Small bowel bacterial overgrowth (SBBO) is defined as an increase in the number of bacteria in the upper gastrointestinal tract, often changing from predominately oropharyngeal to colorectal species (1). Various pathologic alterations in gastrointestinal (GI) tract structure and function result in SBBO. Etiologic factors in the development of SBBO include anatomic abnormalities, abnormalities of intestinal peristalsis, and abnormalities of intestinal immunologic and nonimmune defense mechanisms(2) (Table 1). Normally, small bowel contamination is limited by antegrade peristalsis, bactericidal action of gastric acid, enzymatic digestion, mucus entrapment, low environmental loads, and the ileocecal valve.
Symptoms of SBBO include varying combinations of pain, diarrhea, dyspepsia, and weight loss. When symptoms become severe, treatment is warranted. The nature of the treatment depends on precipitating factors, bacterial species involved, and severity of symptoms. Treatment of SBBO most commonly entails use of antibiotics. Broad-spectrum oral antibiotics, initially given intermittently, are often effective in reducing the number of bacteria. Surgical correction of such localized anatomic abnormalities such stricture, fistulas, or diverticula is often immediately corrective. Treatment in short bowel syndrome, however, may involve simplistic measures such as a decrease in enteral feeding or more drastic measures such as bowel tapering and/or lengthening (3).
Identifying the underlying cause of SBBO is helpful in determining effective treatment. The problem may be persistent despite numerous medical therapies, and surgery is often not curative. Effective treatment strategies with limited side effects are still needed to deal more effectively with this common complication of short bowel syndrome.
CASE REPORTS
Six cases of SBBO related to short bowel syndrome are presented, all with varying clinical manifestation. These patients required additional interventions because of failure of antibiotic treatment. Patient characteristics are summarized in Table 2.
Patient 1 was born at 26 weeks' gestation, and necrotizing enterocolitis developed. After 2 years of home parenteral nutrition, intravenous nutrition constituted 80% and elemental enteral nutrition 20% of total calories. At this time, SBBO developed for the first time, with acidosis and dehydration requiring hospitalization. Various antibiotic regimens based on small bowel culture and sensitivity determination and intermittent intestinal flushes with magnesium citrate were unsuccessful in controlling overgrowth. Ongoing evaluation demonstrated the lack of significant bowel dilatation. It became apparent when interviewing the family in more detail that SBBO began with attempts at toilet training and the patient quickly learned to withhold stools to continue uninterrupted play activities. Short-term therapy with bisacodyl suppositories resulted in good colonic evacuation, which corrected SBBO, abdominal distention, and D-lactic acidosis. Because of concern about the long-term use of this product, a transverse colostomy was performed. Antibiotics were discontinued, permitting progressive discontinuation of parenteral nutrition. Ultimately, nutrition and growth were maintained with an 8-hour nighttime infusion of an elemental formula. After 2 years, the colostomy was reversed. Regular evacuations through the rectum were established and at 1-year follow-up, and the patient has had no reoccurrence of SBBO.
Patient 2 was born after a full-term pregnancy and had a mid-gut volvulus soon after birth. At 4 years of age he was receiving 75% of calories by parenteral nutrition and 25% by gastrostomy with an intact protein formula. At this time, recurrent central catheter infections and encephalopathy related to D-lactic acidosis occurred in association with enterocolitis from SBBO. Numerous antibiotic regimens were tried without prolonged success. Sulfasalazine and corticosteroids were used temporarily to control intestinal inflammation. Because bacteria embedded with mucus were repeatedly identified on the surface of gastric and intestinal biopsy specimens, a 2000 ml of hypertonic PEG-3350 solution (GoLYTELY, Braintree Laboratories, Braintree, MA, U.S.A.) in 10 hours was administered enterally twice a week to control SBBO. This therapy was effective, and during the next year, antiinflammatory therapy was discontinued, and stable antibiotic therapy for SBBO was used without recurrent D-lactic acidosis. During the next 2 years, home parenteral nutrition was discontinued gradually as growth and development were maintained with nighttime enteral nutrition and a high-fat diet by mouth.
