Rationale: Randomized controlled trials do not support a role for continuous positive airway pressure (CPAP) in preventing major cardiovascular events or mortality in patients with obstructive sleep apnea (OSA). However, these trials' setting does not apply to most CPAP-treated patients. Objectives: We aimed to assess the effect of CPAP on mortality in real-world patients. Methods: We performed a population-based longitudinal observational study including all patients with OSA prescribed CPAP during 2011 in Catalonia, Spain, and non-OSA control subjects matched (1:2) by sex, 5-year age group, and region who were followed from 2011 to 2016. Results: A total of 9,317 CPAP-treated patients with OSA and 18,370 control subjects without OSA were included (median age, 67 [57-72] years; 74% male). During a median follow-up of 5.5 years, 2,301 deaths were recorded. After adjustment by a composite of diagnosed comorbidities and previous use of healthcare resources, CPAP-treated patients showed a lower risk of death than control subjects (hazard ratio [HR], 0.67; 95% confidence interval [CI], 0.61-0.74), with the association not being statistically significant in women. Cancer-related deaths were the main drivers of this association (men: HR, 0.44; 95% CI, 0.36-0.54; women: HR, 0.44; 95% CI, 0.28-0.68). No significant associations were found for cardiovascular-related deaths. CPAP-treated women had an increased risk of respiratory-related death (HR, 2.41; 95% CI, 1.37-4.23). Conclusions: CPAP-treated patients had a lower mortality rate than control subjects. This relationship was driven by cancer-related, but not cardiovascular-related, deaths. Results suggest a role for sex when prescribing CPAP, especially considering respiratory-related deaths, and foster a debate on the relationship between OSA and cardiovascular outcomes.
Keywords: CPAP; mortality; obstructive sleep apnea.