Elevated glucose-to-platelet ratio predicts short- term and long-term mortality in critically ill patients with acute...

Acute ischemic stroke (AIS) continues to be a major cause of morbidity and mortality globally. In this study, we investigated whether the glucose-to-platelet ratio (GPR) provides prognostic information in critically ill patients with AIS, and we further tested the robustness of our findings using an independent validation cohort. We conducted a retrospective cohort analysis, using the MIMIC-IV database to derive the model and the eICU Collaborative Research Database (eICU-CRD) as an independent cohort for external validation. Patients were stratified into quartiles according to early ICU GPR. Cox proportional hazards models, restricted cubic spline (RCS) analyses, and Kaplan-Meier survival curves and receiver operating characteristic curve analyses comparing GPR with blood glucose and platelet count alone were used to assess the association between GPR and 28-day and 365-day mortality. In the MIMIC-IV cohort (n = 2,855), higher GPR was significantly associated with increased mortality. In the fully adjusted multivariable Cox model, GPR, analysed as a continuous variable, was associated with higher hazards of both 28-day mortality (HR = 1.06, 95% CI 1.02-1.11; P = 0.009) and 365-day mortality (HR = 1.08, 95% CI 1.04-1.12; P < 0.001). RCS analyses indicated a linear association between GPR and mortality risk (P for non-linearity > 0.05), and Kaplan-Meier curves demonstrated lower survival in patients with higher GPR (log-rank P < 0.001). In ROC analyses, GPR showed higher AUCs than blood glucose or platelet count alone for both 28-day mortality (0.633 vs. 0.620 and 0.557) and 365-day mortality (0.632 vs. 0.625 and 0.573). In the independent eICU-CRD cohort, the relationship between GPR at admission and mortality was confirmed. This association persisted when GPR was analyzed as a continuous measure (HR 1.28, 95% CI 1.19-1.38) and when evaluated in categories (highest vs. lowest group: HR 1.74, 95% CI 1.35-2.24). The restricted cubic spline curves and Kaplan-Meier survival trends closely mirrored those obtained in MIMIC-IV. Moreover, subgroup analyses did not identify any meaningful effect modification across the main clinical subgroups. An elevated early ICU GPR serves as an independent predictor of both short-term and long-term mortality in critically ill AIS patients. Including GPR in clinical evaluations might aid early risk stratification and facilitate timely, targeted interventions.