Circulating Nitric Oxide Is Suppressed in Obstructive Sleep Apnea and Is Reversed by Nasal Continuous Positive Airway Pressure

Abstract

Epidemiological studies have implicated obstructive sleep apnea (OSA) as an independent comorbid factor in cardiovascular and cerebrovascular diseases. The recurrent episodes of occlusion of upper airways during sleep result in pathophysiological changes that may predispose to vascular diseases, and we postulate that nitric oxide may be one of the mediators involved. This study investigates the levels of circulating nitric oxide (NO), measured as serum nitrites and nitrates, in the early morning in OSA subjects compared with control subjects, and the effect of overnight nasal continuous positive airway pressure (nCPAP) in OSA subjects. Thirty men with moderate to severe OSA (age = 41.9 +/- 9.0; apnea-hypopnea index, AHI = 48.0 +/- 18.1) were compared with 40 healthy men (age = 40.6 +/- 5.4; AHI = 1.4 +/- 1.2). Serum nitrite/nitrate levels were significantly lower in OSA subjects (OSA = 38.9 +/- 22.9 microM, control subjects = 63.1 +/- 47.5 microM, p = 0.015). There was significant negative correlation between serum nitrites/nitrates and the following parameters: AHI (r = -0.389, p = 0.001), oxygen desaturation time (r = -0.346, p = 0.004), and systolic blood pressure (BP) (r = -0.335, p = 0.005). Stepwise multiple linear regression with systolic or diastolic BP as the dependent variable identified serum nitrites/nitrates as the only significant correlate. Twenty-two OSA subjects had measurements of serum NO at baseline and after an overnight application nCPAP. There was significant increase in serum NO after nCPAP (baseline = 30.5 +/- 14.4 microM, after nCPAP = 81.0 +/- 82.1 microM, p = 0.01). This study demonstrates, for the first time, that circulating NO is suppressed in OSA, and this is promptly reversible with the use of nCPAP. The findings offer support for nitric oxide being one of the mediators involved in the acute hemodynamic regulation and long-term vascular remodeling in OSA.