Intraoperative Tachycardia and Hypertension Are

Independently Associated with Adverse Outcome in

Noncardiac Surgery of Long Duration
David L. Reich, MD, Elliott Bennett-Guerrero, MD, Carol A. Bodian,
Sabera Hossain, MSc, Wanda Winfree, RN, and Marina Krol, PhD

DrPH,

Department of Anesthesiology, Mount Sinai School of Medicine, New York, New York

Relatively little is known about the influence of intraoperative hemodynamic variables on surgical outcomes. We
drew subjects (n 797) from a study of patients undergoing major noncardiac surgery. The physiological component of the POSSUM (Physiological and Operative Severity Score for the enUmeration of Mortality) operative
risk stratification index was determined, and intraoperative measurements of heart rate (HR), mean arterial blood
pressure, and systolic arterial blood pressure (SAP) were
retrieved from computerized anesthesia records. For every 5-min epoch during the surgery, HR, mean arterial
blood pressure, and SAP were each classified as low, normal, or high. Negative surgical outcome (NSO) was defined as a hospital stay of 10 days with a morbid condition or death during the hospital stay. Statistical analyses

lthough blood pressure and heart rate (HR) are

closely monitored during surgery, relatively
little is known about the associations of abnormalities of HR and blood pressure with outcome indicators, such as morbidity and mortality. Although
numerous studies have examined the relationship between preoperative risk factors and adverse outcomes,
the effect of intraoperative hemodynamic abnormalities on adverse outcomes has been less well defined
(13). In many studies, measurements of intraoperative hemodynamics have been limited to intraoperative blood pressure nadir, steady-state pre- and postoperative measurements, and blood pressure above
and below defined limits of normality for specified
time intervals (1 4).
The results of some previous studies (13) have
been difficult to interpret because they limited their
Accepted for publication April 5, 2002.
Address correspondence and reprint requests to David L. Reich,
MD, Professor of Anesthesiology, Mount Sinai School of Medicine,
One Gustave L. Levy Place, Box 1010, New York, NY 10029-6574.
Address e-mail to david.reich@mssm.edu.
DOI: 10.1213/01.ANE.0000019206.81178.75
2002 by the International Anesthesia Research Society
0003-2999/02

included Mantel-Haenszel tests and multiple logistic regression. There was no significant association between hemodynamic variables and NSO with short operations. In
388 patients with operations longer than the median time
of 220 min, NSO occurred in 15.6%. Controlling for
POSSUM score and operation time beyond 220 min, both
high HR (odds ratio, 2.704; P 0.01) and high SAP (odds
ratio, 2.095; P 0.009) were associated with NSO in longer
operations. Thus, intraoperative tachycardia and hypertension were associated independently with adverse outcomes after major noncardiac surgery of long duration,
over and above the risk imparted by underlying medical
conditions.
(Anesth Analg 2002;95:2737)

statistical analyses to univariate tests. Statistical methods that assess the independent effects of the intraoperative hemodynamic aberrations on the risk of morbidity and mortality after adjusting for the effects of
underlying medical conditions are required. For example, intraoperative hypertension may be a marker
for essential hypertension, such that intraoperative
blood pressure deviations may or may not add additional risk beyond that of the underlying condition of
essential hypertension.
The advent of computerized anesthesia information systems provides the opportunity to record and
store intraoperative hemodynamic data with great
accuracy (5 8). By use of such systems, the independent associations between intraoperative hemodynamic abnormalities and death, stroke, and perioperative myocardial infarction in cardiac surgical
patients have been published (9,10). The purpose of
this investigation was to determine whether intraoperative aberrations of blood pressure or HR were
associated with perioperative mortality or major
morbidity in patients undergoing complex noncardiac surgery while controlling for the influence of
major coexisting medical illness.
Anesth Analg 2002;95:2737

273

274

CARDIOVASCULAR ANESTHESIA REICH ET AL.

