Impact of the Learning Curve on the Outcome of Domino Liver Transplantation

Abstract

Domino liver transplantation (DLT) is a strategy used to increase the number of available grafts. In this procedure, the transplant recipient is a living donor of her own liver. It is mandatory that the graft should be fully functional and the genetic defect should recur with sufficient latency period in the new recipient. Corino-Andrade disease, or familial amyloidotic polyneuropathy (FAP), satisfies these conditions. We retrospectively reviewed our prospective database of DLT. From July 2004 to April 2013, we performed 12 DLTs. We assessed age, sex, real Model for End-Stage Liver Disease (MELD) score, waiting list time, cold and warm ischemia times, intraoperative transfusion requirements, hospital stay, early peritransplantation morbidity (post-reperfusion syndrome, intraoperative cardiac arrest, post-transplantation thrombotic events, and biliary morbidity), acute cellular rejection, retransplantation, mortality, patient and graft survivals. With the intention to study the effect of the learning curve in the global survival results (including both donors and recipients of livers with FAP), we divided our series into 2 periods: the early period (from 2004 to 2008) and the present period (from 2009 to 2013). The crude mortality was 40% vs 0% (P = .042) in the early and present periods, respectively. The cumulative patient survival was also significantly in favor of the present period (P = .049). The graft loss prevalence was 60% vs 7.1% (P = .019) in the early and present periods, respectively. The cumulative graft survival was also significantly in favor of the present period (P = .030; Fig 2). In conclusion, we consider DLT to be a complex procedure, whose initial results are conditioned by the learning curve.