Indomethacin injury to the rat small intestine is dependent upon biliary secretion and is associated with overgrowth of enterococci

Abstract

NSAIDuse is limited due to the drugs' toxicity to the gastrointestinal mucosa, an action incompletely understood. Lower gut injury induced byNSAIDs is dependent on bile secretion and is reported to increase the growth of a number of bacterial species, including an enterococcal species,Enterococcus faecalis This study examined the relationships between indomethacin (INDO)-induced intestinal injury/bleeding, small bowel overgrowth (SBO) and dissemination of enterococci, and the contribution of bile secretion to these pathological responses. Rats received either a sham operation (SO) or bile duct ligation (BDL) prior to administration of two daily subcutaneous doses of saline orINDO, and 24 h later, biopsies of ileum and liver were collected for plating on selective bacterial media. Fecal hemoglobin (Hb) and blood hematocrit (Hct) were measured to assess intestinal bleeding. Of the four treatment groups, onlySO/INDOrats experienced a significant 10- to 30-fold increase in fecal Hb and reduction in Hct, indicating thatBDLattenuatedINDO-induced intestinal injury/bleeding. Ileal enterococcal colony-forming units were significantly increased (500- to 1000-fold) inSO/INDOrats. Of all groups, only theSO/INDOrats demonstrated gut injury, and this was associated with enterococcal overgrowth of the gut and dissemination to the liver. We also demonstrated thatINDO-induced intestinal injury andE. faecalisovergrowth was independent of the route of administration of the drug, as similar findings were observed in rats orally dosed with theNSAID Bile secretion plays an important role inINDO-induced gut injury and appears to support enterococcal overgrowth of the intestine.NSAID-induced enterococcalSBOmay be involved either as a compensatory response to gut injury or with the pathogenic process itself and the subsequent development of sepsis.

Keywords: Bile acid; Enterococcus faecalis; non‐steroidal anti‐inflammatory drugs.

Figure 1

Assessment of GI bleeding due to experimental treatments was determined at the time of euthanasia by measurement of: (A) Hematocrit, and (B) Hemoglobin in feces. Both measures showed a difference only in the SO + INDO group. Values are expressed as the mean ± standard error of the mean. *P < 0.05 versus SO + saline and BDL groups.

Figure 2

Histology of the ileum was performed on tissue samples collected at the time of euthanasia. Representative samples from each of the four treatment groups are shown at the same 40× magnification. Staining was with H&E.

Figure 3

Enterococci were cultured in BEA medium from samples collected at the time of euthanasia. Counts were obtained from: (A) feces, (B) ileum, and (C) liver of animals after parenteral vehicle or indomethacin treatment. The numbers of log₁₀ CFU/g per animal are shown as individual dots with the geometric mean for each group shown as a bar. The differences between groups are indicated by bars with arrows.

Figure 4

Measurements of: (A) Hct, (B) fecal Hb, and CFU of enterococci cultured from; (C) feces, (D) ileum and (E) liver from animals after oral vehicle or indomethacin treatment. *P<0.05 versus Control.