Coronary hemodynamics, myocardial metabolism and left ventricular function at rest and after incremental atrial pacing were evaluated in 12 patients with stress-induced angina and ST segment depression, angiographically normal coronary arteries and no evidence of spasm, generally labeled as syndrome X, and in 10 normal subjects. At baseline study, great cardiac vein flow was comparable in patients and control subjects. During pacing, an equivalent rate-pressure product was reached in the two groups, but the slope of the relation between rate-pressure product and great cardiac vein flow was significantly less steep in patients than in normal subjects (0.0027 vs. 0.0054 ml/mm Hg.beat, p less than 0.001). Nevertheless, the left ventricular ejection fraction was comparable in both groups at rest (66 +/- 6% vs. 71 +/- 7%, p = NS) and during pacing (71 +/- 7% vs. 66 +/- 5%, p = NS). At baseline study, myocardial glucose extraction was more efficient in patients with syndrome X (p less than 0.05), but net myocardial exchange of pyruvate and alanine was, respectively, smaller and greater than in control subjects. Lactate was extracted to a similar extent in the two groups and in no instance was net lactate release observed during pacing or recovery. During pacing and recovery, patients with syndrome X showed net pyruvate release, unlike the control subjects in whom net pyruvate exchange was positive. In addition, patients with syndrome X continued to show net myocardial extraction of alanine during spacing and recovery, whereas normal subjects produced alanine throughout the study. Myocardial carbohydrate oxidation increased significantly during maximal pacing in normal subjects but not in patients, in whom it always remained below (p less than 0.01) the concurrent rate of myocardial uptake of carbohydrate equivalents (glucose, lactate, pyruvate, alanine). Myocardial energy expenditure was significantly lower in patients than in control subjects at maximal rate-pressure product levels (p less than 0.01). The metabolic pattern in patients with syndrome X therefore is not consistent with classic ischemia, although differences in the net exchange of circulating substrates (glucose, pyruvate, alanine) can be demonstrated. Thus, in patients with syndrome X, the symptoms, electrocardiographic signs and impairment in the increase in great cardiac vein flow during pacing coexist with preserved global and regional left ventricular function and myocardial energy efficiency.