Selective intestinal decontamination (SID) has been useful restraining Bacterial translocation (BT) in both animal models and human clinics. The not well known parenteral nutrition-related liver disease is a serious problem associated to short bowel and long-term parenteral nutrition (PN) use, and BT is also frequent in those patients. Germs reach liver through portal vein and activate Kupffer cells, which release cytokines as IL-1 or TNF-alpha. The aim of this study was to test the use of SID restraining BT in a PN undergoing experimental short bowel model, and its possible favourable consequences on hepatic injury determined by IL-1 and TNF-alpha levels. Twenty-five 240-280 g Wistar rats were divided into two groups and maintained in individual metabolic cages for ten days: Resection-PN group (n=15): animals with a bowel resection of the 80% and a continuous PN infusion. Resection-PN-SID (n=10) group: similar to previous group and a daily oral administration of Tobramycine (20mg/kg/day) and Polymyxine-E (25mg/kg/day). Animals were sacrificed and mesenteric lymph nodes (MLN), and both peripheral and portal blood samples were recovered for TB determination in bacterial culture. Determination of both IL-l and TNF-alpha seric levels were carried out by ELISA. Bacterial translocation incidence was higher in RES-NPT group (66.6%) than RES-NPT-SID group (30%) (P>0,05). The relative risk was 2.22 (IC 95% 0,81-6,11) and the number needed to treat was 3 (IC 95% 2-235). Seric levels of IL-1 and TNF-alpha were also higher in RES-NPT group (7,537 and 5,399 pg/ml, respectively) than in RES-NPT-SID group (6,397 and 5,032 pg/ml respectively) (p<0,001). 1. SID reduces TB in a PN undergoing experimental short bowel resection murine model. 2. Parenteral nutrition-related liver disease decreases in DIS receiving animals.