Roger M. Batt
Masterfoods, Mars Inc.
Waltham-on-the-Wolds, Leicestershire, UK
The proximal small intestine normally contains few bacteria.In small intestinal bacterial overgrowth (SIBO) there is proliferation of abnormal numbers of bacteria in the lumen of the upper small intestine. The definition of what is considered an abnormal number of bacteria in the dog is still under discussion. It is classically stated that in normal household pet dogs no more than 104 to 105 bacteria per mL of juice are present in the lumen of the upper small intestine. Although recent reviews have questioned the accuracy of this upper limit of normal, some of the reported variation may reflect inclusion of dogs not from household environments rather than pet dogs. However, it is generally accepted that species normally present in the proximal small intestine of dogs include E. coli, enterococci and lactobacilli, and that obligate anaerobic species are rare. In dogs with SIBO there are not only increased numbers of intraluminal bacteria, but the composition of the flora also changes to a predominantly anaerobic one, resembling that of the colon.
SIBO in the dog has been infrequently reported, probably because of the difficulty in establishing the diagnosis, and initial descriptions were limited to its occurrence in German Shepherd Dogs. However, in recent years it has been described as a common finding in dogs with chronic small intestinal disease, either as a cause or a consequence of their disease. This condition in the dog has been controversial because of difficulties in defining its aetiology and pathogenesis. There have been suggestions that it be renamed antibiotic-responsive diarrhoea (ARD) until more is known about its aetiopathogenesis. However, this does not apply to all cases since it is not always associated with diarrhoea; indeed, weight loss alone can be the only presenting sign.
Accumulated data on clinical cases indicate that SIBO should be considered an important emerging syndrome that may occur in many breeds of dog. It typically presents in young animals as chronic intermittent small bowel diarrhoea, which may be accompanied by loss of body weight or failure to gain weight. Clinical signs are variable and some animals may only exhibit weight loss, while others may have intermittent vomiting or signs suggestive of mild colitis.
SIBO may develop if the normal host defence mechanisms, such as gastric acid secretion, intestinal peristalsis, the ileocaecal valve, intestinal immunoglobulin secretion, and mucus barrier are impaired. In people, SIBO is usually associated with intestinal stasis (blind loop syndrome). Small intestinal dysmotility, as evidenced by reduced migrating motor complex activity, is probably responsible for the prevalence of SIBO in elderly human patients. In dogs, there is rarely evidence for stasis, and the cause of SIBO is often unknown. A naturally developing enteropathy associated with SIBO was first described in German Shepherd Dogs, and it has been postulated that this is related to an apparent relative deficiency of IgA in this breed. SIBO may also develop secondary to exocrine pancreatic insufficiency, and has been reported in asymptomatic laboratory Beagles. We have documented SIBO by culture of duodenal juice in over half of dogs with chronic intestinal disease; dogs of many breeds are affected, although there is a predominance of German Shepherd Dogs. Serum IgA levels in these dogs have been variable. Predisposing conditions usually cannot be identified, although it remains important to rule out causes of intestinal stasis, such as neoplasia and intussusception. Increased numbers of pathogenic E. coli have been demonstrated in the duodenal juice of these dogs, and these may also play a role in the development of this condition. SIBO may furthermore be a secondary complication of many intestinal diseases due to altered intestinal motility and/or local immunity; in addition, malabsorption of nutrients may cause an environment favourable for bacterial proliferation. Conversely, bacterial antigens gaining access to the lamina propria also may cause an inflammatory reaction, although this tends to be milder.
Bacteria or their secreted products can directly damage the mucosa or indirectly impair absorption by competing for nutrients and by changing intraluminal factors such as the concentration of conjugated bile acids. This results in diarrhoea and steatorrhoea, competition with the host for nutrients, and weight loss. Enterocyte damage is often not visible on light microscopy, but may be demonstrated using biochemical or ultrastructural studies, or by measurement of intestinal permeability. Increased mucosal production of interleukin-6, a cytokine that plays a central role in the regulation of inflammatory and immune reactions, has been demonstrated in people with SIBO, suggesting heightened mucosal immune activity.
