Geriatric nutritional risk index is a strong predictor of mortality in hemodialysis patients: data from the Riscavid cohort

Abstract

Malnutrition is a common complication in hemodialysis (HD) patients and it is related to morbidity and mortality. Although a gold standard method for diagnosis of malnutrition is not available, serum albumin, body weight and height are commonly used and are included in the calculation of the Geriatric nutritional risk index (GNRI). Recently the association between GNRI and mortality in chronic HD patients has been documented in Asian populations. Our aim was to examine the relative reliability of the GNRI as a mortality and morbidity predictor in an Italian HD cohort. We prospectively examined the GNRI of 753 maintenance HD patients aged 65.7 ± 14.1 years, 457 males, included in the Riscavid cohort, and followed them up for 84 months. Predictors for all-cause death were examined using Kaplan-Meier and Cox proportional-hazards analyses. Low GNRI was significantly associated with signs of wasting, i.e. low Body mass index, hypoalbuminemia, low normalized protein catabolic rate. Patients within the lowest GNRI quartile had a significantly lower survival rate than those in the 2nd to 4th quartile (p < 0.001). Multivariate Cox proportional-hazards analysis demonstrated that the lowest quartile of GNRI was a significant predictor of case mix adjusted all-cause mortality (HR 1.72; CI 1.35-2.18, p < 0.001). These results demonstrate that low GNRI (<92) is associated with malnutrition and is a strong predictor of overall mortality in HD patients. However, cardiovascular events did not differ among the GNRI quartiles. A low GNRI score can be considered a simple and reliable marker of malnutrition and predictor for mortality risk in Caucasian HD patients.