Sleep-disordered Breathing in Patients with Chronic Obstructive Pulmonary Disease: Prevalence and Outcomes

Rationale: The prevalence of obstructive sleep apnea (OSA) or nocturnal hypoxemia without OSA (i.e., isolated nocturnal hypoxemia [iNH]) and its impact on the natural history of chronic obstructive pulmonary disease (COPD) are unclear. Objective: We determined the prevalences of OSA and iNH in patients with COPD and their contribution to all-cause mortality and COPD exacerbations. Methods: At baseline, we performed home sleep apnea testing in the CHAIN (COPD History Assessment in Spain) study cohort. OSA was defined by an apnea-hypopnea index of ⩾15 events per hour, and iNH was defined by an oxygen saturation as measured by pulse oximetry <90% for >30% of the nocturnal recording time. During an observation time of 6.5 years, we evaluated the association of OSA or iNH with all-cause mortality using multivariable Cox regression models and with COPD exacerbations using negative binomial models. Results: Among 428 patients with COPD, OSA and nocturnal hypoxemia were ruled out in 41%, whereas 27% had iNH and 32% had OSA (i.e., COPD/OSA "overlap syndrome" [OVS]). OVS was independently associated with obesity as defined by a body mass index ⩾30 kg/m² and with severe COPD exacerbations (P < 0.01), whereas iNH was associated with lower forced expiratory volume in 1 second and lower resting arterial oxygen saturation. Compared with patients with COPD without OSA or iNH, those with untreated OVS had greater risks of mortality (hazard ratio, 1.74; 95% confidence interval, 1.03-2.94) and COPD exacerbations (incidence rate ratio, 1.44; 95% confidence interval, 1.05-2.03). Conclusions: OSA and iNH are frequent in patients with COPD, and the prevalences decrease or increase, respectively, with disease severity. Patients with COPD with untreated OVS but not with iNH had a greater risk of all-cause mortality and COPD exacerbations.