Obstructive sleep apnea and hypopnea syndrome (OSAHS) is a common comorbidity of hypertension. The prevalence of OSAHS was from 37% 1 to 56% 2 in hypertensive patients and 83% in patients with hypertension resistant to antihypertensive treatment. 3 However, causal relationship between OSAHS and hypertension has not been well established. Some, 4,5 but not other, 6 hypertension guidelines considered OSAHS as a secondary cause of hypertension. Previous studies often used apnea/hypopnea index (AHI), a direct measure of respiration, instead of oxygen desaturation index, a measure of hypoxemia. 7,8 The latter rather than the former may be a critical pathophysiological mechanism for hypertension.Previous studies also did not sufficiently account for the circadian profile of blood pressure, sympathetic over-activation (i.e, as assessed by tachycardia), or aging. Indeed, sympathetic overactivation may be the mediator between hypoxemia and high blood pressure, and its role may be attenuated with aging because heart rate increase induced by sympathetic activation is less prominent in the elderly.
9-11Our hypothesis was that oxygen desaturation might be closely associated with blood pressure, especially on ambulatory measurement, and that the association might be enhanced in younger subjects by accounting for heart rate as a rough measure of sympathetic activity. We recently performed overnight finger pulse oximetry in outpatients who had established or suspected diagnosis We investigated the relationship between ambulatory blood pressure (BP) and oxygen desaturation index (ODI), while accounting for pulse rate and age. ODI was assessed by overnight finger pulse oximetry in 2342 participants on the day of ambulatory BP monitoring, and calculated as the number of desaturation episodes per sleeping hour. Both BP and pulse rate increased significantly (P ≤ .006) from normal (< 5 events/h) to mildly (5-14), moderately (15-30), and severely (≥ 30 events/h) elevated ODI. The association for BP was substantially attenuated by accounting for pulse rate (partial r² from .003-.012 to .002-.006). In adjusted analysis, the associations of 24-hour diastolic BP and 24-hour pulse rate with ODI were dependent on age (P ≤ .0001) and only significant in younger subjects (< 60 years, P ≤ .0001). In conclusion, the association between ambulatory BP and ODI was partially mediated by pulse rate, a measure of sympathetic activity, and was more prominent in younger subjects.Qi Chen and Yi-Bang Cheng contributed equally to this work.