Normally the bacterial population of the small intestine is controlled by a number of mechanisms (see above). Bacterial overgrowth is the uncontrolled proliferation of these bacteria and, in humans, occurs secondary to a number of underlying disorders that interfere with the control mechanisms. Although the existence of small intestinal bacterial overgrowth in humans is not disputed, the subject is a source of controversy in small animal gastroenterology. In dogs, small intestinal bacterial overgrowth is best considered a clinical sign or a pathogenetic mechanism rather than a diagnosis. Historically, the term idiopathic small intestinal bacterial overgrowth was used to describe an antibiotic-responsive condition of large breed (especially German shepherd) dogs for which no underlying cause could be recognized. However, given concerns as to whether a true overgrowth exists in these cases, the alternative name of antibiotic-responsive diarrhea has been suggested. Although cats might feasibly suffer from secondary SIBO, an idiopathic antibiotic-responsive condition similar to that seen in German shepherds has not been documented in this species.
Genuine bacterial overgrowth is defined by an increase in the absolute number of bacteria in the upper small intestine during the fasted state (i. e., the number of colony-forming units cultured per milliliter of duodenal juice (CFU / mL]). The upper limit for normal bacterial numbers was defined in humans, and that number has been extrapolated to dogs. Controversy exists as to its validity, however, because the original work used a small number of dogs and questionable bacteriologic techniques. Subsequent studies with different collection methods and improved anaerobic culture techniques have demonstrated that a count equal to or greater than 107 CFU / mL total bacteria is commonly found in asymptomatic dogs. Therefore, although a genuine bacterial overgrowth may exist in conditions equivalent to those in humans, defining the presence of small intestinal bacterial overgrowth based on the original numeric limit is flawed. In this chapter, cases with a documented underlying cause are defined as secondary SIBO, and the term idiopathic antibiotic-responsive diarrhea is used for idiopathic antibiotic-responsive conditions without an obvious underlying cause.
Etiology and Pathogenesis
Secondary small intestinal bacterial overgrowth SIBO can occur secondary to (1) diseases that result in excess substrate in the intestinal lumen (e. g., EPI, motility disorder, blind loop), (2) diseases that affect the clearance of bacteria (e. g., partial obstruction, abnormal motility), or (3) morphologic or functional derangement of the mucosa (Box Causes of Secondary Small Intestinal Bacterial Overgrowth). Increased numbers of bacteria in the upper small intestine cause malabsorption and diarrhea through several mechanisms. First, bacteria compete for nutrients; for example, by binding cobalamin and reducing its availability for absorption. Second, bacterial metabolism of nutrients can create products that provoke diarrhea (e. g., hydroxylated fatty acids and deconjugated bile salts), leading to diarrhea from fat malabsorption and stimulation of colonocyte secretion. Finally, the bacterial flora may damage the mucosal brush border, and alterations in enzyme activity, which reverse on antibiotic treatment, can be detected.
Causes of Secondary Small Intestinal Bacterial Overgrowth
- Spontaneous (atrophic gastritis)
- Acid blockers
Exocrine pancreatic insufficiency
Partial intestinal obstruction
- Chronic intussusception
- Surgical resection of ileocolic valve
- Blind loop
- Latent primary pathogens (?)
- Inflammatory bowel disease (cause or effect?)
- Chronic giardiasis
Dietary sensitivity (?)
Idiopothic antibiotic-responsive diarrhea A number of hypotheses exist as to the cause of idiopathic antibiotic-responsive diarrhea. Historically, hypotheses were based on the belief that a genuine increase in small intestine bacterial numbers was present and therefore pathogenesis was related to the mechanisms described above. An underlying defect allowing overgrowth was not obvious, and suggested mechanisms such as abnormal intestinal motility or achlorhydria were not proved.
Given that recent studies have questioned whether a genuine increase in bacterial numbers occurs, recent hypotheses now focus on host-bacterial interactions. As was suggested for mouse models of enteric inflammation, antibiotic-responsive diarrhea may develop secondary to defects in the mucosal barrier, aberrant mucosal immune responses, qualitative changes in the enteric bacterial flora, or a combination of these mechanisms. Defects in the mucosal barrier are supported by studies documenting abnormal permeability and the presence of brush border enzyme defects. A possible underlying selective IgA deficiency in the German shepherd breed has been postulated. German shepherds with intestinal disease have defective small intestinal IgA production, although mucosal IgA+ plasma cell numbers in affected dogs are either normal or increased. The cause of this IgA deficiency is not clear, but a complex defect is likely, involving abnormalities either in the production and release of IgA from the plasma cell or in the pathway of translocation of IgA across the epithelium during secretion.
