhaene M, Goldman M, Stolear JC, Sabot JP,
Waterlot Y, Marchal M. Complications related to subclavian catheters for hemodia-
lysis. Report and review. Am J Nephrol 1986; 6(5):339 345.
92. Silberman H, Berne TV, Escandon R. Prospective evaluation of a double-lumen
subclavian dialysis catheter for acute vascular access. Am Surg 1992; 58(8):
443 445.
93. Schwab S, Besarab A, Beathard G, et al. NKF-DOQI clinical practice guidelines for
vascular access. Am J Kidney Dis 1997; 30(suppl):S150 S190.
94. Caridi JG, West JH, Stavropoulos SW, Hawkins IF Jr. Internal jugular and upper
extremity venous access in interventional radiology: is a postprocedural chest
radiograph necessary? Am J Roentgenol 2000; 174(2):363 366.
95. Lung GB, Trerotola SO, Scheel PF Jr, Savader SJ, Mitchell SE, Venbrux AC,
Osterman FA Jr. Outcome of tunneled hemodialysis catheters placed by radiol-
ogists. Radiology 1996; 198(2):467 472.
96. Beathard GA. Catheter thrombosis. Semin Dial 2001; 14(6):441 445.
97. Skolnick ML. The role of sonography in the placement and management of
jugular and subclavian central venous catheters. Am J Roentgenol 1994; 163(2):
291 295.
98. Dunkman WB, Leinbach RC, Buckley MJ, Mundth ED, Kantrowitz AR, Austen WG,
Sanders CA. Clinical and hemodynamic results of intraaortic balloon pumping and
surgery for cardiogenic shock. Circulation 1972; 46(3): 465477.
99. Hyson EA, Ravin CE, Kelley MJ, Curtis AM. Intraaortic counterpulsation balloon:
radiographic considerations. Am J Roentgenol 1977; 128(6):915 918.
100. Landay MJ, Mootz AR, Estrera AS. Apparatus seen on chest radiographs after
cardiac surgery in adults. Radiology 1990; 174(2):477 482.
101. Tegtmeyer CJ. Roentgenographic assessment of causes of cardiac pacemaker
failure and complications. CRC Crit Rev Diagn Imaging 1977; 9(1):1 50.
102. Hinke DH, Zandt-Stastny DA, Goodman LR, Quebbeman EJ, Krzywda EA, Andris
DA. Pinch-off syndrome: a complication of implantable subclavian venous access
devices. Radiology 1990; 177(2):353 356.
103. Daly BD, Cascade PN, Hummel JD, Kalbeisch S, Strickberger SA, Langberg J,
Morady F. Transvenous and subcutaneous implantable cardioverter debrillators:
radiographic assessment. Radiology 1994; 191(1):273 278.
398 Barack et al.
124. Rohlng BM, Webb WR, Schlobohm RM. Ventilator-related extra-alveolar air in
adults. Radiology 1976; 121(1):25 31.
125. Tagliabue M, Merlini L. Computed tomography in the diagnosis of pulmonary bar-
otrauma associated with adult respiratory distress syndrome. Radiol Med (Torino)
1994; 87(1 2):45 52.
126. Tocino I. Abnormal air and pleural uid collections. Goodman LR, Putman CE,
eds. Critical Care Imaging. Philadelphia: WB Saunders, 1992:137 160.
127. Ovenfor C. Pulmonary interstitial emphysema. Acta Radiol Supple (Stockh) 1964;
224:1.
128. Albelda SM, Gefter WB, Kelley MA, Epstein DM, Miller WT. Ventilator-induced
subpleural air cysts: Clinical, radiographic, and pathologic signicance. Am Rev
Respir Dis 1983; 127(3):360 365.
129. Chon KS, van Sonnenberg E, DAgostino HB, OLaoide RM, Colt HG, Hart E.
CT-guided catheter drainage of loculated thoracic air collections in mechanically
ventilated patients with acute respiratory distress syndrome. Am J Roentgenol
1999; 173(5):1345 1350.
130. Lane EJ, Proto AV, Phillips TW. Mach bands and density perception. Radiology
1976; 121(1):9 17.
131. Grosfeld JL, Kilman JW, Frye TR. Spontaneous pneumopericardium in the
newborn infant. J Pediatr 1970; 76(4):614 616.
