120999 the effect of bisoprolol on perioperative

E F F EC T O F B I S O P R O L O L O N M O R B I D I T Y A N D M O R TA L I T Y I N PAT I E N T S U N D E R G O I N G VAS C U L A R S U R G E RY
THE EFFECT OF BISOPROLOL ON PERIOPERATIVE MORTALITY
AND MYOCARDIAL INFARCTION IN HIGH-RISK PATIENTS
UNDERGOING VASCULAR SURGERY
DON POLDERMANS, PH.D., ERIC BOERSMA, PH.D., JEROEN J. BAX, PH.D., IAN R. THOMSON, PH.D., LOUIS L.M. VAN DE VEN, PH.D., JAN D. BLANKENSTEIJN, PH.D., HUBERT F. BAARS, M.D., TIK-IEN YO, PH.D., GIUSEPPE TROCINO, M.D., CARLO VIGNA, M.D., JOS R.T.C. ROELANDT, PH.D., AND HERO VAN URK, PH.D., FOR THE DUTCH ECHOCARDIOGRAPHIC CARDIAC RISK EVALUATION APPLYING STRESS ECHOCARDIOGRAPHY STUDY GROUP* ABSTRACT
ATIENTS undergoing major vascular surgery Background
are at risk for serious perioperative cardiac most important causes of perioperative morbidity complications, such as nonfatal myocardial in- and mortality among patients undergoing major vas- farction and death.1 A combination of clini- cal assessment and noninvasive cardiac testing can be Methods
used to identify patients at high risk.2 Various inter- trial to assess the effect of perioperative blockade of ventions have been proposed to reduce the risk of car- beta-adrenergic receptors on the incidence of death diac complications in such patients after noncardiac from cardiac causes and nonfatal myocardial infarc- surgery, but none have been found to be efficacious.
tion within 30 days after major vascular surgery in pa- Drugs that block beta-adrenergic receptors prevent tients at high risk for these events. High-risk patients cardiac complications in patients with acute myocar- were identified by the presence of both clinical risk factors and positive results on dobutamine echocar- dial infarction, silent ischemia, and heart failure. Per- diography. Eligible patients were randomly assigned ioperative blockade of beta-adrenergic receptors has to receive standard perioperative care or standard care been proposed to reduce the risk of perioperative car- plus perioperative beta-blockade with bisoprolol.
diac complications.7-9 However, previous trials lacked Results
the statistical power to evaluate the cardioprotective and 846 were found to have one or more cardiac risk effect of beta-blockade on the incidence of serious factors. Of these 846 patients, 173 had positive re- cardiac events, such as death or nonfatal myocardial sults on dobutamine echocardiography. Fifty-nine pa- tients were randomly assigned to receive bisoprolol, We hypothesized that perioperative beta-blockade and 53 to receive standard care. Fifty-three patients with bisoprolol would reduce the perioperative inci- were excluded from randomization because they werealready taking a beta-blocker, and eight were exclud- dence of death from cardiac causes and nonfatal my- ed because they had extensive wall-motion abnor- ocardial infarction in high-risk patients undergoing malities either at rest or during stress testing. Two major vascular surgery. We therefore performed a mul- patients in the bisoprolol group died of cardiac caus- ticenter study of a subgroup of high-risk patients who es (3.4 percent), as compared with nine patients in had clinical predictors of cardiac risk together with the standard-care group (17 percent, P=0.02). Non- positive results on dobutamine echocardiography.
fatal myocardial infarction occurred in nine patientsgiven standard care only (17 percent) and in none of those given standard care plus bisoprolol (P<0.001).
