Abstract 152: Predictors of Intracranial Hemorrhage Among Anticoagulated Patients with Atrial Fibrillation: Insights from the Rivaroxaban Once-daily oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET AF)

Abstract

Background: For patients with atrial fibrillation (AF) at moderate to high risk of stroke, oral anticoagulation reduces the risk of ischemic stroke but may increase hemorrhagic stroke and offset the intended benefits in certain individuals. Identifying individuals with AF at high risk of stroke who are likely to have an intracranial hemorrhage (ICH) if treated with anticoagulants remains a challenge. Methods: We investigated factors associated with ICH in 14,264 patients with nonvalvular AF randomized to rivaroxaban or dose-adjusted warfarin in ROCKET AF. The endpoint of interest was ICH in the entire intention-to-treat population. Cox proportional hazards modeling was used to identify factors associated with ICH. Results: Over a median follow-up of 1.94 years, 136 patients experienced ICH events (intracerebral hemorrhage [n=98], subarachnoid hemorrhage [n=5], subdural hemorrhage [n=32], extradural hemorrhage [n=1]). The average annual rate of ICH was 0.55 per 100 patient-years. The significant independent predictors of increased risk for ICH were increased age, history of prior stroke or TIA, black or Asian race, decreasing serum albumin, and decreased platelet count below 210 x 10 9 /l (Table, C-index=0.677). Creatinine clearance was not associated with the occurrence of ICH after accounting for other variables in the model (P=0.3181). Aspirin and thienopyridine use at baseline were associated with an increased risk of ICH whereas randomization to rivaroxaban (versus warfarin) was protective. Conclusions: Among patients with AF at moderate to high risk of stroke who were treated with anticoagulation, the average annual rate of ICH was 0.55 per 100 patient-years. Risk of ICH was increased among those who were non-white, older, had prior stroke or TIA, had low platelet counts, low serum albumin, and were taking antiplatelet therapy. Rivaroxaban was associated with a significantly lower risk of ICH compared with warfarin. The external validity of these findings requires testing in other AF populations.