Irritable Bowel Syndrome American College of Gastroenterology Task Force on IBS American College of Gastroenterology Task Force on Irritable Bowel Syndrome

Home , Antispasmodic, Pinaverium bromide, Prifinium bromide, Rociverine

Volume 104 supplement 1 January 2009 www.amjgastro.com

s u p p l e m e n t t o

o f f i c i a l publication o f t h e a m e r i c a n c o l l e g e o f gastroenterology

officialsupplement publication of the american college of gastroenterology

An Evidence-Based Systematic Review on the Management of Irritable Bowel Syndrome American College of Gastroenterology Task Force on IBS American College of Gastroenterology Task Force on Irritable Bowel Syndrome

Lawrence J. Brandt, MD, MACG, Chair Department of Medicine Montefiore Medical Center Albert Einstein School of Medicine

William D. Chey, MD, FACG Department of Gastroenterology University of Michigan Medical Center

Amy E. Foxx-Orenstein, DO, FACG Division of Gastroenterology and Hepatology Department of Internal Medicine Mayo Clinic

Lawrence R. Schiller, MD, FACG Division of Gastroenterology Baylor University Medical Center

Philip S. Schoenfeld, MD, FACG Division of Gastroenterology Veterans Affairs Ann Arbor Healthcare System

Brennan M. Spiegel, MD, FACG VA Greater Los Angeles Healthcare System David Geffen School of Medicine at UCLA UCLA/VA Center for Outcomes Research and Education (CORE)

Nicholas J. Talley, MD, PhD, FACG Department of Internal Medicine Mayo Clinic Jacksonville

Eamonn M.M. Quigley, MD, FACG Department of Medicine Cork University Hospital National University of Ireland at Cork

Paul Moayyedi, BSc, MB ChB, PhD, MPH, FRCP (London), FRCPC, FACG, Statistician-Epidemiologist Department of Medicine, Division of Gastroenterology McMaster University Medical Centre

Unrestricted grants have been provided to the American College of Gastroenterology from Takeda Pharmaceutical Company Ltd. and Salix Pharmaceuticals in support of the work of the ACG IBS Task Force. This monograph was developed on behalf of the American College of Gastroenterology and the ACG Institute for Clinical Research & Education by the ACG IBS Task Force which had complete scientific and editorial control of its content. contents Volume 104 supplement 1 january 2009

o f f i c i a l publication o f t h e a m e r i c a n c o l l e g e o f gastroenterology

Supplement An Evidence-Based Systematic Review on the Management of Irritable Bowel Syndrome

Section 1

S1 An Evidence-Based Position Statement on the Management of Irritable Bowel Syndrome

section 2

An Evidence-Based systematic Review on the Management of Irritable Bowel Syndrome S8 2.1 Methodology for systematic reviews of irritable bowel syndrome therapy, levels of evidence, and grading recommendations S9 2.2 The burden of illness of irritable bowel syndrome S12 2.3 The utility of diagnostic criteria in IBS S12 2.4 The role of alarm features in the diagnosis of IBS S14 2.5 The role of diagnostic testing in patients with IBS symptoms S17 2.6 Diet and irritable bowel syndrome S17 2.7 Effectiveness of dietary fiber, bulking agents, and laxatives in the management of irritable bowel syndrome S18 2.8 Effectiveness of antispasmodic agents, including peppermint oil, in the management of irritable bowel syndrome S19 2.9 Effectiveness of antidiarrheals in the management of irritable bowel syndrome S19 2.10 Effectiveness of antibiotics in the management of irritable bowel syndrome S20 2.11 Effectiveness of probiotics in the management of irritable bowel syndrome S21 2.12 Effectiveness of the 5HT 3 receptor antagonists in the management of irritable bowel syndrome S22 2.13 Effectiveness of 5HT 4 (serotonin) receptor agonists in the management of irritable bowel syndrome S23 2.14 Effectiveness of the selective C-2 chloride channel activators in the management of irritable bowel syndrome S24 2.15 Effectiveness of antidepressants in the management of irritable bowel syndrome S25 2.16 Effectiveness of psychological therapies in the management of irritable bowel syndrome S25 2.17 Effectiveness of herbal therapies and acupuncture in the management of irritable bowel syndrome S26 2.18 Emerging therapies for the irritable bowel syndrome

www.amjgastro.com nature publishing group VOLUME 104 SUPPLEMENT 1 JANUARY 2009

An Evidence-Based Position Statement on the Management of Irritable Bowel Syndrome

American College of Gastroenterology IBS Task Force

Irritable bowel syndrome (IBS) is a common disorder Section 1 Evidence-based position statement on the characterized by abdominal pain and altered bowel habit management of irritable bowel syndrome for at least 3 months. With this publication, an American A m e r i c a n C o l l e ge of Gastroenterology IBS Task Force College of Gastroenterology Task Force updates the 2002 Monograph on IBS in light of new data. A series IBS is characterized by abdominal discomfort associated of systematic reviews were performed to evaluate the with altered bowel function; structural and biochemical abnor- diagnostic yield of investigations and the e+ cacy of malities are absent. e pathophysiology of IBS is multifactorial treatments for IBS. e Task Force recommends that and of intense recent interest, largely because of the possibil- further investigations are unnecessary in young patients ity of developing targeted therapies. As IBS is one of the most without alarm features with the exception of celiac sprue common disorders managed by gastroenterologists and serology, which may be of bene* t in some patients. Further primary care physicians, this monograph was developed investigation such as colonoscopy is recommended in those to educate physicians about its epidemiology, diagnostic over 50 years of age and in patients with alarm features. approach, and treatments. e American College of Gastroen- Trials suggest psyllium * ber, certain antispasmodics, and terology (ACG) IBS Task Force updated the 2002 monograph peppermint oil are e7 ective in IBS patients although the because new evidence has emerged on the bene* t and risks of quality of the evidence is poor. Evidence suggests that drugs used for IBS. Furthermore, new drugs also have been some probiotics may be e7 ective in reducing overall IBS developed and the evidence for e+ cacy of these drugs needed symptoms but more data are needed. Antidiarrheals to be assessed. To critically evaluate the rapidly expand- reduce the frequency of stools but do not a7 ect the overall ing research about IBS, a series of systematic reviews were symptoms of IBS. 5HT 3 antagonists are e+ cacious in performed. Standard criteria for systematic reviews were met, IBS patients with diarrhea and the quality of evidence is including comprehensive literature searching, use of prespeci- good. Patients need to be carefully selected, however, * ed study selection criteria, and use of a standardized and because of the risk of ischemic colitis. 5HT4 agonists are transparent process to extract and analyze data from studies. modestly e7 ective in IBS patients with constipation and Evidence-based statements were developed from these data the quality of evidence is good although the possible risk of by the entire ACG IBS Task Force. Recommendations were cardiovascular events associated with these agents may limit graded using a formalized system that quanti* es the strength their utility. Tricyclic antidepressants and selective serotonin of evidence. Each recommendation was classi* ed as strong reuptake inhibitors have been shown to be e7 ective in IBS (grade 1) or weak (grade 2) and the strength of evidence patients of all subtypes. e trials generally are of good classi* ed as strong (level A), moderate (level B), or weak (level quality but the limited number of patients included in C). Recommendations in this position statement may be cross- trials implies that further evidence could change the referenced with the supporting evidence in the accompanying con* dence in the estimate of e7 ect and therefore the article, “ An Evidenced Based Review on the Management of quality of evidence was graded as moderate. Nonabsorbable Irritable Bowel Syndrome ” . antibiotics are e7 ective particularly in diarrhea-predominant IBS and selective C-2 chloride channel activators are e+ cacious in constipation-predominant IBS with a moderate Irritable bowel syndrome: methodology for systematic reviews, quality of evidence. Psychological therapies may also provide levels of evidence and grading of recommendations (see bene* t to IBS patients although the quality of evidence is Section 2.1). poor. The burden of illness of irritable bowel syndrome (see Section 2.2) IBS is a prevalent and expensive condition that is associated with Am J Gastroenterology 2009; 104:S1– S35; doi: 10.1038/ajg.2008.122 a signi cantly impaired health-related quality of life (HRQOL) and reduced work productivity. Based on strict criteria, 7 – 10 %

