In a double-blind parallel group study, 46 patients with chronic stable angina were randomized, after a 2-week placebo washout period, to 1 of 3 treatment groups for an additional 2 weeks. Groups 1 and 2 received nicorandil (5 mg, n = 5; 10 mg, n = 10) twice daily, respectively, increasing to 10 and 20 mg (n = 20) twice daily after 1 week of treatment; group 3 continued to receive placebo. A symptom-limited Bruce protocol exercise test was performed before and 2 hours after the initial dose and, after 2 weeks of treatment, 2 and 12 hours after administration. The following parameters were measured: resting, peak exercise and recovery blood pressure and heart rate, exercise duration, time to onset of angina and time to 1 mm of ST-segment depression. After initial dosing, there were significant increases in exercise duration (16%--n = 5, n = 10 vs -2% [placebo]) and time to onset of angina (20%, n = 5; 26%, n = 10 vs 5% [placebo]) (p less than 0.05). Time to onset of 1 mm of ST-segment depression increased in the nicorandil-treated groups compared with that in the placebo group (27%, n = 5; 25%, n = 10 vs 8% [placebo]). Calculated total exercise work increased in both nicorandil groups compared with exercise work in the placebo group (30%, n = 5; 19%, n = 10 vs 3% [placebo]). A decrease in resting systolic blood pressure (12%) in the 10-mg group was the only significant alteration in the hemodynamic parameters.(ABSTRACT TRUNCATED AT 250 WORDS)