Bacteria and the Role of Antibiotics in

by Mark Pimentel

Although several pathophysiologic models have been suggested, the etiology of irritable bowel syndrome (IBS) is unknown. Emerging evidence suggests that small intestinal bacterial overgrowth (SIBO) may explain some symptoms associated with IBS. The putative contribution of bacteria to IBS pathophysiology and symptoms provides a rationale for investigating the efficacy and safety of antibiotics in IBS. Although few controlled studies have been conducted, evidence has suggested that antibiotics may reduce the presence of SIBO and improve symptoms of IBS. Several antibiotics, includ- ing and neomycin, have been shown to effectively reduce bacterial over- growth and improve bowel symptoms in patients with SIBO and/or IBS. Rifaximin may be well suited for the treatment of IBS due to its favorable safety profile. Further research is warranted to more fully elucidate the role of bacteria in IBS and the thera- peutic benefits of antibiotics in the treatment of IBS.

KEY WORDS: small intestinal bacterial overgrowth, rifaximin, neomycin

INTRODUCTION ceptual models of IBS pathogenesis have been sug- rritable bowel syndrome (IBS) is a functional gas- gested, including visceral hypersensitivity, abnormal trointestinal (GI) disorder characterized by symp- intestinal motility, immune activation, and altered Itoms, such as abdominal pain, bloating, , and brain-gut interactions, none fully explain the hetero- , that are not attributable to structural or geneous range of symptoms associated with IBS (1,2). biochemical abnormalities (1). Although several con- Emerging evidence suggests that bacteria, specifically small intestinal bacterial overgrowth (SIBO), may be Mark Pimentel, M.D., F.R.C.P.(C), Director, GI Motility involved in producing some IBS symptoms (3,4). Program, Cedars-Sinai Medical Center and UCLA These findings provide a rationale for the potential Geffen School of Medicine, Los Angeles, CA. therapeutic benefit of antibiotic therapy. This article

PRACTICAL • SEPTEMBER 2007 25 Bacteria and the Role of Antibiotics in IBS

reviews clinical data suggesting the involvement of direct bacterial assessment of intestinal aspirate cul- bacteria in IBS as well as data available for the efficacy tures provides a better method for detecting SIBO and safety of antibiotics in treating SIBO and IBS. (3–5) despite technical limitations, including lack of accessibility to the distal and potential for contamination during sampling (4). In direct sam- EVIDENCE FOR BACTERIAL pling studies using jejunal aspirates, SIBO was INVOLVEMENT IN IBS observed in 4%–12% of patients with IBS (Table 1) Small intestinal bacterial overgrowth involves abnor- (17,18), with similar results reported for healthy vol- mal colonization and growth in the small intestine of unteers (18). However, because bacterial levels >5 x endogenous bacteria resembling those normally found 103 colony-forming units (CFU)/mL were observed in in the colon (5–8). A potential link between SIBO and the cultures of a significantly higher percentage of IBS has been suggested not only by similarities in patients with IBS (43%) compared with healthy volun- symptoms (e.g., diarrhea, abdominal discomfort, teers (12%; p = 0.002) (18), the standard definition of bloating, and ) (6,8,9) but also by the SIBO in culture studies (>105 CFU/mL [8]) may not reported prevalence of SIBO in patients with IBS accurately reflect bacterial overgrowth in patients with (10–16). In studies using lactulose or glucose breath IBS. Although controversies and discrepancies exist testing, SIBO was detected in up to 84% of patients with regard to the most accurate method for detecting who met Rome I (10,11) or Rome II (12–16) criteria and defining SIBO (6,8), these findings suggest that for IBS (Table 1). Because breath tests indirectly mea- intestinal bacteria may play a role in producing bowel sure bacteria and are associated with relatively low symptoms in a subset of patients. Furthermore, sensitivity and specificity (8), many authors believe although few controlled clinical studies have been published, reports that antibiotic therapy normalized breath test results (10–13,19–21) and improved IBS Table 1 symptoms (10,11,20,22,23) provide support for the Prevalence of SIBO in Patients With IBS potential role of bacteria in IBS in some patients. Study N Prevalence (%)

