Patients with coronary artery disease exhibit a reduced coronary vasodilator reserve in response to exercise testing. Drugs which block coronary beta adrenergic receptors could exacerbate this abnormality leaving the vasoconstrictor alpha tone unopposed and/or counteracting the beta 2-mediated vasodilation elicited by the increase in myocardial oxygen demand. To test this hypothesis we administered propranolol 40 mg qid and placebo, using a cross over randomized single blind protocol, to 14 patients each with effort angina and critical coronary stenosis (greater than or equal to 75%). We performed computer-assisted multistage bicycle ergometer testings (25 W increments at 2 min intervals) after 2 weeks open label placebo (control) and at 2 week intervals following daily administration of propranolol and placebo. Compared to placebo, propranolol reduced significantly (p less than 0.001) peak heart rate (x +/- SD: 114 +/- 6 vs 150 +/- 11 beats/min) and rate pressure product (20.1 +/- 2.1 vs 28.0 +/- 3.9 X 10(-3)) and increased exercise duration (462 +/- 91 vs 355 +/- 85 sec). Conversely 0.1 mV ST segment depression was observed at lower heart rate (106 +/- 9 vs 127 +/- 8 beats/min, p less than 0.001) and rate pressure product (16.9 +/- 3.6 vs 22.4 +/- 2.4 X 10(-3), p less than 0.001). No significant differences were found between placebo and control. Moreover, we assessed the regression lines of the relationship between ST segment depression (ST) and heart rate (HR) during exercise. These have been shown to be shifted to the right after surgical revascularization and are an indirect measure of coronary reserve.(ABSTRACT TRUNCATED AT 250 WORDS)