All-Cause Mortality and Obstructive Sleep Apnea Severity Revisited

Abstract

Objective Obstructive sleep apnea syndrome (OSAS) is a pervasive problem that affects millions worldwide. It is strongly linked to hypertension, coronary artery disease, and stroke. However, its association with mortality is not clearly quantified. A large database of patients who underwent sleep testing was explored for associations with all-cause mortality. Study Design Database study. Setting Community-based use of a portable sleep study device. Subjects and Methods More than 77,000 patients who underwent a validated, portable sleep study were matched to the Social Security Death File to establish mortality. Measures of OSAS severity and other confounding factors were correlated to all-cause mortality using survival analysis with multivariate Cox proportional hazards regression. Results As expected, increasing age (adjusted hazard ratio [HR], 1.080; 95% confidence interval [CI], 1.074-1.086; P < .0001), body mass index (HR, 1.042; 95% CI, 1.033-1.051; P < .0001), and male sex (HR, 1.378; 95% CI, 1.190-1.595; P < .001) were associated with increased all-cause mortality. Epworth sleepiness score was also associated with mortality (HR, 1.015; 95% CI, 1.005-1.025; P = .002). Apnea-hypopnea index (AHI) was not associated with mortality after adjustment for age (HR, 1.001; 95% CI, 0.998-1.004; P = .416). However, within 10-year age subgroups, desaturation index (ages 41-50 years; adjusted HR, 1.217; 95% CI, 1.014-1.461; P = .035), apnea index (ages 21-30 years; HR, 1.632; 95% CI, 1.053-2.532; P = .028), and AHI (ages 31-40 years; HR, 1.222; 95% CI, 1.010-1.478; P = .039) were significantly associated with all-cause mortality in younger patients. In patients older than 50 years, age, sex, and body mass index were dominantly associated with mortality. Conclusion Increasing OSAS severity, measured by a validated home sleep test and quantified by AHI, the apnea index, and the desaturation index, is independently associated with modestly increased all-cause mortality in patients younger than 50 years after adjustment for major confounding factors.