In 1996, a randomized trial of 200 patients undergoing major noncardiac surgery with known coronary artery disease (CAD) or multiple risk factors for it found that those who received atenolol beginning immediately before surgery and while hospitalized had reduced mortality at 2 years [1]. However, there were concerns regarding the randomization, lack of intention-to-treat analysis, and the absence of effect in the immediate postoperative period. The Dutch Echocardiographic Cardiac Risk Evaluation Applying Stress Echocardiography (DECREASE) study group performed a small, unblinded study in 1999 in which perioperative treatment with bisoprolol was compared to usual care in patients with a positive dobutamine stress echocardiogram undergoing major vascular surgery [2]. Dramatic reductions in both postoperative cardiac death and nonfatal myocardial infarction were reported. These two trials led to numerous additional studies of perioperative beta-blockade in a variety of surgical populations, most of which showed no benefit. The DECREASE group published several additional influential perioperative beta-blocker trials. However, after extensive investigation, this family of trials has been discredited due to major flaws including the fabrication of data [3–6].

In 2008, the POISE trial results were published. More than 8,000 beta-blocker naïve patients from 190 hospitals and 23 countries with atherosclerotic vascular disease, or multiple risk factors for it, undergoing major noncardiac surgery were randomly assigned to a regimen of high-dose oral and/or intravenous metoprolol started immediately before and continued after surgery [7]. The risk of cardiovascular death, nonfatal MI, or nonfatal cardiac arrest was decreased in the metoprolol succinate-treated group. However, this benefit was offset by an increased risk of stroke and all-cause mortality possibly due to beta-blocker-induced bradycardia and hypotension. A recent meta-analysis including only “secure” trials (i.e., excluding the DECREASE trials), and dominated by the POISE trial, reported similar findings [6]. Although the dose (100 mg of metoprolol succinate) and the timing (initiated 2–4 h before surgery) of beta-blocker used in the POISE trial may have contributed to the adverse outcomes, the available evidence from the remaining available “secure” trials does not allow identification of specific patients who may benefit or beta-blocker regimens that may be beneficial.

Given concerns about the safety of perioperative beta-blockade, a large, propensity-matched, retrospective cohort study using Veterans Health Administration databases was recently conducted [8]. In patients treated with beta-blockers undergoing noncardiac, nonvascular surgery, overall mortality and cardiac morbidity were reduced without increased risk of stroke. Most patients were receiving long-term beta-blocker therapy, but similar outcomes were also observed in those in whom the beta-blocker was initiated within 30 days of surgery; there was insufficient data to assess the outcomes of patients in whom a beta-blocker was started within 7 days of surgery.