Edward J. Hall, MA, VetMB, PhD, DECVIM-CA, MRCVS
Department of Clinical Veterinary Science, University of Bristol, Langford House, Langford
The gastrointestinal (GI) tract contains a resident microflora that is crucial in maintaining normal digestive and immune function. The proximal GI tract is relatively sterile, with a flora that is influenced by what the animal has just swallowed. Bacterial numbers gradually increase towards the ileocolic valve, so that the colon contains a large and remarkably stable population, and indeed approximately 50% of faecal bulk is bacteria. The composition of the flora also changes along the tract with a progressively increasing proportion of Gram negative and obligate anaerobe bacteria.
There is considerable variation in the microflora in normal individuals, and the population may be affected by a number of factors such as environment, diet, scavenging and coprophagy. The normal microflora has a generally symbiotic relationship with the GI tract, being regulated by host factors (e.g., gastric acid, intestinal peristalsis, pancreatic antibacterial products, the mucosal barrier and the innate immunity) and microbial interactions (e.g., substrate depletion, growth inhibitor production) that aid exclusion of pathogens.
Perturbations in the microflora are likely to be associated with clinical signs, and may be caused by colonization by specific enteropathogens that overcome the normal regulatory mechanisms, or abnormalities in the resident flora.
Known bacterial enteropathogens include Salmonella spp., Campylobacter spp. and enteropathic E. coli. However, they represent just three out of an estimated 100-200 species that colonise the normal GI tract. We currently have very little knowledge of the role of these other bacteria and whether they can be harmful. It is only in the past twenty years that gastric Helicobacter spp. have been shown to be pathogenic in humans. Although their existence in animals has been known for over 100 years, we still don't know their pathogenic potential in dogs and cats, and related Helicobacter organisms have now been found in the biliary tree and intestine, and may be incriminated in biliary and intestinal inflammation.
Many enteric pathogens cause acute clinical disease, but the ability to adhere to or invade the intestinal mucosa can promote long-term colonization, causing chronic disease or carrier status. Enteropathogens tend to be disseminated in faeces of infected animals and their duration of excretion can last for several months, or the animal may even become a carrier. They most likely contaminate water or food, but may be able to survive weeks in the environment under the right conditions. Once they have gained access to the GI tract they survive (at least temporarily) against the host defenses and the established microflora. The ability to colonise depends on mechanisms such as flagella and adherence factors.
These pathogens also express virulence factors, such as secretion of cytotoxins, enterotoxins and resistance to phagocytosis, that allow them to cause damage:
Invasive bacteria such as Salmonella spp., Campylobacter spp., and enteroinvasive E. coli can invade the mucosa and cause dysentery (i.e., diarrhoea with blood), and potentially can cause septicaemia. Salmonellosis in cats in the spring may follow ingestion of wild birds, an illness termed 'songbird fever'.
Organisms secreting cytotoxins include enterohemorrhagic E. coli (EHEC) and cytotoxic necrotising factor (CNF)-secreting E. coli. Cytotoxins, such as heat-stable toxin, shiga-like toxin, verocytotoxin and CNF-1 are lethal to intestinal epithelium and may cause haemorrhagic diarrhoea.
Enterotoxins have a specific biochemical effect and do not cause direct intestinal damage. The classic example in man is Vibrio cholerae, which stimulates intestinal hypersecretion leading to illness and even death from dehydration. Similar enterotoxigenic E. coli (ETEC) do infect animals.
Enteropathogenic E. coli (EPEC) cause attaching-effacing damage to the microvilli and consequently significant malabsorption.
The diagnosis of bacterial intestinal infections is often based solely on the isolation of a potential pathogen by culture. However, such a diagnosis may be incorrect as many of these organisms can also be isolated from the faeces of clinically healthy animals, and the significance of any culture needs to be evaluated with respect to the animal's age, nutritional and immune status and its clinical signs, and any environmental stresses such as housing conditions and hospitalisation. Positive faecal culture results are more likely to be relevant in animals thought to be exposed to infection, if associated with outbreaks of diarrhoea in more than one animal per household, or after kenneling. Evidence of sepsis or the presence of faecal neutrophils is indications for faecal culture.