Lactulose breath test ROLE OF SIBO IN PRODUCING IBS SYMPTOMS Nucera, et al (13) 200 75 Bloating and Excess Gas Production Nucera, et al (12) 98 65 Gas-related symptoms (e.g., bloating, abdominal pain Pimentel, et al (11) 202 78 and distention, and/or increased flatulence) have been Pimentel, et al (10) 111 84 reported in the majority (82%–92%) of patients with Walters, et al (14) 39 10 IBS (24,25) and did not appear to be dependent on the predominant IBS symptom (i.e., constipation, diar- Glucose breath test rhea, or alternating) (26). Bacterial gas production may McCallum, et al (16) 143 38 provide a possible explanation for abdominal bloating Lupascu, et al (15) 65 31 and distention associated with IBS. Intestinal hydro- gen gas, the production of which takes place primarily Jejunal aspirate cultures in the terminal and colon, is a normal product of bacterial fermentation of food not easily digested and Simrén, et al (17) 33 12 absorbed in the proximal small intestine (3). When Posserud, et al (18) 162 4 SIBO is present, abnormal concentrations of colonic bacteria in the proximal small intestine allow for IBS = irritable bowel syndrome; SIBO = small intestinal bacterial overgrowth. fermentation of easily and poorly digested food, thereby increasing gas production in this region (3).

26 PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 Bacteria and the Role of Antibiotics in IBS

The presence of excess intestinal gas in patients with antibiotic treatment resulted in greater improvement in IBS is supported by two radiographic studies of 37 and overall IBS symptoms for patients who produced high 60 patients, respectively, that showed significantly levels of methane compared with those who produced higher levels of abdominal gas in patients with IBS only high levels of hydrogen (p < 0.05) (34). Findings compared with healthy participants (p < 0.01 and from two other breath test studies suggested that consti- p < 0.001, respectively) (24,27). Increased levels of pation in patients with IBS may be related to the type of gas were observed in patients with IBS regardless gas produced by intestinal bacteria (10,35). High of the predominant bowel symptom (27). methane excretion was associated with constipation- Patients with IBS frequently report worsening of predominant IBS but not diarrhea-predominant IBS. All symptoms, such as excess gas and abdominal pain, fol- 18 patients in the two studies combined who excreted lowing food ingestion (3,28). Foods high in carbohy- high levels of methane alone were diagnosed with drates and starch, which provide good substrates for constipation-predominant IBS. In contrast, none of the bacterial fermentation, may be especially likely to 145 total patients with diarrhea-predominant IBS cause these symptoms (28). However, most patients excreted high levels of methane alone (10,35). Patients with IBS are usually unable to identify specific food who excreted high levels of methane also reported triggers, suggesting that increases in the concentration greater constipation severity compared with patients of proximal intestinal bacteria, rather than a specific who did not excrete high levels of methane (p < 0.01) food type, may be responsible for postprandial bloat- (10). However, other researchers observed no correla- ing in IBS (3,28). In an investigation of postprandial tion between methane excretion and predominant bowel gas production, gas excretion and production follow- symptoms or symptom severity in 78 patients with IBS ing consumption of a standardized meal were 2 and 4 (36), suggesting that additional research is needed to times higher, respectively, in six patients with IBS determine the role of methane in IBS. compared with six healthy participants (p < 0.05) (29). Because no evidence of malabsorption was observed, the authors attributed increased gas production to alter- ANTIBIOTIC TREATMENT OF SIBO AND IBS ations in the fermentation process (29). Data suggest that gas-related IBS symptoms, such Rifaximin as bloating, may originate in the small intestine. In two Rifaximin is an oral, broad-spectrum antibiotic with studies of 22 and 40 patients, respectively, in which localized action in the GI tract and low (<0.4%) sys- gas was infused into the jejunum, patients with IBS or temic absorption (37). First marketed in Italy in 1987, bloating exhibited significant impairment of gas tran- rifaximin has been approved in 21 countries for a vari- sit and clearance compared with healthy participants ety of indications (37–39). Although rifaximin is cur- (p < 0.05 and p < 0.01 for each study, respectively) rently approved in the United States only for the (30,31). In contrast, no impairment of colonic transit treatment of travelers’ diarrhea (39), data suggest it may was observed following gas infusion (30). Several be effective in the treatment of SIBO (19–22,40–44) investigators have suggested that gas-handling alter- and IBS (22,23,44). In prospective and retrospective ations may be due to reduced intestinal motility (2,4), studies, rifaximin 600–1,600 mg/day for 5–14 days which, as discussed later, may be affected by bacterial reduced hydrogen excretion compared with pre- gas production. treatment levels (20–22) and normalized breath test results in 17%–80% of patients (Table 2) (19,20,41–44), Constipation. Methane in the distal gut has been with higher doses producing higher normalization rates. shown to slow intestinal transit and has been associated Rifaximin 800–1,200 mg/day for 5–14 days has also with reduced plasma levels of serotonin, a neurotrans- been shown to improve symptoms associated with mitter that stimulates peristalsis (32,33). The role of SIBO (20,40,43,44). In a placebo-controlled study of methane in constipation was suggested by a study of 39 124 patients with gas-related symptoms who received patients with constipation-predominant IBS that showed rifaximin 800 mg/day for 10 days, a reduction in

PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 27 Bacteria and the Role of Antibiotics in IBS

Table 2 Studies of Rifaximin for the Treatment of SIBO and/or IBS

Breath test normalization Authors Study type Diagnosis N Treatment (%) Symptom improvement

Lauritano, Prospective, SIBO* 90 Rifaximin 600, 800, or 1,200 17, 27, 60** NR et al (19) randomized mg/day for 7 days DiStefano, Prospective, SIBO*** 21 Rifaximin 1,200 mg/day 70 (R) vs Rifaximin reduced et al (20) randomized, or chlortetracycline 27 (C)**** cumulative**** double-blind 1 g/day for 7 days and diarrhea***** symptom scores vs baseline Corazza, Prospective, SIBO****** 12 Rifaximin 800 or 1,200 67, 67 Symptom improvement et al (43) open-label mg/day for 5 days observed in 83% of patients Baidoo, Prospective, SIBO****** 14 Rifaximin 800 mg/day NR Symptom improvement et al (40) open-label for 14 days observed in 93% of patients; 86% achieved complete remission Scarpellini, Prospective, SIBO* and 80 1,200 or 1,600 mg/day for 58, 80 NR et al (41) randomized IBS 7 days Gabrielli, Prospective, SIBO* 120 Rifaximin 1,200 mg/day 62 (R) vs NR et al (42) randomized or 750 mg/day 45 (M) vs or levofloxacin 500 mg/day 60 (L) for 7 days DiStefano, Prospective, Functional 34 Rifaximin 800 mg/day NR Rifaximin reduced overall et al (21) randomized, bowel or activated charcoal symptom severity vs double-blind disorders 800 mg/day for 7 days baseline******* Sharara, Prospective, Gas-related 63 Rifaximin 800 mg/day or NR Improvement in overall et al (22) randomized, bowel placebo for 10 days symptom severity reported double-blind, symptoms in 41% of patients in placebo-controlled rifaximin group vs 23% with placebo******** Yang, Retrospective SIBO and 84 Rifaximin 1,200 mg/day 56 Symptom improvement et al (44) chart review IBS********* for 10 days observed in 69% of patients Pimentel, Prospective, IBS********* 87 Rifaximin 1,200 mg/day NR Greater global et al (23) randomized, or placebo for 10 days improvement with double-blind, rifaximin**********; placebo-controlled sustained for 10 weeks posttreatment

C = chlortetracycline; IBS = irritable bowel syndrome; L = levofloxacin; M = metronidazole; NR = not reported; R = rifaximin; SIBO = small intestinal bac- terial overgrowth. *Based on glucose breath test results; **p < 0.001 vs 600 mg/day and p < 0.01 vs 800 mg/day; assessed 1 month posttreatment; ***Based on symptoms or glucose breath test results; ****p < 0.01; *****p < 0.05; ******Based on lactulose breath test results; *******p < 0.02; ********p = 0.03; sustained for 10 days posttreatment; *********Based on Rome I criteria; **********p = 0.02.

28 PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 Bacteria and the Role of Antibiotics in IBS

Placebo (n = 44) Rifaximin (n = 43)

