BACKGROUND Successful coronary thrombolysis involves a risk for reocclusion that cannot be prevented by invasive strategies. Therefore, we studied the effects of three antithrombotic regimens on the angiographic and clinical courses after successful thrombolysis. RESULTS Patients treated with intravenous thrombolytic therapy followed by intravenous heparin were eligible when a patent infarct-related artery was demonstrated at angiography < 48 hours. Three hundred patients were randomized to either 325 mg aspirin daily or placebo with discontinuation of heparin or to Coumadin with continuation of heparin until oral anticoagulation was established (international normalized ratio, 2.8-4.0). After 3 months, in which conservative treatment was intended, vessel patency and ventricular function were reassessed in 248 patients. Reocclusion rates were not significantly different: 25% (23 of 93) with aspirin, 30% (24 of 81) with Coumadin, and 32% (24 of 74) with placebo. Reinfarction was seen in 3% of patients on aspirin, in 8% on Coumadin, and in 11% on placebo (aspirin versus placebo, p < 0.025; other comparison, p = NS). Revascularization rate was 6% with aspirin, 13% with Coumadin, and 16% with placebo (aspirin versus placebo, p < 0.05; other comparisons, p = NS). Mortality was 2% and did not differ between groups. An event-free clinical course was seen in 93% with aspirin, in 82% with Coumadin, and in 76% with placebo (aspirin versus placebo, p < 0.001; aspirin versus Coumadin, p < 0.05). An event-free course without reocclusion was observed in 73% with aspirin, in 63% with Coumadin, and in 59% with placebo (p = NS). An increase of left ventricular ejection fraction was only found in the aspirin group (4.6%, p < 0.001). CONCLUSIONS At 3 months after successful thrombolysis, reocclusion occurred in about 30% of patients, regardless of the use of antithrombotics. Compared with placebo, aspirin significantly reduces reinfarction rate and revascularization rate, improves event-free survival, and better preserves left ventricular function. The efficacy of Coumadin on these end points appears less than that of aspirin. The still-high reocclusion rate emphasizes the need for better antithrombotic therapy in these patients.