Association between the geriatric nutritional risk index and all-cause mortality in elderly critically ill patients with ischemic stroke

Background Malnutrition is prevalent among elderly patients with ischemic stroke (IS) and is associated with poor outcomes. The Geriatric Nutritional Risk Index (GNRI), an objective tool incorporating serum albumin, weight, and height, may offer a practical assessment of nutritional risk. This study aimed to investigate the association between GNRI at ICU admission and 28-day, 90-day, and 180-day all-cause mortality in elderly critically ill patients with IS. Methods A retrospective cohort study was conducted using data from the MIMIC-IV database (2008–2022). Patients aged ≥ 65 years admitted to the ICU with a primary diagnosis of IS were included. Those with ICU stays ≤ 24 hours, missing admission data (height, weight, albumin), or non-first ICU admissions were excluded. The GNRI was calculated at admission and categorized into four groups: Q1 (high risk, GNRI < 82), Q2 (moderate risk, 82 ≤ GNRI < 92), Q3 (low risk, 92 ≤ GNRI ≤ 98), and Q4 (normal, GNRI > 98). The primary outcomes were 28-day, 90-day, and 180-day mortality. Kaplan-Meier survival analysis, log-rank tests, and multivariable Cox proportional hazards models were used to assess associations. Restricted cubic splines and subgroup analyses were performed to examine nonlinear relationships and effect modification. Results Among 2,771 included patients (median age 80 ± 8 years, 51.8% male), the 28-day, 90-day, and 180-day mortality rates were 18.6%, 28.5%, and 33.8%, respectively. Lower GNRI groups were associated with higher disease severity scores and increased mortality. Kaplan-Meier analysis demonstrated significantly worse survival in lower GNRI groups across all time points (log-rank P < 0.001). In multivariable Cox models, higher GNRI (as a continuous variable) was independently associated with a reduced risk of mortality at 28 days (HR 0.95, 95% CI 0.93–0.97), 90 days (HR 0.94, 95% CI 0.93–0.95), and 180 days (HR 0.97, 95% CI 0.96–0.99). Categorical analysis showed patients in Q4 had a significantly lower mortality risk compared to Q1 (for 28-day mortality: HR 0.22, 95% CI 0.17–0.27). A linear inverse relationship between GNRI and mortality risk was confirmed. Subgroup analyses indicated consistent associations across most patient strata, with diabetes being a significant effect modifier. Conclusion The GNRI at ICU admission is an independent predictor of short- and long-term all-cause mortality in elderly critically ill patients with ischemic stroke. Its objective and simple calculation makes it a potentially valuable tool for nutritional risk stratification and prognosis assessment in this vulnerable population, facilitating early intervention. Further prospective studies are warranted to validate these findings.