Survival Predictors in Liver Transplantation: Time-Varying Effect of Red Blood Cell Transfusion

Background. Many attempts have been undertaken to better predict outcome after liver transplantation. The aim of this study was to identify the pre- and intraoperative variables that may influence the survival after liver transplantation, at a single institution. Methods. Anesthetic records from 543 consecutive patients who underwent liver transplantation from June 2006 to June 2014 were reviewed in this retrospective study. Patients undergoing retransplantation were excluded from the analysis, as were patients with familial amyloid polyneuropathy. Preoperative variables studied were age, sex, Model for End-Stage Liver Disease score, primary diagnosis, cold ischemia time, preoperative international normalized ratio, serum albumin, and and hemoglobin levels. Intraoperative variables included were norepinephrine consumption, blood loss, red blood cell transfusion, and surgical time. Variables significant in the univariate analysis with a P value of <.2 were included in a multivariate Cox regression model. Results. Only red blood cell transfusion (hazard ratio [HR], 1.16; 95% confidence interval [CI], 1.04-1.29) and female sex (HR, 1.71; 95% CI, 1.10-2.65) were identified as significant independent predictors for survival after liver transplantation. Because of proportionality assumption violation, the multivariate Cox regression model was subsequently upgraded by adding a time-varying interaction between red blood cell transfusion and time since liver transplantation. As a result, we found that at 3 months after liver transplantation, the rate of dying increased 14% (95% CI, 2%-26%) for each unit transfused, and at 6 months it increased 12% (95% CI, 0.3%-24%). Conclusions. Red blood cell transfusion ceased to influence survival from 1 year onward.