Bleeding Complications to Long-Term Oral Anticoagulant Therapy

Objective: To evaluate the incidence of bleeding complications in recent randomized trials on oral anticoagetlant treatment for prevention of arterial thromboembolism. Data sources: International publications on studies of prevention. of arterial thromboembolism by oral anticoagulant therapy. Study selection and data extraction: Randomized trials an oral anticoagulant therapy in patients with atrial fibrillation, recent myocardial infarction, and prosthetic heart valves were selected. For comparison older nonrandomized studies were studied. Background: Oral anticoagulant drugs are recommended for primary prevention of thromboembolic events in patients with chronic atrial fibrillation, recent myocardial infarction, and prosthetic heart valves. Still many physicians hesitate to prescribe anticoagulant drugs, presumably for fear of bleeding complications. Results: In six recent trials of warfarin in patients with atrial fibrillation, the highest annual incidence of fatal and major bleeding was 0.8% and 2.0%, respectively. In patients treated with warfarin after a recent myocardial infarction, the incidence of fatal and major bleeding was 0.2% and 0.5% per year, respectively. The annual incidence of fatal and major bleeding in patients with prosthetic heart valves on warfarin treatment was found to be 1.4% and 5.2%, respectively. The mean incidence of fatal and major bleeding in patients on warfarin in these eight trials was 0.5% and 1.7% per year, respectively. The mean incidence of fatal and major bleeds in patients on placebo was 0.1% and 0.7% per year, respectively. In three randomized trials evaluating aspirin versus warfarin, the respective mean incidences of fatal and major bleeding during aspirin treatment were 0.2% and 0.8% per year. A remarkable decrease in the incidence of major bleeding complications to oral anticoagulant therapy is revealed by these trials as compared to previous studies. Reasons for this decline may be less intensive anticoagulant regimes, better control of anticoagulant therapy due to the introduction of the international normalized ratio, and careful pretreatment evaluation of risk factors for bleeding. In alI prospective trials of oral anticoagulation, the risk of bleeding was more than over-weighed by the beneficial effect on the incidence of stroke and peripheral thromboemboli.