Beta Blockers for High-Risk Vascular Surgery

Source: Poldermans D, et al. N Engl J Med 1999;341:1789-1794.

Although earlier studies have shown a reduction in perioperative cardiac events with prophylactic beta-blocker use, they lacked power to assess such serious events as perioperative death or nonfatal myocardial infarction (MI). Thus, Poldermans and colleagues randomized 112 major vascular surgery patients with positive dobutamine stress echocardiograms (DSE) at multiple centers to one week of bisoprolol (5-10 mg/d) preoperation (pre-op) followed by 30 days post-operation (post-op) vs. usual care. The primary end points were death from cardiac causes or MI for 30 days post-op. Bisoprolol resulted in lower heart rates, but other patient characteristics were similar in the two groups. Perioperative mortality was less in the bisoprolol group (3% vs 17%; P = 0.02), as was nonfatal MI (0 vs 17%; P < 0.001). All but one of the nine MIs occurred post-operatively; the one pre-op MI resulted in cancellation of the surgery. The combined end point was also reduced (3% vs 34%; P < 0.001). The relative risk of the combined end point on bisoprolol was 0.09 (95% CI 0.02-0.37). Because of these results, the trial was stopped prematurely. Among 53 patients who were excluded from the study because they were already on beta blockers, 7.5 died and there were no MIs. Among eight patients excluded because of extensive wall motion abnormalities on DSE (high risk), four underwent bypass surgery and two of these died, but the two survivors later had successful vascular surgery. The remaining four underwent vascular surgery on beta blockers; one had an MI, but none died. Poldermans et al conclude that bisoprolol reduces perioperative death and MI in high-risk patients undergoing major vascular surgery.

Comment by Michael H. Crawford, MD

Many have resisted the move to liberally use beta blockers for major noncardiac surgery in patients with risk factors for or clinical suspicion of coronary artery disease in favor of preoperative testing to select the appropriate candidate for revascularization or beta blockers. Why the reluctance? The mechanism of action of beta blockers for this purpose is not known and if the myocardium lacks adequate blood flow, it seems more reasonable to return blood flow, especially if it can be accomplished percutaneously. Most agree coronary bypass surgery should be reserved for those who need it anyway. Also, the previous large randomized trial of atenolol prophylactically by Mangano and associates was not very impressive.1 Perioperative deaths and MI were not reduced, but long-term mortality was. However, Mangano et al studied a lower risk group undergoing various surgeries and their study was underpowered for serious perioperative events. The 10-fold decrease in perioperative death or MI in this study (34% vs 3.4%; P < 0.001), even though it was interrupted when half the planned patients were enrolled, is impressive. On the other hand, these were truly high-risk patients with clinical risk factors, such as age older than 70 years, prior MI, heart failure history, undertreatment for ventricular arrhythmias, diabetes, limited exercise ability, and a positive DSE. The latter suggests that they have flow limiting coronary lesions that perhaps could have benefited from a percutaneous intervention (PCI).

Because of the lack of evidence that PCI improves perioperative outcomes in randomized trials and the cost of such an approach, Lee, in an accompanying editorial, suggests that we should de-emphasize testing and revascularization and increase beta blocker use in appropriate patients.2 The question is how liberal to be? Clearly not everyone should receive beta blockers, but who are the appropriate candidates? The ACC/AHA guidelines3 suggest that intermediate-risk patients undergoing major surgery should have noninvasive testing, but Lee believes that, based upon this new evidence, this should be modified to read beta-blocker therapy.

There are other issues as well. What about the lower risk patient (i.e., 1 or 2 risk factors and a negative DSE)? Poldermans et al would not treat such a patient because of the high negative predictive value of DSE. Is one week of therapy enough or necessary? Must selective beta blockers be used like bisoprolol or will nonselective agents work? Also, there are limitations to this study such as the lack of blinding. Of course, it is almost impossible to blind a study using beta blockers because of the heart rate response to the drug. In addition, Poldermans et al eliminated very high-risk patients and the intermediate and lower risk patients. Thus, these data only strictly refer to high-risk patients. However, it will be hard to argue that a perioperative major cardiac event rate of 3.4% on beta blockers can be bested by any other therapy or management approach. Finally, would dipyridamole thallium stress testing provide the same predictive power as DSE for the application of beta blockers?


1. Mangano DT, et al. N Engl J Med 1996;335:1713-1720.

2. Lee TH. N Engl J Med 1999;341:1838-1840.

3. Eagle KA, et al. J Am Coll Cardiol 1996;27:910-948.