TACHYCARDIA, HYPERTENSION, AND OUTCOME

Methods
The study was institutionally approved as a retrospective investigation. Subjects (n 797) were drawn from
a group of 1056 patients who had participated in an
IRB-approved anesthesia outcome study of patients
undergoing major elective noncardiac surgery at one
institution (11). The IRB waived the requirement for
informed consent.
Patients undergoing the following elective surgical
procedures were enrolled: major orthopedic (e.g., revision hip arthroplasty, fusion/instrumentation of
multiple lumbar or thoracic vertebrae); major general
(e.g., any laparotomy expected to exceed 2 h, including partial hepatectomy, pancreatic surgery, and colon
surgery); major urological (e.g., radical cystectomy,
radical nephrectomy); major vascular (e.g., abdominal
aortic aneurysm repair); and major gynecological (e.g.,
cancer debulking procedure, abdominal hysterectomy
with oophorectomy). These procedures were selected
for several reasons: 1) they are routinely performed
surgeries, 2) they represent a diverse group of procedure types, and 3) a previous study performed at
Duke University Medical Center found these procedures to be associated with prolonged hospitalization
and postoperative complications (12). Study patients
received routine anesthetic care and surgical management. Each patient was treated after surgery according
to standard institutional surgical care maps; only
patients undergoing surgical procedures enabling
them to be targeted for discharge from the hospital
before the 10th postoperative day were enrolled.
To quantify perioperative risk of morbidity and
mortality due to underlying medical conditions, the
physiological component of the POSSUM (Physiological and Operative Severity Score for the enUmeration
of Mortality) operative risk stratification index (13)
was determined for each patient. The POSSUM criteria have been cited as the most appropriate scoring
system available for assessing risk in noncardiac surgical patients (14). The POSSUM physiological score
includes 12 preoperative factors (including age, preoperative blood pressure, cardiac disease, and renal
function). A point value of 1, 2, 4, or 8 was assigned for
10 of the 12 factors, depending on the severity of the
abnormality (e.g., 1 point for no dyspnea and 8 points
for dyspnea at rest). The POSSUM score was obtained
by prospective evaluation in the operating room before surgery by a trained anesthesia research nurse.
A subset of patients from the original study had
undergone surgery in operating rooms with computerized anesthesia information systems (CompuRecord;
Philips, Andover, MA). Every patient with a valid computerized anesthesia record (n 797) from the original
study (n 1056) was included in this study. Intraoperative hemodynamic data were derived from these computerized anesthesia records that automatically stored

ANESTH ANALG
2002;95:2737

hemodynamic values every 15 s. HR, mean arterial

blood pressure (MAP), systolic arterial blood pressure
(SAP), and diastolic arterial blood pressure were extracted from the computerized anesthesia records. The
raw data obtained every 15 s may contain artifactual
values because of intermittent electrocautery interference and transducer flushing, among other causes.
Therefore, the data were filtered for such artifacts by
using the median value in consecutive 5-min epochs.
With Microsoft Access (Microsoft, Inc., Redmond,
WA), we quantified the extent and duration of abnormal hemodynamic states during the anesthesia. For
every 5-min epoch, the median HR, MAP, and SAP
was classified as low, normal, or high, according to
criteria derived from the results of a survey of 39
anesthesiologists (Table 1) (15). For each hemodynamic variable, the proportion of 5-min epochs in
which the median value was classified into an abnormal hemodynamic category was calculated. Each hemodynamic variable was evaluated at every 5-min
epoch in an independent fashion. Thus, it was possible
for patients to experience epochs that were classified
as high at one interval and low at another interval in
the same surgical procedure.
The primary composite end point, negative surgical
outcome (NSO), was a postoperative hospital length
of stay of 10 days with a morbid condition or death
during the same hospitalization after surgery. A research nurse reviewed the chart of every patient who
remained an inpatient on the 11th postoperative day.
At that time, the research nurse noted the presence or
absence of postoperative morbidity of the following
types: gastrointestinal, pulmonary, renal, infectious,
wound complication, pain, cardiovascular, neurological, or hematological. The criteria for defining these
morbid states have been published previously (10,16).
The patients were observed until discharge, and all
deaths during hospitalization were noted.
The composite outcome variable (NSO) was the sole
outcome variable analyzed. Data for the derived hemodynamic abnormalities were summarized to identify outliers and to characterize the distributions of
these variables. Bivariate contingency tables were created to examine the association between outcome and
each hemodynamic measure. Significance levels were
determined by 2 tests for trend or by Fishers exact
test, as appropriate. Each derived hemodynamic variable with a P value of 0.20 was considered a potential independent predictor. Mantel-Haenszel tests and
multiple logistic regression were then used to test the
independent influence of these associations while controlling for risk due to medical conditions.