132. Felson B. The mediastinum. Semin Roentgenol 1969; 4:40.
133. Levin B. The continuous diaphragm sign. A newly recognized sign of pneumome-
diastinum. Clin Radiol 1973; 24:337.
134. Crylak D, Milne EN, Imray TJ. Pneumomediastinum: a diagnostic problem.
Crit Rev Diagn Imaging 1984; 23(1):75 117.
135. Hammond DI. The ring-around-the-artery sign in pneumomediastinum.
Can Assoc Radiol J 1984; 35(1):88 89.
136. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R, Tagliabue M.
Lung structure and function in different stages of severe adult respiratory distress
syndrome. J Am Med Assoc 1994; 271(22):1772 1779.
137. de Latorre FJ, Tomasa A, Klamborg J, Leon C, Soler M, Rius J. Incidence of pneu-
mothorax and pneumomediastinum in patients with aspiration pneumonia requiring
ventilatory support. Chest 1977; 72(2):141 144.
138. Steier M, Ching N, Roberts EB, Nealon TF Jr. Pneumothorax complicating
continuous ventilator support. J Thorac Cardiovasc Surg 1974; 67(1):17 23.
139. Tocino IM. Pneumothorax in the supine patient: Radiographic anatomy. Radio-
graphics 1985; 5:557 586.
140. Tocino IM, Miller MH, Fairfax WR. Distribution of pneumothorax in the supine
and semirecumbent critically ill adult. Am J Roentgenol 1985; 144(5):901 905.
141. Kollef MH. Risk factors for the misdiagnosis of pneumothorax in the intensive care
unit. Crit Care Med 1991; 19(7):906 910.
142. Lams PM, Jolles H. The effect of lobar collapse on the distribution of free intra-
pleural air. Radiology 1982; 142(2):309 312.
143. Moskowitz PS, Griscom NT. The medial pneumothorax. Radiology 1976;
120(1):143 147.
144. Markowitz RI. The anterior junction line: A radiographic sign of bilateral pneu-
mothorax in neonates. Radiology 1988; 167(3):717 719.
400 Barack et al.
202. Rudikoff JC. Early detection of pleural uid. Chest 1980; 77:109 111.
203. Weiss W, Boucot KR, Gefter WI. Localized interlobar effusion in congestive heart
failure. Ann Intern Med 1953; 38:1177 1186.
204. Feder BH, Wilk SP. Localized interlobar effusion in heart failure: phantom lung
tumor. Dis Chest 1956; 30:289 297.
205. Greenspan RH, Ravin CE, Polanski SM, McLoud TC. Accuracy of the chest radio-
graph in diagnosis of pulmonary embolism. Invest Radiol 1982; 17(6):539 543.
206. Elliott CG, Goldhaber SZ, Visani L, DeRosa M. Chest radiographs in acute
pulmonary embolism. Results from the International Cooperative Pulmonary
Embolism Registry. Chest 2000; 118(1):33 38.
207. Westermark N. On roentgen diagnosis of lung embolism. Acta Radiol 1938;
19:357 372.
208. Palla A, Donnamaria V, Petruzzelli S, Rossi G, Riccetti G, Giuntini C. Enlargement
of the right descending pulmonary artery in pulmonary embolism. Am J Roentgenol
1983; 141(3):513 517.
209. Worsley DF, Alavi A, Arondick JM, Chen JT, Greenspan RH, Ravin CE. Chest
radiographic ndings in patients with acute pulmonary embolism: observations
from the PIOPED Study. Radiology 1993; 189(1):133 136.
210. Jacoby CG, Mindell HJ. Lobar consolidation in pulmonary embolism. Radiology
1976; 118:287 290.
211. Hampton AO, Castleman B. Correlation of postmortem chest teleroentgenograms
with autopsy ndings, with special reference to pulmonary embolism and infarc-
tion. Am J Roentgenol 1940; 43:305 326.
212. Qanadli SD, Hajjam ME, Mesurolle I, Barre O, Bruckert F, Joseph T, Mignon F,
Vieillard-Baron A, Dubourg O, Lacombe P. Pulmonary embolism detection:
prospective evaluation of dual-section helical CT versus selective pulmonary
angiography in 157 patients. Radiology 2000; 217:447 455.
