PD036 - AGREEMENT BETWEEN GLIM CRITERIA USING PHYSICAL EXAMINATION OR CALF CIRCUMFERENCE AND PG-SGA IN HOSPITALIZED PATIENTS WITH CANCER

PD036

AGREEMENT BETWEEN GLIM CRITERIA USING PHYSICAL EXAMINATION OR CALF CIRCUMFERENCE AND PG-SGA IN HOSPITALIZED PATIENTS WITH CANCER

Y. Lee^1,*, J. Kim¹, K.-A. Ryu¹, J. Kim¹, J. M. Oh¹, W. Y. Park¹, S. Park¹, D. Yoo¹, H. Ahn¹, S. Yi¹, Y. Jung¹, G. Oh¹, S. Y. Kim¹

¹Department of Clinical Nutrition, National Cancer Center Korea, Goyang, Korea, Republic Of

Rationale: This study compared the Global Leadership Initiative on Malnutrition (GLIM) using physical examination (GLIM-PE) and calf circumference (GLIM-CC) with the Patient-Generated Subjective Global Assessment (PG-SGA).

Methods: This prospective study included 260 hospitalized patients with cancer at the National Cancer Center, Korea (February 13–24, 2026). Pediatric, non-malignant, and hospice patients were excluded. Nutritional status was assessed using PG-SGA and GLIM. GLIM-PE was based on physical examination, and GLIM-CC used Asian Working Group for Sarcopenia 2019 calf circumference cut-offs (<34 cm for men, <33 cm for women). Agreement (kappa), sensitivity, specificity, and receiver operating characteristic (ROC) curve analyses were performed, with optimal cut-offs determined by the Youden index. Subgroup analyses were conducted using the Malnutrition Screening Tool for Cancer Patients (MSTC), a screening tool developed at our institution that classifies patients into low- and high-risk groups.

Results: Malnutrition prevalence was 37.3% by PG-SGA and 51.9% by GLIM-PE, and 56.5% by GLIM-CC. Agreement with PG-SGA was higher for GLIM-PE than for GLIM-CC (κ=0.634 vs. 0.449). Sensitivity and specificity were 94.8% and 73.6% for GLIM-PE, and 87.6% and 62.0% for GLIM-CC, respectively. Calf circumference showed acceptable discrimination for GLIM-PE-defined malnutrition (AUC 0.805, 95% CI 0.753–0.858), with optimal cut-offs of 34.5 cm for men and 33.0 cm for women. For PG-SGA-defined malnutrition, AUC was 0.728. In MSTC subgroup analysis, GLIM sensitivity and specificity were 90.2% and 72.7% in the high-risk group, and 98.2% and 73.7% in the low-risk group.

Conclusion: GLIM-PE showed higher agreement with PG-SGA than GLIM-CC. Calf circumference demonstrated acceptable discrimination and may serve as a practical alternative for assessing muscle mass when standard methods are limited.

Disclosure of Interest: None declared