Results
The median operation length in this cohort was
220 min. Patient characteristics and surgical outcomes

ANESTH ANALG
2002;95:2737

CARDIOVASCULAR ANESTHESIA
REICH ET AL.
TACHYCARDIA, HYPERTENSION, AND OUTCOME

Table 1. Criteria Used to Define Abnormal Hemodynamics

Variable

Low

High

Mean arterial blood pressure (mm Hg)

Systolic arterial blood pressure (mm Hg)
Heart rate (bpm)

55
80
45

100
160
110

Table 2. Perioperative Patient Characteristics

Characteristic

Data

No. Patients
797
Age (yr)
60 (4373)
Sex (% male)
54
Race (% Caucasian)
71
Weight (kg)
70 (6080)
Preoperative serum creatinine (mg/dL)
1.0 (0.801.2)
Operative duration (min)
220 (160285)
Intraoperative IV fluid administration (L)
3.5 (2.55.0)
Temperature at the end of surgery (C)
36 (35.536.5)
Negative surgical outcome (%)a
15.6
In-hospital death (%)
1.6
Postoperative hospitalization (days)
7 (610)
Continuous data are presented as median (interquartile range).
a
In-hospital death or hospitalization for 10 days with a morbid condition after surgery.

are described in Table 2. One-hundred-twenty-four

patients had an NSO, including 13 who died in the
hospital up to Day 10. Of the patients who remained
hospitalized on the 11th postoperative day, 111 of 154
(72.1%) had a morbid condition. There was a strong
association between POSSUM score and NSO in patients undergoing operations lasting 220 min and
220 min (P 0.001 for each). These results are displayed in Table 3.
Preliminary analyses using bivariate contingency
tables did not reveal any consistent trends in rate of
NSO with increasing frequency of hemodynamic aberrations. Thus, in all further analyses, the hemodynamic variables (MAP, SAP, and HR) were included
as binary variablesany patient who experienced any
of these aberrant conditions for any amount of time
was considered positive for that variable.
We used Mantel-Haenszel tests to look for an association between the occurrence of each hemodynamic
condition and NSO, controlling for the influence of
POSSUM scores. There were no strong associations
between hemodynamic abnormalities and NSO in the
cohort undergoing operations 220 min in duration,
and this group was not analyzed further. In 388 patients with operations 220 min, increased HR (P
0.044; Table 4) and increased SAP (P 0.034; Table 5)
were each associated with NSO. The proportion of
patients experiencing NSO was larger for patients experiencing low MAP in all but the highest quartile of
POSSUM scores (Table 6). We further noted that there
was a strong association between decreased MAP and
increased HR.

275

Table 3. Incidence of Negative Surgical Outcome

Stratified by Operation Length, Stratified on Physiological
POSSUM Score
Physiological
POSSUM score

Operations
220 min*

Operations
220 min

15
1618
1923
23
Total

5/128 (3.9%)
7/94 (7.5%)
9/92 (9.8%)
21/95 (22.1%)
42/409 (10.3%)

17/142 (12.0%)
13/76 (17.1%)
18/81 (22.2%)
34/89 (38.2%)
82/388 (21.1%)

* P 0.001; P 0.001.

Table 4. Incidence of Negative Surgical Outcome (NSO)

by High Heart Rate, Stratified on POSSUM Physiological
Score in Long Operations (220 minutes)
Physiological
POSSUM score

No high HR

High HR

15
1618
1923
23

15/125 (12%)
10/66 (15.2%)
16/74 (21.6%)
26/77 (33.8%)

2/17 (11.8%)
3/10 (30.0%)
2/7 (28.6%)
8/12 (66.7%)

Overall P 0.044 (Mantel-Haenszel test of association between high HR

and NSO); HR heart rate.

Table 5. Incidence of Negative Surgical Outcome (NSO)

by High Systolic Arterial Blood Pressure, Stratified on
POSSUM Physiological Score in Long Operations
(220 minutes)
Physiological
POSSUM score

No high SAP

High SAP

15
1618
1923
23

9/95 (9.5%)
5/33 (15.2%)
7/41 (17.1%)
10/34 (29.4%)

11/47 (17.0%)
8/43 (18.6%)
11/40 (27.5%)
24/55 (43.6%)

Overall P 0.034 (Mantel-Haenszel test of association between high SAP

and NSO); SAP systolic arterial blood pressure.

Table 6. Incidence of Negative Surgical Outcome (NSO)

by Low Mean Arterial Blood Pressure, Stratified on
POSSUM Physiological Score in Long Operations
(220 minutes)
Physiological
POSSUM score

No low MAP

Low MAP

15
1618
1923
23

12/117 (10.3%)
10/65 (15.4%)
14/70 (20%)
29/71 (40.9%)

5/25 (20%)
3/11 (27.3%)
4/11 (36.4%)
5/18 (27.8%)

Overall P 0.34 (Mantel-Haenszel test of association between low MAP

and NSO); MAP mean arterial blood pressure.

