de Miranda Bruna Luisa Gomes, Silva Flavia Moraes, de Sousa Iasmin Matias, Bertuleza Liliane Nunes, Xavier Jadson Gomes, Rüegg Rodrigo Albert Baracho, Gonzalez Maria Cristina, Fayh Ana Paula Trussardi
Graduate Program in Nutrition, Health Sciences Center, Federal University of Rio Grande do Norte, Natal, Brazil.
Nutrition Science Graduate Program of Federal University of Health Science of Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil.
Nutr Clin Pract. 2025 Oct;40(5):1166-1177. doi: 10.1002/ncp.11295. Epub 2025 Apr 16.
The Global Leadership Initiative on Malnutrition (GLIM) criteria for diagnosing malnutrition were established to provide a standardized approach for diagnosing malnutrition in clinical practice using a nutrition screening tool (NST) as the first step for this process. This study aimed to compare the complementarity of NSTs with the GLIM criteria for malnutrition diagnosis in patients with cancer.
Hospitalized patients with different cancer types were evaluated in a prospective cohort study in which they were initially screened using the Patient-Generated Subjective Global Assessment (PG-SGA), Protocol for Nutritional Risk in Oncology (PRONTO), Malnutrition Universal Screening Tool, Nutritional Risk Screening 2002, Malnutrition Screening Tool, and NutriScore for nutrition risk. Malnutrition diagnosis involved phenotypic and etiological criteria as proposed by the GLIM. Complementarity of NST to GLIM criteria was evaluated by calculating accuracy metrics and investigating association with 12-month mortality.
Nutrition risk ranged from 14.8% (NutriScore) to 82.8% (PRONTO) and frequency of malnutrition from 13.8% (with NutriScore) to 88.9% (with PG-SGA). NutriScore presented the lowest negative predictive value (25.1%) whereas PG-SGA presented the highest (58.32%). Regardless of the NST applied, the risk of malnutrition and diagnosis of malnutrition according to the GLIM criteria, combined or isolated, increased the risk of 12-month mortality.
All NSTs presented low negative predictive value when their complementarity to GLIM criteria for malnutrition diagnosis was tested. Indeed, patients "at risk" presented similar increased risk of 12-month after discharge mortality in comparison with those at risk and malnourished by the GLIM criteria when all NSTs were applied.
全球营养不良领导倡议（GLIM）制定的营养不良诊断标准，旨在为临床实践中使用营养筛查工具（NST）作为该过程的第一步来诊断营养不良提供标准化方法。本研究旨在比较NST与GLIM标准在癌症患者营养不良诊断中的互补性。
在一项前瞻性队列研究中，对不同癌症类型的住院患者进行评估，最初使用患者主观整体评定法（PG-SGA）、肿瘤营养风险评估方案（PRONTO）、营养不良通用筛查工具、营养风险筛查2002、营养不良筛查工具和营养评分来评估营养风险。营养不良诊断采用GLIM提出的表型和病因学标准。通过计算准确性指标并调查与12个月死亡率的关联，评估NST与GLIM标准的互补性。
营养风险范围从14.8%（营养评分）到82.8%（PRONTO），营养不良发生率从13.8%（营养评分）到88.9%（PG-SGA）。营养评分的阴性预测值最低（25.1%），而PG-SGA的阴性预测值最高（58.32%）。无论应用哪种NST，根据GLIM标准联合或单独诊断的营养不良风险和诊断均增加了12个月死亡率的风险。
当测试所有NST与GLIM营养不良诊断标准的互补性时，它们的阴性预测值均较低。事实上，与根据GLIM标准有风险和营养不良的患者相比，当应用所有NST时，“有风险”的患者出院后12个月的死亡风险相似。