Patient 3 was born at 35 weeks' gestation with gastrochisis. During the first 2 years of life, she was maintained at home on parenteral nutrition and bolus feedings of Pregestimil (Mead-Johnson, Evansville, IN, U.S.A.). Because of an increased concentration of liver enzymes, an effort was made to discontinue parenteral nutrition using continuous enteral feedings. During this time, abdominal distention, vomiting, and D-lactic acidosis resulted from SBBO. Radiographic studies revealed dilated small bowel, and a bowel lengthening (Bianchi) procedure was performed. The patient was discharged receiving nighttime parenteral nutrition and continuous enteral feeding by gastrostomy button. Total parenteral nutrition was discontinued 3 months after surgery with no evidence of SBBO. Growth and development were maintained with a continuous enteral infusion of an elemental formula. Two years after the initial surgery, D-lactic acidosis occurred; however, an upper gastrointestinal-small bowel series failed to demonstrate a recurrence of bowel dilation. Antibiotic therapy was unsuccessful in controlling overgrowth. Therapy on alternate days with 500 ml hypertonic saline enemas was begun and proved effective in alleviating overgrowth complications for the next year and a half. Subsequently, the enemas were reduced to a frequency of once a week for 3 months and then were discontinued. Special instructions for frequent voluntary stool evacuations were given to the child, and the school was enlisted to assist in facilitating restroom availability.
Patient 4 was born full-term with gastroschisis and mid-gut volvulus. She required daily intravenous nutrition and fluid replacement along with continuous enteral nutrition. Problems with SBBO began around the age of 4 years and was suspected because of recurrent Escherichia coli sepsis. External contamination caused by poor catheter technique was repeatedly evaluated and was determined not to be a contributing factor. Antibiotic therapy was not beneficial, and hypertonic saline (GoLYTELY) flushes were not tolerated because of vomiting during the infusion. Therefore, a bowel lengthening (Bianchi) procedure was performed, which markedly decreased the incidence of catheter sepsis. Total parenteral nutrition requirements were subsequently reduced by 50%.
Patient 5 had mid-gut volvulus as an infant. He was maintained on TPN for approximately 3 years, during which time enteral feeding by gastrostomy tube was slowly advanced and continued until the patient was 7 years old. After enteral nutrition was discontinued, abdominal distention and malodorous flatus was noted. SBBO was confirmed by culture of small intestinal fluid and a positive glucose breath hydrogen test. Antibiotic therapy with trimethoprim-sulfamethoxazole and metronidazole was initiated with resolution of these symptoms. Because of the caregiver's desire to discontinue medication, the patient was switched to a probiotic, 1010 cfu ofLactobacillus plantarum 299V once a day (CAG Nutrition, Omaha, NE, U.S.A.). Within 2 to 3 weeks of initiating therapy, there was improvement in stool consistency, primarily in reduction of water content. Probiotic therapy was discontinued after 2 months to validate effect. Approximately 10 days after discontinuation, the patient experienced a relapse of SBBO, characterized by loose, watery stools and clinically apparent vasculitis confirmed by punch biopsy in a lower extremity. The probiotic was not reinstituted at the parents' request.
Patient 6 experienced a mid-gut volvulus at age 5. After that time, he was dependent on total parenteral nutrition, receiving only 25% of his nutrition enterally. A Bianchi procedure was performed at 11 years of age for problems with SBBO caused by significant bowel dilatation identified on an upper gastrointestinal-small bowel series. The patient improved and was slowly tapered off intravenous nutrition during the next year. Three years after the Bianchi procedure, recurrent SBBO occurred with watery, occasionally bloody diarrhea, abdominal distention, and arthritis in interphalangeal joints, necessitating reinstitution of intravenous nutrition, and use of continuous antibiotics and corticosteroid therapy. Total parenteral nutrition was eventually reduced to 1000 calories every other day, along with a high-fat oral diet. Symptoms were controlled for 2 years with this regimen; however, arthritis recurred. Treatment with sulfasalazine provided some symptomatic relief. Probiotic therapy using 1010 cfu ofLactobacillus plantarum 299V daily was initiated and after a few weeks, antibiotic therapy was discontinued, as was intravenous nutrition and medication therapy for arthritis. The patient was ultimately switched to 1010 cfu of Lactobacillus GG (CAG Nutrition), a probiotic available in capsule form, with an equally good therapeutic response.
