Detrimental graft survival of size-mismatched graft for high model for end-stage liver disease recipients in liver transplantation.

BACKGROUND: In orthotopic liver transplantation (OLT) size-mismatch may cause adverse outcomes. We previously reported on a method to predict donor-recipient size-mismatch using the body surface area index (BSAi). In this study, we hypothesized that graft survival of size-mismatch transplantation deteriorates with higher model for end-stage liver disease (MELD) score at transplantation. METHODS: We evaluated non-parametrically the association of BSAi and MELD with 1-year graft survival with a generalized additive model. For derivation, transplantations performed between 2005 and 2010 were used. The associations were then validated by comparing Kaplan-Meier estimates between patient groups stratified according to estimated risk, using transplantations from 2011-2013. RESULTS: A total of 30,870 OLT were included in the study with 16,466 in the validation group. The derivation model revealed that graft survival significantly decreased with higher or lower BSAi, and with higher MELD (P < 0.0001). Validation confirmed the correlation of observed graft survival with estimated risk categories. CONCLUSIONS: We found that there is an interactive effect between MELD score and size-mismatch. Also high MELD recipient has a narrower safety margin for size-mismatched graft. The risk calculated from our nonparametric model with MELD and BSAi well predicts outcome in liver transplantation.