of people have IBS worldwide. Community-based data indicate providers should remain alert for signs of somatization in IBS, that diarrhea-predominant IBS (IBS-D) and mixed IBS (IBS-M) and aggressively treat or refer somatization patients to an expe- subtypes are more prevalent than constipation-predominant IBS rienced specialist rather than performing potentially unneces- (IBS-C), and that switching among subtype groups may occur. sary diagnostic tests. IBS is 1.5 times more common in women than in men, is more In addition to direct costs of care, IBS patients engender common in lower socioeconomic groups, and is more commonly signi* cant indirect costs of care as a consequence of both diagnosed in patients younger than 50 years of age. Patients missing work and su7 ering impaired work performance while with IBS visit the doctor more frequently, use more diagnostic on the job. Compared with IBS patients who exhibit normal tests, consume more medications, miss more workdays, have work productivity, patients with impaired productivity have lower work productivity, are hospitalized more frequently, and more extraintestinal comorbidities and more disease-speci* c consume more overall direct costs than patients without IBS. fears and concerns. In contrast, the speci* c pro* le of individual Resource utilization is highest in patients with severe symptoms, bowel symptoms does not undermine work productivity, sug- and poor HRQOL. Treatment decisions should be tailored to gesting that enhancing work productivity in IBS may require the severity of each patient ’ s symptoms and HRQOL decrement. treatments that improve both gastrointestinal (GI) and non- GI symptom intensity, while also modifying the cognitive and Prevalence estimates of IBS range from 1 % to more than 20% . behavioral responses to bowel symptoms and the contexts in When limited to unselected population-based studies, the which they occur. pooled prevalence of IBS in North America is 7% . Community- based data indicate that IBS-D and IBS-M subtypes are more The utility of diagnostic criteria in irritable bowel syndrome prevalent than IBS-C, and that switching may occur among (see Section 2.3) subtype groups. IBS is 1.5 times more common in women IBS is de ned by abdominal pain or discomfort that occurs in than in men, although IBS is not simply a disorder of women. association with altered bowel habits over a period of at least In fact, IBS is now recognized to be a key component of the three months. Individual symptoms have limited accuracy for Gulf War Syndrome, a multi-symptom complex a7 ecting diagnosing IBS and, therefore, the disorder should be considered soldiers (a predominantly male population) deployed in the as a symptom complex. Although no symptom-based diagnos- 1991 Gulf War. IBS is diagnosed more commonly in patients tic criteria have ideal accuracy for diagnosing IBS, traditional under the age of 50 years than in patients older than 50 years. criteria, such as Kruis and Manning, perform at least as well as ere is a graded decrease in IBS prevalence with increasing Rome I criteria; the accuracy of Rome II and Rome III criteria income. has not been evaluated. Patients with IBS have a lower HRQOL compared with non-IBS cohorts. It is possible that patients with IBS develop IBS is a chronic illness of disordered bowel function and abdo- HRQOL decrements due to their disease, and also possible that minal pain or discomfort that is distinguished by the absence of some patients with diminished HRQOL subsequently develop biochemical markers or structural abnormalities. As individual IBS. Although the precise directionality of this relationship symptoms have imperfect accuracy in diagnosing IBS, crite- may vary from patient to patient, it is clear that IBS is strongly ria have been developed to identify a combination of symp- related to low HRQOL, and vice versa. e HRQOL decre- toms to diagnose the condition. Manning et al. promulgated ment can, in some cases, be so severe as to increase the risk of the original account of this approach. Two of four studies that suicidal behavior. Because HRQOL decrements are common in have evaluated the accuracy of the Manning criteria suggested IBS, we recommend that clinicians perform routine screening they perform well, with a sensitivity of 78% and speci* city of for diminished HRQOL in their IBS patients. Treatment should 72 % . Kruis et al developed another set of criteria; three of four be initiated when the symptoms of IBS are found to reduce studies that examined the accuracy of the Kruis symptom score functional status and diminish overall HRQOL. Furthermore, suggested it provides an excellent positive predictive value clinicians should remain wary of potential suicidal behavior in with a high sensitivity (77 % ) and speci* city (89 % ). e Rome patients with severe IBS symptoms, and should initiate timely criteria subsequently were developed and have undergone three interventions if suicide indicators are identi* ed. iterations. One study has evaluated the accuracy of Rome I Patients with IBS consume a disproportionate amount of criteria, and determined it had a sensitivity of 71 % and spe- resources. IBS care consumes over $20 billion in both direct ci* city of 85% ; Rome II and Rome III have not yet been evalu- and indirect expenditures. Moreover, patients with IBS ated. None of the symptom-based diagnostic criteria have an consume over 50 % more health care resources than matched ideal accuracy, and the Rome criteria, in particular, have been controls without IBS. Resource utilization in IBS is driven partly inadequately evaluated. e ACG Task Force believes that a by the presence of comorbid somatization —a trait found in up practical de* nition, i.e., one that is simple to use and incor- to one-third of IBS patients that is characterized by the propen- porates key features of previous diagnostic criteria would be sity to overinterpret normal physiologic processes. ere is a clinically useful. erefore, we have de* ned IBS as abdominal highly signi* cant relationship between levels of somatization pain or discomfort that occurs in association with altered bowel and the amount of diagnostic testing in IBS, suggesting that habits over a period of at least 3 months.

S2 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Position Statement on the Management of IBS

The role of alarm features in the diagnosis of IBS (see random biopsies should be considered to rule out microscopic Section 2.4) colitis (Grade 2C). Overall, the diagnostic accuracy of alarm features is disappointing. Rectal bleeding and nocturnal pain o; er little discriminative As IBS is a disorder of heterogeneous pathophysiology for value in separating patients with IBS from those with organic which speci* c biomarkers are not yet available, diagnostic tests diseases. Whereas anemia and weight loss have poor sensitivity are performed to exclude organic diseases that may masquer- for organic diseases, they o; er very good speci city. As such, in ade as IBS and, in so doing, reassure both the clinician and the patients who ful ll symptom-based criteria of IBS, the absence patient that the diagnosis of IBS is correct. Historically, IBD, of selected alarm features, including anemia, weight loss, and a colorectal cancer, diseases associated with malabsorption, family history of colorectal cancer, in= ammatory bowel disease, systemic hormonal disturbances, and enteric infections are of or celiac sprue, should reassure the clinician that the diagnosis the greatest concern to clinicians caring for patients with IBS of IBS is correct. symptoms. When deciding on the necessity of a diagnostic test in a patient with IBS symptoms, one should * rst consider the Patients with typical IBS symptoms also may exhibit so-called pretest probability of the disease in question. Based on cur- “ alarm features” that increase concern organic disease may rently available evidence, the Task Force feels that patients be present. Alarm features include rectal bleeding, weight who ful* ll the symptom-based diagnostic criteria for IBS and loss, iron de* ciency anemia, nocturnal symptoms, and a fam- who have no alarm features require little formal testing before ily history of selected organic diseases including colorectal arriving at the diagnosis of IBS. e likelihood of uncovering cancer, inK ammatory bowel disease (IBD), and celiac sprue. important organic disease by a complete blood count, serum Usually, it is recommended that patients who exhibit alarm chemistries, and thyroid function studies is low and no greater features undergo further investigation, particularly with in IBS patients than in healthy controls. Similarly, the yield of colonoscopy to rule out organic disease, e.g., colorectal stool ova and parasite examination and abdominal ultrasound cancer. is low. For these reasons, the routine use of these tests in IBS Based on a review of the literature, the accuracy of such alarm patients without alarm features is not recommended. ere is features is disappointing. Rectal bleeding and nocturnal pain emerging evidence, however, to suggest that the prevalence of o7 er little discriminative value in separating patients with IBS celiac sprue is higher among patients with IBS than in controls. from those with organic diseases. Whereas anemia and weight Based on this evidence and decision analytic modeling data that loss have poor sensitivity for organic diseases, they o7 er very suggest cost e7 ectiveness, the Task Force recommends routine good speci* city. As such, in patients who ful* ll symptom-based serologic screening for celiac sprue in patients with IBS-D or criteria of IBS, the absence of selected alarm features, includ- IBS-M. Evidence also suggests that the prevalence of lactose ing anemia, weight loss, and a family history of colorectal can- maldigestion is higher among IBS patients than in healthy con- cer, IBD, or celiac sprue, should reassure the clinician that the trols. Furthermore, the clinical response to lactose maldigestion diagnosis of IBS is correct. may be exaggerated in IBS patients compared with controls. For these reasons, the Task Force suggests that providers ques- The role of diagnostic testing in patients with IBS symptoms tion patients about a link between lactose ingestion and their (see Section 2.5) IBS symptoms. If, aM er a careful history and review of a food Routine diagnostic testing with complete blood count, serum diary, questions remain regarding the presence of lactose mal- chemistries, thyroid function studies, stool for ova and parasites, digestion, performance of a lactose hydrogen breath test can be and abdominal imaging is not recommended in patients with considered. A great deal of attention has been focused on the typical IBS symptoms and no alarm features because of a low potential role of small intestinal bacterial overgrowth (SIBO) likelihood of uncovering organic disease (Grade 1C). Routine in the pathogenesis of IBS symptoms. e available data on serologic screening for celiac sprue should be pursued in patients this topic have yielded conK icting results. On a practical level, with IBS-D and IBS-M (Grade 1B). Lactose breath testing can currently there is no available gold standard test to diagnose be considered when lactose maldigestion remains a concern SIBO. For these reasons, the Task Force feels that there is insuf- despite dietary modi cation (Grade 2B). Currently, there are * cient evidence to recommend the performance of lactulose insu? cient data to recommend breath testing for small intestinal or glucose breath tests to identify SIBO in patients with IBS. bacterial overgrowth in IBS patients (Grade 2C). Because of the Colonic imaging in an IBS patient with no alarm features is low pretest probability of Crohn’ s disease, ulcerative colitis, and unlikely to reveal structural disease that might explain the colonic neoplasia, routine colonic imaging is not recommended patient’ s symptoms. Studies suggest that the prevalence of in patients younger than 50 years of age with typical IBS symp- structural disease identi* ed by colonic imaging is less than toms and no alarm features (Grade 1B). Colonoscopic imaging 1.3% . For this reason, the Task Force recommends that patients should be performed in IBS patients with alarm features to rule younger than 50 years of age who do not have alarm features out organic diseases and in those over the age of 50 years for need not undergo routine colonic imaging. Patients with IBS the purpose of colorectal cancer screening (Grade 1C). When symptoms who have alarm features such as anemia or weight colonoscopy is performed in patients with IBS-D, obtaining loss or those who are older than 50 years of age should undergo

