Renal-Metabolic Reserve Index for Cardiorenal Risk Stratification in Critically Ill Patients With Sepsis

Objective: To evaluate whether the admission renal-metabolic reserve index (RMRI), defined as (blood urea nitrogen + glucose)/albumin, was associated with in-hospital mortality in intensive care unit (ICU) patients with sepsis and to compare its prognostic performance with BAR and GAR.Methods: This retrospective cohort study used MIMIC-IV v3.1. Adults in the first ICU stay who met Sepsis-3 criteria near ICU admission and had blood urea nitrogen, glucose, and albumin measured from 6 hours before to 24 hours after ICU admission were eligible. Patients with end-stage renal disease or chronic dialysis were excluded. The primary outcome was in-hospital mortality. Logistic regression, restricted cubic splines, AUROC analysis, DeLong testing, subgroup analyses, and sensitivity analyses were performed.Results: The locked analytic cohort included 5,736 patients, and 1,251 (21.8%) died in hospital. RMRI was higher in non-survivors than survivors (64.7 [47.9-91.7] vs 52.1 [40.3-70.5]). RMRI remained associated with in-hospital mortality after multivariable adjustment (adjusted odds ratio per 1-SD, 1.33; 95% CI, 1.25-1.43; P < 0.001). Quartile 4 had an adjusted odds ratio of 2.35 (95% CI, 1.91-2.90; P < 0.001) versus quartile 1. Restricted cubic splines showed non-linearity (P = 0.001). AUROC increased from 0.723 for the clinical base model to 0.735 after RMRI addition, to 0.729 after GAR addition, and to 0.744 after BAR addition.Conclusion: Admission RMRI was independently associated with in-hospital mortality in ICU patients with sepsis and improved discrimination beyond the clinical base model and GAR. BAR remained the strongest comparator.