Studies demonstrate that dogs with antibiotic-responsive diarrhea have increased lamina propria CD4+ T cells and increased expression of certain cytokines. It therefore is tempting to speculate that this represents immune dysregulation and perhaps a loss of tolerance toward endogenous bacterial antigens. Such a hypothesis is supported by the fact that antibacterials lead to resolution of clinical signs and decreased cytokine expression but not to a decline in bacterial numbers. The fact that the most effective antibacterials are those with immune-modulating properties (e. g., oxytetracycline, metronidazole, tylosin) may support this hypothesis.
An alternative hypothesis is that an unidentified pathogen is involved; candidates include intestinal Helicobacter spp. or enteropathogenic E. coli. The predisposition of German shepherds to this syndrome could therefore be explained by genetic susceptibility to infection as a result of MHC II antigen expression.
Idiopathic antibiotic-responsive diarrhea Idiopathic antibiotic-responsive diarrhea is most commonly recognized in young German shepherds, although cases have been reported in other dog breed? (but not in cats). Affected dogs show signs of chronic intermittent diarrhea accompanied by weight loss and / or stunting. Intermittent, watery diarrhea, often associated with excessive gas production (manifested as borborygmi and flatus), is seen most frequently. However, vomiting and signs of colitis are sometimes reported, and occasionally dogs are stunted yet do not have diarrhea. Activity levels are normal, and appetite is variable; most affected dogs have polyphagia, pica, or coprophagia, but few are anorectic. A positive response to antibiotics is expected, and the clinical condition may deteriorate if corticosteroids are given. The major differential diagnoses are exocrine pancreatic insufficiency and inflammatory bowel disease, both of which are common in German shepherds.
Secondary small intestinal bacterial overgrowth SIBO may occur secondary to numerous primary conditions (see Box Causes of Secondary Small Intestinal Bacterial Overgrowth), and clinical signs usually relate to the underlying condition. However, signs of secondary small intestinal bacterial overgrowth also can been seen and are indistinguishable from those of idiopathic antibiotic-responsive diarrhea, with diarrhea predominating. When small intestinal bacterial overgrowth develops secondary to a partial obstruction or focal dysmotility, bacterial numbers can exceed 109 CFU / mL. Clinical signs may be noted intermittently, because recurrent diarrhea can temporarily flush out the overgrowth.
Using the historical numeric cutoff of 105 CFU / mL total bacteria, secondary small intestinal bacterial overgrowth was considered common in chronic enteropathies. In reality, true secondary small intestinal bacterial overgrowth is uncommon, with the exception of small intestinal bacterial overgrowth secondary to EPI. An increase in small intestine bacterial numbers has been documented in experimentally induced EPI, although bacterial numbers decrease upon treatment of exocrine pancreatic insufficiency with enzyme replacement. Therefore in many cases the small intestinal bacterial overgrowth itself is of no significance. However, a proportion of naturally occurring exocrine pancreatic insufficiency cases respond suboptimally to pancreatic enzyme supplementation alone and may require concurrent antibiotic therapy. Given that the majority of dogs affected with exocrine pancreatic insufficiency are German shepherds, it is not clear whether this is the result of secondary small intestinal bacterial overgrowth or of a concurrent idiopathic antibiotic-responsive diarrhea.
The diagnosis of small intestinal bacterial overgrowth and antibiotic-responsive diarrhea is controversial. In all cases it is critical that a thorough investigation be conducted to eliminate causes of secondary small intestinal bacterial overgrowth before the patient is treated with antibacterials. In this regard, diagnostic imaging, especially ultrasonography, is useful for ruling out partial intestinal obstructions. Systemic disorders should be ruled out with a minimum database, and exocrine pancreatic insufficiency is eliminated by serum TLI assay. Fecal examination for parasitic and bacterial diseases is also mandatory.
Idiopathic antibiotic-responsive diarrhea Although both direct and indirect tests were previously advocated for idiopathic SIBO, recent studies have suggested that they are all of limited value. One recent study demonstrated that neither indirect biochemical markers (folate, cobalamin, unconjugated bile acids) nor quantitative bacterial culture could reliably identify cases of antibiotic-responsive diarrhea. Therefore the only available diagnostic test for antibiotic-responsive diarrhea is response to an antibacterial trial. However, such a diagnostic modality is appropriate only after thorough diagnostic investigations have eliminated all other causes of an antibacterial responsiveness.
