Agreement between the Global Leadership Initiative on Malnutrition (GLIM) criteria and the Patient-Generated Subjective Global Assessment (PG-SGA) for diagnosing malnutrition in patients with solid tumors admitted for treatment
Clinical Nutrition ESPEN · 2026-06 · 73 · 103286
DOI: 10.1016/j.clnesp.2026.103286 ↗Kurz gefasst
Fachartikel in der Zeitschrift Clinical Nutrition ESPEN, 2026-06. Verfasst von Micheline Tereza Pires de Souza und 6 weiteren Autor:innen. Beteiligte Einrichtung: LIM 16, Serviço de Nefrologia, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil. Electronic address: micheline.pires@gmail.com. Hauptthemen laut MEDLINE: Malnutrition, Nutrition Assessment, Neoplasms. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: State of Sao Paulo Research Foundation.
Themen: Mangelernährung
Abstract (Originalsprache)
BACKGROUND: Malnutrition in patients with cancer is linked to worse clinical outcomes, including reduced treatment tolerance, prolonged hospital stays, diminished quality of life, and increased mortality. Although outpatient cancer treatments are prevalent (90%), comprehensive nutritional screening is often inadequate in these settings. The Patient-Generated Subjective Global Assessment (PG-SGA) is an effective tool for evaluating nutritional status. The Global Leadership Initiative on Malnutrition (GLIM) offers new diagnostic criteria, yet it is less explored in outpatient settings. This study aims to compare the diagnostic agreement of GLIM and PG-SGA for identifying malnutrition in outpatients with solid tumors. METHODS: This cross-sectional study was conducted at the Instituto do Cancer do Estado de São Paulo (ICESP). We included adults with histologically confirmed solid tumors, an ECOG Performance Status (ECOG-PS) of ≤3, and no recent treatment who began cancer treatment and underwent nutritional assessment within 14 days. Patients already in follow-up were excluded. The collected data included clinical information and laboratory tests (albumin, C-reactive protein). Nutritional assessments comprised anthropometry, multifrequency electrical bioimpedance (BIA), PG-SGA (gold standard), and GLIM criteria. Cachexia was diagnosed using Fearon's criteria. Statistical analyses included medians, interquartile ranges, Wilcoxon, Kruskal-Wallis, chi-square, Fisher's exact tests, and kappa statistics to evaluate the agreement between GLIM and PG-SGA, with significance set at p < 0.05. RESULTS: From April 22, 2015, to September 15, 2017, 13,386 patients were screened, 1200 completed the study, and 689 underwent comprehensive nutritional. The cohort had a median age of 60 years (IQR: 51-67), with 50.4% of the participants being male. Common cancer types included breast (26.6%), male genital (21.6%), gastrointestinal (18.7%), and head and neck (10.2%). Most patients (53.7%) were in stages I or II, and 94.9% had ECOG scores of 0 or 1. Nutritional assessments indicated that 58.4% were at nutritional risk (PG-SGA score ≥4), with malnutrition detected in 56.2% by PG-SGA and 35.6% by GLIM. Malnourished patients by GLIM and PG-SGA exhibited more advanced disease stages, poorer clinical performance, lower albumin levels, and higher CRP levels. There was moderate agreement between GLIM and PG-SGA (kappa 0.439). GLIM showed low sensitivity, 55.8 (50.7-60.8), but high specificity, 90.4 (86.5-93.5) for diagnosing malnutrition. Patients identified as malnourished by GLIM had higher ECOG status and weight loss, while those identified by PG-SGA had higher body mass index (BMI). CONCLUSION: PG-SGA and GLIM criteria presented moderate agreement in this cohort of adult outpatients with cancer initiating anticancer treatment. PG-SGA had greater sensitivity for detecting early nutritional impairment among outpatients with cancer, highlighting the need to refine GLIM applications to improve early malnutrition detection.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- LIM 16, Serviço de Nefrologia, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil. Electronic address: micheline.pires@gmail.com.
- Post-graduate Program in Nutrition and Food, Federal University of Pelotas, Pelotas, Brazil.
- LIM 66, Serviço de Geriatria, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil.
- LIM 16, Serviço de Nefrologia, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil.
- Serviço de Nefrologia, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brazil.
- LIM 12, Faculdade de Medicina da Universidade de São Paulo, São Paulo, SP, Brazil.
Förderung
- State of Sao Paulo Research Foundation
Zitieren
APA
de Souza, M. T. P., Gonzalez, M. C., Gil, L. A., Cardenas, T. C., Caires, R. A., Burdmann, E. A., & E Silva, V. T. D. C. (2026). Agreement between the Global Leadership Initiative on Malnutrition (GLIM) criteria and the Patient-Generated Subjective Global Assessment (PG-SGA) for diagnosing malnutrition in patients with solid tumors admitted for treatment. Clinical Nutrition ESPEN, 73, 103286. https://doi.org/10.1016/j.clnesp.2026.103286
Vancouver
de Souza MTP, Gonzalez MC, Gil LA, Cardenas TC, Caires RA, Burdmann EA, et al. Agreement between the Global Leadership Initiative on Malnutrition (GLIM) criteria and the Patient-Generated Subjective Global Assessment (PG-SGA) for diagnosing malnutrition in patients with solid tumors admitted for treatment. Clinical Nutrition ESPEN. 2026;73:103286. doi:10.1016/j.clnesp.2026.103286
BibTeX
@article{deSouza2026,
author = {de Souza, Micheline Tereza Pires and Gonzalez, Maria Cristina and Gil, Luiz A and Cardenas, Thais de Campos and Caires, Renato A and Burdmann, Emmanuel A and E Silva, Verônica Torres da Costa},
title = {Agreement between the Global Leadership Initiative on Malnutrition (GLIM) criteria and the Patient-Generated Subjective Global Assessment (PG-SGA) for diagnosing malnutrition in patients with solid tumors admitted for treatment},
journal = {Clinical Nutrition ESPEN},
year = {2026},
volume = {73},
pages = {103286},
doi = {10.1016/j.clnesp.2026.103286}
}
EUVIMED