A double-blind study of the effects of isosorbide dinitrate, 10 mg given orally four times a day, propranolol, 40 mg four times a day, and the combination of these two drugs was performed on 21 patients with angina pectoris. Each patient received placebo, isosorbide, propranolol, and the combination of the two drugs for 1 month each in a random sequence over 4 months. The number of anginal pains and nitroglycerin tablets used were recorded, and a multistage treadmill ECG exercise test was performed after each treatment period. Frequency of anginal pains was reduced significantly with propranolol (7.0±2.5 pains/week) and the combination of propranolol and isosorbide dinitrate (9.9±2.9) as compared with placebo (21.0±6.4). Similarly, the number of nitroglycerin tablets was reduced with propranolol-containing regimens. Isosorbide was not significantly better than the placebo. Symptomatic improvement with propranolol could be related to a reduction in heart rate, and the product of heart rate and systolic blood pressure during exercise ( P <0.001). The capacity of these patients to perform a multistage exercise test, however, was not improved significantly, and the ischemic ST-segment changes were not altered by the treatments. Thus, propranolol appeared to be effective on the symptoms of angina pectoris, but it did not significantly improve exercise performance, and it did not prevent the ischemic patterns in the exercise ECG in this group of patients. Isosorbide dinitrate, alone or in combination with propranolol, was ineffective in this study.