Abstract Background The safety and efficacy of triple therapy (TT; warfarin with dual antiplatelet therapy DAPT) in post–percutaneous coronary intervention (PCI) patients with atrial fibrillation (AF) are unclear. We aimed to determine whether TT is associated with a decreased stroke rate and an acceptable bleeding rate in this population. Methods This was a single-centre, retrospective study. Primary composite outcome was death, ischemic stroke, or transient ischemic attack. Secondary outcomes included components of primary outcome, bleeding, and blood transfusion rates. Results Of 602 post-PCI patients with AF between 2000 and 2009, 382 received TT, 220 DAPT. Mean follow-up post PCI was 5.9 ± 5.0 months. The TT group had a higher CHADS2 score (2.6 vs 2.1, P < 0.001), older age (72.9 vs 70.5 years, P = 0.039), more heart failure (72.3% vs 36.9%, P = 0.010), and more strokes (14.4% vs 6.4%, P = 0.010). Neither primary outcome, major bleeding, nor blood transfusion rates differed between treatment groups, but more gastrointestinal bleeding occurred with TT use (2.6% vs 0.5%, P = 0.045). Net clinical benefit was −5.2 (CHADS2 ≤ 2), 0.9 (CHADS2 > 2), and −3.2 (overall) per 100 patient-years. Conclusions Although we found no association with TT usage and a reduction in cerebrovascular ischemic or major bleeding events in post-PCI patients with AF regardless of CHADS2 score vs DAPT, the study was likely underpowered to demonstrate a clinically relevant reduction. TT was associated with a 5-fold increase in gastrointestinal bleeding vs DAPT. Net clinical benefit calculations suggest benefits of TT in patients with CHADS2 > 2. Stratification with CHADS2 might be useful to determine the optimal antithrombotic therapy post PCI.