The confounding effect of decreased MAP and increased HR was apparent in the multiple logistic regression analyses. Although each contributed significantly when considered without the other, increased
HR was the stronger predictor. The independent influence of increased HR and increased SAP in operations 220 min is shown in the multiple regression

276

CARDIOVASCULAR ANESTHESIA REICH ET AL.

TACHYCARDIA, HYPERTENSION, AND OUTCOME

ANESTH ANALG
2002;95:2737

Table 7. Multivariate Analysis of Negative Surgical Outcome in Long Operations (220 minutes)
Variable

Odds ratio

P value

Operation duration 220 min (per minute)

POSSUM physiological score (per point of score)
High heart rate
High systolic arterial blood pressure

1.003
1.096
2.704
2.095

0.02
0.0001
0.01
0.009

results (Table 7). These results are consistent with

those noted in the Mantel-Haenszel tests.

Discussion
This study investigated associations between hemodynamic variables measured in the course of complex
noncardiac surgery, and poor postoperative outcome.
Preoperative risk assessment with the POSSUM physiological score was included in the model to examine
the additional effects of intraoperative hemodynamic
aberrations on poor postoperative outcome, over and
above the influence of the factors that existed when
patients were brought to surgery. The POSSUM physiological score reflects the presence of preoperative
hypotension, hypertension, tachycardia, and bradycardia. Thus, our findings suggest that intraoperative
tachycardia and hypertension during long, complex
noncardiac surgery may partially explain the variability in outcome seen among patients with similar degrees of preoperative risk.
There are very few reports of independent associations of intraoperative hemodynamic aberrations
with complications. Jain et al. (9) reported that SAP
90 mm Hg after cardiopulmonary bypass was an
independent predictor of perioperative myocardial infarction. In a previous study at two institutions (including the authors), we identified independent associations between various hemodynamic aberrations,
including increased pulmonary artery diastolic pressure, with death, stroke, and myocardial infarction
after coronary artery bypass surgery (9).
A potential limitation of our study relates to the
selection of the primary end point as a composite of
mortality and prolonged postoperative hospitalization
with morbidity. Mortality is the most important end
point, but it is relatively infrequent compared with
morbidity in this group of surgical patients. We chose
prolonged postoperative hospitalization with morbidity because the presence of both strongly suggests that
the prolonged hospitalization was due to the morbid
condition (12). Furthermore, if the morbidity resulted
in a prolonged hospitalization, it also has a greater
economic effect.
A major question not addressed by this study is the
reason for the association of brief periods of hemodynamic abnormalities with complications that are overwhelmingly noncardiac. The majority of the morbidity

observed in patients with prolonged hospitalization in

our study involved organ systems unrelated to the
type or site of surgery. Infectious, pulmonary, and
renal complications were observed frequently. Hemodynamic predictors of adverse outcome identified in
this study may not be the primary causes. They may
be markers of pathophysiological states that caused
the complications but are not reflected in the POSSUM
physiological score or the anesthesia record. For example, high HR could represent hypovolemic states
with inadequate tissue perfusion, due to causes such
as hemorrhage, endotoxemia, or exaggerated inflammatory responses to surgery (16). The confounding
effect noted between decreased MAP and increased
HR further suggests that hypovolemia may be the
underlying cause of the hemodynamic aberrations
that were associated with complications.
This observational study allowed us to identify hemodynamic abnormalities associated with adverse
outcomes in patients undergoing treatment by experienced teams of anesthesiologists and surgeons. The
hemodynamic changes observed in these patients occurred despite attempts to maintain normal hemodynamic status during surgery. It does not necessarily
follow, however, that more aggressive therapy aimed
at normalizing hemodynamics would improve outcome. A randomized clinical trial would be required
to test the latter hypothesis.
Another limitation of the study was the categorization of hemodynamic variables into normal and abnormal groups on the basis of limits defined by a
survey of clinicians. The choice of absolute limits
rather than relative changes from baseline was based
on our concern that we could not establish the true
baseline values in these patients in the perioperative
period.
In conclusion, specific intraoperative hemodynamic
aberrations were independently associated with poor
postoperative outcome, as reflected by mortality or
prolonged postoperative hospitalization with morbidity. These findings demonstrate the prognostic relevance of intraoperative hemodynamic status over and
above the effects of certain preoperative risk factors.
The question of whether outcomes could be improved
by greater control of intraoperative hemodynamic
variables awaits prospectively designed studies. These
additional studies would be justifiable on the basis of
the expense of current health care resources that must

ANESTH ANALG
2002;95:2737

CARDIOVASCULAR ANESTHESIA
REICH ET AL.
TACHYCARDIA, HYPERTENSION, AND OUTCOME

be devoted to caring for patients with poor postoperative outcomes.