colonic imaging to exclude organic disease. ere is emerging No placebo-controlled, randomized study of laxatives in evidence to suggest that microscopic colitis can masquerade IBS has been published. Laxatives have been studied mostly as IBS-D, and therefore, when patients with IBS-D undergo in patients with chronic constipation. A single small sequen- colonoscopy, performance of random mucosal biopsies should tial study compared symptoms before and with PEG laxative be considered. In patients whose symptoms are consistent with treatment in adolescents with IBS-C. Stool frequency improved IBS and who also have alarm features, the nature and severity of from an average of 2.07 ± 0.62 bowel movements per week to the symptoms as well as the patient’ s expectations and concerns 5.04 ± 1.51 bowel movements per week (P < 0.05), but there was inK uence the choice of diagnostic testing. no e7 ect on pain intensity.

Diet and irritable bowel syndrome (see Section 2.6) Effectiveness of antispasmodic agents, Including peppermint Patients oC en believe that certain foods exacerbate their IBS symp- oil, in the management of irritable bowel syndrome (see toms. B ere is, however, insu? cient evidence that food allergy Section 2.8) testing or exclusion diets are e? cacious in IBS and their routine Certain antispasmodics (hyoscine, cimetropium, pinaverium, use outside of a clinical trial is not recommended (Grade 2C). and peppermint oil) may provide short-term relief of abdominal pain / discomfort in IBS (Grade 2C). Evidence for long-term Approximately 60% of IBS patients believe that food exacer- e? cacy is not available. Evidence for safety and tolerability is bates their symptoms, and research has suggested that allergy to limited (Grade 2C). certain foods could trigger IBS symptoms. A systematic review identi* ed eight studies that assessed a symptomatic response to ere is evidence for the e+ cacy of antispasmodics as a class exclusion diets in 540 IBS subjects. Studies reported a positive and some peppermint oil preparations (which also may act as response in 12.5 – 67 % of patients, but the absence of control antispasmodics) in IBS. ere are, however, signi* cant varia- groups makes it is unclear whether these rates simply reK ect a tions in the availability of these agents in di7 erent countries; placebo response. ere is no correlation between foods that little of the data is recent; early trials vary considerably in terms patients identify as a cause of their IBS symptoms and the results of inclusion criteria, dosing schedule, duration of therapy, and of food allergy testing. One randomized trial suggested that study endpoints; and many are of poor quality and frequently patients with IBS can identify foods that cause symptoms, but fail to di7 erentiate between the e7 ects of these agents on global two subsequent trials have not con* rmed this. symptoms and individual symptoms, such as pain. Further- more, the adverse event pro* le of these agents has not been Effectiveness of dietary ﬁ ber, bulking agents, and laxatives in de* ned adequately. the management of irritable bowel syndrome (see Section 2.7) Psyllium hydrophilic mucilloid (ispaghula husk) is moderately Effectiveness of antidiarrheals in the management of irritable e; ective and can be given a conditional recommendation (Grade bowel syndrome (see Section 2.9) 2C). A single study reported improvement with calcium poly- B e antidiarrheal agent loperamide is not more e; ective than carbophil. Wheat bran or corn bran is no more e; ective than placebo at reducing pain, bloating, or global symptoms of IBS, placebo in the relief of global symptoms of IBS and cannot be but it is an e; ective agent for the treatment of diarrhea, reduc- recommended for routine use (Grade 2C). Polyethylene glycol ing stool frequency, and improving stool consistency (Grade 2C). (PEG) laxative was shown to improve stool frequency —but not Randomized controlled trials comparing loperamide with other abdominal pain— in one small sequential study in adolescents antidiarrheal agents have not been performed. Safety and toler- with IBS-C (Grade 2C). ability data on loperamide are lacking.

Dietary * ber supplements studied in patients with IBS Patients with IBS-D display faster intestinal transit compared include wheat and corn bran. Bulking agents include psyllium with healthy subjects and, therefore, agents that delay intestinal hydrophilic mucilloid (ispaghula husk) and calcium polycar- transit may be bene* cial in reducing symptoms. Loperamide is bophil. Most trials of these agents are suboptimal and had small the only antidiarrheal agent su+ ciently evaluated in randomized sample sizes, short duration of follow-up, and were conducted controlled trials (RCTs) for the treatment of IBS-D. Of the before modern standards for study design were established. two RCTs evaluating the e7 ectiveness of loperamide in the Neither wheat bran nor corn bran reduced global IBS symp- treatment of IBS with diarrhea-predominant symptoms, there toms. Psyllium hydrophilic mucilloid improved global IBS were no signi* cant e7 ects in favor of loperamide compared with symptoms in four of the six studies reviewed. Meta-analysis placebo. e trials were both double-blinded, but the propor- showed that the relative risk of IBS symptoms not improving tion of women in each trial was unclear and neither reported with psyllium was 0.78 (95 % CI = 0.63 – 0.96) and the number adequate methods of randomization or adequate concealment needed to treat (NNT) was six (95% CI = 3 – 50). A single study of allocation. Each trial used a clinical diagnosis of IBS sup- of calcium polycarbophil showed bene* t. Adverse events in plemented by negative investigations to de* ne the condition. these studies were not reported systematically. Bloating may be Loperamide had no e7 ect on symptoms of bloating, abdomi- a risk with these agents. nal discomfort or global IBS symptoms. ere was a bene* cial

S4 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Position Statement on the Management of IBS

e7 ect to improve stool frequency and consistency, although the compared with placebo-treated patients (43 vs. 23 %, p < 0.05). overall impact of loperamide on IBS symptoms was not statis- In a single RCT, clarithromycin was not signi* cantly better than tically signi* cant. Both trials reported that all subjects in the placebo. In one report, metronidazole-treated patients demon- loperamide group had improved stool consistency compared strated signi* cant improvement over placebo-treated patients, with controls. Based on these results, loperamide is considered but data from this study were not presented in an extractable an e7 ective therapy for diarrhea. Inadequate data on adverse form. Overall adverse event data were not available for these events was reported. trials, but no severe adverse events were reported.