Study Protocol
Thus, the primary study end point of death from car-diac causes or nonfatal myocardial infarction oc- Between 1996 and 1999, we prospectively screened all patients curred in 2 patients in the bisoprolol group (3.4 per- undergoing elective abdominal aortic or infrainguinal arterial re- cent) and 18 patients in the standard-care group (34 construction at seven participating centers. The ethics committeeat each center approved the study protocol. The patients were Conclusions
incidence of death from cardiac causes and nonfatalmyocardial infarction in high-risk patients who areundergoing major vascular surgery. (N Engl J Med From Erasmus Medical Center, Rotterdam, the Netherlands (D.P., E.B., J.J.B., L.L.M.V., J.R.T.C.R., H.U.); the University of Manitoba, Winnipeg, Canada (I.R.T.); University Hospital, Utrecht, the Netherlands (J.D.B.); 1999, Massachusetts Medical Society.
Twee Steden Ziekenhuis, Tilburg, the Netherlands (H.F.B.); Sint ClaraZiekenhuis, Rotterdam, the Netherlands (T.-I.Y.); San Gerardo Hospital,Monza, Italy (G.T.); and Istituto di Ricovero e Cura a Carattere ScientificoHospital, San Giovanni Rotondo, Italy (C.V.). Address reprint requests toDr. Poldermans at the Department of Vascular Surgery, Erasmus UniversityMedical Center, Dr. Molewaterplein 40, 3015 GD Rotterdam, the Neth-erlands, or at [email protected].
Other authors were Paolo M. Fioretti, Ph.D., Istituto di Cardiologia, Azienda Ospedaliera, Santa Maria della Misericordia, Udine, Italy; and Ber-nard Paelinck, M.D., Ziekenhuis Middelheim, Antwerp, Belgium.
*The members of the study group are listed in the Appendix.
Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne screened for the following cardiac risk factors: age over 70 years;angina; prior myocardial infarction on the basis of history or a TABLE 1. BASE-LINE CHARACTERISTICS OF THE PATIENTS.*
finding of pathologic Q waves on electrocardiography; compen-sated congestive heart failure or a history of congestive heart fail-ure; current treatment for ventricular arrhythmias; current treat- BISOPROLOL
STANDARD-CARE
ment for diabetes mellitus; or limited exercise capacity, defined as the inability to perform most normal daily activities. Any patient CHARACTERISTIC
with a risk factor underwent dobutamine echocardiography. Pa- tients who had a positive result on dobutamine echocardiography Patients were excluded if they had extensive wall-motion abnor- malities (wall-motion index more than 1.70 at rest), asthma, or Previous myocardial infarction — no. (%) strong evidence during stress testing of left main or severe three- vessel coronary artery disease. Patients were also excluded if they were already taking a beta-adrenergic–receptor antagonist. Thesepatients either continued their usual medication or were switched to bisoprolol. They were treated similarly to patients assigned to the bisoprolol group, but their data were analyzed separately.
After giving written informed consent, eligible patients were *There were no significant differences between the two groups.
randomly assigned to receive either standard perioperative care orstandard care plus bisoprolol. A computer algorithm was used ateach center to assign patients randomly, in a one-to-one ratio, toone of the two study groups. Treatment with bisoprolol, a selec-tive b -adrenergic–receptor antagonist, was started at least 1 week TABLE 2. RESULTS OF DOBUTAMINE ECHOCARDIOGRAPHY.*
before surgery and continued for 30 days postoperatively. The initial dose of bisoprolol was 5 mg orally once a day. Ap- proximately one week after starting bisoprolol, the patients were BISOPROLOL
STANDARD-CARE
reassessed, and the dose was increased to a maximum of 10 mg once daily if the heart rate remained above 60 beats per minute.
CHARACTERISTIC
The same dose of bisoprolol was continued postoperatively ex- cept in patients who were unable to take medication orally or by nasogastric tube postoperatively. In these patients, the heart rate was monitored continuously in the intensive care unit or hourly on the ward, and intravenous metoprolol was administered at a dose sufficient to keep the heart rate below 80 beats per minute.
The heart rate and blood pressure were measured immediately No. of ischemic segments — no. of patients before each scheduled dose of bisoprolol. Bisoprolol was withheld if the heart rate was under 50 beats per minute or the systolic *There were no significant differences between the two groups.
