Effects of starvation and haemorrhage on the large bowel coliform flora with special reference to bacterial mucosal adherence and translocation

Abstract

Effects of starvation and haemorrhage on the large bowel coliformflora with special reference to bacterial mucosal adherence and translocation

by Carl-Gustaf Nettelbladt, MD Department of Surgery, Karolinska Hospital andInstitute, S-171 76 Stockholm, Sweden

Intestinal bacteria translocating to extraintestinal sites have been suggestedto play a role in the etiology of posttraumatic infections and multiple organ failure.In the present study, effects of haemorrhagic stress and /or starvation on the gutflora and bacterial translocation were studied in the rat.

Hyperosmotic glucose infusion dunng haemorrhage in starved rats improved plasmarefill but did not reduce bacterial translocation. In this experiment four groupsof rats were subjected to haemorrhage. Two groups were given an i.v.infusion of hyperosmoticglucoseduring moderate (38%bloodloss;n=12) or severe (43%bloodloss; n=19) haemorrhage withoutreinfusion. Control groups received an i.v. infusion of saline during moderate (n=12)or severe (n=18) haemorrhage. No difference in bacterial translocation was observedbetween groups infused with glucose or saline.

Starvation increased the number of coliform bacteria in caecal contents and inducedadherence of coliform bacteria to caecal epithelium. Rats in a control group (n=19)were given their regular food. Six rats were starved for 24 hours and another 15for 48 hours, with free access to water. Six rats underwent non-lethal haemorrhage(mean arterial pressure=55 mm Hg). These animals were only allowed water until sampling24 hours later. Twentyfour hours starvation increased the number of bacteria in caecalcontents 25-fold (p<0.05). 48 hours starvation further increased the number (p<0.001)and induced a marked increase in bacteria adherent to caecal epithelium (p<0.001).In the haemorrhaged group similar changes were observed and bacterial translocationincreased (p<0.05) as compared to control rats.

In order to characterise and compare coliform bacteria in caecal contents, oncaecal epithelium and in mesenteric Iymph nodes of 57 rats subjected to differentdegrees of stress a biochemical finger printing method was used. A total of 291 biochemicalphenotypes were found in caecal contents of all rats. Out of these, 108 were detectedon caecal epithelium and only 19 of these appeared in mesenteric Iymph nodes of thecorresponding rat. A total of 36 biochemical phenotypes were found in mesentericIymph nodes of all rats. Twentyone of these belonged to four common phenotypes. Theprevalence of these four phenotypes in mesenteric Iymph nodes did not differ significantly,while some of them had a low and some a high prevalence in caecal contents and oncaecal epithelium. The phenotypes found in mesenteric Iymph nodes were also, mostly,found adherent to the caecal epithelium of the corresponding rat.

Orally inoculated translocating strains of E. coli were able to colonise the gut.Thus, inoculation increased bactenal translocation in rats lacking these strainsin their indigenous gut flora. Two groups of rats were inoculated with two translocatingstrains of E. coli. One group (n=l l ) was starved for 24 hours and the other (n=20)underwent non-lethal haemorrhage (mean arterial pressure=50 mm Hg) and was starvedfor 24 hours thereafter. Two non-inoculated groups of similar size, subjected tothe same treatments served as controls. In the inoculated groups, bacterial translocationincreased both after 24 hours starvation (p<0.05) and haemorrhage (p<0.01)as compared to non-inoculated control rats.

Ingestion of bulking fibre prevented mucosal adherence and translocation of atranslocating strain of E. coli. Four groups of rats were inoculated with a translocatingstrain of E coli. Animals in a control group (n=8) were given their regular food.Eight rats were starved for 48 hours and eight given only bulking fibre for 48 hours.An additional group (n=8) ingested bulking fibre only for 24 hours before and afterhaemorrhage (mean arterial pressure=50 mm Hg). Bulking fibre for 48 hours reducedboth mucosal adherence (p<0.05) and translocation (p<0.001) of the inoculatedbacteria as compared to starved rats.

It is concluded that haemorrhagic stress and starvation have marked effects onthe gut flora. Glucose infusion given during haemorrhage, previously shown to protectthe animal during circulatory shock, can not prevent bacterial translocation. Briefstarvation increases the number of coliform bacteria in caecum and induces bacterialmucosal adherence. Only a limited number of strains of coliform bacteria translocateafter stress. Adherence to caecal epithelium was associated with bacterial translocation.Inoculation with translocating strains of E. coli increases bacterial translocation.Bulking fibre prevents bacterial mucosal adherence and translocation. Detailed knowledgeof the gut E. coli flora is of great importance when interpreting results of studieson bactenal translocation.

Key words: Bacterial translocation, plasma refill, starvation, haemorrhage, coliforms,Escherichia coli, biochemical fingerprinting, bulking fibre.

ISBN 91-628-2783-9 Stockholm 1998