Background: Previous studies have reported larger myocardial perfusion defects with exercise as compared with dipyridamole. The aim of this study was to assess the hypothesis that this mismatch may reflect differences in epicardial and microvascular coronary vasomotor function.
Methods and results: The response to intracoronary acetylcholine, nitroglycerin, and adenosine was studied in 36 patients with suspected angina and normal or near-normal coronary angiography findings who underwent both exercise and dipyridamole perfusion imaging. Of the patients, 27 (75%) had reversible defects with exercise (group I) and 9 had normal scans or nonreversible defects (group II). Repeated imaging with dipyridamole showed significant improvement or disappearance of perfusion defects in group I patients. The mean summed difference score (SDS) decreased from 5.52 +/- 3.19 with exercise to 1.11 +/- 1.60 with dipyridamole (P = .0001) in group I and did not change in group II. An abnormal epicardial response to acetylcholine, reflecting endothelial dysfunction, occurred in 93% of group I patients compared with only 33% of group II patients (chi(2) = 9.46, P = .002) and was significantly related to exercise SDS (r = 0.49, P = .002) but not to dipyridamole SDS. By contrast, most patients showed normal epicardial and microvascular responses to the mainly non-endothelium-dependent vasodilators nitroglycerin and adenosine with no differences in coronary flow reserve between groups (2.91 +/- 0.72 vs 2.98 +/- 0.52, P = .79).
Conclusions: Exercise-dipyridamole perfusion mismatch may reflect differences in epicardial endothelial and microvascular dysfunction.