GLIM-defined malnutrition and overall survival in cancer patients: A meta-analysis

BackgroundMalnutrition defined by the Global Leadership Initiative on Malnutrition (GLIM) has been associated with cancer mortality, but the effect is limited and inconsistent. We performed this meta-analysis aiming to assess this relationship in patients with cancer. MethodsWe systematically searched Embase, PubMed, Web of Science, Cochrane, CINAHL, CNKI, Wanfang, and VIP databases from January 1, 2019, to July 1, 2022. Studies evaluating the prognostic effect of GLIM-defined malnutrition on cancer survival were included. A fixed-effect model was fitted to estimate the combined hazard ratio (HR) with a 95% CI. Heterogeneity of studies was analyzed using the I-2 statistic. Quality assessment were performed using the Newcastle-Ottawa Scale (NOS) and the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) tool. ResultsThe search strategy identified 4378 articles in all databases combined. Nine studies (8829 patients) meeting the inclusion criteria were included for quantitative analysis. Meta-analysis revealed significant associations between GLIM-defined pooled malnutrition (HR = 1.75; 95% CI, 1.43-2.15), moderate malnutrition (HR = 1.44; 95% CI, 1.29-1.62), and severe malnutrition (HR = 1.79; 95% CI, 1.58-2.02) with all-cause mortality. Sensitivity analysis supported the robustness of these associations. The between-study heterogeneity was low (all I-2 < 50%), and study quality assessed with NOS was high (all scores > 6). The evidence quality according to the GRADE tool was very low. ConclusionsOur meta-analysis suggests a significant negative association of malnutrition, as defined by the GLIM, with overall survival in patients with cancer. However, definitive conclusions cannot be made, owing to the low quality of the source data.

Automated summary

This meta-analysis found a significant negative association between malnutrition defined by GLIM and overall survival in cancer patients based on analysis of 9 studies. However, due to the low quality of the source data, definitive conclusions cannot be made.