Dobutamine Echocardiography
Dobutamine echocardiography was performed as described pre- viously.10 The left ventricle was divided into 16 segments, and wallmotion was scored on a 5-point scale (with a score of 1 indicating members of the adverse-events committee. The primary end normal, 2 mild hypokinesis, 3 severe hypokinesis, 4 akinesis, and points were death from cardiac causes or nonfatal myocardial in- 5 dyskinesis). The results were considered positive if wall motion farction during the perioperative period. The cause of death was in any segment decreased by one or more grades during testing.
determined by reviewing the clinical presentation, including serum The extent and location of ischemia were evaluated in every pa- cardiac isoenzyme levels, 12-lead electrocardiograms, and, when tient, and a wall-motion index was calculated at rest.
available, autopsy results. Nonfatal myocardial infarction was de-fined by a serum creatine kinase level of more than 110 U per liter, Perioperative Management
with an MB isoenzyme fraction of more than 10 percent, and afinding of new Q waves lasting more than 0.03 second on the Anesthetic management, monitoring, surgical technique, and other aspects of perioperative management were at the discretionof the attending physicians. The attending physicians were aware Statistical Analysis
of the patients’ echocardiographic results and treatment groups.
The calculation of sample size was based on a previous study If symptoms or signs of perioperative myocardial ischemia accom- in which we noted a 28 percent incidence of serious perioperative panied by tachycardia developed in a patient in the standard-care cardiac events in patients who had clinical risk factors as well as group, the attending physician was permitted to administer beta- positive results on dobutamine echocardiography.11 We calculated that the inclusion of 266 patients would allow us to detect a re-duction in the incidence of the primary end point from 30 per- Follow-up
cent to 15 percent with an alpha level of 0.05 and a statistical The patients were monitored for cardiac events for 30 days power of 0.80. As part of the study design, an interim analysis by after surgery. Twelve-lead echocardiography was performed and an independent safety committee was planned after enrollment of the serum creatine kinase level (with the MB fraction) was deter- the first 100 patients. In accordance with the O’Brien and Flem- mined one, three, and seven days after surgery. Additional meas- ing criteria,12 the protocol specified that the trial would be urements were performed at the discretion of the attending phy- stopped if there was a significant difference in the rate of the pri- sician. Outpatient follow-up was performed at 30 days if a patient mary end point between the bisoprolol group and the control had been discharged from the hospital. All data were collected by the participating centers and evaluated in a masked fashion by Continuous data are presented as median values and corre- Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved. E F F EC T O F B I S O P R O L O L O N M O R B I D I T Y A N D M O R TA L I T Y I N PAT I E N T S U N D E R G O I N G VAS C U L A R S U R G E RY
sponding 25th and 75th percentiles, whereas dichotomous dataare presented as percentages. Differences between the groups of pa- TABLE 3. CHARACTERISTICS OF SURGICAL PROCEDURES.*
tients in clinical and surgical characteristics and in the results ofdobutamine echocardiography were evaluated by Wilcoxon’s non-parametric test or Fisher’s exact test, whichever was appropriate.
BISOPROLOL STANDARD-CARE
Differences between the groups in the rates of occurrence of the primary end point were evaluated by Fisher’s exact test. Event rates CHARACTERISTIC
were further examined by the Kaplan–Meier method and a corre- sponding log-rank test. Cox proportional-hazards regression analy- sis was performed to estimate the relative risk of the end point.
Analyses were performed according to the intention to treat.
Characteristics of the Patients
A total of 1351 patients were screened, and 846 were found to have one or more cardiac risk factors.
Stress-induced ischemia occurred in 173 of the 846 patients (20 percent) during dobutamine echocardi- ography. Of these patients, 112 underwent random- ization: 59 were assigned to receive bisoprolol plus standard care and 53 to receive standard care alone.
