Preoperative nutritional status defined by GLIM criteria and postoperative delirium in older patients receiving general anesthesia.

Abstract

Background: Preoperative malnutrition may reflect reduced physiological reserve and increased vulnerability to postoperative delirium (POD) in older surgical patients. However, evidence based on standardized diagnostic criteria remains limited. This study aimed to evaluate the association between Global Leadership Initiative on Malnutrition (GLIM)-defined malnutrition and POD among older patients undergoing surgery under general anesthesia.

Methods: This retrospective observational cohort study included consecutive patients aged ≥65 years who underwent elective or emergency surgery under general anesthesia between January 2022 and December 2025. Preoperative malnutrition was retrospectively classified according to the GLIM framework using available phenotypic and etiologic criteria. Patients with and without GLIM-defined malnutrition were matched 1:2 using propensity score matching. POD documented during hospitalization was identified from Confusion Assessment Method assessments, Confusion Assessment Method for the Intensive Care Unit assessments, or explicit physician diagnoses. The association between GLIM-defined malnutrition and POD was examined using conditional logistic regression stratified by matched set. Subgroup and sensitivity analyses were also performed.

Results: The matched cohort comprised 495 patients, including 165 with GLIM-defined malnutrition and 330 without malnutrition. All propensity score covariates achieved acceptable post-matching balance, with absolute standardized mean differences <0.10. POD occurred in 42 patients with malnutrition (25.5%) and 37 patients without malnutrition (11.2%; p < 0.001). GLIM-defined malnutrition was associated with higher odds of POD after multivariable adjustment (adjusted odds ratio, 2.36; 95% confidence interval, 1.36-4.10; p = 0.002). No significant interactions were identified across age, surgical category, American Society of Anesthesiologists class, or immediate postoperative intensive care unit admission subgroups. Comparable associations were observed in the unmatched cohort, alternative matched analyses, fixed observation windows, restricted-length-of-stay cohorts, and analyses using alternative POD definitions.

Conclusion: GLIM-defined malnutrition was associated with higher odds of POD during hospitalization in older patients undergoing surgery under general anesthesia. GLIM-based nutritional assessment may provide clinically useful information for perioperative delirium risk stratification.