DISCUSSION
At birth, the intestine is sterile. Detectable colonization begins within 3 days after birth by essentially the same organisms as were carried by the infant's birth mother. Within the first 3 years of life, all protective microbial flora-that is, nonpathogenic, normal resident bacteria-are obtained(4,5). The normal intestinal tract has 10 times more microbial organisms than all the remaining cells in the human body(6). The number of commensal organisms in the proximal small bowel is much smaller than in the colon and includes both facultative anaerobic and aerobic bacteria. Up to 100,000 primarily aerobic organisms per milliliter are found in the proximal small intestine increasing to more than 100 billion predominantly strict and facultative anaerobes per milliliter in the colon (2). The type of bacterial flora varies along the length of the normal gastrointestinal tract. Diet influences colonization patterns: Facultative anaerobes are more commonly found in cow milk-fed than breast-fed infants. There are at least 500 bacterial species throughout the intestinal tract.
It is presumed that enteric flora in a normal intestinal tract ensure a stable environment for physiologic digestion and absorption. When SBBO occurs, regardless of cause, it results in histologic changes of the small bowel, including subtotal villous atrophy and mucosal inflammation, which ultimately may alter function. Mucosal inflammation and atrophy in SBBO may vary from mild to severe. Facultative anaerobes injure the intestinal surface by direct adherence and production of enterotoxins. Aerobic bacteria produce enzymes and metabolic products also capable of causing epithelial cell injury(2).
In short bowel syndrome, SBBO causes malabsorption of fat from bacterial deconjugation of bile acids and bile salt injury to the mucosal surface. Bacterial competition for host substrate, impaired brush-border membrane transport, endogenous nutrient losses, and malnutrition-induced pancreatic protease deficiency contribute to protein malabsorption(4). Carbohydrate malabsorption related to decreased brush-border hydrolase activities, disaccharidase enzyme deficiencies, and impaired monosaccharide absorption also occurs (7). Many bacteria compete for uptake of vitamin B12, causing deficiency.
Symptoms of SBBO result mainly from nutrient malabsorption. Such symptoms include abdominal distention, diarrhea, cramping, and weight loss(8). Anemia may result from malabsorption aggravated by occult blood loss, vitamin B12 deficiency, or both. Ataxia and delirium may occur as a result of assimilation of neurotoxic fermentation products such as D-lactate. Systemic distribution of bacterial antigen-antibody complexes may cause rashes, arthritis, and nephritis. Controversy exists about whether abnormal intestinal motility is a result of SBBO, but it certainly appears to be a related causative factor(9).
Diagnosis of SBBO usually begins with clinical suspicion in the at-risk patient. The most direct and accurate means of confirmation is by quantitative culture of upper small bowel fluid, generally obtained by aspiration. The presence of 105 or more colony forming units of non-pharyngeal bacteria, generally coliforms, suggest the diagnosis of proximal SBBO (2). Identification of unconjugated bile acids and short-chain fatty acids in duodenal fluid may also be helpful in the diagnosis of SBBO (4). In view of the expense and technical difficulty of intestinal aspiration, numerous noninvasive screening studies have been helpful in replacing aspiration and culture or in more clearly identifying those patients who may benefit from aspiration. These tests include urine indicans, serum D-lactic acid, and glucose breath hydrogen test (Table 3) (10,11,12). An early increase in hydrogen production in the breath after glucose consumption indicates significant small bowel bacterial fermentation of carbohydrates. Glucose is a preferable substrate, because lactose and lactulose are poorly absorbed in patients with short bowel syndrome and yield false-positive results of small bowel bacterial overgrowth because of rapid transit into the colon. A recent study has confirmed that lactose is less effective than glucose as a substrate in testing for hydrogen production associated with bacterial overgrowth (13). An upper gastrointestinal radiograph with small bowel series may be done in cases in which partial bowel obstruction, bowel dilatation, or other mechanical factors are suspected as causative factors of SBBO.