All but one patient underwent the planned vascular surgical procedure after randomization. Sixty-one pa- *There were no significant differences between the two groups.
tients did not undergo randomization: 53 were al-ready taking a beta-blocker at enrollment and under-went the planned vascular surgical procedure whilecontinuing to take this medication, and 8 had exten-sive wall-motion abnormalities. TABLE 4. MEAN HEART RATES OF RANDOMIZED PATIENTS AT BASE
The clinical characteristics of the randomized LINE AND ONE, THREE, AND SEVEN DAYS POSTOPERATIVELY.* patients are presented in Table 1. The results of do-butamine echocardiography, details of the surgical BISOPROLOL GROUP
STANDARD-CARE GROUP
procedure, and important aspects of perioperative MEASUREMENT
management are presented in Tables 2 and 3. There were no significant differences between the groupswith respect to any of these variables. Patients in the bisoprolol group had significantly lower heart rates before surgery and during the perioperative period than did those given standard care (Table 4).
Beta-Blockade
In the bisoprolol group, bisoprolol was started an average of 37 days (range, 7 to 89) before surgery.
The daily dose was 10 mg in 15 patients and 5 mg in 44. All these patients received bisoprolol orally onthe morning of surgery. The day after surgery, biso- *The mean heart rate was measured in the morning before the next dose prolol was administered either orally or by nasogas- tric tube in 31 patients and intravenously in 28 pa-tients. By the fourth day after surgery, all patientshad resumed oral therapy. No patient in this grouphad a preoperative exacerbation of underlying pe- prolol group (P<0.001). Eight of the infarctions occurred postoperatively, and one occurred preoper-atively. The one preoperative infarction was consid- Cardiac Events in Randomized Patients
ered a primary end point, even though the surgery Nine patients in the standard-care group (17 per- was canceled. The overall rate of the combined end cent) died of cardiac causes during the perioperative point of death from cardiac causes or nonfatal myo- period, as compared with two patients (3.4 percent) cardial infarction was 34 percent (95 percent confi- in the bisoprolol group (P=0.02). There were no dence interval, 21 to 48 percent) in the standard-care deaths from other causes. Nine patients in the stand- group, as compared with 3.4 percent (95 percent con- ard-care group (17 percent) had a nonfatal myocar- fidence interval, 0 to 8.0 percent; P<0.001) in the bis- dial infarction, as compared with none in the biso- oprolol group. The estimated relative risk of death Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne Figure 1. Kaplan–Meier Estimates of the Cumulative Percentages of Patients Who Died of Cardiac Caus-
es or Had a Nonfatal Myocardial Infarction during the Perioperative Period.
I bars indicate standard errors. The difference between groups was significant (P<0.001 by the log-rank test).
from cardiac causes or nonfatal myocardial infarction vascularization but with perioperative administration during the perioperative period among patients re- of beta-blockers. None of these four patients died, ceiving bisoprolol therapy, as compared with those but one had a perioperative myocardial infarction.
receiving standard care, was 0.09 (95 percent confi-dence interval, 0.02 to 0.37). Four patients in the DISCUSSION
standard-care group received beta-blockers perioper- In this randomized, multicenter study, we found atively because of severe intraoperative myocardial is- that perioperative administration of bisoprolol re- chemia or postoperatively because of unstable angina.
duced the perioperative incidence of both death from The significant difference between groups in the cardiac causes and nonfatal myocardial infarction in incidence of serious cardiac events prompted the safe- high-risk patients undergoing major vascular surgery.
