Stroke prevention strategies for cardiac surgery: a systematic review and meta-analysis of randomized controlled trials.

Abstract

Stroke is a much-feared complication of cardiac surgery, but existing literature on preventive strategies is fragmented. Hence, a systematic review and meta-analysis of stroke prevention strategies for cardiac surgery was conducted. An electronic literature search was conducted to retrieve randomized controlled trials (RCTs) investigating perioperative interventions for cardiac surgery, with stroke as an outcome. Random-effects meta-analyses were conducted to generate risk ratios (RRs), 95% confidence intervals (95% CI), and forest plots. Descriptive analysis and synthesis of literature was conducted for interventions not amenable to meta-analysis, focusing on risks of stroke, myocardial infarction and study-defined major adverse cardiovascular events (MACE). Fifty-six RCTs (61 894 patients) were retrieved. Many included trials were underpowered to detect differences in stroke risk. Among pharmacological therapies, only preoperative amiodarone was shown to reduce stroke risk in one trial. Concomitant left atrial appendage closure (LAAC) significantly reduced stroke risk (RR = 0.55, 95% CI = 0.36-0.84, P = 0.006) in patients with preoperative atrial fibrillation, and there was no difference in on-pump versus off-pump coronary artery bypass grafting (CABG) (RR = 0.94, 95% CI = 0.64-1.37, P = 0.735). Much controversy exists in literature on the timing of carotid endarterectomy relative to CABG in patients with severe carotid stenosis. The use of preoperative remote ischemic preconditioning was not found to reduce rates of stroke or MACE. This review presents a comprehensive synthesis of existing interventions for stroke prevention in cardiac surgery, and identifies gaps in research which may benefit from future, large-scale RCTs. LAAC should be considered to reduce stroke incidence in patients with preoperative atrial fibrillation.