Association between the glucose-to-albumin ratio and short-term in-hospital mortality in critically ill patients with premature acute myocardial infarction: A retrospective study from the eICU-CRD.

Premature acute myocardial infarction (PAMI) is a high-risk condition among critically ill patients, accompanied by metabolic and inflammatory disturbances. The glucose-to-albumin ratio (GAR) reflects both stress hyperglycemia and nutritional-inflammatory status, but evidence regarding its prognostic relevance in critically ill patients with PAMI remains limited. This study aimed to investigate the prognostic value of GAR for short-term in-hospital mortality in this population. We conducted a retrospective cohort study using data from the eICU Collaborative Research Database. Patients with PAMI were divided into 2 groups according to the X-tile-derived exploratory cutoff of 5.80. Cox proportional hazards regression models were constructed with stepwise adjustment for potential confounders to assess the association between GAR and 28-day in-hospital mortality. Restricted cubic spline analysis was used to evaluate the dose-response relationship between GAR and mortality. Kaplan-Meier survival curves were generated to compare survival between groups, and subgroup and interaction analyses were performed across major clinical strata. Receiver operating characteristic curve analysis was used to assess the exploratory discriminatory ability and adjunctive prognostic information of GAR. A total of 622 patients with PAMI were included. Patients in the high GAR group (>5.80) had a higher 28-day in-hospital mortality rate than those in the low GAR group (≤5.80; 11.9% vs 1.6%). In fully adjusted Cox regression models, each 1-unit increase in GAR was associated with an increased risk of 28-day in-hospital mortality (hazard ratio 1.19, 95% confidence interval 1.07-1.33; P = .001). Restricted cubic spline analysis showed a significant nonlinear association between GAR and short-term mortality. Subgroup analyses showed no significant interactions between GAR and other covariates for 28-day in-hospital mortality. Receiver operating characteristic analysis suggested that GAR had exploratory discriminatory ability for 28-day in-hospital mortality, with an area under the curve of 0.779 (95% confidence interval 0.718-0.839). The combination of GAR and the Sequential Organ Failure Assessment score further improved overall adjunctive prognostic information. Elevated GAR was associated with increased 28-day in-hospital mortality in critically ill patients with PAMI. GAR may provide adjunctive prognostic information for short-term in-hospital mortality, and its combination with the Sequential Organ Failure Assessment improved the accuracy of prognostic assessment.