213. Goodman LR, Curtin JJ, Mewissen MW, Foley WD, Lipchik RJ, Crain MR, Sagar
KB, Collier BD. Detection of pulmonary embolism in patients with unresolved
clinical and scintigraphic diagnosis: helical CT versus angiography. Am J
Roentgenol 1995; 164:1369 1374.
214. van Rossum AB, Pattynama PM, Ton ER, Treurniet FE, Arndt JW, van Eck B, Kieft
GJ. Pulmonary embolism: validation of spiral CT angiography in 149 patients.
Radiology 1996; 201(2):467 470.
215. Patel S, Kazerooni EA, Cascade PN. Pulmonary embolism: optimization of small pul-
monary artery visualization at multi-detector row CT. Radiology 2003; 227:455460.
216. Baile EM, Gregory GK, Muller NL, DYachkova Y, Coche EE, Pare PD, Mayo JR.
Spiral computed tomography is comparable to angiography for the diagnosis of
pulmonary embolism. Am J Respir Crit Care Med 2000; 161:1010 1015.
217. Garg K, Sieler H, Welsh CH, Johnston RJ, Russ PD. Clinical validity of helical CT
being interpreted as negative for pulmonary embolism: implications for patient
treatment. Am J Roentgenol 1999; 172(6):1627 1631.
218. Kavanagh EC, OHare A, Hargaden G, Murray JG. Risk of pulmonary embolism
after negative MDCT pulmonary angiography ndings. Am J Roentgenol 2004;
182(2):499 504.
219. Tillie-Leblond I, Mastora I, Radenne F, Paillard S, Tonnel AB, Remy J, Remy-
Jardin M. Risk of pulmonary embolism after a negative spiral CT angiogram in
404 Barack et al.
235. Silingardi V, Canossi GC, Torelli G, Russo G. The radiology target sign of septic
pulmonary embolism in a case of acute myelogenous leukemia. Respiration 1981;
42:61 66.
236. Kuhlman JE, Fishman EK, Teigen C. Pulmonary septic emboli: diagnosis with CT.
Radiology 1990; 174:211 213.
237. Iwasaki Y, Nagata K, Nakanishi M, Natuhara A, Harada H, Kubota Y, Yokomura I,
Hashimoto S, Nakagawa M. Spiral CT ndings in septic pulmonary emboli.
Eur J Radiol 2001; 37(3):190 194.
238. Schwartz EE, Teplick JG, Onesti G, Schwartz AB. Pulmonary hemorrhage in
renal disease: Goodpastures syndrome and other causes. Radiology 1977; 122:
39 46.
239. Crawford ES. The diagnosis and management of aortic dissection. J Am Med Assoc
1990; 264(19):2537 2541.
240. Svensson LG, Crawford ES, Hess KR, Coselli JS, Sa HJ. Dissection of the aorta
and dissecting aortic aneurysms: improving early and long-term surgical results.
Circulation 1990; 82(suppl 5):IV24 38.
241. Nienaber CA, von Kodolitsch Y, Nicolas V, Siglow V, Piepho A, Brockhoff C,
Koschyk DH, Spielmann RP. The diagnosis of thoracic aortic dissection by
noninvasive imaging procedures. N Engl J Med. 1993; 328:1 9.
242. Sommer T, Fehske W, Holzknecht N, Smekal AV, Keller E, Lutterbey G, Kreft B,
Kuhl C, Gieseke J, Abu-Ramadan D, Schild H. Aortic dissection: a comparative
study of diagnosis with spiral CT, multiplanar transesophageal echocardiography,
and MR imaging. Radiology 1996; 199(2):347 352.
243. Andresen J, Baekgaard N, Allermand H. Evaluation of patients with thoracic
aortic dissection by intraarterial digital substraction angiography. Vasa 1992;
21(2):167 170.
244. Ledbetter S, Stuk JL, Kaufman JA. Helical (spiral) CT in the evaluation of emer-
gent thoracic aortic syndromes. Traumatic aortic rupture, aortic aneurysm, aortic
dissection, intramural hematoma, and penetrating atherosclerotic ulcer. Radiol
Clin N Am 1999; 37(3):1 17.
245. Movsowitz HD, Lampert C, Jacobs LE, Kotler MN. Penetrating atherosclerotic
aortic ulcers. Am Heart J 1994; 128(6 Pt 1):1210 1217.
246. Harris JA, Bis KG, Glover JL, Bendick PJ, Shetty A, Brown OW. Penetrating ather-
osclerotic ulcers of the aorta. J Vasc Surg 1994; 19(1):90 98.
