Hypertension is a common finding among obstructive sleep apnea (OSA) patients, and is thought to be caused by sympathetic hyperactivity. The present study compares the contributions of the respiratory disturbance index (RDI) as a reflection of sleep fragmentation, and the magnitude of oxygen desaturation, to sympathetic activation as indexed by urinary norepinephrine concentrations, as well as to morning and evening blood pressure in sleep apnea syndrome patients. Data (polysomnography, blood pressure [BP], and urine catecholamines) of 38 consecutive OSA patients (age, 46+/-14.5 years) were analyzed. Stepwise logistic regression analysis revealed that minimal oxygen saturation level (SaO2min) was a significant predictor of both morning and evening norepinephrine levels, and that 37% of morning systolic BP variance could be accounted for by a combination of age and norepinephrine, while 20% of the diastolic BP variance was accounted for by SaO2min alone. In contrast, RDI entered the prediction equation only when minimal oxygen saturation was rejected first. Our results indicate that the degree of nocturnal hypoxia is more closely associated with the level of sympathetic activation and with daytime level of blood pressure than with sleep fragmentation.