The species of bacteria in duodenal juice of dogs with SIBO varies markedly, with coliforms, staphylococci, enterococci, and Clostridium and Bacteroides spp. predominating. Anaerobic overgrowth is most common, found in approximately 70% of dogs with SIBO. This is of clinical significance, since anaerobic bacteria have a much greater potential to damage the intestinal brush borderand cause malabsorption; in addition, anaerobes, especially Bacteroides, are the major cause of bile salt deconjugation resulting in fat malabsorption and steatorrhoea.
Symptomatic SIBO typically presents in young animals as chronic intermittent small bowel diarrhoea, which may be accompanied by loss of body weight or failure to gain weight. Diarrhoea often has been present since puppyhood, and gradually worsens. Some dogs also may have signs of a mild colitis, due to colonic irritation by bacterial metabolites, and these dogs may be erroneously diagnosed as having primary colitis. Weight loss may be severe, and is in some dogs the only sign. Appetite is often reduced. Vomiting is not typically associated with bacterial overgrowth; its presence suggests concurrent inflammatory bowel disease. Some dogs with SIBO are presented because of excessive intestinal gas.
CBC and biochemical profile should be performed to rule out systemic disease. Faeces should be examined for parasites and cultured for enteric pathogens. Abdominal radiography and especially ultrasound can be helpful to rule out partial obstruction, particularly in young (intussusception) or older (neoplasia) animals. Subsequently, exocrine pancreatic insufficiency (EPI) should be ruled out by assay of serum TLI activity.
Serum Folate and Cobalamin
Assays of serum folate and cobalamin appear to be the most helpful aids to the diagnosis of SIBO in the dog for use in general practice, although they have poor sensitivity (i.e., many affected dogs do not have abnormal test results). Normal serum vitamin concentrations do not exclude the possibility of SIBO, because alterations depend on the type and numbers of organisms present, the severity of any secondary mucosal damage that may interfere with folate absorption despite high intraluminal concentration, and depletion of body stores. If pancreatic function is normal (i.e., serum TLI is normal) then finding a decreased serum cobalamin concentration or increased serum folate is supportive of SIBO. If both of these are found together, SIBO is extremely likely; however, this combination occurs infrequently. High serum folate may also be a consequence of high folate intake, such as a high-folate diet or coprophagia. Demonstration of low serum cobalamin is the more useful finding, since it is less influenced by diet and coprophagia and appears to relate more to the severity of clinical disease
Measurement of intestinal permeability is a sensitive tool for the detection of mucosal damage, but it does not tell you about the underlying cause. However, these tests are useful to detect and assess the severity of mucosal damage in dogs with overgrowth. Increased intestinal permeability can be demonstrated using a differential sugar absorption test in 50–60% of clinical cases with SIBO, even when there are no histologic abnormalities. In addition, changes in intestinal permeability following antibiotics may be used to monitor response to treatment. Normalization of intestinal permeability following antibiotic therapy suggests successful treatment, and antibiotics may be discontinued. Antibiotics possibly should be continued longer if permeability remains high despite apparent response to treatment; in addition, other causes of intestinal disease should be suspected and investigated (e.g., dietary sensitivity). Persistent high permeability in dogs with a poor clinical response should prompt one to look for underlying disease, such as a primary inflammatory bowel disease.