ty committee to interrupt the study after the planned The combined incidence of these cardiac events was interim analysis. While the safety committee was eval- 34 percent in the standard-care group, as compared uating the results, six additional patients underwent with 3.4 percent in the bisoprolol group; this differ- randomization. The number of patients who reached ence caused the safety committee to interrupt the the end point is presented in Figure 1. The majority study after the first planned interim analysis. The ob- of events occurred during the first seven days after served treatment effect is not attributable to differ- ences between the two groups with respect to clinicalcharacteristics, results on dobutamine echocardiog- Cardiac Events in Nonrandomized Patients
raphy, surgical procedures, anesthetic or analgesic Among the 53 patients with positive results on technique, or the duration of hospitalization in the dobutamine echocardiography who were already tak- ing a beta-blocker, 4 (7.5 percent) died of cardiac At first glance, the 34 percent rate of serious per- causes during the perioperative period and none had ioperative events in the standard-care group seems a myocardial infarction. The eight patients who did high. However, this rate is consistent with the results not undergo randomization because of extensive rest- of a previous study at one of our institutions. In a ing or stress-induced wall-motion abnormalities (or group of 300 consecutive patients, the incidence of both) had a median wall-motion index of 1.84 and serious cardiac events was 28 percent in patients with had new wall-motion abnormalities in seven seg- one or more risk factors and positive results on dobu- ments. Four of these patients underwent coronary- tamine echocardiography.11 Other investigators have artery bypass grafting, and two of them died. The two noted high rates of perioperative cardiac events in surviving patients subsequently underwent unevent- patients in whom ischemia can be induced on peri- ful vascular surgery. The other four patients under- went vascular surgery without prior myocardial re- The cause of the perioperative cardiac events re- Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved. E F F EC T O F B I S O P R O L O L O N M O R B I D I T Y A N D M O R TA L I T Y I N PAT I E N T S U N D E R G O I N G VAS C U L A R S U R G E RY
mains undefined. Such events occur almost exclusively Preoperative evaluation offers a unique opportu- among patients with positive results on dobutamine nity to screen patients for underlying coronary ar- echocardiography,11 suggesting that hemodynamical- tery disease and cardiac risk factors that affect the ly significant coronary-artery stenosis has an essential perioperative outcome. A task force of the American role in the pathogenesis of perioperative coronary is- College of Cardiology and the American Heart As- chemic syndromes. Coronary stenosis may increase a sociation has published guidelines for the preopera- patient’s vulnerability to ischemia or infarction caused tive cardiovascular evaluation of patients scheduled by perioperative hemodynamic instability, anemia, hy- for noncardiac surgery.2 According to these guide- poxemia, hypercoagulability, or rupture of coronary lines, patients undergoing vascular surgery who have intermediate risk factors (Canadian Cardiovascular The mechanism by which beta-blockade reduces Society class I or II angina, myocardial infarction, perioperative cardiac events is also unclear. Several congestive heart failure, or diabetes mellitus) or lim- mechanisms have been proposed. Myocardial oxygen ited exercise capacity should undergo a noninvasive balance may be improved by decreases in the heart cardiac evaluation before surgery. We prefer dobuta- rate and myocardial contractility, thus preventing my- mine echocardiography for this purpose, because of ocardial ischemia, reducing the size of the infarct, or its extremely high negative predictive value and be- both. Beta-blockade may reduce myocardial oxygen cause it can be used in patients who cannot exercise.
consumption by suppressing lipolysis and thus causing Patients with no clinical risk factors and without the myocardium to metabolize more glucose in rela- stress-induced ischemia have a rate of serious peri- tion to free fatty acids. Beta-blocking drugs may in- operative cardiac events that is close to zero.12 They crease the stability of coronary atherosclerotic plaques can undergo surgery without extra evaluation or ex- or increase the threshold for ventricular fibrillation tensive monitoring. Patients with a positive test re- in the presence of ischemia.6 Antagonists of beta-adre- sult are at high risk and are candidates for perioper- nergic receptors have also been shown to improve the outcome among patients with acute myocardial in- On the basis of our results, we recommend that farction, silent ischemia, and heart failure.3-5 high-risk surgical patients receive beta-blockers per- The only previous randomized, controlled study ioperatively, beginning one to two weeks before sur- evaluating the cardioprotective effect of beta-adrener- gery. The goal should be to reduce the heart rate to gic antagonists in patients undergoing major surgery less than 70 beats per minute preoperatively and to was performed by Mangano and colleagues.7,8 In this less than 80 beats per minute in the immediate post- study, 200 patients who had or were at risk for coro- operative period. Therapy should be continued for nary artery disease were randomly assigned to receive at least two weeks postoperatively. An alternative to atenolol or placebo during the perioperative period.
this approach would be to omit preoperative nonin- The patients were monitored perioperatively for cardi- vasive cardiac testing and prescribe a beta-blocker per- ac events and then followed for two years after surgery.
ioperatively for all patients with clinical risk factors During the first 48 hours after surgery, the incidence who are undergoing high-risk surgery. Although our of myocardial ischemia, as detected by continuous results applied to patients who were undergoing ma- three-lead Holter monitoring, was reduced by 50 per- jor vascular surgery, we recommend that high-risk cent in patients given atenolol. During the two-year patients undergoing other types of noncardiac sur- follow-up period, the mortality rate was 10 percent in gery receive a beta-blocker perioperatively.
patients given atenolol and 21 percent in the controls.
