Background <p>Emergence delirium (ED) is a common complication in elderly surgical patients. While nutritional risk is prevalent, its relationship with ED remains unclear. This study evaluated the association between preoperative nutritional risk, assessed by the objective Geriatric Nutritional Risk Index (GNRI), and ED in elderly patients undergoing non-cardiac surgery.</p> Methods <p>This secondary analysis of a prospective observational study included patients aged 65–90 years undergoing elective non-cardiac surgery. Preoperative nutritional risk was assessed using GNRI calculated from serum albumin and body weight. Nutritional risk was defined as GNRI ≤ 98. The primary outcome was ED assessed by the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) in the Post-Anesthesia Care Unit. Multivariable logistic regression was used to identify independent risk factors.</p> Results <p>Of 915 patients, 20.0% (183/915) were at nutritional risk based on GNRI. The incidence of ED was significantly higher in the nutritional risk group compared to the non-nutritional-risk group (49.7% vs. 33.9%, <i>P</i> &lt; 0.001). After adjusting for confounders, nutritional risk assessed by GNRI was identified as an independent risk factor for ED (OR 1.80, 95% CI 1.25–2.59, <i>P</i> = 0.002) and was associated with prolonged postoperative hospital stay (HR 0.83, <i>P</i> = 0.026). No significant independent association was found between nutritional risk and postoperative delirium or non-delirium complications.</p> Conclusions <p>Preoperative nutritional risk identified by the objective GNRI is independently associated with an increased risk of emergence delirium and prolonged hospitalization in elderly patients. GNRI serves as a valuable, specific screening tool for risk stratification in this population.</p> Clinical trial registrations <p>Chinese Clinical Trial Registry (chictr.org.cn, ChiCTR-OOC-17012734, September 19, 2017)</p>