<![CDATA[Background: Malnutrition significantly impacts outcomes in elderly critically ill patients. The Geriatric Nutritional Risk Index (GNRI) and Prognostic Nutritional Index (PNI) are two established tools to assess nutritional status and predict mortality. However, their comparative prognostic performance in this population remains unclear. Objective: This study aimed to compare the predictive ability of GNRI and PNI for mortality among elderly ICU patients. Methods: We conducted a systematic search in PubMed, Scopus, and Web of Science for studies assessing GNRI and/or PNI in relation to mortality in ICU patients aged ≥60 years. We extracted mean and standard deviation values for survivors and non-survivors. Meta-analyses were conducted to calculate pooled mean differences (MD) with 95% confidence intervals (CI), and heterogeneity was evaluated using the I² statistic. Results: Eight studies involving 6,217 ICU patients were included. Both GNRI and PNI scores were significantly lower in non-survivors. The pooled MD for GNRI was −8.99 [95% CI −9.71 to −8.27] (I² = 86%), and for PNI was −4.45 [95% CI −4.94 to −3.96] (I² = 47%). GNRI showed a larger effect size but greater heterogeneity, while PNI results were more consistent. Most studies had low to moderate risk of bias based on the ROBINS-E tool. Conclusion: GNRI and PNI are valid prognostic tools for predicting mortality in elderly ICU patients. GNRI may provide stronger predictive value, whereas PNI offers more consistent prognostic performance.]]>
