RESUMO
BACKGROUND & AIMS: Acutely ill older adults are at higher risk of malnutrition. This study aimed to explore the applicability and accuracy of the GLIM criteria to diagnose malnutrition in acutely ill older adults in the emergency ward (EW). METHODS: We performed a retrospective secondary analysis, of an ongoing cohort study, in 165 participants over 65 years of age admitted to the EW of a Brazilian university hospital. Nutrition assessment included anthropometry, the Simplified Nutritional Assessment Questionnaire (SNAQ), the Malnutrition Screening Tool (MST), and the Mini-Nutritional Assessment (MNA). We diagnosed malnutrition using GLIM criteria, defined by the parallel presence of at least one phenotypic [nonvolitional weight loss (WL), low BMI, low muscle mass (MM)] and one etiologic criterion [reduced food intake or assimilation (RFI), disease burden/inflammation]. We used the receiver operating characteristic (ROC) curves and Cox and logistic regression for data analyses. RESULTS: GLIM criteria, following the MNA-SF screening, classified 50.3% of participants as malnourished, 29.1% of them in a severe stage. Validation of the diagnosis using MNA-FF as a reference showed good accuracy (AUC = 0.84), and moderate sensitivity (76%) and specificity (75.1%). All phenotypic criteria combined with RFI showed the best metrics. Malnutrition showed a trend for an increased risk of transference to intensive care unit (OR = 2.08, 95% CI 0.99, 4.35), and severe malnutrition for in-hospital mortality (HR = 4.23, 95% CI 1.2, 14.9). CONCLUSION: GLIM criteria, following MNA-SF screening, appear to be a feasible approach to diagnose malnutrition in acutely ill older adults in the EW. Nonvolitional WL combined with RFI or acute inflammation were the best components identified and are easily accessible, allowing their potential use in clinical practice.