Effectiveness of antibiotics in the management of irritable Effectiveness of probiotics in the management of irritable bowel bowel syndrome (see Section 2.10) syndrome (see Section 2.11) A short-term course of a nonabsorbable antibiotic is more In single organism studies, lactobacilli do not appear e; ective e; ective than placebo for global improvement of IBS and for for patients with IBS; bi dobacteria and certain combinations of bloating (Grade IB). B ere are no data available to support the probiotics demonstrate some e? cacy (Grade 2C). long-term safety and e; ectiveness of nonabsorbable antibiotics for the management of IBS symptoms. Probiotics possess a number of properties that may prove of bene* t to patients with IBS. Interpretation of the available lit- Rifaximin, a nonabsorbable antibiotic, has demonstrated e+ - erature on the use of probiotics in IBS, however, is hampered cacy in three RCTs evaluating 545 patients with IBS. All of these by di+ culties in comparing studies using probiotics that varied RCTs demonstrated statistically signi* cant improvement in widely in terms of species, strains, preparations, and doses. Fur- global IBS symptoms, bloating symptoms, or both in rifaximin- thermore, and reK ecting limitations in study design, the data treated patients compared with placebo-treated patients. More- are conK icting: the dichotomous data suggest that all probiotic over, these three RCTs were well designed, meeting all criteria therapies have a trend for being e+ cacious in IBS, whereas the for appropriately designed RCTs (i.e., truly randomized stud- continuous data indicate that Lactobacilli have no impact on ies with concealment of treatment allocation, implementation symptoms; probiotic combinations improve symptoms; and of masking, completeness of follow-up, and intention-to-treat there is a trend for Bi dobacteria to improve IBS symptoms. analysis) and meeting most criteria of the Rome committee for Another de* ciency in study design is that most studies were of design of treatment trials of functional GI disorders. Rifaximin short-term, so we lack information on long-term use. Available is not Food and Drug Administration (FDA)-approved for safety data indicate that these preparations are well tolerated treatment of IBS, although it is FDA-approved for treatment and free from serious adverse side e7 ects in this population. of traveler ’ s diarrhea at a dose of 200 mg twice daily for three days; IBS trials used higher doses of rifaximin for longer peri- Effectiveness of the 5HT 3 receptor antagonists in the manage- ods: 400 mg three times daily for 10 days, 400 mg twice daily ment of irritable bowel syndrome (see Section 2.12) for 10 days, and 550 mg twice daily for 14 days. Also, the larg- B e 5-HT3 receptor antagonist alosetron is more e; ective than est RCT (n = 388 patients) only examined patients with IBS-D. placebo at relieving global IBS symptoms in male (Grade 2B) Rifaximin-treated patients were 8 – 23 % more likely to experience and female (Grade 2A) IBS patients with diarrhea. Potentially global improvement in their IBS symptoms, bloating symptoms, serious side e; ects including constipation and colon ischemia or in both, compared with placebo-treated patients. Rifaximin- occur more commonly in patients treated with alosetron treated patients also demonstrated signi* cant improvement compared with placebo (Grade 2A). B e bene ts and harms in diarrhea compared with placebo-treated patients. Based on balance for alosetron is most favorable in women with severe these results, rifaximin is most likely to be e+ cacious in IBS-D IBS and diarrhea who have not responded to conventional patients or IBS patients with a predominant symptom of bloat- therapies (Grade 1B). B e quality of evidence for e? cacy of ing; the appropriate dosage is approximately 1,100 –1,200 mg / day 5-HT 3 antagonists in IBS is high. for 10– 14 days. Minimal safety data were reported in these trials, but rifaximin-treated patients reportedly tolerated anti- Alosetron remains the only 5-HT3 receptor antagonist biotics without severe adverse events. However, given the oM en approved for the treatment of women with severe IBS-D in the chronic and recurrent nature of IBS symptoms and the theo- United States. In eight large, well-designed clinical trials that retical risks related to long-term treatment with any antibiotic, a evaluated alosetron use in 4,840 patients, this drug has demon- recommendation regarding continuous or intermittent use of strated superiority over placebo for abdominal pain, urgency, this agent in IBS must await further, long-term studies. It must global IBS symptoms, and diarrhea-related complaints. Con- also be stressed that available data on rifaximin is based on phase sidering the primary therapeutic endpoint as “ adequate II studies; phase III studies have yet to be reported. relief ” of abdominal pain and discomfort or urgency, the rela- Neomycin, metronidazole, and clarithromycin also have tive risk of IBS persisting with alosetron treatment was 0.79 been evaluated for the management of IBS. In a single RCT (95% CI = 0.69-0.91 with NNT = 8; 95% CI = 5 – 17). In one of 111 patients, neomycin-treated patients were more likely placebo-controlled study, alosetron demonstrated sustained to experience 50 % improvement in global IBS symptoms relief of abdominal pain and discomfort as well as urgency in

IBS-D patients for up to 48 weeks, with a safety pro* le compa- likely to experience satisfactory improvement of global IBS rable to that of placebo. Another recent randomized, placebo- symptoms than were placebo-treated patients. Tegaserod is also controlled trial found alosetron to be more e7 ective for the only 5-HT4 agonist that has been evaluated in an IBS-M abdominal pain and discomfort than placebo in men with population. In a well-designed RCT, tegasarod-treated patients IBS-D (53 vs. 40 % , P < 0.001). with the IBS-M were 15% more likely to demonstrate improve- In a recent systematic review which included data from seven ment in global IBS symptoms compared with placebo-treated studies, patients randomized to alosetron were statistically patients. In most RCTs, tegaserod-treated patients were signi* - signi* cantly more likely to report an adverse event than those cantly more likely to experience improvement in abdominal randomized to placebo (relative risk (RR) of adverse event = 1.18; discomfort, satisfaction with bowel habits, and bloating than 95 % CI = 1.08 – 1.29). e number needed to harm (NNH) with placebo-treated patients. Diarrhea occurred signi* cantly more alosetron was 10 (95 % CI = 7 – 16). Dose-dependent constipa- oM en in tegaserod-treated patients compared with placebo- tion was the most commonly reported adverse event with treated patients, most trials reporting diarrhea in approximately alosetron (1 mg twice daily = 29 % ; 0.5 mg twice daily = 11 % ). 10 % of tegaserod-treated patients and in approximately 5 % of Another systematic review of the clinical and postmarketing placebo-treated patients. Approximately 1– 2 % of tegaserod- surveillance data from IBS patients and the general population treated patients discontinued tegaserod because of diarrhea. con* rmed a greater incidence of severe complicated constipa- Tegaserod was removed from the market in March of 2007 tion and ischemic colitis in patients taking alosetron, however, aM er examination of the total clinical trial database revealed the incidence of these events was low, with a rate of 1.1 cases of that cardiovascular events were more frequent in tegaserod- ischemic colitis and 0.66 cases of complicated constipation per treated patients (n = 11,614) compared with placebo-treated 1,000 patients-years of alosetron use. patients (n = 7,031; 0.11% vs. 0.01 % ). irteen tegaserod-treated Current use of alosetron is regulated by a prescribing patients had cardiovascular events including myocardial infarc- program set forth by the FDA and administered by the tion (n = 4), unstable angina ( n = 6), and cerebral vascular acci- manufacturer (Prometheus Laboratories, San Diego, CA). e dent (n = 3) whereas one placebo-treated patient had a transient recommended starting dose of alosetron is 0.5 mg twice daily. ischemic attack. Currently, tegaserod is not available under any If aM er four weeks, the drug is well tolerated but the patient’ s treatment investigational new drug protocol, but it is available IBS-D symptoms are not adequately controlled, the dose can from the FDA through an emergency investigational new drug be escalated to 1 mg twice daily. Alosetron should be discon- protocol. tinued if the patient develops symptoms or signs suggestive of Renzapride and cisapride did not produce any statistically severe constipation or ischemic colitis or if there is no clinical signi* cant improvement in global IBS symptoms compared response to the 1 mg twice daily dose aM er four weeks. with placebo.

Effectiveness of 5HT4 (serotonin) receptor agonists in the Effectiveness of the selective C-2 chloride channel activators in management of irritable bowel syndrome (see Section 2.13) the management of irritable bowel syndrome (see Section 2.14) B e 5-HT4 receptor agonist tegaserod is more e; ective than Lubiprostone in a dose of 8 g twice daily is more e; ective than placebo at relieving global IBS symptoms in female IBS-C placebo in relieving global IBS symptoms in women with IBS-C patients (Grade 1A) and IBS-M patients (Grade 1B). B e most (Grade 1B). common side e; ect of tegaserod is diarrhea (Grade 1A). A small number (0.11 % ) of cardiovascular events (myocardial infarc- Lubiprostone (8 g twice daily) is approved by the FDA for tion, unstable angina, or stroke) were reported among patients the treatment of IBS-C in women on the basis of two well- who had received tegaserod in clinical trials. designed, large clinical trials. Based on a conservative endpoint designed to minimize placebo response, lubiprostone improved

Currently, there are no 5-HT4 receptor agonists available for global IBS-C symptoms in nearly twice as many subjects as did use in North America. Tegaserod (6 mg twice daily) has been placebo (18 vs. 10 % , P < 0.001). Lubiprostone also improved approved by the FDA for the treatment of IBS with constipa- individual symptoms of IBS including abdominal discomfort/ tion in women. Tegaserod was evaluated in multiple RCTs that pain, stool constancy, straining, and constipation severity. No were very well designed, meeting all criteria for appropriately one symptom appeared to drive the global improvement. E7 ects designed RCTs (i.e., truly randomized studies with concealment were well maintained for up to 48 weeks in open-label continu- of treatment allocation, implementation of masking, complete- ation studies. Side e7 ects included nausea (8 % ), diarrhea (6 % ) ness of follow-up and intention-to-treat analysis) and meeting and abdominal pain (5 % ), but were less frequent in these IBS-C almost all criteria of the Rome committee for design of treat- studies than in previous studies of patients with chronic consti- ment trials of functional GI disorders. Each of the RCTs assess- pation in which a larger dose (24 g twice daily) of lubiprostone ing the e+ cacy of tegaserod 6 mg twice daily demonstrated that was used. Lubiprostone should not be used in patients with it was superior to placebo for global IBS symptom improve- mechanical bowel obstruction or preexisting diarrhea. Women ment. Based on the de* ned end point of global IBS symptom capable of bearing children should have a documented negative improvement, tegaserod-treated patients were 5 – 19 % more pregnancy test before starting therapy and should be advised to

