Prognostic value of the Geriatric Nutritional Risk Index in sepsis-associated acute kidney injury: a retrospective cohort study

Background The Geriatric Nutritional Risk Index (GNRI) is a simple and objective tool for assessing nutritional status. It has shown prognostic value in patients with acute kidney injury (AKI). However, evidence in sepsis-associated AKI (S-AKI) remains limited, especially among patients receiving continuous renal replacement therapy (CRRT). Methods This retrospective cohort study included 773 critically ill S-AKI patients who received CRRT, using data from the publicly available Dryad database. Patients were stratified into tertiles based on GNRI values. Cox proportional hazards models were employed to assess the association between GNRI and 28-day and 90-day all-cause mortality, with the lowest tertile serving as the reference group. Kaplan–Meier survival analyses were used to compare cumulative mortality across GNRI strata. Results Among 773 patients, the 28-day and 90-day mortality rates were 61.7% and 73.0%, respectively. After adjusting for multiple confounders, higher GNRI scores were independently associated with lower all-cause mortality. Compared to the lowest GNRI group, the highest tertile showed significantly reduced mortality risks (28-day HR = 0.53; 95% CI: 0.47–0.75; p < 0.001; 90-day HR = 0.50; 95% CI: 0.41–0.63; p < 0.001). The association remained robust in subgroup analyses and was particularly pronounced in patients aged ≥65 years. Conclusion GNRI independently predicts short-term mortality in critically ill S-AKI patients on CRRT. It is simple, objective, and integrates nutritional and inflammatory status. It can assist early risk stratification and nutritional assessment.