A total of 243 patients who had reversible ischemic attacks (RIA) were submitted to clinical trial to determine whether dipyridamole (400 mg/day) (D) or aspirin (100 mg/48 hours) + dipyridamole (300 mg/day) (ASA + D) would produce significant reduction in the subsequent occurrence of RIA and completed stroke. One hundred and fifteen were selected for Group ASA + D and 71 were treated with dipyridamole only. The treatment groups were similar in relation to age, sex, risk factors, duration and presumed vascular territory of RIA, incidence of alterations of arterial supra-aortic trunks, cerebral infarct (CT scan), and platelet function. Patients were followed for a mean time of 21 months. At the end of the study, 21.7% of the ASA + D group and 19.7% in the D group had suffered new episodes of RIA or completed stroke (p = 0.88). Frequency of stroke (reversible ischemic neurologic deficit or completed stroke) was 7.8% in the ASA + D patients and 9.8% in the D patients (p = 0.83). Subgroup analysis did not show significant differences either. It is concluded that ASA + D has no significantly greater beneficial effect than that observed with D alone in the secondary prevention of atherothrombotic cerebral ischemia. However, a statistical Type II error cannot be excluded by the reduced number of recurrences.