Omitting the First Step of GLIM Framework Yielded Better Accordance With PG-SGA in Chinese Ambulatory Cancer Patients: a Prospective Cross-sectional Study

Background and Aims: The Global Leadership Initiative on Malnutrition (GLIM) criteria is a new framework for diagnosing malnutrition in combination of phenotypic and etiologic criteria after nutrition screening using validated screening tools. The aim of this study was to evaluate the efficacy of malnutrition screening tool (MST), malnutrition universal screening tool (MUST) and nutritional risk screening 2002 (NRS2002) as the first step of GLIM framework in comparison to Patients-Generated Subjective Global Assessment (PG-SGA) in Chinese ambulatory cancer patients.Methods: A single-center prospective cross-sectional study was conducted. Nutritional screening and assessment were performed within 4h after admission to the hospital using a structured questionnaire including MST, MUST, NRS2002, PG-SGA and GLIM, with supplement information of calf circumference (CC) measurement and body composition measurement using bioelectrical impedance analysis (BIA). Malnutrition diagnosis made by GLIM framework using MST, MUST or NRS2002 as the first step or without screening step were compared to PG-SGA separately. Sensitivity, specificity, positive (PPV) and negative (NPV) predictive values and κ values were used to evaluate performance of the screening tools.Results: Of the 562 included patients, Of the participants 62.8% (355/562) were male and 37.2% (210/562) were female, with a male to female radio of 1.69:1. The median age of the patients was 59.0 years (range, 21-82y; interquels range 52.0-65.0y). From the 562 patients included in the study, 41.8% of patients were evaluated as malnutrition (PG-SGA≥4) and 11.9% were diagnosed as severe malnutrition (PG-SGA D). For GLIM criteria, omitting the screening step yielded fair accordance with PG-SGA in diagnosing malnutrition (κ=0.623) and severe malnutrition (κ=0.515). Using MUST as the first step of GLIM framework has better performance (κ=0.614; κ=0.515) than using MST (κ=0.504, κ=0.496) or NRS2002 (κ=0.363, κ=0.503) as the screening tool regardless of severity gradings.Conclusions: Using PG-SGA as the standard, GLIM framework omitting first step has better performance compared with using MST, MUST or NRS2002 as the screening tool. Among the screening tools validated to be used in the first step of GLIM framework, MUST may be the better choice for ambulatory cancer patients.