To elucidate if an effect of propofol on endothelium-dependent relaxation could contribute to propofol-induced vasodilation, smooth muscle relaxation of isolated human omental artery and vein segments precontracted by endothelin-1 were measured. Substance P induced a concentration-dependent relaxation (mean ±SEM) in both artery (63 ± 8.4% of precontraction, n=9) and vein (60 ± 11%, n=7). The relaxation was enhanced by 10−6 M propofol (artery, 72 a 9.5%, n=9; vein, 81 ± 12%, n=7) but not affected by 10−7, 10−5 and 10−4 M propofol. In the presence of Nω-nitro-L-arginine methyl ester (nitric oxide synthase inhibitor), 10−6 M propofol still enhanced the substance P-induced relaxation in arteries but not veins, whereas 10−4 M propofol inhibited the relaxation in both arteries (rightward shift of the concentration-response curve) and veins (28 ± 7.5%, n=8). In the presence of potassium chloride (to prevent hyperpolarization), the enhancement of substance P-induced relaxation by 10−6 M propofol was abolished in both arteries and veins whereas 10−5 and 10−4 M propofol reduced the relaxation in arteries (38 ± 13% at 10−5 M, n=6; 30 ± 11% at 10−4 M, n=6) but not in veins. These results demonstrate that propofol, at lower, clinically relevant concentrations, promotes endothelium-dependent relaxation mediated via hyperpolarization in human omental arteries and via both nitric oxide and hyperpolarization in human omental veins.