Sildenafil, a selective inhibitor of phosphodiesterase type 5, produces relaxation of isolated epicardial coronary artery segments by causing accumulation of cGMP. Because shear-induced nitric oxide-dependent vasodilation is mediated by cGMP, this study was performed to determine whether sildenafil would augment the coronary resistance vessel dilation that occurs during the high-flow states of exercise or reactive hyperemia. In chronically instrumented dogs, sildenafil (2 mg/kg per os) augmented the vasodilator response to acetylcholine, with a leftward shift of the dose-response curve relating coronary flow to acetylcholine dose. Sildenafil caused a 6. 7 +/- 2.1 mmHg decrease of mean aortic pressure, which was similar at rest and during treadmill exercise (P < 0.05), with no change of heart rate, left ventricular (LV) systolic pressure, or LV maximal first time derivative of LV pressure. Sildenafil tended to increase myocardial blood flow at rest and during exercise (mean increase = 14 +/- 3%; P < 0.05 by ANOVA), but this was associated with a significant decrease in hemoglobin, so that the relationship between myocardial oxygen consumption and oxygen delivery to the myocardium (myocardial blood flow x arterial O(2) content) was unchanged. Furthermore, sildenafil did not alter coronary venous PO(2), indicating that the coupling between myocardial blood flow and myocardial oxygen demands was not altered. In addition, sildenafil did not alter the peak coronary flow rate, debt repayment, or duration of reactive hyperemia that followed a 10-s coronary occlusion. The findings suggest that cGMP-mediated resistance vessel dilation contributes little to the increase in myocardial flow that occurs during exercise or reactive hyperemia.