Endothelial dysfunction in patients with chest pain and normal coronary arteries

A subgroup of patients with chest pain and angiographically normal epicardial coronary arteries have reduced dilator response to metabolic or pharmacological stimuli, but the mechanisms responsible for this reduced dilator response are unknown. In this study, we have investigated whether microvascular endothelial dysfunction is a cause of the observed reduced vasodilator reserve.

The functional response of the microvasculature was studied with rapid atrial pacing at 150 beats per minute. Fifty-one patients, 20 hypertensive and 31 normotensive, with chest pain and normal epicardial coronary arteries (< 10% stenosis) were studied. Endothelial function was tested with incremental infusions of acetylcholine to achieve estimated intracoronary concentrations ranging from 10(-7) M to 10(-5) M. Endothelium-independent smooth muscle vasomotion was measured using intracoronary sodium nitroprusside. Endothelial dysfunction of epicardial coronary arteries, demonstrated as severe (> 50%) constriction with < 10(-5) M acetylcholine concentration, was evident in five patients (10%). In the remaining 46 patients, coronary blood flow increased with acetylcholine (mean, 78 +/- 43%) and atrial pacing (mean, 51 +/- 37%), and coronary vascular resistance decreased by 35 +/- 16% and 29 +/- 14%, respectively, but the responses were heterogeneous. There was a correlation between the coronary resistance change with acetylcholine and the change with atrial pacing: r = 0.68, p < 0.001 in these 46 patients. Thus, patients with depressed dilation with atrial pacing had reduced endothelium-dependent dilation with acetylcholine, and vice versa. However, the microvascular dilation caused by sodium nitroprusside was not significantly different between patients with and those without reduced dilation with atrial pacing, indicating that the vasodilator defect was not caused by smooth muscle dysfunction. There were no differences in the vasodilator responses with atrial pacing, acetylcholine, or nitroprusside between normotensive and hypertensive patients. Multivariate regression analysis was performed to determine whether age, sex, serum cholesterol level, hypertension, presence of mild epicardial vessel atherosclerosis, resting left ventricular function, change in left ventricular ejection fraction with exercise, vasodilation with acetylcholine, and vasodilation with sodium nitroprusside were independently related to the vasodilator response to atrial pacing. Only the change in coronary vascular resistance with acetylcholine was independently correlated with the change in resistance with atrial pacing: R2 = 0.46, p < 0.0001.

Patients with chest pain, normal epicardial coronary arteries, and reduced vasodilation in response to atrial pacing appear to have associated endothelial dysfunction of the coronary microvasculature. Thus, microvascular endothelial dysfunction may contribute to the reduced vasodilator reserve with atrial pacing and anginal chest pain in these patients.