Pharmacotherapy for Patients With Atrial Fibrillation and Cerebral Microbleeds

Background: Patients with cerebral microbleeds have increased risk of intracranial hemorrhage and ischemic stroke. No trial specifically informs antithrombotic therapy for patients with cerebral microbleeds and atrial fibrillation. We investigated the safety of anticoagulation versus no anticoagulation with regard to cerebrovascular outcomes and mortality. Methods: All consecutive atrial fibrillation patients from 2015 to 2018 with MRI evidence of ≥1 cerebral microbleed at time of imaging were reviewed. Patients were treated with warfarin, direct oral anticoagulants, or neither. Primary outcome was all-cause mortality informed by National Death Registry and the composite of ischemic and hemorrhagic stroke. All statistical tests were 2-sided and significant at P < .05. Results: The median interval from patient identification until the end of electronic health record surveillance was 9.93 months (interquartile range, 2.83-19.17 months). We identified 308 atrial fibrillation patients with cerebral microbleeds; 128(41.6%) were on warfarin, 88(28.6%) on direct oral anticoagulants, and 92(29.9%) on neither. Over the surveillance interval, 87 deaths, 51 ischemic strokes, and 14 hemorrhagic strokes occurred. The estimated likelihoods of the composite stroke outcome and ischemic stroke only did not differ significantly among the 3 groups. However, patients taking direct oral anticoagulants had a significantly smaller likelihood of all-cause mortality than patients who were not anticoagulated (adjusted hazard ratio: .44[.23,.83], P=.012). Conclusions: In patients with coprevalent atrial fibrillation and cerebral microbleeds, we did not detect differences in subsequent ischemic stroke, hemorrhagic stroke, or both, comparing warfarin, direct oral anticoagulants, or neither. Patients treated with direct oral anticoagulants had better survival than nonanticoagulated patients.