References
1. Goldman L, Caldera DL. Risks of general anesthesia and elective operation in the hypertensive patient. Anesthesiology 1979;
50:28592.
2. Prys-Roberts C, Meloche R, Foex P. Studies of anaesthesia in
relation to hypertension. I. Cardiovascular responses of treated
and untreated patients. Br J Anaesth 1971;43:11237.
3. Charlson ME, MacKenzie CR, Gold JP, et al. Preoperative characteristics predicting intraoperative hypotension and hypertension among hypertensives and diabetics undergoing noncardiac
surgery. Ann Surg 1990;212:66 81.
4. Mauney FM, Ebert PA, Sabiston DC. Postoperative myocardial
infarction: a study of predisposing factors, diagnosis and mortality in a high risk group of surgical patients. Ann Surg 1970;
172:497502.
5. Hollenberg HP, Pirraglia PA, Williams-Russo P, et al. Computerized data collection in the operating room during coronary
artery bypass surgery: a comparison to the hand-written anesthesia record. J Cardiothorac Vasc Anesth 1997;11:54551.
6. Lerou JG, Dirksen R, van Daele M, et al. Automated charting of
physiological variables in anesthesia: a quantitative comparison
of automated versus handwritten anesthesia records. J Clin
Monit 1988;41:37 47.
7. Cook RI, McDonald JS, Nunziata E. Differences between handwritten and automatic blood pressure records. Anesthesiology
1989;71:38590.

277

8. Reich DL, Wood RK, Mattar R, et al. Arterial blood pressure and
heart rate discrepancies between handwritten and computerized anesthesia records. Anesth Analg 2000;91:612 6.
9. Jain U, Laflamme CJA, Aggarwal A, et al. Electrocardiographic
and hemodynamic changes and their association with myocardial infarction during coronary artery bypass surgery. Anesthesiology 1997;86:576 91.
10. Reich DL, Bodian CA, Krol M, et al. Intraoperative hemodynamic predictors of mortality, stroke and myocardial infarction
following coronary artery bypass surgery. Anesth Analg 1999;
88:814 22.
11. Bennett-Guerrero E, Panah MH, Barclay GR, et al. Decreased
endotoxin immunity is associated with greater mortality and/or
prolonged hospitalization after surgery. Anesthesiology 2001;
94:992 8.
12. Bennett-Guerrero E, Welsby I, Dunn TJ, et al. The use of a
postoperative morbidity survey to evaluate patients with prolonged hospitalization after routine, moderate-risk, elective surgery. Anesth Analg 1999;89:514 9.
13. Copeland GP, Jones D, Walters M. POSSUM: a scoring system
for surgical audit. Br J Surg 1991;78:355 60.
14. Jones HJ, de Cossart L. Risk scoring in surgical patients. Br J
Surg 1999;86:149 57.
15. Reich DL, Pavone L, Bennett-Guerrero E, Krol M. How do
anesthesiologists define abnormal heart rate and blood pressure
[abstract]? Anesth Analg 1999;88:S192.
16. Bennett-Guerrero E, Ayuso L, Hamilton-Davies C, et al. Relationship of preoperative antiendotoxin core antibodies and adverse outcomes following cardiac surgery. JAMA 1997;277:
646 50.

Attention Authors!
Submit Your Papers Online
You can now have your paper processed and reviewed faster by sending it to us through our new,
web-based Rapid Review System.
Submitting your manuscript online will mean that the time and expense of sending papers through the
mail can be eliminated. Moreover, because our reviewers will also be working online, the entire review
process will be significantly faster.
You can submit manuscripts electronically via www.rapidreview.com. There are links to this site from the
Anesthesia & Analgesia website (www.anesthesia-analgesia.org), and the IARS website (www.iars.org).
To find out more about Rapid Review, go to www.rapidreview.com and click on About Rapid Review.