Clinical Outcomes of Anticoagulation for Stroke Prevention in Atrial Fibrillation: A 10 year cohort study.
From January 1990 through July 2000 a cohort of 373 patients without prior thromboembolic events were treated with warfarin for nonvalvular or valvular atrial fibrillation. Patients were managed in an anticoagulation clinic using point-of-care prothrombin time monitoring. A computerized database was used to collect information on warfarin dose, INR results and clinical events. There were 6577 prothrombin time measurements with INR results as shown in the table and 52.1% of measured INRs were within the recommended range of 2.0–3.0. During 424 patient-years there was one stroke and three TIAs. The incidence of thromboembolism was 0.9% per year. Eleven patients sustained major bleeding events: gastrointestinal (8); retroperitoneal (1); subdural hematoma (1) and cebrebrospinal (1). The incidence of major bleeding was 2.6% per year. The lowest combined percentage of both major bleeding and thromboembolism occurred with INR values of 2.0–3.9. Long-term warfarin therapy with unselected outpatients in an anticoagulation clinic produced stroke and bleeding rates comparable to pooled data from controlled studies.