In order to assess the complications of sleep apnea, we have reviewed a data base of 619 consecutive admissions to a university sleep disorders center. Although patients with obstructive sleep apnea (OSA) described more subjective sleepiness than patients with central sleep apnea (CSA) or primary snoring (PS), the multiple sleep latency test (MSLT) indicated similar levels of physiologic sleepiness in the two apneic groups, which was greater than among those with PS. There was no significant relationship between individual subjective estimates of habitual sleepiness and the MSLT values. Among the OSA patients the mean minimum arterial oxygen desaturation during REM sleep accounted for 65 percent of the variance of the mean sleep latency on the MSLT, with an additional, smaller, contribution of the disordered breathing rate per hour. Subjective reports of sleepiness were associated with sleep efficiency and the number of disordered breathing events in NREM sleep. Patients with OSA or CSA had similar diastolic blood pressures and frequencies of history of treatment for hypertension, which were significantly higher in OSA than in the PS group. In the OSA group the absolute minimum arterial oxygen desaturation during NREM sleep was the most significant contributor to waking diastolic blood pressure, with an additional small contribution by weight. A history of treatment for hypertension was most strongly associated with weight, without significant additional contributions by measures of disordered breathing events or oxygen desaturation; however, weight was highly intercorrelated with measures of the apnea/hypopnea index and minimum arterial oxygen desaturation. In summary, these data support recent findings which show a close relation of obesity to a history of hypertension in OSA, and extend to this group a previous observation that in regular heavy snorers, there may be a disparity between levels of physiologic and subjective sleepiness.