Simple muscle assessment improves GLIM-based malnutrition detection and prognostic stratification in hospitalized...

Malnutrition in hospitalized patients is prevalent and associated with adverse clinical outcomes. However, the widely used Nutritional Risk Screening 2002 (NRS-2002) often overlooks a substantial proportion of patients who meet the Global Leadership Initiative on Malnutrition (GLIM) criteria for malnutrition, leading to delayed nutritional intervention. Therefore, this study aims to analyze the characteristic of these overlooked patients and explore a more scientific screening strategy.

This study analyzed data from the China Nutrition Basic Data 2022 (CNBD 2022) project, a nationwide, multicenter observational study. NRS-2002 and the GLIM criteria were performed for all patients. An Enhanced Screening strategy (ESS) which supplemented NRS-2002 with handgrip strength (HGS) and calf circumference (CC) for NRS-2002 negatives was evaluated. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Kappa value of NRS-2002 and ESS against the GLIM criteria were compared. Logistic regression was used to assess associations between NRS-2002 (or ESS) /GLIM combinations and clinical outcomes (including length of hospital stay (LOS) and 30-day mortality).

A total of 51672 patients were included. Among the 8776 malnourished patients overlooked by NRS-2002, 83.4% presented with phenotypes involving low muscle mass. ESS improved screening sensitivity from 54.9% to 73.6%. Compared to the reference group of patients negative by both NRS-2002 and GLIM criteria [NRS-2002(-)/GLIM(-)], the subgroup ESS(+)/GLIM(-) showed a significantly increased risk of LOS> mean (adjusted OR = 1.361). This risk was also higher than that of patients ESS(-)/GLIM(+) (adjusted OR = 1.124).

The limited sensitivity of NRS-2002 for GLIM-defined malnutrition was mainly due to its lack of direct muscle assessment. For patients with NRS-2002 scores<3, integrating simple muscle assessment could significantly improve screening sensitivity and prognostic stratification in adult inpatients.