The Relationship of Sleepiness and Blood Pressure to Respiratory Variables in Obstructive Sleep Apnea

Abstract

As a follow-up to a previous assessment of complications of sleep-disturbed breathing in 265 patients, we have reevaluated measures of sleepiness and hypertension in patients with obstructive sleep apnea (OSA) (n = 518), central sleep apnea (n = 50), and subclinical sleep-disordered breathing (SDB) (n = 107). Both subjective and objective (multiple sleep latency test [MSLT]) measures indicated that OSA patients were sleepier than those with subclinical SDB. The OSA patients weighed significantly more than the patients with central sleep apnea or subclinical SDB. They had a higher proportion of men, described more habitual sleepiness, and had a higher likelihood of feeling unrefreshed in the morning compared with the group with subclinical SDB. Among the OSA patients, there was a significant correlation between subjective and objective assessment of sleepiness, but this relationship was quantitatively very small. A forward stepwise regression analysis revealed that weight, and to a lesser degree waking time after sleep onset, could account for 65.5% of the variance in subjective sleepiness. Seventy-five percent of the variance of the mean sleep latency in the MSLT could be accounted for by the mean minimum arterial oxygen saturation in non-REM sleep and the nocturnal sleep latency. Diastolic BP was significantly higher in OSA patients compared with the patients with central sleep apnea and subclinical SDB. When covarying for weight, age, and gender, this effect lost significance. Among OSA patients taken by themselves, 98.3% of the variance in diastolic blood pressure could be accounted for by the mean minimum arterial oxygen saturation in non-REM sleep, with very small additional contributions of apnea/hypopnea index, weight, and age. In summary, among patients across a spectrum of SDB, differences in diastolic BP were primarily associated with weight, age, and gender. Among OSA patients, perhaps because of a more limited variance in weight, diastolic BP was associated with measures of SDB.