A Cochrane review [Abstract] 1 included 8 studies with a total of 9598 subjects. Oral anticoagulant agent (warfarin or acenocumarol) versus antiplatelet agent (aspirin in dosages ranging from 75 to 325 mg/day, aspirin+clopidogrel or triflusal) was tested in non-valvular AF patients, most without prior stroke or TIA (approximately 90%) with mean follow up of 1.9 years. Oral anticoagulants were associated with lower risk of all stroke (OR 0.68, 95% CI 0.54 to 0.85), ischemic stroke (OR 0.53, 95% CI 0.41 to 0.68) and systemic emboli (OR 0.48, 95% CI 0.25 to 0.90). Assuming an estimated annualized rate of stroke of 4% per year on antiplatelet therapy for primary prevention, about 13 strokes per year would be prevented for every 1000 AF patients given oral anticoagulants instead of antiplatelet therapy. All disabling or fatal strokes (OR 0.71, 95% CI 0.59 to 1.04; 7 studies, n=9562,) and myocardial infarction (OR 0.69, 95% CI 0.47 to 1.01; 7 studies, n=8927) were substantially but not significantly reduced by oral anticoagulants. Vascular death (OR 0.93, 95% CI 0.75 to 1.15; 7 studies, n=8915) and all cause mortality (OR 0.99, 95% CI 0.83 to 1.18; 7 studies, n=8927) were similar with these treatments. Intracranial hemorrhages (OR 1.98, 95% CI 1.20 to 3.28) as well as extracranial bleeds (OR 1.90, 95% CI 1.07 to 3.39; 7 studies, n=2880) were increased by oral anticoagulant therapy. The threshold of absolute benefit that warrants anticoagulation remains controversial and depends on patient's preferences and availability of optimal anticoagulation monitoring.
The following decision support rules contain links to this evidence summary:
Primary/Secondary Keywords