Correlation between the glucose-to-potassium ratio and short-term death rate in patients with acute kidney injury: A retrospective cohort study based on MIMIC-IV database

Abstract

Electrolyte disorders and fluid imbalance are common in acute kidney injury (AKI), but the glucose-to-potassium ratio (GPR)’s link to short-term prognosis remains unclear. This study explored the association between GPR and 30-day mortality in AKI patients using data from the Medical Information Mart for Intensive Care IV database. The potential nonlinear correlation between GPR and death rate during hospitalization was explored via the restricted cubic spline (RCS) models. The consistency of results across different patient subgroups was assessed through subgroup analyses. After screening, 38,744 AKI patients were included. After adjusting for all confounders, the Cox proportional hazard model analysis demonstrated a significant link between continuous GPR values and short-term risk of death in AKI patients (hazard ratio (95% confidence interval) = 1.088 (1.053–1.125), p < 0.001). Moreover, in comparison to the low-GPR group, patients in the high-GPR group had a notably elevated death rate, whereas those in the moderate-GPR group exhibited a significantly reduced risk of death. The RCS analysis unraveled a distinct U-shaped link between GPR and short-term death rate, with a turning point at GPR = 1.493. Additionally, subgroup analyses demonstrated consistent results across all patient subgroups. This study unraveled that GPR was notably linked to the short-term death rate in AKI patients. Both elevated and reduced levels of GPR can increase the risk of death in these patients. As a readily accessible biomarker, GPR can offer insights into the risk stratification and prognosis assessment of AKI patients.