"Model for Early Allograft Function" Outperforms "Early Allograft Dysfunction" as a Predictor of Transplant Survival.

BACKGROUND: The Model of Early Allograft Function (MEAF) grades the severity of liver graft dysfunction. Unlike the categorical early allograft dysfunction (EAD) classification, MEAF is a continuous score, based on bilirubin, international normalized ratio, and alanine aminotransferase within 3 days posttransplant.

METHODS: Multivariable regression models were used to validate the MEAF score in 660 liver-only transplants performed between 2000 and 2014. MEAF performance for prediction of transplant survival was compared with that of EAD in univariable and multivariable models by means of Harrell's c-indices, integrated discrimination improvement, and net reclassification improvement.

RESULTS: Median donor and recipient age was 52 years (interquartile range [IQR], 41-62 years) and 58 years (IQR, 50-64 years), respectively. Model for End-Stage Liver Disease score was 15 (IQR, 11-21); cold ischemia time, 8.0 hours (IQR, 6.4-9.7 hours); MEAF, 4 (IQR, 3-6). EAD occurred in 182 (27.6%) cases. Transplant survival was 93%, 90%, and 88% at 3, 6, and 12 months. Both MEAF and EAD were independent predictors of transplant survival within 3, 6, and 12 months. MEAF outperformed EAD as predictor of transplant survival, either when used as a standalone parameter or when corrected for additional independent predictors of transplant survival.

CONCLUSIONS: MEAF is a more accurate predictor of transplant loss than the commonly used EAD classification. As a continuous score grading graft dysfunction, MEAF provides additional, granulated information that could be used both clinically and as a surrogate endpoint of transplant survival in clinical trials.