S6 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Position Statement on the Management of IBS

use contraception while taking lubiprostone. Studies need to be cognitive behavioral therapy, dynamic psychotherapy, hypno- conducted in men before this agent can be recommended for therapy, and relaxation therapy. use in men. In 20 RCTs that compared various psychological therapies with usual care, there was a bene* t for cognitive beha- Effectiveness of antidepressant agents in the management of vioral therapy, dynamic psychotherapy, and hypnotherapy, irritable bowel syndrome (see Section 2.15) but not relaxation therapy. ere have been more studies of Tricyclic antidepressants (TCAs) and selective serotonin reuptake cognitive behavioral therapy than any other management inhibitors (SSRIs) are more e; ective than placebo at relieving approaches, and a high-quality, large North American trial global IBS symptoms, and appear to reduce abdominal pain. of 12-week duration clearly showed its bene* t. Psycho- B ere are limited data on the safety and tolerability of these logical therapies are not documented to have any serious agents in patients with IBS (Grade 1B). adverse events, although the mechanisms of their bene* t remain unclear. Nine trials were identi* ed that tested TCAs in various doses for IBS. TCAs clearly were superior to placebo (NNT = 4, 95% Effectiveness of herbal therapies and acupuncture in the CI = 3–6). ere is no convincing evidence that the dose needed management of irritable bowel syndrome (see Section 2.17) has to be in the antidepressant range, and most trials tested Available RCTs mostly tested unique Chinese herbal mixtures, low-dose TCAs. In two of the trials, abdominal pain was the and appeared to show a bene t. It is not possible to combine primary endpoint and a signi* cant bene* t was observed. these studies into a meaningful meta-analysis, however, and Five trials that assessed SSRIs also showed a bene* t in IBS overall, any bene t of Chinese herbal therapy in IBS continues over placebo (NNT = 3.5). eoretically, SSRIs should be of to potentially be confounded by the variable components used most bene* t for IBS-C, whereas TCAs should be of greatest and their purity. Also, there are signi cant concerns about bene* t for IBS-D because of their di7 erential e7 ects on intes- toxicity, especially liver failure, with use of any Chinese herbal tinal transit time, but there is a lack of available data from the mixture. A systematic review of trials of acupuncture was clinical trials to assess this clinical impression. inconclusive because of heterogeneous outcomes. Further work e safety of using antidepressants in IBS remains poorly is needed before any recommendations on acupuncture or documented, although data suggest that the SSRIs are tolerated herbal therapy can be made. better than the TCAs. No data on the e+ cacy of SSRIs or other new antidepressant drug classes are available in this literature. Emerging therapies for irritable bowel syndrome (see Section Effectiveness of psychological therapies in the management of 2.18) irritable bowel syndrome (see Section 2.16) Our expanding knowledge of the pathogenesis of IBS has led to Psychological therapies, including cognitive therapy, dynamic the identi cation of a wide variety of novel therapeutic agents. psychotherapy, and hypnotherapy, but not relaxation therapy, Broadly speaking, there are agents in development for IBS with are more e; ective than usual care in relieving global symptoms predominantly peripheral e; ects and some with both peri- of IBS (Grade 1B). pheral and central e; ects. Examples of classes of drugs with predominantly peripheral e; ects include agents that a; ect chloride Among patients with IBS who seek care, particularly in subspe- secretion, calcium channel blockers, opioid receptor ligands, and cialty practice, the majority have anxiety, depression, or features motilin receptor ligands. Drug classes, which exert e; ects both of somatization. e overlap of psychologic disorders with IBS peripherally and centrally, include novel serotonergic agents, has led to studies evaluating the bene* ts of psychological thera- corticotropin-releasing hormone antagonists, and autonomic pies in reducing IBS symptoms. Psychological therapies include m o d u l a t o r s .

An Evidence-Based Systematic Review on the Management of Irritable Bowel Syndrome

American College of Gastroenterology IBS Task Force: Lawrence J. B r a n d t , M D , MACG, Chair 1 , W i l l i a m D . C h e y , 2 M D , , y F E. AC Am F G o xx-Orenstein , DO, FACG 3 , E a m o n n M.M. Q u i g l e y , M D , F AC G 4 , L a w r e n c e R . S c h i l l e r , 5 M , P D h ,i l i Fp AC S. G S c h o e n f e l d , M D , F AC6 , B G r e n n a n M. S p i e g e l , M D , 7 F , AC N i G c h o l a s J . T a l l e y , M D , P h D , F AC G 8 w i t h P a u l M o a y y e d i , E p idemiologist-Statistician, BSc, MB ChB, PhD, MPH, FRCP (London), FRCPC, FACG 9

Section 2.1 Methodology for systematic reviews of irritable (xi) Psychological therapies bowel syndrome therapy, levels of evidence, and grading of (xii) Herbal therapies and acupuncture recommendations (xiii) Emerging therapies We have conducted a series of systematic reviews on the diagnostic criteria, the value of diagnostic tests, and the e+ cacy Subjects needed to be followed up for at least one week. e of therapy in IBS. We also performed a narrative review of trial needed to include one or more of the following outcome the epidemiology of IBS. ere have been several systematic measures: reviews of therapy for IBS (1 – 5) , but these either have not (i) Global assessment of IBS cure or improvement quantitatively combined the data into meta-analyses (1 – 3) (ii) Abdominal pain cure or improvement or have inaccuracies in applying eligibility criteria and data (iii) Global IBS symptom or abdominal pain scores extraction (4,5) . We have, therefore, repeated all systematic reviews of IBS and synthesized the data where appropriate. Search strategy for identi cation of studies Medline (1966 – June 2008), Embase (1988 – June 2008), and the Systematic review methodology Cochrane Controlled Trials Register (Issue 2, 2008) electronic For all reviews, we evaluated manuscripts that studied adults databases were searched. An example of the Medline search is using any de* nition of IBS. is included a clinician-de* ned given below. diagnosis, Manning criteria (6), the Kruis score (7) , or Rome I IBS patients were identi* ed with the terms irritable bowel (8) , II (9) , or III (10) criteria. syndrome a n d functional diseases , colon (both as medical sub- For reviews of diagnostic tests, we included case series and ject heading (MeSH) and free text terms), and IBS , spastic case-control studies that evaluated serologic tests for celiac colon , irritable colon , a n d functional adj5 (adj5 is a term used sprue (anti-gliadin, anti-endomysial, and tissue transglutami- by Medline for words that appear within 5 adjectives of each nase antibodies), lactose hydrogen breath tests, and tests for other) bowel (as free text terms). small bowel bacterial overgrowth (lactulose and glucose hydro- Studies identi* ed from this search were combined with gen breath test or jejunal aspirates). the following terms used to identify therapies for IBS: dietary For reviews of therapies of IBS, we included only parallel ber , cereals , psyllium , sterculia , karaya gum , parasympatho- group RCTs comparing active intervention with either placebo lytics , scopolamine , trimebutine , muscarinic antagonists, and the or no therapy. following free text terms: bulking agent , psyllium ber , ber , e following treatments were considered: husk , bran , ispaghula , wheat bran , spasmolytics , spasmolytic (i) Diet agents , antispasmodics , mebeverine , alverine , pinaverium bro- (ii) Fiber, bulking agents, and laxatives mide , otilonium bromide , cimetropium bromide , hyoscine butyl (iii) Antispasmodics and Peppermint Oil bromide , butylscopolamine , peppermint oil , loperamide and (iv) Antidiarrheal agents colpermin , serotonin antagonists , serotonin agonists , cisapride , (v) Antibiotic therapy receptors (serotonin , 5-HT 3 ), and receptors (serotonin , 5-HT 4 ; (vi) Probiotic therapy both as MeSH terms and free text terms), and the following (vii) 5HT3 antagonists free text terms: 5-HT , 5-HT , alosetron , cilansetron , tegas- (viii) 5HT4 agonists 3 4 (ix) Selective C-2 chloride channel activators erod , and renzapride , psychotropic drugs , antidepressive agents , (Lubiprostone) antidepressive agents (tricyclic) , desipramine , imipramine , (x) Antidepressants trimipramine , doxepin , dothiepin , nortriptyline , amitriptyline ,

1 Division of Gastroenterology, Monteﬁ ore Medical Center, Bronx , New York, USA ; 2 Division of Gastroenterology, University of Michigan Health System, Ann Arbor , Michigan , USA ; 3 Division of Gastroenterology and Hepatology, Department of Internal Medicine, Mayo Clinic, Rochester, Minnesota , USA ; 4 Department of Medicine, Clinical Sciences Building, Cork University Hospital, Cork , Ireland ; 5 Digestive Health Associates of Texas, Baylor University Medicine Center, Dallas, Texas , USA ; 6 Veterans Affairs Ann Arbor Healthcare System, Division of Gastroenterology, Ann Arbor , Michigan , USA ; 7 VA Greater Los Angeles Healthcare System, David Geffen School of Medicine at UCLA, Los Angeles, California , USA ; 8 Department of Medicine, Mayo Clinic, Jacksonville , Florida , USA ; 9 Department of Medicine, Division of Gastroenterology, McMaster University Medical Centre, Hamilton , Canada . Correspondence: Lawrence J. Brandt, MD, MACG, Monteﬁ ore Medical Center, 111 East 210 Street, Bronx, New York 10467, USA.

