PM2.5 as a risk factor for sleep oxygen desaturation in COPD: identifying susceptible individuals by smoking and inflammatory phenotypes

Background Nocturnal hypoxia is prevalent in chronic obstructive pulmonary disease (COPD) and contributes significantly to poor disease prognosis. However, its key environmental risk factors and susceptible phenotypes remain poorly understood. Methods This prospective repeated-measure study included 96 patients with COPD with real-time personal monitoring of fine particulate matter (PM 2.5 ) and oxygen saturation (SpO 2 ) during sleep (84 971 observations), aiming to investigate the association of PM 2.5 with oxygen desaturation and identify susceptible individuals by smoking and inflammatory phenotypes. Inflammatory phenotypes were determined by exhaled biomarkers (nitric oxide and hydrogen sulfide) and blood leucocytes (neutrophils and eosinophils). Generalised linear mixed models were applied to estimate the risk of oxygen desaturation associated with PM 2.5 . Results Personal PM 2.5 exposure was significantly associated with sleep SpO 2 decline within hours. The highest risk of oxygen desaturation associated with PM 2.5 was observed in current smokers (OR=1.53, 95% CI: 1.22 to 1.92), followed by former smokers (OR=1.14, 95% CI: 1.02 to 1.28), with no significant effect in never smokers. Notably, current smokers exhibited higher neutrophilic inflammation, and those with higher airway neutrophilic inflammation were more susceptible to PM 2.5 -related oxygen desaturation (p interaction<0.001). For former smokers, those with higher small-airway eosinophilic inflammation were more susceptible (p interaction <0.001). Both former and current smokers with a systemic mixed granulocyte phenotype were at greater risk of oxygen desaturation. Furthermore, individuals with constant high inflammation are more susceptible to PM 2.5 -related oxygen desaturation than those without inflammation. Conclusions Airborne PM 2.5 is an important risk factor for sleep hypoxia in patients with COPD, especially for individuals with smoking history and specific inflammatory phenotypes. Trial registration number NCT05076630 .