Structured abstractQuestionIn patients undergoing major non-cardiac surgery, what is the efficacy and harm of pharmacological interventions in preventing perioperative myocardial ischemia, nonfatal myocardial infarction (MI), or cardiac mortality?Data sourcesStudies were identified by computerized searches (MEDLINE, EMBASE, Cochrane Library) up to November 2002 and citation review. No language restrictions were applied.Study selectionStudies were selected if they were randomized controlled trials (RCTs) comparing a cardioprotective drug to placebo or another cardioprotective drug in patients undergoing major non-cardiac surgery, followed patients over the entire intraoperative period or longer, enrolled at least ten patients per group, initiated the drug regimen no later than 24 hr after surgery, and did not focus on cardiovascular responses during the induction of anesthesia.Data extractionData were extracted in triplicate on trial design, surgical setting, patient characteristics, drug regimens, length of postoperative follow-up, prevalence of previous MI, and outcomes including adverse events. The main outcomes were intraoperative myocardial ischemia, postoperative myocardial ischemia, perioperative nonfatal MI, and cardiac mortality.Main resultsTwenty-one trials with a total of 3,646 patients met the selection criteria. Studied drugs included β-blockers (11 RCTs, 866 patients), α2-agonists (six RCTs, 2,614 patients), calcium channel blockers (three RCTs, 121 patients), and nitrates (one RCT, 45 patients). All compared a single cardioprotective drug with placebo. Meta-analyses were only performed for drugs tested in at least two trials. β-blockers and α2-agonists reduced intraoperative ischemia; β-blockers also reduced postoperative ischemia (Table). β-blockers reduced the risk of nonfatal MI (odds ratio [OR] 0.19; 95% confidence interval [CI] 0.08–0.48; Table). This effect was not significant when the two trials with extremely high control event rates were excluded. β-blockers and α2-agonists reduced the risk of cardiac mortality (Table). Again, this effect was not significant for β-blockers when the single trial with extremely high baseline risk was excluded. β-blockers and calcium channel blockers increased the risk of bradycardia compared to placebo (Table).ConclusionsCompared to placebo, β-blockers and α2-agonists reduce the risk of perioperative cardiac events in patients undergoing major non-cardiac surgery; the magnitude of risk reduction with β-blockers appears to be dependent on the baseline risk.FundingProsper Grant No. 3233-051939.97/2, Swiss National Research Foundation.