Global Leadership Initiative on Malnutrition criteria for the diagnosis of malnutrition and prediction of mortality in patients awaiting liver transplant: A validation study.

OBJECTIVES: The Global Leadership Initiative on Malnutrition (GLIM) is a framework aiming to standardize malnutrition diagnosis. However, it still needs to be validated, in particular for patients with chronic liver disease. This study aimed to validate the GLIM criteria in patients with liver cirrhosis awaiting liver transplant (LTx). METHODS: This was a retrospective observational study carried out with adult patients on the waiting list for LTx, consecutively evaluated between 2006 and 2021. The phenotypic criteria were unintentional weight loss, low body mass index, and reduced muscle mass (midarm muscle circumference [MAMC]). The etiologic criteria were high Model for End-Stage Liver Disease (MELD) and MELD adjusted for serum sodium (MELD-Na) scores, the Child-Pugh score, low serum albumin, and low food intake and/or assimilation. Forty-three GLIM combinations were tested. Sensitivity (SE), specificity (SP), positive and negative predictive values, and machine learning (ML) techniques were used. Survival analysis with Cox regression was carried out. RESULTS: A total of 419 patients with advanced liver cirrhosis were included (median age, 52.0 y [46-59 y]; 69.2% male; 68.8% malnourished according to the Subjective Global Assessment [SGA]). The prevalence of malnutrition by the GLIM criteria ranged from 3.1% to 58.2%, and five combinations had SE or SP >80%. The MAMC as a phenotypic criterion with MELD and MELD-Na as etiologic criteria were predictors of mortality. The MAMC and the presence of any phenotypic criteria associated with liver disease parameters and low food intake or assimilation were associated with malnutrition prediction in ML analysis. CONCLUSIONS: The MAMC and liver disease parameters were associated with malnutrition diagnosis by SGA and were also predictors of 1-y mortality in patients with liver cirrhosis awaiting LTx.

Different combinations of the GLIM criteria for patients awaiting a liver transplant: Poor performance for malnutrition diagnosis but a potentially useful prognostic tool.

BACKGROUND & AIMS: Studies using the Global Leadership Initiative on Malnutrition (GLIM) criteria for patients with liver cirrhosis are limited. This study aimed to assess the impact of malnutrition according to the GLIM criteria on the outcomes of patients awaiting a liver transplant (LTx) and compare these criteria with Subjective Global Assessment (SGA). METHODS: This retrospective observational study included adult patients awaiting LTx. Patient clinical data, nutritional status according to various tools including SGA, and resting energy expenditure were assessed. The distinct phenotypic and etiologic criteria provided 36 different GLIM combinations. The GLIM criteria and SGA were compared using the kappa coefficient. The variables associated with mortality before and after the LTx and with a longer length of stay (LOS) after LTx (≥18 days) were assessed by Cox regression and logistic regression analyses, respectively. RESULTS: A total of 152 patients were included [median age 52.0 (interquartile range: 46.5-59.5) years; 66.4% men; 63.2% malnourished according to SGA]. The prevalence of malnutrition according to the GLIM criteria ranged from 0.7% to 30.9%. The majority of the GLIM combinations exhibited poor agreement with SGA. Independent predictors of mortality before and after LTx were presence of ascites or edema (p = 0.011; HR:2.58; CI95%:1.24-5.36), GLIM 32 (PA-phase angle + MELD) (p = 0.026; HR:2.08; CI95%:1.09-3.97), GLIM 33 (PA + MELD-Na≥12) (p = 0.018; HR:2.17; CI95%:1.14-4.13), and GLIM 34 (PA + Child-Pugh) (p = 0.043; HR:1.96; CI95%:1.02-3.77). Malnutrition according to GLIM 28 (handgrip strength + Child-Pugh) was independently associated with a longer LOS (p = 0.029; OR:7.21; CI95%:1.22-42.50). CONCLUSION: The majority of GLIM combinations had poor agreement with SGA, and 4 of the 36 GLIM combinations were independently associated with adverse outcomes.