Características do doador, suas aplicações em índices de risco e na avaliação no tempo de vida útil do transplante hepático.

Abstract

The aim of this study was to investigate the effect of the demographic, clinical, and laboratory characteristics of the donor-recipient binomial and of variables related to the procedure on liver graft survival in 30 days, either individually or grouped in previously described risk indices (MELD, D-MELD, DRI, SOFT, BAR and TRI). Additionally, a new risk index was calculated with the variables that were associated with liver graft survival in the population studied and compared to the other indices used in this study. We carried out a retrospective study that used information from 471 liver transplants performed at the Hospital das Clínicas of Universidade Federal de Minas Gerais from 2008 through 2018. The outcome was the survival time of the liver graft (period between the date of the liver transplant and the date of death/re-transplant). The explicative variables were: demographic, clinical, and laboratory characteristics of both recipient and donor, and variables related to the procedure. Kaplan-Meier survival curves and Cox regression models were used. All variables that were associated with graft failure in the multivariate analysis were used to calculate a new risk index (Belo Horizonte Risk Index - BHRI). The ability to predict graft failure within 30 days of the BHRI and each calculated risk index was performed comparing the area under the ROC curve. The multivariate analysis showed that the BMI of the donor (HR: 2.10, 95%CI: 1.19-3.74) and of the recipient (HR: 2.40, 95%CI: 1.45-4.00), hospitalization of the recipient immediately before transplantation (HR: 3.70, 95%CI: 2.25-6.07) and portal vein thrombosis in the recipient before liver transplantation (HR: 2.14, 95%CI: 1.20-3.81) were independently associated with liver graft failure. The risk indices previously described in the literature demonstrated a low predictive power of the hepatic graft failure and the area under the ROC curve for these indices ranged from 0.47 for the DRI to 0.62 for the BAR. A slightly better performance was observed in the index developed in this study (BHRI), which obtained an area under the ROC curve of 0.66. Our results suggest that the characteristics related to the donor-recipient binomial should be taken into account by the surgical team at the transplant, given the possibility of a worse prognosis in relation to graft survival, however this information should not be used to contraindicate the procedure. The low predictive power for liver graft failure obtained using the risk indices previously described in the literature that we found in the present study is probably due to the differences between the populations used in the current analysis from other populations used for the elaboration of the previous indices. Additional studies to promote adaptations of these indices to the characteristics of each population can bring a better donor-recipient matching, allowing the use of expanded criteria donors and reducing the waiting list for the liver graft.