Episodes of myocardial ischemia in patients with coronary artery disease may be due to transient increases in coronary vasomotor tone superimposed on a fixed atherosclerotic obstruction. The purpose of this study was to determine whether identification of the clinical pattern of angina could predict the therapeutic response to the addition of nifedipine to a regimen of beta blockers and/or long-acting nitrates. Seventy-two patients with stable exertional angina were divided into two groups: "classic exertional angina" (17 patients), defined as exertional angina with a stable threshold; and "mixed angina" (55 patients), defined as exertional angina provoked by a variable threshold and/or at least two episodes of rest angina within the 3 months prior to screening. Patients were studied with nifedipine and placebo in a 6-week, double-blind, crossover design that used serial anginal diaries, exercise treadmill tests, and 24-hour ambulatory ECG monitoring. In patients with mixed angina, nifedipine reduced the frequency of angina compared to that during placebo treatment (13.1 vs 9.9 episodes/3 weeks, p less than 0.01) and reduced nitroglycerin consumption (11.7 vs 7.5 tablets/3 weeks, p less than 0.05); while in patients with classic exertional angina, nifedipine had no symptomatic effect (7.9 vs 6.8 anginal episodes/3 weeks, NS; 6.4 vs 5.8 nitroglycerin tablets/3 weeks, NS). Patients in both groups experienced a significant decrease in the manifestations of ischemia during exercise testing. Patients with mixed angina experienced a reduction in the daily frequency of painful episodes of ST segment depression during nifedipine treatment compared to placebo (0.6 vs 0.2 episodes, p less than 0.05), but there was no effect on the frequency of episodes of silent ischemia (4.2 vs 3.4 episodes, NS). In patients with classic exertional angina, the addition of nifedipine had no effect on any measure of ambulatory ischemia. We conclude that patients with mixed angina are more likely to benefit symptomatically from the addition of nifedipine therapy than patients with classic exertional angina. The lack of a consistently preferential response to nifedipine in patients with mixed angina, however, suggests that episodic coronary vasoconstriction may not be the only mechanism responsible for ischemia in these patients, and/or that nifedipine may not necessarily provide additional therapeutic benefit beyond that conferred by a regimen of beta blockers and/or nitrates.