Secondary small intestinal bacterial overgrowth Although numerous tests are available to document secondary overgrowth, in practice it is more important to identify the underlying cause.
Duodenal juice culture Quantitative aerobic and anaerobic culture of duodenal juice previously formed the gold standard for diagnosis. For dogs, increases in either total or anaerobic bacteria above the upper cutoff (105 CFU / mL total bacteria or 104 CFU / mL anaerobic bacteria) were considered diagnostic. However, the validity of these cutoffs has been questioned. A count of 107 CFU / mL total duodenal bacteria has been documented commonly in healthy dogs, and numbers as high as 109 CFU / mL have been found occasionally in cats and asymptomatic dogs. Some of the discrepancies may reflect difficulties and differences in the methodology, because numbers vary widely when individual animals are repeatedly sampled. Use of an inappropriately low cutoff value leads to overdiagnosis of SIBO, which probably explains why it has been reported in 50% of dogs with chronic intestinal disease. Culture of endoscopic biopsies has not been shown to be of greater diagnostic utility.
The flora present in idiopathic antibiotic-responsive diarrhea may be comprised predominantly of either aerobic or anaerobic bacteria, but it tends to be a mixed population, with staphylococci, streptococci, coliforms, enterococci, and corynebacteria and anaerobes such as bacteroids, fusobacteria, and clostridia. These bacteria generally are commensals found normally in the oropharynx, small intestine, and large intestine. However, culture of fecal bacteria cannot be correlated with small intestine bacterial numbers and cannot be used to diagnose this condition.
Indirect tests for small intestinal bacterial overgrowth / antibiotic-responsive diarrhea Indirect tests include serum biochemical markers and breath hydrogen analysis.
Serum folate and cobalamin concentrations Bacteria synthesize folate and bind cobalamin, preventing its absorption. Therefore small intestinal bacterial overgrowth would be predicted to be associated with an increased serum folate concentration or a decreased cobalamin concentration, or both. Although such results would be expected in cases of genuine small intestinal bacterial overgrowth (e. g., secondary SIBO), no studies have specifically assessed the usefulness of these tests in those diseases. Alterations of serum folate and cobalamin noted in exocrine pancreatic insufficiency may reflect pancreatic dysfunction rather than secondary SIBO.
Furthermore, recent studies have demonstrated that these tests are of limited value in the diagnosis of idiopathic antibiotic-responsive diarrhea. This poor performance may be related to dietary factors, the presence of concurrent disease, or the use of drugs that alter serum vitamin concentrations. The other possibility (for idiopathic antibiotic-responsive diarrhea) is that disease pathogenesis is not related to genuine increases in bacterial numbers. Although these tests are often the only tests available to practitioners, they are generally unhelpful in the diagnosis of idiopathic antibiotic-responsive diarrhea.
Serum unconjugated bile acids Bile acids are synthesized and conjugated in the liver and excreted into the intestine via the biliary tract. Some small intestine bacterial species can deconjugate bile acids, which then are absorbed passively by the small intestine. Studies in humans have suggested that increased serum unconjugated bile acid (SUBA) concentrations are an indirect indicator of SIBO. A preliminary study in dogs suggested that SUBA concentrations increased in SIBO. However, these results have been contradicted by another recent study, which demonstrated that SUBA concentrations were neither sensitive nor specific for diagnosis of idiopathic antibiotic-responsive diarrhea. Further work may be required to clarify the diagnostic value of SUBA concentrations in dogs.
Other biochemical tests Measurement of increased amounts of a bacterial product made either naturally or after oral administration of a test substance could be used to diagnose SIBO, but none of these are reliable.
Breath hydrogen excretion Bacterial fermentation in the intestinal tract releases hydrogen which, after systemic absorption, is exhaled and can be measured in breath samples. Theoretically, small intestinal bacterial overgrowth can result in a high resting breath hydrogen concentration or an early (or double) hydrogen peak after ingestion of a test meal. However, increased breath hydrogen can also be seen with carbohydrate malabsorption or decreased orocecal transit time. Given these problems with interpretation and the fact that protocols have not been standardized, breath hydrogen testing has not been widely adopted.