Our study has several limitations. Most important, However, atenolol did not significantly reduce the in- it was not conducted in a blinded fashion. The at- cidence of death from cardiac causes during hospital- tending physicians knew which treatment had been ization or that of perioperative myocardial infarction. prescribed, and bias on their part may have affected The failure of atenolol to alter the perioperative the management and outcome. However, we found outcome significantly reflects the low incidence of se- no significant differences in major aspects of periop- rious perioperative cardiac events in the study popu- erative treatment (e.g., anesthetic or analgesic tech- lation (3 percent). The study included both patients nique or duration of hospitalization in the intensive with known coronary artery disease and those with care unit) between the two groups. The rate of events only coronary risk factors, and the patients under- in the standard-care group was not higher than that went various surgical procedures. In contrast, we stud- predicted on the basis of our previous work. It is un- ied a population of patients who were undergoing likely that factors other than the use of beta-blockers vascular surgery who were identified by clinical screen- can account for the reduced rate of events in the ing and dobutamine echocardiography as being at bisoprolol group. For such factors to have affected high risk, with an anticipated rate of cardiac events the outcome, physicians at eight different institutions of 28 percent. Our results demonstrate the impor- would have had to use — consistently and exclusive- tance of testing risk-reduction strategies in a sub- ly — an unproved, but extremely effective, risk-reduc- tion strategy in the bisoprolol group.
Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne The lack of blinding may also have affected the re- van de Meer and P. Klootwijk; Safety committee — M.L. Simoons and G.A.
porting of events. Systematic screening for myocar- van Es; Participating centers — the Netherlands: Erasmus Medical Center,Rotterdam (D. Poldermans, M. van Nierop, H. van Urk); Sint Clara Ziek- dial infarction with the use of cardiac isoenzyme levels enhuis, Rotterdam (M.G. Scheffer, T.-I. Yo); Twee Steden Ziekenhuis, Til- and electrocardiography was performed only during burg (H.F. Baars, S. Berends, S.E. Kranendonk); Academisch Ziekenhuis,Utrecht (J.D. Blankensteijn, H.W.J. Meijburg, R. Rienks); and Medisch the first week after surgery. Subsequent testing was Centrum, Alkmaar (J.H. Cornel, H.A. van Dijk); Belgium: Ziekenhuis discretionary and may have been influenced by bias Middelheim Antwerp (B. Paelinck); Italy: San Gerardo Hospital, Monza on the part of the investigators. Such bias may have (G. Emanuelli, G. Trocino, A. Virtuani, M. Zerbato); and Istituto di Ri-covero e Cura a Carattere Scientifico Hospital, San Giovanni Rotondo (C.
affected the diagnosis and reporting of myocardial infarctions after the first week. However, most infarc-tions occurred in the first seven days after surgery REFERENCES
and would have been detected by routine screening. 1. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990;72:
Our decision to exclude eight patients with exten- sive resting or stress-induced wall-motion abnormal- 2. Eagle KA, Brundage BH, Chaitman BR , et al. Guidelines for perioper-
ative cardiovascular evaluation for noncardiac surgery: report of the Amer-
ities was arbitrary and may have affected the results.
ican College of Cardiology/American Heart Association Task Force on Our exclusion criteria were established before the Practical Guidelines (Committee on Perioperative Cardiovascular Evalua-tion for Noncardiac Surgery). J Am Coll Cardiol 1996;27:910-48.
study began and reflected our belief that it would be 3. First International Study of Infarct Survival Collaborative Group. Ran-
inappropriate to constrain the perioperative care of domised trial of intravenous atenolol among 16 027 cases of suspected these patients in any way. There is little doubt that acute myocardial infarction: ISIS-1. Lancet 1986;2:57-65.