247. Welch TJ, Stanson AW, Sheedy PF, Johnson CM, McKusick MA. Radiologic
evaluation of penetrating aortic atherosclerotic ulcer. Radiographics 1990;
10:675 685.
248. Yucel EK, Steinberg FL, Egglin TK, Geller SC, Waltman AC. Athanasoulis CA.
Penetrating aortic ulcers: diagnosis with MR imaging. Radiology 1990;
177:799 781.
249. Groskin SA. Selected topics in chest trauma. Semin Ultrasound CT MRI 1996;
17:119 141.
250. Mirvis SE, Bidwell JK, Buddemeyer EU, Diaconis JN, Pais SO, Whitley JE,
Goldstein LD. Value of chest radiography in excluding traumatic aortic rupture.
Radiology 1987; 163:487 493.
251. Goldstein LA, Thompson WR. Esophageal perforation: A 15 year experience. Am J
Surg 1982; 143:495 503.
406 Barack et al.
252. Maroney MJ, Mirvis SE, Shanmuganathan K. Esophageal occlusion caused by thor-
acic spine fracture or dislocation: CT diagnosis. Am J Roentgenol 1996; 167:714
715.
253. Lee S, Mergo PJ, Ros PR. The leaking esophagus: CT patterns of esophageal
rupture, perforation, and stulization. Crit Rev Diagn Imaging 1996; 37(6):
461 490.
254. Cardella JF, Fox PS, Lawler JB. Interventional radiologic placement of peripherally
inserted central catheters. J Vasc Intervent Radiol 1993:4:653 660.
255. Mewissen MW, Crain MR, Paz-Fumagalli R, Beres RA, Wertz RA. Interventional
Procedures in the Intensive Care Unit Patient. Radiol Clin North Am 1996; 34:
157 176.
256. Andrews NC, Parker EF, Shaw RR, Wilson NJ, Webb WR. Management of nontu-
berculous empyema. Am Rev Respir Dis 1962; 85:935 936.
257. Sahn SA. Management of complicated parapneumonic effusions. Am Rev Respir
Dis 1993; 148:813 817.
258. van Sonnenberg E, Nakamonto SK, Mueller PR, Casola G, Neff CC, Friedman PJ,
Ferrucci JT Jr, Simeone JF. CT and ultrasound-guided catheter drainage of
empyemas after chest-tube failure. Radiology 1984; 151(2):349 353.
259. Silverman SB, Muller PR, Saini S, Hahn PF, Simeone JF, Forman BH, Steiner E,
Ferrucci JT. Thoracic empyema: Management with image guided catheter drai-
nage. Radiology 1988; 169(1):5 9.
260. Robinson LA, Mountol AL, Fleming WH, Alonso A, Galbraith TA. Intrapleural
brinolytic treatment of multiloculated thoracic empyemas. Ann Thorac Surg
1994; 57(4):803 814.
261. Pollak JS, Passik CS. Intrapleural urokinase in the treatment of loculated pleural
effusions. Chest 1994; 105:868 873.
262. Ha HK, Kang MW, Park JM, Yan WJ, Shinn KS, Bahk YW. Lung abscess:
Percutaneous catheter therapy. Acta Radiologica 1993; 34:362 365.
263. Yellin A, Yellio EO, Lieberman Y. Percutaneous tube drainge: The treatment of
choice for refractory lung abscess. Ann Thorac Surg 1985; 39:266 170.
264. Vainrub B, Musher DM, Guinn GA, Young EJ, Septimus EJ, Travis LL. Percutaneous
drainage of lung abscess. Am Rev Respir Dis 1978; 117(1):153160.
265. van Sonnenberg E, DAgostino HB, Casola G, Wittich GR, Varney RR, Harker
C. Lung abscess: CT-guided drainage. Radiology 1991; 178(2):347 351.
266. Taavitsainen M, Bondestam S, Mankinen P, Pitkaranta P, Tierala E. Ultrasound
guidance for pericardiocentesis. Acta Radiologica 1991; 32(1):9 11.
267. Girard MS, Kaufman JA, Lau TN, Morris CS, Oliva V, Parr C. Inferior vena cava
ltration. Society of Interventional Radiology 28th Annual Scientic Meeting
Workshop Book, 2003, pp. 349 353.
Index
407
408 Index