To evaluate the role of antithrombotic therapy, on preserving graft patency, we performed a meta-analysis of randomized clinical trials involving aspirin (ASA), dipyridamole (D), anticoagulants (AC) and placebo or nontreatment controls (P). Manual literature searches were performed supplemented by computerized MEDLINE listings complete to July 1991. Saphenous vein graft occlusion was determined by angiography (patients with > or = 1 distal anastomotic occlusion). The trial data were aggregated with the methods of Mantel and Haenszel. The results are reported as odds ratios (OR) +/- 95% confidence intervals (CI). Seventeen trials were evaluated. Aspirin strongly influenced graft occlusion [ASA +/- D vs P: OR 0.60, 95% CI 0.51, 0.71, P < 0.0001], but dipyridamole provided no additional benefit [ASA+D vs ASA: OR 0.94, 95% CI 0.72, 1.24, P = 0.71]. Anticoagulants reduced graft occlusion [AC vs P: OR 0.56, 95% CI 0.33, 0.93, P = 0.025] and the results were similar to that achieved with aspirin [ASA vs AC: OR 0.95, 95% CI 0.62, 1.44, P = 0.87]. The combination of aspirin and anticoagulants was superior to anticoagulants alone in two limited trials [ASA+AC vs AC: OR 0.55, 95% CI 0.33, 0.88, P = 0.01]. A low (100 mg) to medium (325 mg) daily aspirin dosage was more effective than a high dose (975 mg). Early postoperative treatment (< or = 6 h) strongly influenced graft occlusion while preoperative administration provided no additional benefit. No mortality advantage was identified for any antithrombotic therapy. Aspirin or anticoagulants enhance saphenous vein graft patency following aortocoronary bypass surgery, and a combination thereof deserves further investigation in a trial large enough to detect the effects of these treatments with respect to clinical events.