The U-shaped relationship between admission peripheral oxygen saturation and all-cause hospital mortality in acute exacerbation of chronic obstructive pulmonary disease: a retrospective analysis using the MIMIC III database

Background: Chronic obstructive pulmonary disease (COPD) and acute exacerbation of chronic obstructive pulmonary disease (AECOPD) pose global challenges, with oxygen saturation (SpO2) levels crucial in evaluating mortality. This study explored the correlation between admission SpO2 levels and allcause hospital mortality in patients with AECOPD, assessing whether SpO2 can serve as an independent risk factor for predicting in-hospital mortality in these patients. Methods: This study involved 996 AECOPD patients sourced from the Medical Information Mart for Intensive Care (MIMIC) III database (version 1.3), with 134 fatalities. Patients were categorized into a death group (n=134) and a survival group (n=862). The average admission SpO2 value was recorded for all 996 AECOPD patients. Subsequently, a generalized additive model (GAM) curve was employed to examine the association between admission SpO2 levels and all-cause hospital mortality. Following this, Cox regression analysis and survival analysis were conducted to further investigate the link between admission SpO2 and all-cause hospital mortality. Results: The GAM curve demonstrated a non-linear, U-shaped relationship between admission SpO2 and all-cause hospital mortality in AECOPD patients. The nadir of all-cause hospital mortality was associated with an SpO2 of 89.5%. Notably, an SpO2 of 89.5% served as the optimal cutoff for predicting all-cause hospital mortality. Cox regression analysis identified SpO2 as a risk factor for all-cause hospital mortality in AECOPD patients. Patients with SpO2 >= 89.5% exhibited independently lower death risk compared to those Conclusions: Admission SpO2 level is an independent risk factor for predicting all-cause hospital mortality in AECOPD patients and can serve as a prognostic indicator. A U-shaped relationship was observed, with an admission SpO2 level of 89.5% associated with the lowest mortality, suggesting an optimal range for improved prognosis.