PD265
GLOBAL LEADERSHIP INITIATIVE ON MALNUTRITION (GLIM)-DEFINED MALNUTRITION SHOWED BETTER DISCRIMINATION THAN BODY MASS INDEX (BMI) IN PREDICTING IN-HOSPITAL MORTALITY: A RECEIVER OPERATING CHARACTERISTIC (ROC) ANALYSIS IN HOSPITALIZED ADULTS
V. Fuchs Tarlovsky1,*, A. Loza2, K. Alvarez Altamirano1, R. Flores-Mejia3
1Nutrition and metabolism Research Unit, 2Clinical Nutrition, Hospital General de Mexico, Dr. Eduardo Liceaga, 3Escuela Superior de Medicina, Instituto Politécnico Nacional, Mexico City, Mexico
Rationale: Rationale: Body mass index (BMI) is commonly used to assess nutritional status but has limited ability to discriminate mortality risk in acute inflammatory conditions. The Global Leadership Initiative on Malnutrition (GLIM) may better capture metabolic vulnerability.
Objective: To evaluate the association between GLIM-defined malnutrition and in-hospital mortality, and to compare its prognostic performance with BMI.
Methods: Methods: Retrospective cohort of adults hospitalized with COVID-19. Malnutrition was assessed using GLIM criteria within 72 hours of admission. Age-adjusted logistic regression and ROC curve analysis were performed.
Results: Results: Among 89 patients, 79.8% were classified as malnourished by GLIM, while only 9% had low BMI. GLIM-defined malnutrition was independently associated with in-hospital mortality (OR 39.3, 95% CI 4.61–334.63; p=0.0008), whereas BMI was not. Weight loss was also independently associated with mortality (OR 4.16, 95% CI 1.45–11.92; p=0.008). The GLIM-based model showed better discrimination (AUC 0.877) than BMI (AUC 0.769) and weight loss alone (AUC 0.808).
Conclusion: Conclusion: GLIM-defined malnutrition outperformed BMI in predicting in-hospital mortality, supporting the use of multidimensional nutritional assessment in acute inflammatory disease.
Disclosure of Interest: None declared