Salmonella spp. and Campylobacter jejuni are the most commonly isolated enteropathogens in dogs, and can cause acute and chronic disease. Acute enterocolitis typically results in diarrhoea that often contains blood and mucus; vomiting, inappetence, and abdominal pain with pyrexia are less common. Although a tentative diagnosis may be made on the isolation of these organisms in an animal demonstrating these signs, it must be remembered that these organisms may occur in association with underlying diseases. Even with pathogenic E. coli, identification of genes encoding virulence determinants by PCR does not necessarily indicate that the organism is responsible for the clinical signs. Concurrent infection of Campylobacter with canine parvovirus results in more severe disease than with single infection by either organism.
Clostridium perfringens and C. difficile may cause diarrhoea as a consequence of enterotoxin production under certain environmental conditions. However, they are sometimes described as nosocomial infections as clostridial-associated diarrhoea is typically seen in hospitalized animals. Whilst infection with C. difficile may be a true nosocomial event, C. perfringens is generally considered part of the normal resident microflora of dogs and cats. 'Flare factors' that may trigger these organisms to produce enterotoxin are believed to include hospitalization and sudden diet change, two events that frequently occur together.
Clostridium difficile can produce two toxins (A & B), and is responsible for antibiotic-associated pseudomembranous colitis in humans. It has been incriminated as a cause of chronic diarrhoea in dogs but pseudomembranous colitis is not a typical feature.
Enterotoxin-producing C. perfringens has been associated with acute and chronic GI signs. A change in intestinal pH was believed to trigger sporulation and enterotoxin production, and hence acute disease. Examination of a Diff-Quik stained faecal smear for "safety-pin" shaped C. perfringens-containing spores has been suggested as a simple screening test as enterotoxin secretion was believed to coincide with sporulation. However, this has now been disproved by measuring C. perfringens enterotoxin in faeces in the absence of spore formation. Indeed, enterotoxin can be detected in the faeces of healthy dogs. The presence of a large numbers of clostridial endospores (> 5 per oil field) on smears may be suggestive, but a positive faecal enterotoxin assay (ELISA or reverse passive latex agglutination) is likely to be more significant. Most affected dogs exhibit signs of colitis but a causal relationship to diarrhoea remains to be proven.
Haemorrhagic gastroenteritis (HGE) is the term given to a syndrome that is characterized by acute haemorrhagic diarrhoea accompanied by marked haemoconcentration. The cause of the syndrome is unknown. It may potentially be a consequence of C. perfringens enterotoxin production.
Abnormalities of the flora
Whilst clinically healthy animals live in relative harmony with their enteric flora, disease may occur not only because of invasion by specific pathogens or proliferation of opportunistic pathogens as described above, but also because imbalances can occur within the normal flora. Such imbalances may be the result of alterations in the normal physiological mechanisms that control the flora, which may lead to quantitative or qualitative alterations in the flora.
Small intestinal bacterial overgrowth (SIBO) describes proliferation of abnormal numbers of bacteria in the small intestinal although what actually constitutes abnormal numbers remains controversial. Classically more than 105 organisms (or 104 anaerobes) per ml of duodenal juice has been defined as SIBO, but this cutoff is almost certainly too low and it has been shown that numbers fluctuate widely within individual dogs. Furthermore, cats probably naturally have a numerically larger flora yet no disease.
True SIBO probably does occur secondary to problems such as stagnant bowel ('blind loop') and exocrine pancreatic insufficiency, where bacterial fermentation and proliferation can occur. In contrast, the spontaneous syndrome, seen most typically in German shepherds and termed idiopathic SIBO, perhaps represents qualitative changes in the flora, as it is often accompanied by the presence of a predominantly anaerobic small intestinal flora. Alternatively, it has been suggested that the cause is not a change in the flora, but a change in the host's response to that flora. Accordingly the number and nature of the bacterial flora may not be important and the broader term 'antibiotic-responsive diarrhoea' (ARD) has been applied. Whilst ARD would include those cases previously termed idiopathic SIBO, it will also encompass those cases where there is infection with an unidentified enteropathogen that is responsive to the antibiotic.
