Repeat liver retransplantation: rationale and outcomes

Background: Liver retransplantation remains the only option for recurrent graft failure. The aim of our study is to identify predictive factors involved in patients and graft survival for patients undergoing repeat retransplantation (RRT). Methodology: From January 1985 to December 2012, 2940 liver transplantations were performed in 2477 patients at Paul Brousse Hospital, Villejuif, France. All patients who underwent third, fourth, and fifth transplantation were included in the study and retrospectively analyzed. Results: In the univariate analysis, the factors that were associated with 90-d patient post-operative survival were pre-operative vasopressors support, pre-operative extra hepatic sepsis, primary non-function (PNF) as indication of RRT, recipient's model of end stage liver disease (MELD), urgent RRT, creatinine value at RRT, and prothrombin ratio. The multivariate logistic regression confirmed the role of systemic septic status (OR = 12.8, p = 0.01) and vasopressor drug support (OR = 4.7, p = 0.05) as predictors of post-operative mortality. In the univariate analysis, the factors that were associated with patient 10 yr long-term survival (were vasopressor support, systemic septic patient, PNF as indication of RRT, RRT occurred between 1985 and 1999, recipient's MELD, creatinine value at RRT, and prothrombin ratio. The multivariate logistic regression confirmed the role of systemic septic patient (OR = 6.4, p = 0.03) and the RRT between 1985 and 1999 (OR = 3.6, p = 0.05) as predictors of long-term mortality. Conclusion: RRT represent a valid alternative in selected patients. Selection should be oriented on patients needing third transplant without extra hepatic sepsis and vasoactive drug support at moment of RRT. If necessary, fourth and fifth RRT could be performed with a decision made on case-by-case basis, despite a high post-operative mortality.