Graft weight integration in the early allograft dysfunction formula improves the prediction of early graft loss after liver transplantation

The role of the graft-to-recipient weight ratio (GRWR) in adult liver transplantation (LT) has been poorly investigated so far. The aim is to evaluate the contribution of the GRWR to the well-recognized early allograft dysfunction (EAD) model (i.e., Olthoff model) for the prediction of 90-day graft loss after LT in adults. Three hundred thirty-one consecutive adult patients undergoing LT between 2009 and 2018 at Tor Vergata and Sapienza University in Rome, Italy, served as the Training-Set. The Validation-Set included 123 LTs performed at the Erasmus Medical Center, Rotterdam, the Netherlands. The mEAD model for 90-day graft loss included the following variables: GRWR ≤\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\le $$\end{document} 1.57 = 2.5, GRWR ≥\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge $$\end{document} 2.13 = 2.5, total bilirubin ≥ 10.0 mg/dL = 2.0, INR ≥ 1.60 = 2.3, and aminotransferase > 2000 IU/L = 2.2. The mEAD model showed an AUC = 0.74 (95%CI = 0.66–0.82; p < 0.001) and AUC = 0.68 (95%CI = 0.58–0.88; p = 0.01) in the Training-Set and Validation-Set, respectively, outperforming conventional EAD in both cohorts (Training-Set: AUC = 0.64, 95%CI = 0.57–0.72; p = 0.001; Validation-Set: AUC = 0.52, 95%CI = 0.35–0.69, p = 0.87). Incorporation of graft weight in a composite multivariate model allowed for better prediction of patients who presented an aminotransferase peak > 2000 IU/L after LT (OR = 2.39, 95%CI = 1.47–3.93, p = 0.0005). The GRWR is important in determining early graft loss after adult LT, and the mEAD model is a useful predictive tool in this perspective, which may assist in improving the graft allocation process.