BACKGROUND AND PURPOSE: There is growing enthusiasm for coronary artery bypass grafting (CABG) without cardiopulmonary bypass (CPB). Although deleterious effects of CPB are known, it remains to be proven that avoiding CPB will result in reduction in morbidity. We sought to determine whether off-pump surgery is associated with reduced occurrence of adverse outcomes as compared with CABG with CPB. METHODS: Studies were identified by searching the MEDLINE, EMBASE and the Cochrane Register 1980 to 2006 (February). We also searched the reference lists of randomized clinical trials (RCT) and reviews to look for additional studies. STUDY SELECTION: RCTs comparing off-pump surgery to CABG with CPB. No restriction applied on the size of the trial or end point reports. DATA EXTRACTION: 2 reviewers independently searched for studies, read abstracts and abstracted all data. DATA SYNTHESIS: combined estimates were obtained using fixed or random effect meta-analyses. Relative risks and risk differences were calculated. Heterogeneity was assessed using chi(2) and I(2) values. RESULTS: There were 3996 patients enrolled in 41 RCTs (mean age 62, 22% female). No study reported information on race. Off-pump CABG was associated with a 50% reduction in the relative risk of stroke (95% CI, 7% to 73%), 30% reduction in atrial fibrillation (AF; 95% CI, 16% to 43%) and 48% reduction in wound infection (95% CI, 26% to 63%) with no heterogeneity among RCTs. This translated into avoidance of 10 strokes, 80 cases of AF and 40 infections per 1000 CABG. Fewer distal grafts were performed and there was evidence for >10 reinterventions per 1000 with off-pump CABG. Long-term follow-up is not yet reported in the trials. CONCLUSIONS: Off-pump CABG is associated with reduced risk of stroke, AF and infections as compared with CABG with CPB. Evidence should be generalized taking into account RCT enrollment limitations, drawbacks related to training requirements, propensity to perform fewer grafts and likely reinterventions after off-pump surgery.