ABSTRACT
BACKGROUND & AIMS: In 2018, the Global Leadership Initiative on Malnutrition (GLIM) presented the criteria for malnutrition diagnosis; reports about the proportion of malnutrition patients meeting these GLIM criteria in acute care hospitals, however, remain very limited. The relationship between GLIM criteria and existing nutritional indicators, patterns for hospitalization, and malnutrition severity is also unclear. This study aims to investigate this relationship. METHODS: Our study included 490 patients (aged 69.5 ± 16.0 years, 45% women) who had been hospitalized during a specific month in an acute care hospital, and for whom we could assess nutritional status according to GLIM criteria. We analyzed the cut-off value on the MNA-SF score and grip strength (GS) for GLIM criteria-defined malnutrition severity grading with receiver operating characteristic (ROC) analysis. We extracted factors relating to malnutrition by multivariate logistic regression analysis. RESULTS: In all, 33% of patients met the GLIM criteria for malnutrition. Malnutrition severity was correlated with age, GS and emergency hospitalization (p < 0.001, respectively). For the MNA-SF score, we determined a cut-off value of point 9 for severe malnutrition [area under curve (AUC) 0.92, p < 0.001], and of point 11 for moderate malnutrition [range 0-14 (AUC 0.90, p < 0.001)]. We were able to identify 98% of patients defined on GLIM criteria as malnourished, with the MNA-SF score. Using the HG, we could also evaluate the malnutrition grading in men younger than 70 years and women older than 70 years (men younger than 70 years: cut off for severe malnutrition, 20 kg, AUC 0.82; for moderate malnutrition, 29 kg, AUC 0.83; women older than 70 years: for severe malnutrition, 11 kg, AUC 0.78; for moderate malnutrition, 14.5 kg, AUC 0.72; p < 0.001, respectively). We extracted emergency hospitalization as an independent factor relating to malnutrition, adjusted for age and sex (odds ratio: 2.99; 95% CI: 2.00-4.47; p < 0.001). CONCLUSIONS: Using the MNA-SF for GLIM criteria screening, we identified malnourished patients with high accuracy, and GS was also a reliable nutritional assessment. Emergency hospitalization patients were at a high risk for malnutrition.
