Abstract
BACKGROUND: Previous studies have demonstrated a significant correlation between obstructive sleep apnea (OSA) and frailty. However, the association of mean pulse oxygen saturation (MSpO(2)) with frailty among OSA patients remains unconfirmed. This study aimed to explore this potential association using data from a multicenter, prospective cohort. METHODS: A total of 1006 elderly patients diagnosed with OSA through polysomnography (PSG) from January 2015 to October 2017 were enrolled. Patients were stratified into four groups according to their MSpO(2) levels to assess differences in frailty onset. Multivariate Cox regression analysis, Kaplan-Meier curves, restricted cubic splines, and subgroup analyses were employed to evaluate variations in frailty onset across different MSpO(2) levels. RESULTS: Over a median follow-up period of 52 months, 275 patients developed frailty. Analysis using restricted cubic splines revealed a U-shaped trend between MSpO(2) and frailty risk (non-linear p-value = 0.028). Patients in the lowest quartile (MSpO(2) < 91.6%) exhibited a higher risk of frailty (hazard ratio [HR] = 1.43, 95% confidence interval [CI] 1.03-1.97, P = 0.029) compared to those in the third quartile (MSpO(2) 93-95%). Subgroup and sensitivity analyses confirmed the robustness of the U-shaped relationship. CONCLUSION: There is a U-shaped association between MSpO(2) and frailty among patients with OSA. Enhancing MSpO(2) levels may mitigate the risk of frailty and improve prognosis in this population.