Idiopathic ARD is most commonly seen in young, large-breed dogs, especially German shepherds, a breed in which abnormalities of mucosal immunity are suspected. It causes chronic diarrhoea and weight loss or failure to thrive. Coprophagy is a common feature and indeed may reinforce the perturbations of the intestinal flora. In practice a tentative diagnosis can be made from the age, breed, lack of other abnormalities and of course the response to antibiotics. Definitive diagnosis by the supposed gold standard of quantitative duodenal juice culture is flawed, even if the technique can be standardized, as absolute bacterial numbers are not the only factor. Indirect diagnosis by measurement of serum folate and cobalamin concentrations, unconjugated bile acids or breath hydrogen are problematic, as all of these tests correlate poorly with bacterial numbers and response to treatment, Yet logically they should reflect some alteration of the bacterial flora that may have a functional consequence. In SIBO, diarrhoea is believed to be a consequence of direct bacterial damage of the intestinal mucosa plus hydroxylation of fatty acids and deconjugation of bile salts, both of which stimulate colonic secretory diarrhoea.
The condition is treated by administration of oral antibiotic. Oxytetracycline, metronidazole or tylosin are frequently chosen as they are relatively safe, cheap and effective. Curiously response to treatment persists after prolonged use, although development of bacterial resistance would be predicted, and sometimes lower dosages can be effective. Furthermore, these antibiotics do not lead to a reduction in small intestinal numbers. Therefore it has been suggested that they either work by exerting a selective pressure in the flora, or actually alter the host response to that flora.
Idiopathic inflammatory bowel disease (IBD) is believed to be a loss of the host tolerance to the indigenous flora, and successful treatment usually lies with immunosuppression rather than antibacterial therapy. An alternative approach would be to give probiotics. These are live micro-organisms administered orally, that alter the intestinal microflora and are proposed to have a beneficial effect on health. Lactobacilli and Bifidobacteria are frequently claimed to have probiotic activity, but E. coli, Enterococcus and non-bacterial Saccharomyces have also been used. Possible mechanisms for probiotic activity include: production of antimicrobial metabolites; competitive interactions (receptor binding); interaction with epithelial function, e.g., improved intestinal permeability; immune modulation. They have been shown to decrease intestinal permeability and have other effects on mucosal immune responses in experimental rodent models of IBD. Clinical and there is emerging evidence in dogs and cats. However, the commercial products currently available are not canine or feline in origin, and yet it has been shown in experimental models of IBD, that their performance varies with the individual organism given. Thus it appears that strain-specific probiotics may be required not only for the species but also perhaps even for the type of intestinal disease being treated. Thus probiotics may become a simple adjunct to conventional treatment of intestinal disease in dogs and cats as soon we can give the appropriate strain(s) at the right dose and at the right time.
1. Batt RM, et al.: Enteric bacteria-friend or foe? J Sm Anim Pract 37:261, 1996.
2. McDonough PL and Simpson KW: Diagnosing emerging bacterial infections: Salmonellosis, Campylobacteriosis, clostridial toxicosis and Helicobacteriosis. Sem Vet Med Surg 11:1, 1996.
3. Fox JG. Campylobacter infections. In Greene CE (ed): Infectious Diseases of the Dog and Cat, Philadelphia, WB Saunders, 1998, p. 226.Weese JS, et al.: The roles of Clostridium difficile and enterotoxigenic Clostridium perfringens in diarrhea in dogs. J Vet Intern Med 15:374, 2001.
4. Marks SL, et al.: Evaluation of methods to diagnose Clostridium perfringens-associated diarrhea in dogs. J Am Vet Med Assoc 214:357, 1999.
5. Sasaki J, et al.: Hemorrhagic enteritis associated with Clostridium perfringens type A in a dog. J Vet Med Sci 61:175, 1999.
6. Batt RM and McLean L: Comparison of the biochemical changes in the jejunal mucosa of dogs with aerobic and anaerobic bacterial over-growth. Gastroenterol 93:986, 1987.
7. Batt RM, et al.: Response of the jejunal mucosa of dogs with aerobic and anaerobic bacterial overgrowth to antibiotic therapy. Gut 29: 473, 1988.
8. German AJ, et al.: Cytokine mRNA expression in mucosal biopsies from German shepherd dogs with small intestinal enteropathies. Digest Dis Sci 45:7, 2000.
9. German AJ, et al.: Comparison of direct and indirect tests for small intestinal bacterial over-growth and antibiotic-responsive diarrhea in dogs. J Vet Intern Med 17:33, 2003.
10. Baillon MLA, et al.: Effects of probiotics Lactobacillus acidophilus DSM 13241 in healthy adult dogs. Am J Vet Res 65:338, 2004.