Xanthinoxidase (XO) derived radical species are involved in bacterial translocation (BT) in cholestatic rats. The mechanism by which XO influences remains unclear. It has been shown recently that nuclear factor-kappa B (NF-κB), a ubiquitous transcription factor, can be activated by oxidative stress and thereby promote the process of BT. We investigated the effects of NF-κB inactivation on the incidence of BT in cholestatic rats. Sprague-Dawley rats were randomly assigned to one of eight groups: groups 1–4 were sham laparotomized rats either untreated (S1) or treated for 5 days with thalidomide (S2), curcumin (S3), or Inchin-ko (ICK; S4); groups 5–8 underwent common bile duct ligation (CBDL) for 5 days and were either untreated (C1) or treated with thalidomide (C2), curcumin (C3), or ICK (C4). After 5 days bacteriological cultures were performed from portal blood and V. cava, from the central mesenteric lymph node complex (MLN), spleen, and liver. The intensity of the activated NF-κB-subunit p65/p50 in the ileum mucosa was estimated by light microscopy and a scoring system from 1 to 20. Malondialdehyde (MDA) and myeloperoxidase activity (MPO) in the ileum were evaluated and expressed as U/g dry weight. Thalidomide and ICK reduced in CBDL-rats significantly the BT rate (63% vs. 18%, 63% vs. 30%, P<0.01). Enzyme estimations (MDA, MPO, and GSH) in sham operated animals showed no significant changes in the untreated groups compared with the treated groups. CBDL-rats pre-treatment with all three compounds caused a significant increase of MDA levels if groups were compared with the untreated C1-group (C1 31.6±7.7, C2 54.5±12.2, C3 53.3±11.2, and C4 47.2±9.4). GSH was reduced after the pre-treatment by all compounds but only significantly after curcumin pre-treatment (C1 vs. C3: 13.9±1.8 vs. 7.1±1.8; P<0.05). MPO estimations were significantly higher in the untreated C1-group if compared with groups C2, C3, and C4 (C1 1036.4±340.9, C2 709.9±125.9, C3 545.2±136.6, and C4 556.7±247.4; P<0.05). Thalidomide inhibited significantly the activation of NF-κB (C2 vs. C1: 6.0±4.5 vs. 12.7±5.3; P<0.01). Likewise, Curcumin and ICK suppressed NF-κB activation, but this did not reach significance in this experiment. NF-κB is involved in the process of BT in cholestatic rats and may be activated by XO derived ROS. We assume that the activated NF-κB initiates transcription of target genes inducing cytokine production, which in turn disrupts the tight junctions leading to BT from the intestinal lumen to the MLNs and circulation.