Association of leuko-glycemic index with mortality in ICU patients with Acute kidney injury: A retrospective multicenter cohort study

Abstract

Background Acute kidney injury (AKI) complicates the course of critical illness and carries high short- and long-term mortality rates; however, reliable early prognostic markers remain limited. The leuko-glycemic index (LGI), the product of white blood cell counts and blood glucose, has shown prognostic value in other acute conditions; however, its role in intensive care unit (ICU) patients with AKI remains unexplored. Methods In this multicenter retrospective cohort study, we analyzed 112,235 adult ICU admissions from the MIMIC-IV (n = 54,840) and eICU-CRD (n = 57,395) databases. Patients with multiple ICU stays and those aged <18 years were excluded. The LGI was calculated within 24 h of admission and stratified into quartiles (Q1–Q4). The primary endpoints were 30- and 90-day in-hospital mortality, and the secondary endpoint was overall in-hospital mortality. Kaplan–Meier survival curves, multivariable Cox proportional hazards models, and restricted cubic splines (RCS) assess the association between LGI and outcomes, with subgroup and interaction analyses for key comorbidities. Results Higher LGI quartiles were associated with progressively worse survival (p < 0.0001). In the fully adjusted models, each unit increase in LGI conferred a small but significant rise in 30-day (HR 1.01; 95% CI 1.01–1.01) and 90-day mortality (HR 1.01; 95% CI 1.01–1.01), while Q4 versus Q1 yielded HRs of 1.22 (95% CI 1.13–1.31) and 1.21 (95% CI 1.13–1.30), respectively. RCS demonstrated a nonlinear relationship, and the associations persisted across subgroups. Conclusion LGI may be an independent biomarker associated with short-term mortality in ICU patients with AKI. Its potential clinical relevance warrants further investigation and validation in larger cohorts.