Figure 1. The mean global symptom improvement following a 10-day course of rifaximin 1,200 mg/day (n = 43) was sustained for up to 10 weeks posttreatment compared with placebo (n = 44; p = 0.02 for the main group effect). From Pimentel, et al (23). With permission. hydrogen excretion from baseline levels was correlated chronic antibiotic treatment may not be necessary for with improvement in bloating and overall symptom sustained clinical benefit and that rifaximin may affect scores (p = 0.01 for both) (22). Interestingly, none of the a causal mechanism of IBS (23). patients in this study had abnormal breath test results at baseline, suggesting that rifaximin improves symptoms Safety profile. Because rifaximin is minimally in the absence of SIBO (22). (<0.4%) absorbed, it is not likely to be associated with To investigate the efficacy of rifaximin in treating systemic adverse effects. In the studies presented in IBS, a subanalysis of a randomized, placebo- this review that reported safety data, rifaximin admin- controlled study (22) was conducted for the 70 patients istered for up to 14 days was well tolerated (19–23,40). who met Rome II criteria for IBS (22). A significantly In many of these studies, no drug-related adverse higher percentage of patients who received rifaximin effects were observed (20–22,40). The adverse effects 800 mg/day reported symptom improvement com- that were reported (e.g., abdominal pain, weakness, pared with those who received placebo (40% of 37 vs headache, constipation) were mild (19,23), and the 18% of 33 patients, respectively); this positive effect incidence did not differ from placebo (23). Rifaximin was sustained for at least 10 days posttreatment (22). has also exhibited favorable safety and tolerability in In addition, in a randomized, double-blind study of 87 short-term studies of travelers’ diarrhea (38) as well as patients with IBS, rifaximin 1,200 mg/day for 10 days in longer duration studies of remission main- significantly improved bloating (p < 0.001) and global tenance during which patients received daily rifaximin IBS symptoms (p = 0.02) compared with placebo treatment for at least 3 months (45). (Table 2) (23). These benefits were sustained up to 10 Unlike systemic antibiotics, rifaximin has not weeks posttreatment (Figure 1), suggesting that been associated with clinically relevant antimicrobial

PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 29 Bacteria and the Role of Antibiotics in IBS

resistance during approximately 20 years of availability Safety profile. In the study conducted by Pimentel, et in Europe (37), and a study of patients with SIBO has al (10), no drug-related adverse effects were reported shown rifaximin to be associated with fewer adverse following 10 days of treatment with neomycin. How- effects than levofloxacin or metronidazole (42). ever, neomycin may be associated with potentially Because rifaximin is only minimally absorbed, drug serious adverse effects, including delayed-onset oto- interactions with systemically absorbed medications toxicity that may lead to hearing loss, muscle-related are unlikely. Rifaximin does not induce or inhibit key neurotoxicity, and nephrotoxicity. Neomycin has also cytochrome P450 enzymes in vitro and has shown no been shown to inhibit the absorption of some drugs clinical evidence of interacting with drugs metabolized (e.g., penicillin, digoxin, and methotrexate) and by these enzyme systems (38). Rifaximin 600 mg/day enhance the effect of others (e.g., anticoagulants). As has also shown a lack of systemic accumulation with with other systemic antibiotics, antimicrobial resis- repeated administration for up to 3 days (38). tance and development of Clostridium difficile are potential concerns. Neomycin Neomycin is an oral aminoglycoside antibiotic that is Other Systemic Antibiotics absorbed into the systemic circulation. The efficacy of Several other systemic antibiotics, including metro- neomycin has been shown in a randomized, double- nidazole, ciprofloxacin, and chlortetracycline, have blind study with 111 patients who met Rome I criteria been evaluated for efficacy in treating SIBO or IBS, for IBS (10). In a subanalysis of the 93 patients (84%) but the limited published data available concern pri- with abnormal breath test results at baseline, 46% of marily patients with SIBO. In a study of 21 patients patients who received neomycin 1 g/day for 10 days with blind loop syndrome and SIBO, metronidazole (n = 46) had a clinical response compared with 15% of 500 mg/day for 7 days significantly reduced total those who received placebo (n = 47; p < 0.01). Only 8 hydrogen excretion compared with baseline levels (p < (20%) of the 41 patients treated with neomycin who 0.001) and significantly improved symptom severity had abnormal breath test results at baseline and com- (p = 0.0011) (46). Ciprofloxacin for the treatment of plete study data achieved a normal breath test result. SIBO was reported in an open-label study of 12 Of the 84 patients with an abnormal breath test result patients with nonalcoholic , six of whom at baseline and complete study data, the percent had positive breath test results (47). Ciprofloxacin improvement in composite symptom scores was great- 1 g/day for 5 days normalized glucose breath test results est among patients who had breath test result normal- for five of these six patients (p = 0.025) (47). In stud- ization with neomycin (62% ± 9%), followed by those ies of patients with IBS and SIBO, antibiotic treatment whose breath test results were not normalized with with metronidazole, ciprofloxacin, neomycin, or doxy- neomycin (34% ± 6%) and those who received placebo cycline (n = 47) (11) or metronidazole, rifaximin, or (4% ± 12%; p = 0.01) (10). In a subanalysis of patients fluoroquinolones (n = 64) (12) significantly improved with constipation-predominant IBS from the Pimentel, breath test results (p < 0.00001 and p < 0.01, respec- et al (10) study, neomycin 1 g/day for 10 days resulted tively) (11,12) and reduced the number of patients with in greater global symptom improvement compared abdominal pain (p < 0.001) and diarrhea (p < 0.001) with placebo (44% ± 8% vs 6% ± 4%, respectively; (11). However, specific benefits of individual anti- p < 0.001) (34). The percent improvement in symptoms biotics were not reported in these studies; therefore, no was greater in patients who excreted high levels of conclusions can be drawn about the efficacy and safety methane in lactulose breath testing compared with of metronidazole or ciprofloxacin (11,12). In a study those who excreted high levels of hydrogen (68% ± of patients with SIBO, chlortetracycline 1 g/day for 7 6% vs 33% ± 10%, respectively; p < 0.05), supporting days did not improve breath test results for fasting, the potential involvement of increased methane pro- peak, or total hydrogen excretion compared with base- duction in patients with constipation (34). line measures and did not improve symptom scores for