S8 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

selective serotonin re-uptake inhibitors , paroxetine , sertraline , of evidence and the strength of recommendation. is system, = uoxetine , citalopram , venlafaxine , cognitive therapy , psycho- which is outlined in Table 1 , includes the assessment of quality therapy , behaviour therapy , relaxation techniques , and hypnosis of evidence and bene* t-risk pro* le in the graded recommenda- (both as MeSH terms and free text terms), and the following tion. e grading scheme classi* es recommendations as strong free text terms: behavioral therapy , relaxation therapy , and (Grade 1) or weak (Grade 2) according to the balance of bene- hypnotherapy. Saccharomyces , Lactobacillus , Bi dobacterium , * ts, risks, burdens, and sometimes costs, based on evaluation Escherichia coli or probiotics (MeSH and free text terms). by experts. Also, this system classi* es the quality of evidence as An RCT * lter was applied to the electronic searches. e high (Grade A), moderate (Grade B), or low (Grade C) accord- searches were limited to humans. A recursive search of the ing to the quality of study design, the consistency of results bibliography of relevant articles also was conducted. among individual studies, and directness and applicability of DDW abstract books were hand searched between 2000 and study endpoints. With this graded recommendation, the clini- 2008; UEGW abstract books were hand searched between 2000 cian receives guidance about whether or not recommenda- and 2007. Authors of trial reports that did not give enough tions should be applied to most patients and whether or not detail for adequate data extraction were contacted and asked recommendations are likely to change in the future aM er to contribute full datasets. Experts in the * eld were contacted production of new evidence. Grade 1A recommendations for leads on unpublished studies and no language restrictions represent a “strong recommendation that can apply to most were applied. patients in most circumstances and further evidence is unlikely Trials were assessed for risk of bias according to three to change our con dence in the estimate of treatment e; ect.” characteristics: method of randomization, method of conceal- In the opinion of the Task Force, a Grade 1A recommenda- ment of treatment allocation, and implementation of masking. tion can only be justi* ed by data from thousands of patients. In addition the quality of studies were graded according to Currently-available IBS therapies have not been studied in the Jadad scale (11) with a score of ≥ 4 being considered a thousands of appropriate patients. erefore, no currently high quality. available IBS therapy has received a Grade 1A recommenda- Eligibility, quality, and outcome data were extracted by the tion. e system is most appropriate for IBS management lead reviewer (Paul Moayyedi) and by a masked second reviewer strategies and is less relevant for de* nitions and epidemio- (Alex Ford) on specially developed forms. Any discrepancy was logic data, so statements in the epidemiologic section are not resolved by consensus using a third reviewer (Nicholas Talley). g r a d e d . Data were extracted as intention-to-treat analyses, and dropouts were assumed to be treatment failures. Section 2.2 The burden of illness of irritable bowel syndrome IBS is a prevalent and expensive condition that is associated with D a t a s y n t h e s i s a signi cantly impaired HRQOL and reduced work productivity. Whenever possible, any improvement of global IBS symp- Based on strict criteria, 7– 10 % of people have IBS worldwide. toms as a binary outcome was taken as the primary outcome Community-based data indicate that IBS-D and IBS-M sub- measure. If this was not available, improvement in abdominal types are more prevalent than IBS-C, and that switching among pain was used. e impact of interventions was expressed as subtype groups may occur. IBS is 1.5 times more common in RR of IBS symptoms not improving together with 95 % con* - women than in men, is more common in lower socioeconomic dence intervals. If there were su+ cient data, relative risks were groups, and is more commonly diagnosed in patients younger combined using the DerSimonian and Laird random e7 ects than 50 years of age. Patients with IBS visit the doctor more fre- model (12) . Tests of heterogeneity also were reported (13) . quently, use more diagnostic tests, consume more medications, When the test of heterogeneity was signi* cant (p < 0.10 and / or miss more workdays, have lower work productivity, are hospital- I 2>25 % ), the reasons for this were explored by evaluating di7 er- ized more frequently, and consume more overall direct costs than ences in study population, study design, or study endpoints in patients without IBS. Resource utilization is highest in patients subgroup analyses. Publication bias or other causes of small study with severe symptoms, and poor HRQOL. Treatment decisions e7 ects were evaluated using tests for funnel plot asymmetry (14) . should be tailored to the severity of each patient ’ s symptoms and e NNT was calculated as the inverse of the risk di7 erence HRQOL decrement. from the meta-analysis and checked using the formula: IBS is a prevalent condition that can a7 ect patients physically, psychologically, socially, and economically. Awareness of and knowledge about this burden of illness serves several purposes. (RRR = relative risk reduction, BR = baseline risk). For patients, it emphasizes that many others have IBS, and that people su7 ering from IBS should not feel alone with their diag- Methodology for assessing levels of evidence and grading nosis or disease-related experiences. For healthcare providers, recommendations it highlights that IBS is a large part of both internal medicine A commonly used system for grading recommendations in evi- and gastroenterology practices. Moreover, it allows providers dence-based guidelines (15) was employed to assess the quality to improve their understanding of the impact of IBS on their

Table 1 . Grading recommendations

Grade of Recommenda- Beneﬁ t vs. risk and burdens Methodological quality of Implications tion/description supporting evidence

1A. Strong recom- Benefi ts clearly outweigh risk RCTs without important limitations Strong recommendation, can apply to most pa- mendation, high-quality and burdens, or vice versa or overwhelming evidence from tients in most circumstances. Further evidence evidence observational studies is unlikely to change our confi dence in the estimate of effect 1B. Strong recommen- Benefi ts clearly outweigh risk RCTs with important limitations Strong recommendation, can apply to most dation, moderate-quality and burdens, or vice versa (inconsistent results, methodologi- patients in most circumstances. Higher quality evidence cal fl aws, indirect, or imprecise) evidence may well change our confi dence in or exceptionally strong evidence the estimate of effect from observational studies 1C. Strong recommen- Benefi ts clearly outweigh risk Observational studies or case Strong recommendation can apply to most dation, low-quality or and burdens, or vice versa series patients in most circumstances. Higher quality very low-quality evidence evidence is very likely to change our confi - dence in the estimate of effect 2A. Weak recommen- Benefi ts closely balanced with RCTs without important limitations Weak recommendation, best action may differ dation, high-quality risks and burden or overwhelming evidence from depending on circumstances or patients’ or evidence observational studies societal values. Further evidence is unlikely to change our confi dence in the estimate of effect 2B. Weak recommenda- Benefi ts closely balanced with RCTs with important limitations Weak recommendation, best action may differ tion, moderate-quality risks and burden (inconsistent results, methodologi- depending on circumstances or patients’ or evidence cal fl aws, indirect, or imprecise) societal values. Higher quality evidence may or exceptionally strong evidence well change evidence our confi dence in the from observational studies estimate of effect 2C. Weak recommenda- Uncertainty in the estimates Observational studies or case Very weak recommendations; other alterna- tion, low-quality or very of benefi ts, risks, and burden; series tives may be equally reasonable. Higher quality low-quality evidence benefi ts, risk, and burden may evidence is likely to change our confi dence in be closely balanced the estimate of effect

RCT, randomized controlled trial.

