We conducted a multicenter, double-blind, placebo-controlled randomized trial of aspirin treatment (324 mg in buffered solution daily) for 12 weeks in 1266 men with unstable angina (625 taking aspirin and 641 placebo). The principal end points were death and acute myocardial infarction diagnosed by the presence of creatine kinase MB or pathologic Q-wave changes on electrocardiograms. The incidence of death or acute myocardial infarction was 51 per cent lower in the aspirin group than in the placebo group: 31 patients (5.0 per cent) as compared with 65 (10.1 per cent); P = 0.0005. Nonfatal acute myocardial infarction was 51 per cent lower in the aspirin group: 21 patients (3.4 per cent) as compared with 44 (6.9 per cent); P = 0.005. The reduction in mortality in the aspirin group was also 51 per cent--10 patients (1.6 per cent) as compared with 21 (3.3 per cent)--although it was not statistically significant; P = 0.054. There was no difference in gastrointestinal symptoms or evidence of blood loss between the treatment and control groups. Our data show that aspirin has a protective effect against acute myocardial infarction in men with unstable angina, and they suggest a similar effect on mortality.