30 PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 Bacteria and the Role of Antibiotics in IBS

diarrhea, borborygmi, or lassitude compared with 7. Husebye E, Skar V, Hoverstad T, Iversen T, Melby K. Abnormal intestinal motor patterns explain enteric colonization with Gram- baseline levels (20). Chlortetracycline normalized negative bacilli in late radiation . Gastroenterology, breath test results in only 3 (27%) of 11 patients (20). 1995;109:1078-1089. 8. Simren M, Stotzer P-O. Use and abuse of hydrogen breath tests. Gut, 2006;55:297-303. 9. Di Stefano M, Corazza GR. Treatment of small intestine bacterial CONCLUSIONS overgrowth and related symptoms by rifaximin. Chemotherapy, 2005;51(suppl 1):103-109. Emerging evidence suggests that bacteria, specifically 10. Pimentel M, Chow EJ, Lin HC. Normalization of lactulose breath overgrowth of colonic bacteria in the small intestine, testing correlates with symptom improvement in irritable bowel syndrome. a double-blind, randomized, placebo-controlled study. may play a pathogenic role in IBS. The hypothesized Am J Gastroenterol, 2003;98:412-419. involvement of SIBO in IBS provides a potential 11. Pimentel M, Chow EJ, Lin HC. Eradication of small intestinal bac- terial overgrowth reduces symptoms of irritable bowel syndrome. explanation for symptoms, such as bloating and Am J Gastroenterol, 2000;95:3503-3506. increased gas production, experienced by many 12. Nucera G, Gabrielli M, Lupascu A, et al. Abnormal breath tests to lactose, fructose and sorbitol in irritable bowel syndrome may be patients with IBS. The efficacy of antibiotics for explained by small intestinal bacterial overgrowth. Aliment Phar- reducing gas excretion and relieving symptoms associ- macol Ther, 2005;21:1391-1395. ated with gas production supports the involvement of 13. Nucera G, Lupascu A, Gabrielli M, et al. Sugar intolerance in irri- table bowel syndrome: the role of small bowel bacterial over- intestinal bacteria in IBS. Among antibiotics currently growth [abstract]. Gastroenterology, 2004;126(4 suppl 2):A-511. 14. Walters B, Vanner SJ. Detection of bacterial overgrowth in IBS available, rifaximin has been most extensively evalu- 14 using the lactulose H2 breath test: comparison with C-D-xylose ated for the treatment of SIBO and IBS. In contrast to and healthy controls. Am J Gastroenterol, 2005;100:1566-1570. systemic antibiotics, no clinically relevant anti- 15. Lupascu A, Gabrielli M, Lauritano EC, et al. Hydrogen glucose breath test to detect small intestinal bacterial overgrowth: a preva- microbial resistance has been reported with rifaximin, lence case-control study in irritable bowel syndrome. Aliment and the low (<0.4%) absorption profile of rifaximin Pharmacol Ther, 2005;22:1157-1160. suggests a minimal risk of systemic adverse effects. 