patients’ well being, and then act on this insight by selecting the studies. To re* ne the prevalence estimate, we performed treatments tailored to each patient ’ s symptoms and HRQOL an updated systematic review to target studies that only used decrement. For research funding and drug-approval authori- Rome de* nitions, drew upon patients from the general adult ties, it shows that IBS is far more than a mere nuisance, and is community (i.e., not exclusively from primary or secondary instead a condition with a prevalence and HRQOL impact that care), and included patients who were not selected speci* - matches other major diagnoses such as diabetes, hypertension, cally (e.g., not evaluating IBS in subjects with reK ux symptoms or kidney disease (16,17) . For employers and healthcare insur- or in twins). We identi* ed four eligible studies evaluating ers, it reveals the overwhelming direct and indirect expendi- 32,638 North American subjects and found that IBS prevalence tures related to IBS, and provides a business rationale to ensure varied between 5 and 10 % with a pooled prevalence of 7 % that IBS is treated e7 ectively. e objective of this section is to (95 % CI = 6 – 8 % ) (20 – 23) . Although previous reviews indicated review key data regarding the burden of illness of IBS, includ- that IBS patients are divided evenly among the three major ing: (1) the prevalence of IBS and its subtypes; (2) the age of subgroups (IBS-D, IBS-C, and IBS-M) (24), the true prevalence onset and gender distribution of IBS; (3) the e7 ect of IBS on of IBS subtypes in North America remains unclear; one study HRQOL; and (4) the economic burden of IBS, including direct suggested that IBS with diarrhea is the most common subtype and indirect expenditures and their clinical predictors. (21) , whereas another indicated that mixed-type IBS is most Previous systematic reviews have measured the prevalence common (22) . of IBS in both North American and European nations (18,19) . ere are several demographic predictors of IBS, including Prevalence estimates range from 1 % to over 20 %. is wide gender, age, and socioeconomic status. e odds of having range indicates that IBS prevalence, like prevalence of all dis- IBS are higher in women than in men (pooled OR = 1.46; 95 % eases, depends on several variables, including the case-* nding CI = 1.13 – 1.88) (20 – 23) , although IBS is not simply a disorder de* nition employed (e.g., Manning criteria vs. Rome criteria), of women. In fact, IBS is now recognized to be a key component the characteristics of the source population (e.g., primary vs. of the Gulf War Syndrome, a multi-symptom complex a7 ecting secondary care), and the methodology and sampling frame of soldiers (a predominantly male population) deployed in the

S10 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

1991 Gulf War (25 – 27) . IBS is diagnosed more commonly in Indeed, several studies have identi* ed predictors of HRQOL patients under the age of 50 years than it is in patients older than in IBS (31 – 35) , the most consistent of which is the severity of 50 years, although 2 –6 % of the latter group also su7 er from the the predominant bowel symptom. Data from several studies disorder (20 – 22) . ese data suggest that the pretest likelihood indicate that in patients with IBS, HRQOL decreases in paral- for IBS is higher in younger patients, but that patients of all ages lel with increasing symptom severity (29,31,33). It is therefore may be diagnosed with IBS. Our review identi* ed two stud- important not only to identify the predominant symptom of ies that reported IBS prevalence by income strata (21,23) , both patients with IBS, but also to gauge its severity. Additional data of which revealed a graded decrease in IBS prevalence with indicate that physical HRQOL in IBS is related to the dura- increasing income: 8 –16 % of people earning less than $ 20,000 tion of symptom K ares (≥ 24 h vs. < 24 h) and the presence of annually carry the diagnosis, compared with only 3 – 5 % of abdominal pain (as opposed to “ discomfort ” ) and that mental people earning more than $ 75,000 (21,23) . HRQOL is associated with abnormalities in sexuality, mood, and Several studies have compared HRQOL in IBS patients anxiety (35) . Perhaps more importantly, both domains share a and in healthy controls or controls with non-IBS medical common association with symptoms of chronic stress and vital disorders; these have been summarized in a previous system- exhaustion, including tiring easily, feeling low in energy, and atic review (16) . Data consistently demonstrate that patients experiencing sleep di+ culties (35) . Patients acknowledge that with IBS score lower on all eight scales of the SF-36 HRQOL these symptoms adversely inK uence their ability to function questionnaire compared with “ normal ” non-IBS cohorts. by prompting avoidance of socially vulnerable situations (e.g., Patients with IBS have the same physical HRQOL as patients being away from restrooms) and activities (e.g., eating out for with diabetes, and a lower physical HRQOL compared with dinner). In contrast, HRQOL is not strongly determined by the patients who have depression or gastroesophageal reK ux presence of speci* c gastrointestinal symptoms (e.g., diarrhea, disease (16,17) . Perhaps more surprisingly, mental HRQOL constipation, bloating, dyspepsia), extent of previous gastroin- scores on the SF-36 were lower in patients with IBS than in testinal evaluation (e.g., previous K exible sigmoidoscopy or those with chronic renal failure— an organic condition marked colonoscopy), or common demographic characteristics (e.g., by considerable physical and mental disability. is HRQOL gender, age, marital status) (33). decrement can, in some cases, be so severe as to raise the risk e above * ndings suggest that rather than focusing on of suicidal behavior (28,29) . e relationship between IBS and physiologic epiphenomena to gauge HRQOL (e.g., stool suicidality is independent of comorbid psychiatric diseases frequency, stool characteristics, subtype of IBS), it may be more such as depression (28,29) . Many of these studies, however, e+ cient to assess HRQOL by gauging global symptom severity, were performed in tertiary-care referral populations, and the addressing symptom-related fears and concerns, and identify- HRQOL decrement and suicidality risk documented in these ing and eliminating factors contributing to vital exhaustion in cohorts may not be applicable to community-based popu- IBS. In short, treating bowel symptoms in IBS is necessary, but lations. It is possible that patients with IBS develop HRQOL may not be su+ cient, to inK uence overall HRQOL. In addi- decrements due to their disease, and also possible that some tion to treating symptoms, providers should attempt to modify patients with diminished HRQOL subsequently develop IBS positively the cognitive interpretation of IBS symptoms— i.e., (30). Although the precise directionality of this relationship acknowledge and address the emotional context in which may vary from patient to patient, it is clear that IBS is strongly symptoms occur (36 – 38) . related to low HRQOL, and vice versa. Nonetheless, IBS is also Patients with IBS consume a disproportionate amount of likely to cause a negative impact on HRQOL, and failing to rec- resources. Burden of illness studies estimate that there are 3.6 ognize this impact could undermine the physician– patient rela- million physician visits for IBS in the United States annually, tionship and lead to dissatisfaction with care. Because HRQOL and that IBS care consumes over $20 billion in both direct and decrements are common in IBS, we recommend that clinicians indirect expenditures (39) . Moreover, patients with IBS consume perform routine screening for diminished HRQOL in their IBS over 50 % more health care resources than matched controls with- patients. Treatment should be initiated when the symptoms of out IBS (40,41) . ese data suggest that the economic burden of IBS are found to reduce functional status and diminish overall IBS stems not only from the high prevalence of the disease, but HRQOL. Furthermore, clinicians should remain wary of poten- also from the disproportionate use of resources it causes. tial suicidal behavior in patients with severe IBS symptoms, and It is unclear why patients with IBS consume a disproportionate should initiate timely interventions if suicide forerunners are amount of resources, especially in the light of data that diagnos- identi* ed. tic tests and procedures in IBS rarely detect alternative underly- A practical limitation of determining HRQOL in busy out- ing conditions (see Section 2.5). Despite the dissemination and patient settings is that its accurate measurement requires a use of guidelines reinforcing these data (24) , much of the cost thorough and oM en time-consuming evaluation of biologic, of care in IBS arises from sequential diagnostic tests, invasive psychologic, and social health domains. To help providers procedures, and abdominal operations (39,42) . For example, gain better insight into their patients’ HRQOL, a concise list patients with IBS are three times more likely than matched con- of factors known to predict HRQOL in IBS might be helpful, trols to undergo cholecystectomy (42) despite knowledge that which providers could then use to question patients routinely. IBS symptoms almost invariably persist following the surgery.

