Liver Retransplantation: A Model for Determining Long-Term Survival

Abstract

BACKGROUND: Because of the worse results from retransplantation in relation to the initial liver transplantation, there is a need to refine the indication for retransplantation, such that fair distribution of this benefit is obtained. METHODS: This was a study of 139 patients who underwent liver retransplantation. Thirty variables were studied: 18 relating to the recipient and 12 to the donor. All the independent variables were initially compared with the length of survival using univariate analyses. Variables presenting significance were compared with the dependent variable of length of survival, to determine which factors were related to longer survival among patients, when evaluated together. RESULTS: A multivariate model for determining long-term survival among patients with retransplants was built up using the following variables: recipient's age, creatinine, urgency of retransplantation and early failure of the first graft. Through this multivariate model it was possible to determine a score that was categorized according to tertile distributions (below the 33rd percentile, score <24; 33rd to 66th percentile, 24 < or = score < or = 32; above the 66th percentile, score > 32). One-year, 3-year, and 5-year patient survival rates following retransplantation were respectively 85%, 82%, and 77% for scores <24; 69%, 66%, and 61% for scores between 24 and 32; and 21%, 19%, and 16% for scores >32 (P < 0.0001). CONCLUSION: The variables of recipient's age, creatinine, urgency of retransplantation, and early failure of the initial transplantation were factors that were independently related to the long-term survival of patients with liver retransplants.