Background Patients with essential hypertension have abnormal endothelium-dependent vascular relaxation, largely related to reduced bioactivity of nitric oxide (NO). The purpose of the present investigation was to determine whether this defect is due to a deficit at the specific intracellular signal-transduction pathway level or is a consequence of a more generalized endothelial abnormality. Methods and Results The responses of the forearm vasculature to acetylcholine and bradykinin (endothelium-dependent agents that act through different signal transduction pathways) and to sodium nitroprusside (a direct dilator of vascular smooth muscle) were studied in 10 hypertensive patients (5 men, 5 women; aged 48+/-9 years old [mean+/-SD]) and 12 control subjects (6 men, 6 women; aged 48+/-7 years old). To determine the contribution of NO to bradykinin-induced vasodilation, the vascular responses to bradykinin were also measured after administration of N-G-monomethyl-L-arginine, an arginine analogue that inhibits the synthesis of NO. Drugs were infused into the brachial artery, and forearm blood flow was measured by strain-gauge plethysmography. The response to acetylcholine was significantly blunted in hypertensive patients (maximal blood flow, 7.5+/-2 versus 16.6+/-8 mL . min(-1). 100 mL(-1) in control subjects [mean+/-SD]; P<.005). Similarly, the vasodilator effect of bradykinin was significantly reduced in hypertensive patients compared with control subjects (maximal blood flow, 8.7+/-2 versus 15.8+/-6 mL . min(-1). 100 mL(-1) in control subjects; P<.005). A significant correlation was found between the maximal blood flow with acetylcholine and that with bradykinin (r=.89). No significant differences were found between the two groups for vascular response to sodium nitroprusside. N-G-monomethyl-L-arginine significantly blunted the response to bradykinin in control subjects (maximal blood flow decreased from 15.8+/-6 to 10.1+/-2 mL . min(-1). 100 mL(-1), P<.003). In contrast, inhibition of NO synthesis did not modify the response to bradykinin in hypertensive patients (maximal blood flow, 8.7+/-2 and 8.5+/-3 before and during infusion of N-G-monomethyl-L-arginine, respectively; P=NS). As a consequence, the response to bradykinin after inhibition of NO synthesis was not significantly different between the two groups. Conclusions Patients with essential hypertension have impaired endothelium-dependent vasodilator responses to both acetylcholine and bradykinin. These findings indicate that the endothelial dysfunction in this condition is not related to a specific defect of a single intracellular signal-transduction pathway and suggest a more generalized abnormality of endothelial vasodilator function.