Model-based assessment of autonomic control in obstructive sleep apnea syndrome.

Abstract

STUDY OBJECTIVES: To quantitatively assess autonomic cardiovascular control in patients with obstructive sleep apnea syndrome (OSAS) using a mathematical model that relates changes in R-R interval (RRI) to respiration and changes in systolic blood pressure (SBP), and to compare the results obtained with conventional techniques. DESIGN: Respiration, RRI, and arterial blood pressure were monitored noninvasively in awake subjects in the supine and standing postures. A mathematical model was used to partition the fluctuations in RRI into a component ("RSA") correlated with respiration and a component ("baroreflex") correlated with fluctuations in SBP. SETTING: Sleep disorders laboratory in a hospital setting. PATIENTS OR PARTICIPANTS: 11 middle-aged male patients with untreated OSAS (apnea-hypopnea index = 75.9 +/- 11.1 (SE) events h-1) and 11 age-matched normal controls (10 males + 1 female). INTERVENTIONS: The subjects were monitored while breathing spontaneously in both supine and standing postures. Each subject also had to perform a battery of 5 standard autonomic stress tests (AST). MEASUREMENTS AND RESULTS: Four of the 5 ASTs did not indicate any difference between controls and OSAS subjects. There were also no differences in the indices derived from power spectral analysis of RRI and blood pressure variability; however, these spectral indices were sensitive to postural changes (orthostatic stress). Both RSA and baroreflex gains estimated from the model were substantially depressed in OSAS (P < 0.02) Changes in posture affected RSA gain but not baroreflex gain. The time-courses of the dynamics of these model components were not significantly different between subject groups. CONCLUSIONS: OSAS leads to abnormal parasympathetic and sympathetic control of heart rate, as reflected in depressed RSA and baroreflex gains. Model-based assessment was more sensitive in detecting abnormal autonomic function, compared to standard autonomic testing and power spectral analysis. The model-based approach represents a relatively simple and nonintrusive means of quantifying the key aspects of autonomic control in spontaneously breathing OSAS patients during wakefulness.