4. Pepine CJ, Cohn PF, Deedwania PC, et al. Effects of treatment on out-
the randomized patients still constitute an appropri- come in mildly symptomatic patients with ischemia during daily life: the ate high-risk group in which to test the value of per- Atenolol Silent Ischemia Study. Circulation 1994;90:762-8.
5. CIBIS-II Investigators and Committees. The Cardiac Insufficiency
Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999;353:9-13.
There were doubtless minor variations between 6. Kjekshus J, Gullestad L. Heart rate as a therapeutic target in heart fail-
centers in the interpretations of the results of dobu- ure. Eur Heart J 1999;1:Suppl H:H64-H69.
7. Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol on mor-
tamine echocardiography, but the outcome in the tality and cardiovascular morbidity after noncardiac surgery. N Engl J Med standard-care group shows that the presence of one 1996;335:1713-20. [Erratum, N Engl J Med 1997;336:1039.] or more risk factors plus a positive test result does 8. Wallace A, Layug B, Tateo I, et al. Prophylactic atenolol reduces post-
operative myocardial ischemia. Anesthesiology 1998;88:7-17.
define a high-risk population. Furthermore, since ran- 9. Mangano DT, Browner WS, Hollenberg M, et al. Association of peri-
domization was performed at each participating cen- operative myocardial ischemia with morbidity and mortality in men under- ter, differences in the results of dobutamine echocar- going noncardiac surgery. N Engl J Med 1990;323:1781-8.
10. McNeill AJ, Fioretti PM, el-Said EM, Salustri A, Forster T, Roelandt
diography between centers would not have biased JRTC. Enhanced sensitivity for detection of coronary artery disease by ad- dition of atropine to dobutamine stress echocardiography. Am J Cardiol 1992;70:41-6.
In conclusion, we found that bisoprolol reduces the 11. Poldermans D, Arnese M, Fioretti PM, et al. Improved cardiac risk
perioperative incidence of death from cardiac causes stratification in major vascular surgery with dobutamine-atropine stress and of nonfatal myocardial infarction in high-risk pa- echocardiography. J Am Coll Cardiol 1995;26:648-53.
12. O’Brien PC, Fleming TR. A multiple testing procedure for clinical tri-
tients undergoing abdominal aortic or infrainguinal 13. Lalka SG, Sawada SG, Dalsing MC, et al. Dobutamine stress echocar-
diography as a predictor of cardiac events associated with aortic surgery.
APPENDIX
J Vasc Surg 1992;15:831-42.
14. Lane RT, Sawada SG, Segar DS, et al. Dobutamine stress echocardi-
The members of the Dutch Echocardiographic Cardiac Risk Evaluation ography as a predictor of perioperative cardiac events. Am J Cardiol 1991; Applying Stress Echocardiography study group and the participating cen- ters were as follows: Steering committee — D. Poldermans, I.R. Thomson, 15. Davila-Roman VG, Waggoner AD, Sicard GA, Geltman EM, Schecht-
H. van Urk, J.J. Bax, P.M. Fioretti, A.J. Man in ’t Veld, L.L.M. van de Ven, man KB, Perez JE. Dobutamine stress echocardiography predicts surgical and J.R.T.C. Roelandt; Statistical analysis — E. Boersma; Data-base man- outcome in patients with an aortic aneurysm and peripheral vascular dis- agement — V.C. Poldermans and A. Schyns; Adverse-events committee — P.
ease. J Am Coll Cardiol 1993;21:957-63.
Downloaded from www.nejm.org at NSW HEALTH AND HEALTH TECHNOLOGY on January 14, 2009 . Copyright 1999 Massachusetts Medical Society. All rights reserved.

Source: http://dbh.wikispaces.com/file/view/Poldermans.pdf