To study the mechanism of chest pain in patients with insignificant epicardial coronary artery disease, 50 patients underwent great cardiac vein (GCV) flow, oxygen content and lactate determinations at rest and during pacing, and left ventricular end-diastolic pressure (LVEDP) measurements at rest and after pacing. Twenty-four patients having typical chest discomfort during pacing demonstrated significantly lower increase in flow from baseline (36 +/- 18% versus 86 +/- 24%, p less than 0.001) and decrease in coronary resistance (-17 +/- 12% versus -43 +/- 7%, p less than 0.001) compared with 26 patients without pacing-induced chest pain, despite no significant difference in myocardial oxygen consumption (MVO2) between the 2 groups. Lactate consumption at a heart rate (HR) of 150 beats/min was significantly less (28.3 +/- 21.5 versus 51.3 +/- 35.8 mM X ml/min, p less than 0.001) and the increase in LVEDP from rest to after pacing was significantly greater (5 +/- 2 versus 1 +/- 2 mm Hg, p less than 0.001) in the chest pain group. After administration of ergonovine, 0.15 mg intravenously, to 46 of these patients, 31 had typical pain either at rest (1 patient) or during pacing. This group had significantly lower increase in flow (38 +/- 20% versus 107 +/- 38%, p less than 0.001), and decrease in coronary resistance (-16 +/- 12% versus -45 +/- 11%, p less than 0.001) compared with the 15 patients not having chest pain, despite no significant difference in MVO2 between the 2 groups. Patients with chest pain also had lower lactate consumption at a HR of 150 beats/min (39.2 +/- 23.6 versus 65.3 +/- 46.3 mM X ml/min, p less than 0.01), greater arterial-GCV oxygen difference (12.5 +/- 1.3 versus 11.6 +/- 1.0 ml O2/100 ml, p less than 0.05), and a more marked increase in LVEDP from rest to after pacing (11 +/- 3 versus 5 +/- 2 mm Hg, p less than 0.001). Quantitative coronary arteriography demonstrated no significant luminal narrowing of the epicardial coronary arteries in response to ergonovine. These data are consistent with the hypothesis that some patients with chest pain and angiographically normal epicardial coronary arteries have dynamic abnormalities of the small coronary arteries or coronary microcirculation that cause abnormal vasodilator reserve or vasoconstriction, resulting in myocardial ischemia and angina pectoris.