Traditional therapy of SBBO involves decreasing bacterial replication with antibiotics and, when possible, correcting factors that caused or promoted bacterial overgrowth. Ideally, antibiotic agents are selected on the basis of specific sensitivities. Antibiotic control of bacterial overgrowth generally involves suppression of overgrowing strict anaerobes and facultative anaerobes. Against the later, metronidazole (20 mg/kg per day) has achieved widest popularity. Facultative anaerobes may be attacked with trimethoprim-sulfamethoxazole (40-50 mg/kg per day), aminoglycosides, such as gentamicin given orally, and extended spectrum penicillins and cephlasporins. Intermittent monitoring of serum gentamicin levels to avoid ototoxicity and nephrotoxicity may be necessary, although our experience suggests that absorption is insignificant.
Frequency of antibiotic usage in SBBO is variable. Antibiotics may be used the first 5 days of each month. Frequency of administration may be increased if the patient becomes symptomatic. Multiple antibiotics, variable rotation schedules, or alternative antibiotics are often necessary because of the development of resistant strains. When significant intestinal inflammation is present, antiinflammatory therapy with sulfasalazine or corticosteroids may be used concurrently. The continued use of antibiotics is highly empiric and is based on symptom response. Recalcitrant patients with small bowel bacterial overgrowth may require repeated duodenal culture and sensitivity determination to improve control of symptoms.
Other medical therapies for SBBO include those that are used to correct depleted micronutrients such as B12 deficiency. Gastric acid-suppressing medications are avoided. Gastric acid normally helps decrease pathogenic bacterial proliferation and if suppressed, may result in alteration of intestinal microflora. Poor motility may be dealt with by using such prokinetic agents as cisapride or erythromycin. Segments of dilated, poorly peristaltic bowel may be corrected with various operations, including resection and tapering. Tapering or lengthening surgery corrects the"cesspool" effect that can occur in SBBO caused by extreme bowel dilatation. Increasing intestinal length with a Bianchi procedure creates a segment of bowel twice the length and half the diameter of the originally dilated bowel segment (14). This means of treating SBBO maintains the total intestinal surface area and corrects the stasis effect of dilated intestinal segment causing overgrowth. various treatment options for SBBO are summarized in Figure 1.
Unfortunately, poor bowel motility may not respond to drug therapy or be amenable to surgical correction. Furthermore, antibiotic therapy carries significant limitations, including development of resistant strains and toxic and allergic reactions. For these reasons, there has been interest in devising new treatment strategies for SBBO. Such methods include reducing bacterial numbers through periodic gastrointestinal irrigation and altering diet to contain fewer fermentable items by reducing carbohydrates and increasing fat calories.
The newest therapy on the horizon is bacterial substitution-that is, replacing existing bacterial strains with so-called probiotic flora. These bacteria, with little or no pathogenicity, have functions of importance to the health and well-being of the host. It is increasingly accepted that probiotic bacteria are effective tools for controlling or eliminating growth of pathogenic bacterial, viral, and fungal infections. Probiotic bacteria can control various enteric pathogens such as Salmonella typhimurium, Shigella sp, Clostridium difficile, Campylobacter jejuni, and E. coli. They may also provide protection against urogenital pathogens such as Gardnerella vaginalitis, Bacteroides bivius, Candida albicans, and Chlamydia trachomatis (6).
The increasing interest in infection control through the use of nonpathogenic bacteria has arisen for several reasons. There has been a recognition that antibiotic therapy has not been as successful as expected. The reported mortality rate for deaths caused by Gram-negative bacteremia remain essentially unchanged from the preantibiotic rate(6). In addition, antibiotic treatment deranges protective flora and antibiotic resistant microbial strains emerge. There is widespread public interest in more ecological methods of disease control. The World Health Organization has recognized antibiotic overuse as a significant and serious threat to human health and has recommended global programs to reduce the use of antibiotics and has increased efforts to prevent disease through use of vaccines and therapies using bacterial interference and macrophages (15).
