Arterial blood pressure patterns in 12 men with sleep apnea and arterial hypertension were studied at baseline and after 6 months' therapy with nasal continuous positive airway pressure (nCPAP). Preexisting antihypertensive medication was discontinued 1 week before baseline measurements. Weight did not change during the study period; body mass index was 29.3 (range, 25.4-38.5) vs. 29.3 (25.0-38.5). During therapy the apnea index decreased from 58 (range 30-73) to 2 (range 0-7) apneic episodes per hour (p less than 0.01). Intra-arterial systolic (BP sys.) and diastolic (BP dias.) blood pressure and heart rate decreased during therapy (p less than 0.001). Mean values +/- 95% confidence intervals were as follows: BP sys., 147.1 (+/- 1.6) mm Hg vs. 126.4 (+/- 1.5) mm Hg; BP dias., 81.6 (+/- 0.8) mm Hg vs. 69.4 (+/- 0.6) mm Hg; heart rate, 68.8 (+/- 0.7) beats/min vs. 65.4 (+/- 0.7) beats/min. Furthermore, the variability of these parameters decreased during therapy: variability BP sys., 53.8 (+/- 1.1) mm Hg vs. 25.6 (+/- 1.1) mm Hg; variability BP dias., 35.6 (+/- 0.7) mm Hg vs. 17.9 (+/- 0.7) mm Hg; variability of heart rate, 28.1 (+/- 0.7) beats/min vs. 14.9 (+/- 0.7) beats/min (p less than 0.001). During treatment we found that blood pressure scores already dropped during the awake phase, with a further decrease during non-REM and REM sleep (p less than 0.001). Our results, which demonstrate the reversibility of high blood pressure upon treatment of sleep apnea, indicate that sleep apnea can be an etiological factor in hypertension. Sleep apnea should therefore be considered in the differential diagnosis of arterial hypertension.