Impact of glycemic variability on all-cause mortality in critically ill elderly patients with hyperlipidemia

Glycemic variability (GV) is a critical indicator of acute glucose fluctuations. We aimed to assess the association between GV and 28-day and 365-day all-cause mortality in critically ill elderly patients with hyperlipidemia. A retrospective multicenter study was performed on 4494 patients from MIMIC‑IV databases and 435 patients from NWICU databases as the external validation cohort. Survival rates were compared by using Kaplan–Meier curves, and multivariable Cox regression models were used to assess the association between GV with mortality. Restricted cubic spline (RCS) analysis was used to explore non-linear relationships, and subgroup analyses were performed to examine potential variations across sex, race, and comorbidities. Sensitivity analyses were also used to confirm the findings on 180-day mortality, and all adult patients with hyperlipidemia. Furthermore, the NWICU database was utilized as an external validation cohort for additional sensitivity testing. Kaplan–Meier analysis showed higher mortality in the high GV group. In the fully adjusted cox regression model, higher GV was significantly linked to an increased risk of 365-day mortality [1.62 (95% CI, 1.39–1.88; P < 0.001)] and 28-day mortality [1.56 (95% CI, 1.28–1.89; P < 0.001)] in elderly patients with hyperlipidemia, subgroup and sensitivity analysis further confirmed the findings. High GV is associated with increased 28-day and 365-day mortality in critically ill elderly patients with hyperlipidemia.