Prediction of graft dysfunction based on extended criteria donors in the model for end-stage liver disease score era

To explain the influence of recipient status combined with the accumulation of extended criteria donor (ECD) variables on the appearance of severe ischemia-reperfusion injury and graft survival in a model for end-stage liver disease (MELD)-based system, we analyzed our most recent consecutive liver transplantations (LTs), dividing them into two periods: 400 LTs (1992-2002; pre-MELD era) and 275 LTs (2002-2007; post-MELD era).

Primary dysfunction (PD) was defined as primary graft failure that required emergency retransplantation or as initial poor function. Donor variables were included in a regression model to assess the probability of PD.

Donor age, macrovesicular steatosis more than 30%, and cold ischemia time were associated with allograft dysfunction. Mean probability of PD was 14.8%, 19.2%, 27.5%, and 37.4% for ECD 0, 1, 2, and more than or equal to 3, respectively (P=0.003). Distribution of no-mild, moderate, and severe ischemia-reperfusion injuries among MELD categories was 72.53%, 24.17%, and 3.30% (MELD group=12-19); 56.52%, 36.96%, and 6.5% (MELD group=20-28); and 23.91%, 54.35%, and 21.74% (MELD group >or=29), respectively (P=0.043). The development of PD according to ECD variables was 18.8%, 18.1%, 28.0%, and 35.3% for ECD 0, 1, 2, and more than or equal to 3, respectively (P=0.047). These variables were independent predictors of PD (Cox proportional regression model): ECD 2 (relative risk [RR]=1.59; 95% confidence interval [CI]=1.25-1.62), ECD 3 (RR=2.74; 95% CI=2.38-3.13), MELD 21 to 30 (RR=1.89; 95% CI=1.32-2.06), and MELD more than or equal to 30 (RR=3.38; 95% CI=2.43-3.86). Graft survival decreased, whereas MELD and the number of ECD variables increased.

The combination of three or more ECD variables and an MELD more than or equal to 29 is the worst scenario for graft success after LT.