Association of international normalized ratio stability and bleeding outcomes among atrial fibrillation patients undergoing percutaneous coronary intervention: Insights from the PIONEER AF-PCI trial

Background: Among atrial fibrillation patients undergoing percutaneous coronary intervention enrolled in PIONEER AF-PCI (An Open-Label, Randomized, Controlled, Multicenter Study Exploring Two Treatment Strategies of Rivaroxaban and a Dose-Adjusted Oral Vitamin K Antagonist Treatment Strategy in Subjects With Atrial Fibrillation Who Undergo Percutaneous Coronary Intervention), it is unclear if the observed reduction in bleeding events with rivaroxaban regimens is consistent across a range of the international normalized ratio (INR) among subjects administrated Vitamin K antagonist (VKA)-triple therapy. This analysis compares the occurrence of clinically significant bleeding between rivaroxaban and VKA strategies, according to INR stability of subjects administrated VKA. Methods and Results: A total of 2124 atrial fibrillation patients undergoing percutaneous coronary intervention were randomized to 3 groups: Rivaroxaban 15 mg od plus a P2Y₁₂ inhibitor (group 1, n=709); rivaroxaban 2.5 mg bid plus dual antiplatelet therapy (group 2, n=709); and warfarin plus dual antiplatelet therapy (group 3, n=706). Subjects assigned to the VKA group were stratified according to time in therapeutic range and time spent with an INR >3. Kaplan-Meier estimates were calculated for clinically significant bleeding through 1 year and hazard ratios were derived using Cox Proportional Hazards models. Among group 3, 93.4% of the participants had a time in therapeutic range available (mean time in therapeutic range=65.0±24.8%). Both groups 1 and 2 were associated with a reduction in clinically significant bleeding compared with subjects in group 3, regardless of the time in therapeutic range (hazard ratio ranges=0.53-0.71 and 0.57-0.76; respectively, P<0.05 for all). Rivaroxaban strategies were associated with a reduction in clinically significant bleeding compared with VKA regardless of the proportion of time spent with an INR >3 (hazard ratio ranges=0.59-0.67 and 0.42-0.69; P<0.05 for all). Conclusions: Among atrial fibrillation patients undergoing percutaneous coronary intervention, rivaroxaban-based therapy was superior to warfarin plus dual antiplatelet therapy in lowering bleeding outcomes regardless of the INR stability.