Abstract
PURPOSE: The association between glycemic variability (GV) and the clinical outcomes of patients with chronic obstructive pulmonary disease (COPD) remains uncertain. This study aimed to assess the connection between GV and hospital mortality among hospitalized COPD patients. PATIENTS AND METHODS: Data from the MIMIC-IV database were used in this study. GV was analyzed as a continuous variable and in quartiles. Cox proportional hazards modeling with stepwise adjustment was used to assess the link between GV and in-hospital mortality, and HRs and 95% CIs were calculated. A restricted cubic spline model was applied to explore non-linear associations. Survival curves and Log rank tests were performed to assess survival differences. Subgroup analyses and interaction tests were conducted to determine the relationship across diverse populations. RESULTS: Our analysis encompassed 11873 patients, among whom 838 individuals died during their hospitalization. A significant association between GV and in-hospital mortality in COPD patients was found across all Cox proportional hazards models: the unadjusted model (HR [95% CI]: 1.849 [1.409, 2.427], P < 0.001), the partially adjusted model (HR [95% CI]: 1.880 [1.432, 2.468], P < 0.001), and the fully adjusted model (HR [95% CI]: 1.551 [1.110, 2.167], P = 0.01). The restricted cubic spline (RCS) models revealed a significant non-linear relationship (P-nonlinear < 0.001). Survival curves demonstrated significant differences in in-hospital survival rates across varying GV levels (log-rank P < 0.0001). CONCLUSION: GV exhibited a strong association with in-hospital mortality among COPD patients, and significant differences were observed in in-hospital survival rates across varying GV levels.