16. McCallum R, Schultz C, Sostarich S. Evaluating the role of small intestinal bacterial overgrowth (sibo) in diarrhea predominant irri- Although further research is needed, the sustained table bowel syndrome (IBS-D) patients utilizing the glucose breath clinical benefit reported following a short course of test (GBT) [abstract]. Gastroenterology, 2005;128(4 suppl 2):A- 460. rifaximin suggests that antibiotic therapy may address 17. Simren M, Ringstrom G, Agerforz P, Bjornsson ES, Abrahamsson a causative mechanism underlying some manifesta- H, Stotzer P-O. Small intestinal bacterial overgrowth is not of major importance in the irritable bowel syndrome [abstract]. Gas- tions of IBS. Additional studies are warranted to more troenterology, 2003;124(4 suppl 1):A-163. fully elucidate the putative role of intestinal bacteria in 18. Posserud I, Stotzer P-O, Bjornsson ES, Abrahamsson H, Simren the pathophysiology of IBS and the potential therapeu- M. Small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Gut, 2007;56:802-808. tic benefit of antibiotics in treating IBS. ■ 19. Lauritano EC, Gabrielli M, Lupascu A, et al. Rifaximin dose- finding study for the treatment of small intestinal bacterial over- growth. Aliment Pharmacol Ther, 2005;22:31-35. 20. Di Stefano M, Malservisi S, Veneto G, Ferrieri A, Corazza GR. Rifaximin versus chlortetracycline in the short-term treatment of small intestinal bacterial overgrowth. Aliment Pharmacol Ther, References 2000;14:551-566. 1. Malagelada J-R. A symptom-based approach to making a positive 21. Di Stefano M, Strocchi A, Malservisi S, Veneto G, Ferrieri A, diagnosis of irritable bowel syndrome with constipation. Int J Clin Corazza GR. Non-absorbable antibiotics for managing intestinal Pract, 2006;60:57-63. gas production and gas-related symptoms. Aliment Pharmacol 2. Barbara G, De Giorgio R, Stanghellini V, Cremon C, Salvioli B, Ther, 2000;14:1001-1008. Corinaldesi R. New pathophysiological mechanisms in irritable 22. Sharara AI, Aoun E, Abdul-Baki H, Mounzer R, Sidani S, ElHajj bowel syndrome. Aliment Pharmacol Ther, 2004;20(suppl 2):1-9. I. A randomized double-blind placebo-controlled trial of rifaximin 3. Lin HC. Small intestinal bacterial overgrowth: a framework for in patients with abdominal bloating and flatulence. Am J Gas- understanding irritable bowel syndrome. JAMA, 2004;292:852- troenterol, 2006;101:326-333. 858. 23. Pimentel M, Park S, Mirocha J, Kane SV, Kong Y. The effect of a 4. Lee H-R, Pimentel M. Bacteria and irritable bowel syndrome: the nonabsorbed oral antibiotic (rifaximin) on the symptoms of the irri- evidence for small intestinal bacterial overgrowth. Curr Gastroen- table bowel syndrome: a randomized trial. Ann Intern Med, terol Rep, 2006;8:305-311. 2006;145:557-563. 5. Husebye E. The pathogenesis of gastrointestinal bacterial over- 24. Chami TN, Schuster MM, Bohlman ME, Pulliam TJ, Kamal N, growth. Chemotherapy, 2005;51(suppl 1):1-22. Whitehead WE. A simple radiologic method to estimate the quan- 6. Singh VV, Toskes PP. Small bowel bacterial overgrowth: presen- tity of bowel gas. Am J Gastroenterol, 1991;86:599-602. tation, diagnosis, and treatment. Curr Treat Options Gastroenterol, 25. Dotevall G, Svedlund J, Sjodin I. Symptoms in irritable bowel syn- 2004;7:19-28. drome. Scand J Gastroenterol, 1982;79:16-19.

PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007 31 Bacteria and the Role of Antibiotics in IBS

26. Houghton LA, Lea R, Agrawal A, Reilly B, Whorwell PJ. Rela- 37. Koo HL, DuPont HL. Current and future developments in travel- tionship of abdominal bloating to distention in irritable bowel syn- ers’ diarrhea therapy. Expert Rev Anti Infect Ther, 2006;4:417-427. drome and effect of bowel habit. Gastroenterology, 38. Scarpignato C, Pelosini I. Experimental and clinical pharmacology 2006;131:1003-1010. of rifaximin, a gastrointestinal selective antibiotic. Digestion, 27. Koide A, Yamaguchi T, Odaka T, et al. Quantitative analysis of 2006;73(suppl 1):13-27. bowel gas using plain abdominal radiograph in patients with irrita- 39. Baker DE. Rifaximin: a nonabsorbed oral antibiotic. Rev Gas- ble bowel syndrome. Am J Gastroenterol, 2000;95:1735-1741. troenterol Disord, 2005;5:19-30. 28. Simren M, Mansson A, Langkilde AM, et al. Food-related gas- 40. Baidoo L, Kundu R, Berenbaum PL, et al. Rifaximin is effective trointestinal symptoms in the irritable bowel syndrome. Digestion, therapy for small bowel bacterial overgrowth [abstract]. Gastroen- 2001;63:108-115. terology, 2005;128(4 suppl 2):A-672. 29. King TS, Elia M, Hunter JO. Abnormal colonic fermentation in 41. Scarpellini E, Gabrielli M, Lauritano CE, et al. High dosage rifax- irritable bowel syndrome. Lancet, 1998;352:1187-1189. imin for the treatment of small intestinal bacterial overgrowth. Ali- 30. Salvioli B, Serra J, Azpiroz F, et al. Origin of gas retention and ment Pharmacol Ther, 2007;25:781-786. symptoms in patients with bloating. Gastroenterology, 42. Gabrielli M, Lauritano C, Scarpellini E, et al. Rifaximin versus sys- 2005;128:574-579. temic antibiotics for the treatment of small bowel bacterial over- 31. Serra J, Azpiroz F, Malagelada J-R. Impaired transit and tolerance growth [abstract]. Gastroenterology, 2007;132(4 suppl 2):A-187. of intestinal gas in the irritable bowel syndrome. Gut, 2001;48:14- 43. Corazza GR, Ventrucci M, Strocchi A, et al. Treatment of small 19. intestine bacterial overgrowth with rifaximin, a non-absorbable 32. Lin HC, Pimentel M, Chen JH. Intestinal transit is slowed by lunial rifamycin. J Int Med Res, 1988;16:312-316. methane [abstract]. Neurogastroenterol Motil, 2002;14:437. 44. Yang J, Lee H-R, Low K, Chatterjee S, Pimentel M. Rifaximin ver- 33. Pimentel M, Kong Y, Park S. IBS subjects with methane on lactu- sus other antibiotics in the primary treatment and retreatment of lose breath test have lower postprandial serotonin levels than sub- bacterial overgrowth in IBS. Dig Dis Sci, 2007. E-pub ahead of jects with hydrogen. Dig Dis Sci, 2004;49:84-87. print. 34. Pimentel M, Chatterjee S, Chow EJ, Park S, Kong Y. Neomycin 45. Shen B, Fazio V, Remzi F, et al. Oral rifaximin as maintenance improves constipation-predominant irritable bowel syndrome in a therapy for antibiotic-dependent pouchitis [abstract]. Am J Gas- fashion that is dependent on the presence of methane gas: sub- troenterol, 2006;101(suppl):S441-S442. analysis of a double-blind randomized controlled study. Dig Dis 46. Di Stefano M, Miceli E, Missanelli A, Mazzocchi S, Corazza GR. Sci, 2006;51:1297-1301. Absorbable vs. non-absorbable antibiotics in the treatment of small 35. Pimentel M, Mayer AG, Park S, Chow EJ, Hasan A, Kong Y. intestine bacterial overgrowth in patients with blind-loop syn- Methane production during lactulose breath test is associated with drome. Aliment Pharmacol Ther, 2005;21:985-992. presentation. Dig Dis Sci, 2003;48:86-92. 47. Sajjad A, Mottershead M, Syn WK, Jones R, Smith S, Nwokolo 36. Di Stefano M, Moroni F, Tana P, et al. Methane breath excretion is CU. Ciprofloxacin suppresses bacterial overgrowth, increases fast- not accompanied by a different pattern of clinical presentation in ing insulin but does not correct low acylated ghrelin concentration IBS patients [abstract]. Gastroenterology, 2007;132(4 suppl 2):A- in non-alcoholic steatohepatitis. Aliment Pharmacol Ther, 194–A-195. 2005;22:291-299.

32 PRACTICAL GASTROENTEROLOGY • SEPTEMBER 2007