The diagnostic and prognostic implication of exaggerated blood pressure response to exercise have been controversial, with opinions ranging from a benign process to a harbinger of potential cardiovascular morbidity. Endothelial dysfunction has been demonstrated in patients with atherosclerosis and as a risk factor for coronary artery disease. However, whether the cause of exercise-induced hypertension might be related to endothelial dysfunction has not been well elucidated. We evaluated endothelial function in patients who showed a systolic blood pressure ≥ 210 mmHg in males and ≥ 190 mmHg in females during treadmill exercise test. We measured the endothelial function of the brachial artery in 35 patients with exercise-induced hypertension, and in 35 age- and gender-matched normal control subjects, by a high resolution ultrasound technique, and the concentration of NO₂^-/NO₃^- and cyclic guanosine monophosphate (GMP). Endothelial-dependent vasodilation was impaired in patients with hypertension compared to normal controls (3.14±0.61 vs. 6.5±0.76%, p < 0.05). The extent of vasodilation was significantly correlated with age (r=-0.28, p < 0.05) and systolic blood pressure difference (r=-0.36, p < 0.05). The levels of NO₂^-/NO₃^- and cyclic GMP at maximal exercise were significantly higher than those at rest and recovery in both controls and the hypertensive group (p < 0.05). Although there was no significant difference in the increment of NO₂^-/NO₃^- during maximal exercise between the controls and hypertensive group (55±17 vs. 56±12µMol/L, p=NS), cyclic GMP level during maximal exercise was significantly higher in the control group than the hypertensive group (10±1.8 vs. 8.3±2.5 pmol/ml, p 0.05). Patients with exercise-induced hypertension have poor endothelium-dependent vasodilation due to an impaired nitric oxide/cyclic GMP pathway, which may play a significant role in increasing blood pressure during exercise with inadequate peripheral adjustment to changing cardiac output.