BackgroundThe optimal antithrombotic therapy for atrial fibrillation (AF) patients undergoing coronary stenting is unknown. The present meta-analysis sought to investigate the efficacy and safety of triple therapy (TT; warfarin, clopidogrel and aspirin) vs dual antiplatelet therapy (DAPT; clopidogrel plus aspirin) in those patients.MethodsPubMed and Cochrane Library were searched for studies enrolling AF patients undergoing coronary stenting on TT and DAPT up to September 2016, and fourteen studies were included. Efficacy outcomes included ischemic stroke, stent thrombosis, major adverse cardiovascular event (MACE), all-cause mortality and myocardial infarction (MI); safety outcome was major bleeding. We conducted meta-analysis and used odds ratio (OR) with 95% confidence intervals (CI) to compare TT and DAPT. Meta-regression, sensitivity and subgroup analysis were taken to investigate the source of heterogeneity in the outcome of major bleeding.Results14 eligible observational studies with 11,697 subjects were identified. Compared with DAPT, TT had decreased the risk of ischemic stroke [OR = 0.74, 95% CI (0.59, 0.93), P = 0.009] and stent thrombosis [OR = 0.40, 95% CI (0.18, 0.93), P = 0.033]. While, there was an increased risk of major bleeding [OR = 1.55, 95% CI (1.16, 2.09), P = 0.004] associated with TT. The risk of MACE, all-cause mortality and MI had no significant statistical difference between TT and DAPT. Furthermore, the results of univariate and multivariate meta-regression analysis implicated that there were no obvious correlations between certain baseline characteristics (age, gender, race, hypertension, study design) and risk of major bleeding. Also of major bleeding, the findings of sensitivity analysis were generally robust, and a prespecified subgroup analysis of race demonstrated that the source of heterogeneity might attribute to Asian studies mostly.ConclusionsTT reduced the risk of ischemic stroke and stent thrombosis with an acceptable major bleeding risk compared with DAPT, and TT was considered as a valid alternative in AF patients undergoing coronary stenting. Further prospective randomized trials are needed to ensure the reliability of these data and find the optimal therapeutic strategy in this setting of patients.