Post-Stroke Mortality in ICU Patients with Serum Glucose-Potassium Ratio: An Analysis of MIMIC-IV Database

Zhen Yuan Aoli Chen Yunqing Zeng Jiwei Cheng ^*

• Putuo District Central Hospital, Shanghai, China

Introduction:Acute ischemic stroke (AIS) patients admitted to the intensive care unit (ICU) have a high mortality rate, necessitating the early identification of those at risk of a poor prognosis. This study investigated the association between the blood glucose-to-potassium ratio (GPR) and the prognosis of AIS patients.Methods:We conducted a retrospective cohort study using data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. The primary outcomes were 28-day, 90-day, and 1-year mortality rates following ICU admission. Multivariate Cox proportional hazards regression models were used to estimate adjusted hazard ratios (HRs) with 95% confidence intervals (CIs). Subgroup analyses, Kaplan‒Meier survival curves, and restricted cubic spline models were employed to further evaluate the relationship between the GPR and mortality in AIS patients.Results:A total of 2,636 AIS patients were included in the study, with a mean age of 69.4 ± 15.6 years. The 1-year mortality rate was 36.8% (n = 969). After adjusting for confounders, compared with the first quartile (Q1, GPR ≤ 1.39), the 1-year mortality risks for the second quartile (Q2, 1.39 < GPR ≤ 1.74), third quartile (Q3, 1.74 < GPR ≤ 2.25), and fourth quartile (Q4, GPR ≥ 2.25) were HR = 1.17 (95% CI: 0.95–1.43, p = 0.132), HR = 1.42 (95% CI: 1.17–1.73, p < 0.001), and HR = 1.61 (95% CI: 1.33–1.96, p < 0.001), respectively. Similar trends were observed for 28-day and 90-day mortality. Kaplan‒Meier (KM) analysis revealed that groups with higher GPRs had higher mortality rates at 28 days, 90 days, and 1 year. Nonlinear analysis further confirmed the presence of an inflection point in the association between the GPR and 365-day mortality, which was identified at GPR = 2.75. At ratios less than this threshold, the risk of mortality increased significantly with increasing GPR (HR: 1.466; 95% CI: 1.239–1.735; p < 0.001). However, above this ratio, the association plateaued and was no longer statistically significant (HR: 0.899; 95% CI: 0.726–1.113; p = 0.095).