Optimizing repeat liver transplant graft utility through strategic matching of donor and recipient characteristics.

Repeat liver transplantation (LT) is controversial because of inferior outcomes versus primary LT. A minimum 1-year expected post-re-LT survival of 50% has been proposed. We aimed to identify combinations of Model for End-Stage Liver Disease (MELD), donor risk index (DRI), and recipient characteristics achieving this graft survival threshold. We identified re-LT recipients listed in the United States from March 2002 to January 2010 with > 90 days between primary LT and listing for re-LT. Using Cox regression, we estimated the expected probability of 1-year graft survival and identified combinations of MELD, DRI, and recipient characteristics attaining >50% expected 1-year graft survival. Re-LT recipients (n = 1418) had a median MELD of 26 and median age of 52 years. Expected 1-year graft survival exceeded 50% regardless of MELD or DRI in Caucasian recipients who were not infected with hepatitis C virus (HCV) of all ages and Caucasian HCV-infected recipients <50 years old. As age increased in HCV-infected Caucasian and non-HCV-infected African American recipients, lower MELD scores or lower DRI grafts were needed to attain the graft survival threshold. As MELD scores increased in HCV-infected African American recipients, lower-DRI livers were required to achieve the graft survival threshold. Use of high-DRI livers (>1.44) in HCV-infected recipients with a MELD score > 26 at re-LT failed to achieve the graft survival threshold with recipient age ≥ 60 years (any race), as well as at age ≥ 50 years for Caucasians and at age < 50 years for African Americans. Strategic donor selection can achieve >50% expected 1-year graft survival even in high-risk re-LT recipients (HCV infected, older age, African American race, high MELD scores). Low-risk transplant recipients (age < 50 years, non-HCV-infected) can achieve the survival threshold with varying DRI and MELD scores.