A double-blind crossover trial comparing diltiazem (360 mg/day) and nifedipine (120 mg/day) for treatment of stable angina was conducted in 21 of 27 patients with proven coronary artery disease who completed the trial. All patients started with a 2 week placebo period followed by a random assignment to either drug treatment for 3 weeks and subsequent crossover to the other treatment. The two drug treatment periods were separated by a 1 week placebo washout phase and the study was completed with a 1 week placebo phase. There were no significant differences between patients' responses to diltiazem and nifedipine in relation to time to onset of angina, ST depression responses to exercise, heart rate or systolic or diastolic blood pressure. A total of 37 adverse effects were reported with nifedipine compared with 9 with diltiazem in the 22 patients in whom drug safety was analyzed. Additionally, two patients treated with nifedipine were withdrawn from study participation before crossover. There was a significant (p less than 0.05) difference with respect to incidence of edema (7 of 22 patients taking nifedipine, 1 of 22 taking diltiazem) and dizziness (7 of 22 patients taking nifedipine, 0 of 22 taking diltiazem). The most frequent adverse effect reported with diltiazem was rash (3 of 22 patients). Severe adverse effects were reported in four patients: in one with diltiazem (rash) and in three with nifedipine (palpitation in two and headache in one). A reduction in prescribed dosage was required in 37% of nifedipine-treated compared with 6% of diltiazem-treated patients. Efficacy measures were significantly improved above placebo levels by both diltiazem and nifedipine.(ABSTRACT TRUNCATED AT 250 WORDS)