Abstracts
1997 Digestive Disease Week

Bacterial translocation via the thoracic duct in a porcine model of hepatic ischemia/reperfusion injury.

LCJM Lemaire*, BA Van Wagensveld*, TM van Gulik*, J Dankert§, DJ Gouma*, JJB van Lanschot*. Departments of *Surgery and §Microbiology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands.

Intestinal bacterial translocation has been postulated as an important risk factor for the development of multiple organ failure (in the absence of a definite focus of infection). Research on this topic has been focused on detection of bacteria and endotoxins in (portal) blood or mesenteric lymph nodes (MLN's). Lymphatic translocation beyond the level of MLN's, i.e. the thoracic duct, has not been studied, although this could be of significant importance because it implicates bypass of the reticuloendothelial system of the liver and direct discharge of bacteria and endotoxins into the systemic circulation.

Aim: To determine whether bacterial translocation occurs in a (clinically relevant) model of severe non-bacterial tissue damage and to assess the various routes of bacterial translocation.

Materials and Methods: In a porcine model of 2 hrs ischemia and 5 hrs reperfusion of the liver, the thoracic duct and the portal and subclavian vein were cannulated (n=5 pigs). Lymph was drained through siphoning and collected in a plastic bag. During the ischemic period, a hemihepatectomy was performed. To prevent venous congestion of the gut, a portocaval shunt was made prior to hepatic ischemia. MLN's were cultured before and after ischemia and after 5 hrs reperfusion. Lymph, portal blood and systemic blood were collected before and after ischemia and after 1, 2, 4 and 5 hrs of reperfusion, and analyzed for the presence of bacteria and endotoxins. In addition, TNF-alpha concentrations were measured (ELISA) at different timepoints.

Results: Culture-positive MLN's were present in 2/5 pigs after 2 hrs of ischemia and also in 2/5 pigs after 5 hrs reperfusion. Thoracic duct lymph cultures were positive in 3/5 pigs after 2 hrs of ischemia; lymph cultures were positive in 4/5 pigs 1 hr after reperfusion and in all pigs 2, 4 and 5 hrs after reperfusion (p< 0.05 compared to the preischemic state). During the ischemic and reperfusion period, the incidence of positive portal blood cultures did not increase significantly in comparison to the preischemic state. The incidence of positive systemic blood cultures was significantly increased 4 hrs after reperfusion (4/5 pigs compared to 0/5 pigs at preischemia, p< 0.05). Median endotoxin levels increased threefold in the thoracic duct lymph from 0.04 Endotoxin Units (EU)/mL (= 3.3 pg/mL) at preischemia to 0.136 EU/mL (= 11.3 pg/mL), 5 hrs after reperfusion (p<0.001, ANOVA). Portal and systemic blood endotoxin levels did not increase significantly during the experiment. TNF-alpha concentrations in lymph and portal and peripheral blood were not detectable (< 38.4 pg/mL) during the experiment except 1 hr after reperfusion in lymph, the median TNF-alpha concentration peaked to 50.7 pg/ml.

Conclusion: This study shows that the thoracic duct is one of the routes for bacterial translocation to the systemic circulation.