Low geriatric nutritional risk index predicts poor prognosis in patients with cirrhosis: a retrospective study

AIM: Malnutrition, which increases the risk of liver disease-related events and mortality, is a serious complication in cirrhosis. This study aimed to investigate whether the geriatric nutritional risk index (GNRI) could predict the long-term prognosis in patients with cirrhosis. METHODS: We retrospectively evaluated 266 patients with cirrhosis and classified them into two groups based on baseline GNRI scores: risk (≤98, n = 104) and no-risk groups (>98, n = 162). The cumulative survival rates were compared between the two groups in patients with compensated and decompensated cirrhosis, respectively. Cox proportional hazards regression analysis was used to identify significant and independent factors associated with mortality. RESULTS: The median observation period was 54.9 (33.6–61.7) months and 65 (24.4%) liver disease-related deaths occurred during the follow-up period. The GNRI scores significantly and inversely correlated with Child-Pugh score (r = −0.579), model for end-stage liver disease score (r = −0.286), and Mac-2 binding protein glycosylation isomer (r = −0.494). Multivariate analysis identified low GNRI as a significant and independent factor associated with mortality [overall cohort: hazard ratio (HR), 0.926; p < 0.001; compensated cirrhosis: HR, 0.947; p = 0.003; decompensated cirrhosis: HR, 0.923; p < 0.001]. The risk group demonstrated significantly lower cumulative survival rates than the no-risk group in overall cohort, and patients with compensated and decompensated cirrhosis (p < 0.001, <0.001, and = 0.013, respectively). CONCLUSION: Low GNRI was associated with poor long-term prognosis in both patients with compensated and decompensated cirrhosis. Therefore, the GNRI is a simple and useful tool for predicting prognosis and modifying the nutritional status in patients with cirrhosis.