Abstract
OBJECTIVES: Geriatric Nutritional Risk Index (GNRI) is a new and simple index recently introduced to assess nutritional status, and its predictive value for clinical outcomes has been demonstrated in patients with chronic kidney disease. However, the association between the GNRI and prognosis has not been evaluated so far in patients with acute kidney injury (AKI), especially in those receiving continuous renal replacement therapy (CRRT). METHODS: A total of 1096 patients with severe AKI initiating CRRT were identified for inclusion in this retrospective observational study. Patients were divided into three groups according to GNRI tertiles, with tertile 1 as the reference. The outcomes of interest were the 28- and 90-days of all-cause mortality. The associations between GNRI and clinical outcomes were estimated using multivariate Cox proportional hazards model analysis. RESULTS: The overall mortality rates at 28- and 90-days were 61.6% (675/1096) and 71.5% (784/1096), respectively. After adjusting for multiple confounding factors, GNRI was identified as an independent prognostic factor for 28-days all-cause mortality (HR, 0.582; 95% CI, 0.467-0.727; p < .001 for tertile 3 vs. tertile 1) as well as 90-days all-cause mortality (HR, 0.540; 95% CI, 0.440-0.661; p < .001 for tertile 3 vs. tertile 1). The observed inverse associations were robust across subgroup analysis, and were more pronounced in elderly patients over 65 years of age. Finally, incorporating GNRI in a model with established risk factors might significantly improve its predictive power for the short-term death. CONCLUSIONS: GNRI is considered to be a useful prognostic factor in patients with severe AKI initiating CRRT, especially in elderly patients.