Hermann Blessberger, Juergen Kammler, Hans Domanovits, Oliver Schlager, Brigitte Wildner, Danyel Azar, Martin Schillinger, Franz Wiesbauer, Clemens Steinwender

Randomized controlled trials have yielded conflicting results regarding the ability of beta-blockers to influence perioperative cardiovascular morbidity and mortality. Thus routine prescription of these drugs in unselected patients remains a controversial issue. The objective of this review was to systematically analyse the effects of perioperatively administered beta-blockers for prevention of surgery-related mortality and morbidity in patients undergoing any type of surgery while under general anaesthesia. We identified trials by searching the following databases from the date of their inception until June 2013: MEDLINE, Embase , the Cochrane Central Register of Controlled Trials (CENTRAL), Biosis Previews, CAB Abstracts, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Derwent Drug File, Science Citation Index Expanded, Life Sciences Collection, Global Health and PASCAL. In addition, we searched online resources to identify grey literature. We included randomized controlled trials if participants were randomly assigned to a beta-blocker group or a control group (standard care or placebo). Surgery (any type) had to be performed with all or at least a significant proportion of participants under general anaesthesia. Two review authors independently extracted data from all studies. In cases of disagreement, we reassessed the respective studies to reach consensus. We computed summary estimates in the absence of significant clinical heterogeneity. Risk ratios (RRs) were used for dichotomous outcomes, and mean differences (MDs) were used for continuous outcomes. We performed subgroup analyses for various potential effect modifiers. We included 88 randomized controlled trials with 19,161 participants. Six studies (7%) met the highest methodological quality criteria (studies with overall low risk of bias: adequate sequence generation, adequate allocation concealment, double/triple-blinded design with a placebo group, intention-to-treat analysis), whereas in the remaining trials, some form of bias was present or could not be definitively excluded (studies with overall unclear or high risk of bias). Outcomes were evaluated separately for cardiac and non-cardiac surgery.CARDIAC SURGERY (53 trials)We found no clear evidence of an effect of beta-blockers on the following outcomes.• All-cause mortality: RR 0.73, 95% CI 0.35 to 1.52, 3783 participants, moderate quality evidence.• Acute myocardial infarction (AMI): RR 1.04, 95% CI 0.71 to 1.51, 3553 participants, moderate quality evidence.• Myocardial ischaemia: RR 0.51, 95% CI 0.25 to 1.05, 166 participants, low quality evidence.• Cerebrovascular events: RR 1.52, 95% CI 0.58 to 4.02, 1400 participants, low quality evidence.• Hypotension: RR 1.54, 95% CI 0.67 to 3.51, 558 participants, low quality evidence.• Bradycardia: RR 1.61, 95% CI 0.97 to 2.66, 660 participants, low quality evidence.• Congestive heart failure: RR 0.22, 95% CI 0.04 to 1.34, 311 participants, low quality evidence.Beta-blockers significantly reduced the occurrence of the following endpoints.• Ventricular arrhythmias: RR 0.37, 95% CI 0.24 to 0.58, number needed to treat for an additional beneficial outcome (NNTB) 29, 2292 participants, moderate quality evidence.• Supraventricular arrhythmias: RR 0.44, 95% CI 0.36 to 0.53, NNTB five, 6420 participants, high quality evidence.• On average, beta-blockers reduced length of hospital stay by 0.54 days (95% CI -0.90 to -0.19, 2450 participants, low quality evidence).NON-CARDIAC SURGERY (35 trials)Beta-blockers significantly increased the occurrence of the following adverse events.• All-cause mortality: RR 1.25, 95% CI 1.00 to 1.57, 11,413 participants, low quality of evidence, number needed to treat for an additional harmful outcome (NNTH) 167.• Hypotension: RR 1.50, 95% CI 1.38 to 1.64, NNTH 16, 10,947 participants, high quality evidence.• Bradycardia: RR 2.23, 95% CI 1.48 to 3.36, NNTH 21, 11,033 participants, moderate quality evidence.We found a potential increase in the occurrence of the following outcomes with the use of beta-blockers.• Cerebrovascular events: RR 1.59, 95% CI 0.93 to 2.71, 9150 participants, low quality evidence.Whereas no clear evidence of an effect was found when all studies were analysed, restricting the meta-analysis to low risk of bias studies revealed a significant increase in cerebrovascular events with the use of beta-blockers: RR 2.09, 95% CI 1.14 to 3.82, NNTH 265, 8648 participants.Beta-blockers significantly reduced the occurrence of the following endpoints.• AMI: RR 0.73, 95% CI 0.61 to 0.87, NNTB 76, 10,958 participants, high quality evidence.• Myocardial ischaemia: RR 0.51, 95% CI 0.34 to 0.77, NNTB nine, 978 participants, moderate quality evidence.• Supraventricular arrhythmias: RR 0.73, 95% CI 0.57 to 0.94, NNTB 112, 8744 participants, high quality evidence.We found no clear evidence of an effect of beta-blockers on the following outcomes.• Ventricular arrhythmias: RR 0.68, 95% CI 0.31 to 1.49, 476 participants, moderate quality evidence.• Congestive heart failure: RR 1.18, 95% CI 0.94 to 1.48, 9173 participants, moderate quality evidence.• Length of hospital stay: mean difference -0.45 days, 95% CI -1.75 to 0.84, 551 participants, low quality evidence. According to our findings, perioperative application of beta-blockers still plays a pivotal role in cardiac surgery, as they can substantially reduce the high burden of supraventricular and ventricular arrhythmias in the aftermath of surgery. Their influence on mortality, AMI, stroke, congestive heart failure, hypotension and bradycardia in this setting remains unclear.In non-cardiac surgery, evidence shows an association of beta-blockers with increased all-cause mortality. Data from low risk of bias trials further suggests an increase in stroke rate with the use of beta-blockers. As the quality of evidence is still low to moderate, more evidence is needed before a definitive conclusion can be drawn. The substantial reduction in supraventricular arrhythmias and AMI in this setting seems to be offset by the potential increase in mortality and stroke.