Much preliminary evidence supports the expectation that probiotic bacteria such as various lactobacilli species can be effective weapons in preventing and treating SBBO. Lactic acid bacteria are particularly suitable for microbial interference treatment as they inhibit many pathogenic Gram-negative bacteria. Live adhesive lactobacilli, such asLactobacillus GG and Lactobacillus plantarum 299V, unlike many lactobacillus species, have protective effects for preventing pathogenic bacteria from adhering and contributing to luminal nutrient depravation. Lactobacilli are known to have at least four functions in the colon: to inhibit growth of potentially pathogenic microorganisms; to produce nutrients, mainly short-chain fatty acids and antimicrobial products such as pyroglutamate; to remove potentially toxic substances from the intestine; and to stimulate the immune system of the intestine(16). The indigenous microflora of the gastrointestinal tract protect against colonization by pathogenic invading bacteria through mechanisms of antagonism, barrier effects, and colonization resistance. All of these factors lead to the observation that probiotic therapy in SBBO may be effective in reducing the use of antibiotic therapy and in controlling symptoms related to overgrowth.
CONCLUSION
Diagnostic measures proving SBBO are helpful in confirming the existence of SBBO but determining the cause is essential for effective treatment. Treatments are aimed at medically or surgically reducing or eliminating sluggish bowel motility and cleansing the bowel of pathogenic bacteria. Several alternative measures may be used in the treatment of SBBO. Regardless of measures used, most therapies are either not without complications or become ineffective as time passes. Preliminary experience with probiotics to change the flora to nonpathogenic organisms is promising and may demonstrate greater effectiveness and results in fewer long-term complications. However, carefully controlled studies on clearly defined probiotic strains, are needed to target specific disease states and conditions.
REFERENCES
1. Vanderhoof JA. Short bowel syndrome in children and small intestinal transplantation. Pediatr Clin North Am 1996;43:533-50.
2. Forstner G, Sherman P, Lichtman S. Bacterial overgrowth. In: Walker W, Durie P, Hamilton J, Walker-Smith J, Watkins J.Pediatric gastrointestinal diseases. St. Louis: Mosby, 689-700.
3. Thompson JS, Pinch LW, Murray N, Vanderhoof, JA. Experience with intestinal lengthening procedures. J Pediatr Surg 1991;26:721-4.
4. Sherman P, Lichtman S. Small bowel bacterial overgrowth syndrome. Dig Dis 1987;5:157-74.
5. Bishop RF, Anderson CM. Bacterial flora of the stomach and small intestine in children with intestinal obstruction. Arch Dis Child 1960;35:487-91.
6. Bengmark S. Econutrition and health maintenance-A new concept to prevent GI inflammation, ulceration and sepsis. Clin Nutr 1996;15:1-10.
7. Sherman P, Wesley A, Forstner G. Sequential disaccharidase loss in rat intestinal blind loops: Impact of malnutrition.Am J Psychol 1985;248:6626-32.
8. Taylor S, Sondheimer J, Sokol R. Noninfectious colitis associated with short gut syndrome in infants. J Pediatr 1991;119:24-9.
9. Jonas MM. Bacterial overgrowth in the small bowel. In: Wyllie R, Hyams J. eds. Pediatric gastrointestinal diseases. Philadelphia: W. B. Saunders, 528-35.
10. Aarbakke J, Schjonsby H. Value of urinary simple phenol and indican determinations in the diagnosis of the stagnant loop syndrome.Scand J Gastroenterol 1976;11:409-14.
11. Scully TB, Kraft SC, Carr WC, et al. D-lactate-associated encephalopathy after massive small bowel resection.J Clin Gastroenterol. 1989;11:448-51.
12. Kerlin P, Wong L. Breath hydrogen testing in bacterial overgrowth of the small intestine. Gastroenterology 1988;95:982-8.
13. Riordan S, McIver C, Walker B, Duncombe V, Bolin T, Thomas C. The lactulose breath hydrogen test and small intestinal bacterial overgrowth. Am J Gastroenterol 1996;119:24-9.
14. Thompson JS. Recent advances in the surgical treatment of the short bowel syndrome. Surg Annu 1990;22:107-27.
15. World Health Organization. Scientific working group on monitoring and management of bacterial resistance to antimicrobial agents. Geneva: World Health Organization, 1994.
16. Fuller R. Probiotics in human medicine. Gut 1991;32:439-42.