Aspirin, alone or in combination with dipyridamole, is known to prevent occlusion of aortocoronary vein grafts. The benefit of dipyridamole in addition to aspirin remains controversial, and the efficacy and safety of oral anticoagulants for prevention of vein-graft occlusion have not been established. We assessed one-year angiographic vein-graft patency after aortocoronary-bypass surgery in 948 patients assigned to receive aspirin, aspirin plus dipyridamole, or oral anticoagulants in a prospective, randomised trial. The design was double-blind and placebo-controlled for the aspirin groups, but open for oral anticoagulant treatment. Dipyridamole (5 mg/kg per 24 h intravenously for 28 h, followed by 200 mg twice daily) and oral anticoagulants (desired prothrombin time range 2.8-4.8 international normalised ratio) were started before surgery, and aspirin (50 mg per day) was started after surgery. Clinical outcome was assessed by the incidence of myocardial infarction, thrombosis, major bleeding, or death. Occlusion rate of distal anastomoses was 11% in the aspirin plus dipyridamole group versus 15% in the aspirin group (relative risk 0.76, 95% CI 0.54-1.05) and 13% in the oral anticoagulants group. Clinical events occurred in 20.3% of patients receiving aspirin plus dipyridamole compared with 13.9% of the aspirin group (relative risk 1.46, 95% CI 1.02-2.08) and 16.9% of the oral anticoagulants group. Our data provide no convincing evidence that addition of dipyridamole to 50 mg aspirin per day improves aortocoronary vein-graft patency. Moreover, there is evidence that the combination increases the overall clinical-event rate. Compared with aspirin, oral anticoagulants provided no benefit.