Bleeding Complications Associated with Warfarin Treatment in Ischemic Stroke Patients with Atrial Fibrillation: A Population-Based Cohort Study

BackgroundBleeding events are the major obstacle to the widespread use of warfarin for secondary stroke prevention. Previous studies have not examined the use of risk stratification scores to estimate lifetime bleeding risk associated with warfarin treatment in a population-based setting. The purpose of this study is to determine the lifetime risk of bleeding events in ischemic stroke patients with atrial fibrillation (AF) undergoing warfarin treatment in a population-based cohort and to evaluate the use of bleeding risk scores to identify patients at high risk for lifetime bleeding events.MethodsThe resources of the Rochester Epidemiology Project Medical Linkage System were used to identify acute ischemic stroke patients with AF undergoing warfarin treatment for secondary stroke prevention from 1980 to 1994. Medical information for patients seen at Mayo Clinic and at Olmsted Medical Center was used to retrospectively risk-stratify stroke patients according to bleeding risk scores (including the HAS-BLED and HEMORR2HAGES scores) before warfarin initiation. These scores were reassessed 1 and 5 years later and compared with lifetime bleeding events.ResultsOne hundred patients (mean age, 79.3 years; 68% women) were studied. Ninety-nine patients were observed until death. Major bleeding events occurred in 41 patients at a median of 19 months after warfarin initiation. Patients with a history of hemorrhage before warfarin treatment were more likely to develop major hemorrhage (15% versus 3%, P = .04). Patients with baseline HAS-BLED scores of 2 or more had a higher lifetime risk of major bleeding events compared with those with scores of 1 or less (53% versus 7%, P < .01), whereas those with HEMORR2HAGES scores of 2 or more had a higher lifetime risk of major bleeding events compared with those with scores of 1 or less (52% versus 16%, P = .03). Patients with an increase in the HAS-BLED and HEMORR2HAGES scores during follow-up had a higher remaining lifetime risk of major bleeding events compared with those with no change.ConclusionsOur findings indicate high lifetime bleeding risk associated with warfarin treatment for patients with ischemic stroke. Risk stratification scores are useful to identify patients at high risk of developing bleeding complications and should be recalculated at regular intervals to evaluate the bleeding risk in anticoagulated patients with ischemic stroke.