Bacterial translocation during liver transplantation: a randomized trial comparing conventional with venovenous bypass vs. piggyback methods.

The aim of this study was to evaluate the bacterial translocation in liver transplantation (LT), comparing the conventional and the piggyback methods. A total of 32 patients were randomized into the 2 groups. Samples of blood were collected from the radial artery, portal vein (PV) and hepatic vein (HV), in up to 120 minutes postreperfusion. The samples were sent for endotoxin level, as well as samples up to 2 minutes post-perfusion were sent to culture. Hepatic artery and PV blood flows were measured at postreperfusion collection times. The results analyzed were: endotoxin concentration, its quantity, and hepatic clearance. The statistical treatment consisted of analyzing each group's mean profile. The analysis for endotoxin concentration in the radial artery was the deviation related to presurgery measure, and in the PV the deviation related to preclamping (PC) measure. The overall mean level of endotoxin concentration was 0.99 EU/mL in the artery, 1.30 EU/mL in the PV, and 1.22 EU/mL in the HV. The deviation was significant in the portal (P = 0.0031), but not in the artery samples (P = 0.2092). We detected a significant quantity of endotoxin in the artery and in the portal and the HVs (P < 0.001). There was no difference between the 2 groups and no hepatic clearance of endotoxin was detected either (P = 0.1515). All the cultures were negative. In conclusion, the study detected a significant translocation of endotoxin, but not of bacteria. The study also detected the absence of endotoxin hepatic clearance in both the piggyback and the conventional methods without any difference between them.