Breath Hydrogen Testing
Breath tests measure the breath excretion of CO2 or hydrogen (H2) produced by intraluminal bacterial metabolism of an administered substrate. They appear to be the one of the most sensitive and specific tests available for the diagnosis of SIBO, although they are not yet technically feasible in most veterinary practices. The H2 breath test has been used most often in both human and veterinary medicine. It has been used not only for diagnosis of SIBO but also for detection of carbohydrate malassimilation and measurement of oro-caecal transit time. The time after ingestion of the test substrate at which increased breath H2 concentrations are first detected is used to distinguish between SIBO and carbohydrate malabsorption. In SIBO, elevated breath H2 concentrations occur within 1 to 2 hours after ingestion of the test substrate. An H2 breath test using a multiple sugar solution has been used successfully for detection of SIBO in dogs and has the advantage that it simultaneously allows for quantification of intestinal permeability. A limitation of breath H2 tests in people is that 15–20% of the human population have intestinal flora that does not produce hydrogen and therefore cannot demonstrate a positive test result if bacterial overgrowth develops. The same probably applies to the dog, since there are significant numbers of dogs with culture-proven overgrowth but persistently negative breath tests.
The H2 breath test is more sensitive than serum folate and cobalamin assay, and has been useful to identify cases of SIBO with a falsely negative duodenal juice culture. A positive breath H2 test is very suggestive of SIBO, and there is no need to culture duodenal juice in these cases. However, a negative test does not rule it out, and culture of duodenal juice remains necessary in these patients.
Culture of Duodenal Juice
Definitive diagnosis of SIBO is based on results of microbiologic culture of duodenal juice, obtained usually at endoscopy or alternatively via intra-operative permucosal aspiration. Juice culture is still the gold standard for the diagnosis of SIBO, but it is technically difficult, time-consuming and expensive, and it may still not identify all cases of SIBO (for example when this is in the more distal portions of the small intestine or in isolated pockets). However, intestinal biopsies can be taken at the same time as the juice collection, and these are useful to rule out primary mucosal disease as the cause of malabsorption. Duodenal biopsy in SIBO is often normal. Over 75% of clinical cases with SIBO will have no histologic abnormalities, whereas mild to moderate lymphocytic infiltrates occur in up to 25%. Mild lymphocytic-plasmacytic enteritis can occur as a consequence of SIBO, and may resolve following appropriate antibiotic treatment.
Duodenal bacterial counts may be influenced by environmental factors, such as housing conditions (kennelled dogs tend to have higher bacterial numbers, perhaps associated with coprophagia) and infective load (such as endoparasites and naturally occurring enteropathogens in hot climates). This should be taken into account when defining bacterial levels deemed diagnostic of bacterial overgrowth.
Bacterial deconjugation of bile salts may result in increased serum concentrations of unconjugated bile acids. Unlike the conjugated bile acids normally present in the small intestinal lumen, these unconjugated bile acids (UBA) diffuse across the intestinal mucosa into the blood. Dogs with SIBO have been shown to have significantly higher serum concentrations of UBA. This test has also proven useful to identify dogs with culture proven SIBO that did not have abnormal serum vitamin concentrations. Until now, this test was technically too complicated for routine use, but new developments should lead to this becoming more available in the near future. It may therefore become a useful addition to the battery of diagnostic tests required to diagnose SIBO.
Response to treatment with antibiotics may also help in the tentative diagnosis of SIBO. However, lack of response does not rule it out, since prolonged treatment may be required in some dogs before clinical improvement is manifest.
SIBO can be a subclinical intestinal abnormality, as has been reported in man, German Shepherd dogs and laboratory Beagles. Development of clinical signs in these individuals probably depends on the nature of the bacterial population (for instance, colonization with anaerobes is more likely to result in signs) and the effect of the overgrowth flora on the local immune system. These patients may be identified on basis of abnormalities in serum folate and/or cobalamin concentrations, a positive hydrogen breath test, or by culture of duodenal juice aspirated in the course of other investigations. Treatment is not required as long as they are asymptomatic; however, they are at risk for developing signs once the delicate balance in their intestinal ecosystem is disturbed. Progressive decreases in serum cobalamin concentration in dogs with asymptomatic SIBO often precede development of clinical signs.
