Mortality in obstructive sleep apnea-hypopnea patients treated with positive airway pressure

Abstract

Study objectives: The aims of this study were to analyze mortality in patients with obstructive sleep apnea-hypopnea syndrome (OSAHS) treated with positive airway pressure (PAP) and to know whether PAP compliance affects survival, as well as to investigate the prognostic value of several pretreatment variables.

Design and patients: A study was made of an historical cohort of 871 patients in whom OSAHS had been diagnosed by sleep study between January 1994 and December 2000 and who had been treated with PAP. Patients were followed up until December 2001. The mean (+/- SD) age of the group was 55.4 +/- 10.6 years, the mean apnea-hypopnea index (AHI) 55.1 +/- 28.7, and 80.9% were men. To assess whether mortality was influenced by PAP therapy compliance, patients were assigned to one of the following compliance categories: < 1 h/d; 1 to 6 h/d; or > 6 h/d. Survival rates were calculated according to the Kaplan-Meier method. Survival curves were compared with the log-rank test and the trend test, when necessary. Univariate and multivariate analyses using a time-dependent Cox model were performed to elicit which variables correlated with mortality.

Setting: Outpatient sleep disorders unit.

Results: By the end of the follow-up period (mean duration, 48.5 +/- 22.7 months), 46 patients had died. The 5-year cumulative survival rates were significantly lower in patients who did not use PAP (compliance < 1 h) than in those who used the device for > 6 h/d (85.5% [95% confidence interval (CI), 0.78 to 0.92] vs 96.4% [95% CI, 0.94 to 0.98; p < 0.00005]) and 1 to 6 h/d (85.5% [95% CI, 0.78 to 0.92] vs 91.3% [ 95% CI, 0.88 to 0.94; p = 0.01]), respectively. A trend in survival rates across the groups was identified (p = 0.0004). The main cause of death in 19 cases was cardiovascular disease (CVD). Variables that independently correlated with mortality in the multivariate analysis were the following PAP use categories: compliance for > 6 h/d (odds ratio [OR], 0.10; 95% CI, 0.04 to 0.29); compliance for 1 to 6 h/d (OR, 0.28; 95% CI, 0.11 to 0.69); arterial hypertension (AHT) [OR, 3.25; 95% CI, 1.24 to 8.54]; age (OR, 1.06; 95% CI, 1.01 to 1.10); and FEV1 percent predicted (OR, 0.96; 95% CI, 0.94 to 0.98).

Conclusion: Mortality rates in OSAHS patients who did not receive PAP therapy were higher compared with those treated with PAP and were moderately or highly compliant with therapy. A trend in survival across compliance categories was found. Patients died mainly from CVD. Categories of PAP compliance, AHT, age, and FEV1 percent predicted were the variables that independently predicted mortality.