Intestinal permeability Intestinal permeability, as measured by 51Cr-EDTA and differential sugar absorption, can be abnormal in small intestinal bacterial overgrowth and can improve after antibiotic treatment. However, such findings are not pathognomonic for either secondary small intestinal bacterial overgrowth or idiopathic antibiotic-responsive diarrhea.
Lack of histologic changes on intestinal biopsy Histopathologic examination of intestinal biopsies is most often normal or demonstrates only subde abnormalities. However, such findings are not pathognomonic, because other conditions yield similar results. Despite the lack of histologic evidence of inflammation, disturbances in immune cell populations have been noted, most notably increases in IgA+ plasma cells and CD4+ cells. However, such techniques have not been applied for routine diagnostic purposes.
Empiric response to antibiotics Currendy, the best diagnostic test for idiopathic antibiotic-responsive diarrhea is, logically, the response to empirical therapy. However, a response to antibacterials is not specific and indeed may be beneficial in inflammatory bowel disease, infectious diarrhea, and even a range of nonenteric diseases such a portovascular anomalies. Furthermore, response to antibiotic therapy does not discriminate idiopathic antibiotic-responsive diarrhea from secondary SIBO. Therefore an empirical response to antibiotics is valid only after thorough diagnostic investigations have eliminated other possible causes.
The suggested criteria for a diagnosis of idiopathic antibiotic-responsive diarrhea are (1) a positive response to the antibiotic trial based on resolution of relevant clinical signs; (2) relapse of signs upon withdrawal of treatment; (3) remission on reintroduction of antibiotics; and (4) elimination of other etiologic causes based on the results of other diagnostic tests and histopathologic assessment.
Secondary small intestinal bacterial overgrowth Although antibacterial therapy improves clinical signs, appropriate treatment for the underlying condition is preferable. For EPI, pancreatic enzyme supplementation can reduce bacterial numbers because exogenous proteases have antibacterial properties.
Idiopathic antibiotic-responsive diarrhea No cure is available for idiopathic antibiotic-responsive diarrhea, but signs can be controlled with antibacterials. A broad-spectrum antibiotic is indicated; suitable choices include oxytetracycline (10 to 20 mg / kg given orally every 8 hours), metronidazole (10 to 20 mg / kg given orally every 8 hours), and tylosin (20 mg / kg given orally every 8 or every 12 hours). Oxytetracycline (OTC) is cheap, and because systemic absorption is not required, it can be given with food. It cannot be used before permanent tooth eruption because it causes staining of tooth enamel. Some authors have criticized the use of oxytetracycline because it is associated with rapid development of plasmid-mediated antibiotic resistance. However, given that long-term efficacy is maintained in most cases, oxytetracycline may not be acting through its antibacterial properties, because it does not significantly reduce small intestine bacterial numbers. Rather, it may provide a selective pressure on the intestinal flora, encouraging the establishment of less harmful bacteria, or it may exert immunomodulatory effects, which this antibiotic group has. Immunomodulatory activity has also been suggested for other antibacterials commonly used to treat antibiotic-responsive diarrhea, namely metronidazole and tylosin.
Whichever antibacterial is chosen, a 4- to 6-week course is appropriate initially, although the antibiotic should be changed after 2 weeks if the response has been suboptimal. In some cases, premature cessation of treatment can lead to relapse, and prolonged therapy often is necessary. In some animals a delayed relapse occurs several months after cessation of antibiotics, and such cases require either repeated courses or indefinite therapy. Efficacy is often maintained despite a reduction in dosage frequency from three times to even once daily. Dogs may also “outgrow” the problem with age, either as a result of a decrease in caloric intake or because of developing maturity of the mucosal immune system. It has also been suggested that idiopathic antibiotic-responsive diarrhea in German shepherds may predispose some of these dogs to inflammatory bowel disease later in life, but currently no direct evidence supports this supposition.
Ancillary treatments Dietary manipulation can be a useful adjunct to the treatment both of idiopathic antibiotic-responsive diarrhea and of secondary SIBO. In general, a highly digestible, low-fat diet is desired to reduce the substrate available for bacterial use. Addition of FOS has been advocated to reduce small intestine bacterial numbers, although evidence of efficacy is conflicting. Administration of probiotics has not been thoroughly assessed in idiopathic antibiotic-responsive diarrhea. Finally, if low cobalamin concentrations are documented, parenteral cobalamin therapy is warranted.