An attempt should be made to identify and correct an underlying cause, such as partial obstruction due to intussusception, tumours or foreign bodies. Detection of dysmotility is more difficult and often not feasible; however, motility modifying agents such as cisapride or low-dose erythromycin may empirically be used in refractory patients. In many dogs with SIBO a cause cannot be found, and long-term oral antibiotic treatment is required. Oxytetracycline (10–20 mg/kg TID for 28 days) is used initially, and may need to be continued for extended periods if clinical signs recur on withdrawal of medication. Its mechanism of action may involve more than just pure antibacterial action (e.g., direct influence on the mucosa), although this is not certain. Metronidazole (10–20 mg/kg TID) and tylosin (20 mg/kg BID) are good alternative choices and are used if dogs fail to respond to oxytetracycline. Broad-spectrum bactericidal antibiotics tend to be less effective.
Dietary management with a low fat diet may also be valuable, because this can minimize the secretory diarrhoea, which is a consequence of bacterial metabolism of fatty acids and bile salts. Since intestinal permeability is often increased in SIBO, a restricted antigen diet may be of value to reduce the incidence of secondary dietary sensitivities. Dietary supplementation with fructo-oligosaccharides has been suggested as a means of modifying bacterial counts in the small intestine in German Shepherd Dogs with asymptomatic naturally occurring bacterial overgrowth. However, since these compounds are more likely to affect the large rather than the small intestine, further studies in clinical cases are required to assess the efficacy of prebiotics in the management of canine SIBO.
Probiotics are a mixture of non-pathogenic bacteria, often containing Lactobacillus, which can change intestinal pathobiology by preventing enteric infections, modifying metabolic actions of intestinal bacteria, and promoting nutrition. They also may promote local mucosal and systemic immune response. Probiotics are extensively used in large animals, and have also been advocated as a means of modulating gut flora in people with gastrointestinal disease.
Parenteral cobalamin (e.g., 500 µg/month for 6 months) may help dogs with apparent cobalamin deficiency. It may have to be given more frequently if serum cobalamin levels remain subnormal. Persistently low serum cobalamin levels are often associated with a poor clinical response to treatment.
Prolonged antibiotic therapy is often required in treatment of dogs with idiopathic SIBO, and serial measurement of intestinal permeability and breath H2 testing are helpful in monitoring response to treatment. Some dogs with SIBO relapse as soon after antibiotics are discontinued. In these patients long-term antibiotic treatment will be required, but empiric reduction of the dose to well below the recommended level may be effective in controlling signs.
In dogs with moderate to marked inflammatory bowel disease, corticosteroids should be added to the treatment regimen if response to antibiotics alone is inadequate. Corticosteroids are not recommended in the initial treatment of dogs with lymphocytic/plasmacytic enteritis and SIBO because in our experience they appear to worsen clinical signs associated with SIBO.
Chronic SIBO may cause permanent functional damage to the intestinal mucosa. This may explain the poor response to treatment of some dogs, and also the need for indefinite dietary management with controlled diets after apparent successful antibiotic therapy in some dogs with chronic SIBO.
1. Rutgers HC, Batt RM, Elwood CM, Lamport A. Small intestinal bacterial overgrowth in dogs with chronic intestinal disease. J Am Vet Med Assoc 1995;206:187-19.
2. Rutgers HC, Batt RM, Proud FJ, et al. Intestinal permeability and function in dogs with small intestinal bacterial overgrowth. J Sm Anim Pract 1996;37:428-434.
3. Bissett SA, Guilford WG, Spohr A. Breath hydrogen testing in small animal practice. Comp Cont Educ 1997;19:916-931.
4. Ludlow CL, Davenport DJ. Small intestinal bacterial overgrowth. In: Bonagura JD, ed. Current Veterinary Therapy XIII. Philadelphia, WB Saunders, 1999: 637-641.
5. Melgarejo T, Williams DA, O'Connell NC, Setchell KD. Serum unconjugated bile acids as a test for intestinal bacterial overgrowth in dogs. Dig Dis Sci 2000; 45:407-414.