RESUMO

Background. Significant amounts of red blood cells (RBCs) transfusions are associated with poor outcome after liver transplantation (LT). We report our series of LT without perioperative RBC (P-RBC) transfusions to evaluate its influence on early and long-term outcomes following LT. Methods. A consecutive series of LT between 2006 and 2011 was analyzed. P-RBC transfusion was defined as one or more RBC units administrated during or ≤48 hours after LT. We divided the cohort in "No-Transfusion" and "Yes-Transfusion." Preoperative status, graft quality, and intra- and postoperative variables were compared to assess P-RBC transfusion risk factors and postoperative outcome. Results. LT was performed in 127 patients ("No-Transfusion" = 39 versus "Yes-Transfusion" = 88). While median MELD was significantly higher in Yes-Transfusion (11 versus 21; P = 0.0001) group, platelet count, prothrombin time, and hemoglobin were significantly lower. On multivariate analysis, the unique independent risk factor associated with P-RBC transfusions was preoperative hemoglobin (P < 0.001). Incidence of postoperative bacterial infections (10 versus 27%; P = 0.03), median ICU (2 versus 3 days; P = 0.03), and hospital stay (7.5 versus 9 days; P = 0.01) were negatively influenced by P-RBC transfusions. However, 30-day mortality (10 versus 15%) and one- (86 versus 70%) and 3-year (77 versus 66%) survival were equivalent in both groups. Conclusions. Recipient MELD score was not a predictive factor for P-RBC transfusion. Patients requiring P-RBC transfusions had worse postoperative outcome. Therefore, maximum efforts must be focused on improving hemoglobin levels during waiting list time to prevent using P-RBC in LT recipients.