Similarly, nearly 25% of colonoscopies performed in patients e ACG Task Force conducted a systematic review of the younger than 50 years of age are for IBS symptoms (43) , regard- accuracy of symptom-based criteria in the diagnosis of IBS less of data that indicate colonoscopy has a low diagnostic yield (51) . Overall, eight studies (52 – 59) were identi* ed involv- in IBS, and that “ negative ” examinations fail to improve intes- ing 2,280 patients. In all studies, IBS was de* ned as a clinical tinal symptoms, do not augment HRQOL, and are unlikely to diagnosis aM er investigations that included either a colono- provide additional reassurance when compared with not per- scopy or a barium enema. e accuracy of individual symp- forming colonoscopy (44) . Resource utilization in IBS also is toms was described in six studies (52 – 57) evaluating 1,077 driven partly by the presence of comorbid somatization —a trait patients. Symptoms such as abdominal pain, loose or frequent found in up to one-third of IBS patients that is characterized by stools associated with pain, incomplete evacuation, mucus per the propensity to overinterpret normal physiologic processes rectum, and abdominal distention all had limited accuracy in (45,46) . Patients with somatization typically report a barrage of diagnosing IBS. Lower abdominal pain had the highest sensi- seemingly unrelated physical complaints (e.g., back pain, tin- tivity (90 % ) but very poor speci* city (32 %), whereas patient- gling, headaches, temporomandibular joint pain, muscle aches) reported visible abdominal distention had the highest speci* city that may, in fact, be linked to underlying psychosocial distress (77% ) but low sensitivity (39% ). A variety of criteria therefore (45,46) . ese patients are sometimes misclassi* ed as hav- have been developed to identify a combination of symptoms ing several underlying organic conditions, and subsequently to diagnose IBS (see Table 2 ) . undergo sequential diagnostic tests in chase of the disparate e * rst description of this approach was by Manning et al . symptoms (47) . ere is a linear and highly signi* cant relation- (52) and there have been four studies evaluating the accuracy ship between levels of somatization and the amount of diag- of Manning ’ s criteria in 574 patients. Two studies (52,58) sug- nostic testing in IBS, suggesting that providers should remain gested these criteria performed well, whereas accuracy was poor alert for somatization in IBS, and aggressively treat or refer in the other two studies (56,57). Overall, Manning’ s criteria had somatization patients to an experienced specialist rather than a pooled sensitivity of 78 % and pooled speci* city of 72% (51) . performing potentially unnecessary diagnostic tests (47) . e next description of symptom criteria was by Kruis et al . , In addition to direct costs of care, IBS patients engender and four studies (53– 55) have described the accuracy of this signi* cant indirect costs of care as a consequence of both approach in 1,166 patients. ree studies (53,54,58) suggested missing work and su7 ering impaired work performance the Kruis symptoms score had an excellent positive predictive while on the job. Employees with IBS are absent 3– 5 % of the value with a pooled sensitivity of 77% and pooled speci* city workweek, and report impaired productivity 26– 31 % of the of 89 % . Subsequently, an international working group deve- week (48 – 50) — rates that exceed those of non-IBS con- loped the Rome criteria, which have undergone three itera- trol employees by 20% (49) ; this is equivalent to 14 hours of tions over 15 years. ese criteria have been heavily promoted, lost productivity per 40-hour workweek. Compared with IBS although there has been only one study in which the accuracy patients who exhibit normal work productivity, patients with of Rome I criteria has been evaluated and none describing impaired productivity have more extraintestinal comorbidi- the accuracy of Rome II or III (50) . In the 602 patients studied, ties (e.g., chronic fatigue syndrome, * bromyalgia, interstitial the Rome I criteria had a sensitivity of 71% and speci* city of cystitis), and more disease-speci* c fears and concerns (50) . In 85% (59) . contrast, the speci* c pro* le of individual bowel symptoms does All studies evaluating the accuracy of diagnostic criteria not undermine work productivity (50), suggesting that enhanc- in patients with IBS face the problem of lack of a reference ing work productivity in patients with IBS may require treat- standard test for this condition. Notwithstanding, none of ments that improve both GI and non-GI symptom intensity, the symptom-based diagnostic criteria have an ideal accuracy while also modifying the cognitive and behavioral responses to and the Rome criteria, in particular, have been inadequately bowel symptoms and the contexts in which they occur. In other evaluated, despite their extensive use in the research setting. words, it may be inadequate to treat bowel symptoms alone e ACG Task Force felt that a pragmatic de* nition that was without simultaneously addressing the emotional context in simple to use and that incorporated key features of previous which the symptoms occur. diagnostic criteria would be clinically useful. We, therefore, de* ned IBS as abdominal pain or discomfort that occurs in Section 2.3 The utility of diagnostic criteria in IBS association with altered bowel habits over a period of at least IBS is de ned by abdominal pain or discomfort that occurs in three months. association with altered bowel habits over a period of at least three months. Individual symptoms have limited accuracy for Section 2.4 The role of alarm features in the diagnosis of IBS diagnosing IBS and, therefore, the disorder should be considered Overall, the diagnostic accuracy of alarm features is disappoint- as a symptom complex. Although no symptom-based diagnosing. Rectal bleeding and nocturnal pain o; er little discriminative tic criteria have ideal accuracy for diagnosing IBS, traditional value in separating patients with IBS from those with organic criteria, such as Kruis and Manning, perform at least as well as diseases. Whereas anemia and weight loss have poor sensitivity Rome I criteria; the accuracy of Rome II and Rome III criteria for organic diseases, they o; er very good speci city. As such, in has not been evaluated. patients who ful ll symptom-based criteria of IBS, the absence

S12 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

Table 2 . Summary of diagnostic criteria used to deﬁ ne irritable bowel syndrome

Diagnostic criteria Symptoms, signs, and laboratory investigations included in criteria

Manning (1978) IBS is defi ned as the symptoms given below with no duration of symptoms described. The number of symptoms that need to be present to diagnose IBS is not reported in the paper, but a threshold of three positive is the most commonly used: 1. Abdominal pain relieved by defecation 2. More frequent stools with onset of pain 3. Looser stools with onset of pain 4. Mucus per rectum 5. Feeling of incomplete emptying 6. Patient-reported visible abdominal distension Kruis (1984) IBS is defi ned by a logistic regression model that describes the probability of IBS. Symptoms need to be present for more than two years. Symptoms: 1. Abdominal pain, fl atulence, or bowel irregularity 2. Description of character and severity of abdominal pain 3. Alternating constipation and diarrhea Signs that exclude IBS (each determined by the physician): 1. Abnormal physical fi ndings and/or history pathognomonic for any diagnosis other than IBS 2. Erythrocyte sedimentation rate >20 mm/2 h 3. Leukocytosis >10,000/cc 4. Anemia (Hemoglobin < 12 for women or < 14 for men) 5. Impression by the physician that the patient has rectal bleeding Rome I (1990) Abdominal pain or discomfort relieved with defecation, or associated with a change in stool frequency or consistency, PLUS two or more of the following on at least 25% of occasions or days for three months: 1. Altered stool frequency 2. Altered stool form 3. Altered stool passage 4. Passage of mucus 5. Bloating or distension Rome II (1999) Abdominal discomfort or pain that has two of three features for 12 weeks (need not be consecutive) in the last one year: 1. Relieved with defecation 2. Onset associated with a change in frequency of stool 3. Onset associated with a change in form of stool Rome III (2006) Recurrent abdominal pain or discomfort three days per month in the last three months associated with two or more of: 1. Improvement with defecation 2. Onset associated with a change in frequency of stool 3. Onset associated with a change in form of stool

IBS, irritable bowel syndrome.

of selected alarm features, including anemia, weight loss, and a Patients with typical IBS symptoms also may exhibit so-called family history of colorectal cancer, in= ammatory bowel disease, “ alarm features ” that increase concerns organic disease may be or celiac sprue, should reassure the clinician that the diagnosis of present. Alarm features include rectal bleeding, weight loss, iron IBS is correct. de* ciency anemia, nocturnal symptoms, and a family history

of selected organic diseases including colorectal cancer, IBD, family history of colorectal cancer, IBD or celiac sprue, should and celiac sprue. Usually, it is recommended that patients who reassure the clinician that the diagnosis of IBS is correct. exhibit alarm features undergo further investigation, particularly with colonoscopy to rule out organic disease, e.g., color- Section 2.5 The role of diagnostic testing in patients with IBS ectal cancer. e utility of this approach has been addressed in symptoms a systematic review of the literature (60) . is review evaluated Routine diagnostic testing with complete blood count, serum all patients presenting with lower gastrointestinal symptoms, as chemistries, thyroid function studies, stool for ova and para- there was no study that speci* cally addressed IBS patients as a sites, and abdominal imaging is not recommended in patients group. Nevertheless, the results of this review are likely to be with typical IBS symptoms and no alarm features because of applicable to IBS patients or may even overestimate the utility a low likelihood of uncovering organic disease (Grade 1C). of alarm symptoms because abdominal pain (a de* ning Routine serologic screening for celiac sprue should be pursued in symptom of IBS) is a negative predictor of serious underlying patients with IBS-D and IBS-M (Grade 1B). Lactose breath testing pathology (60) . In 13 studies evaluating the diagnostic utility can be considered when lactose maldigestion remains a concern of abdominal pain in 19,238 patients, the pooled positive despite dietary modi cation (Grade 2B). Currently, there are likelihood ratio was 0.72 (95 % CI = 0.60 – 0.88), and the pooled insu? cient data to recommend breath testing for small intestinal negative likelihood ratio was 1.21 (95 % CI = 1.11-1.32) for bacterial overgrowth in IBS patients (Grade 2C). Because of the colorectal cancer, i.e., the presence of abdominal pain reduces low pretest probability of Crohn’ s disease, ulcerative colitis, and the likelihood and the absence of pain increases the likelihood colonic neoplasia, routine colonic imaging is not recommended in of colorectal cancer. patients younger than 50 years of age with typical IBS symptoms Our review on the utility of alarm features to diagnose color- and no alarm features (Grade 1B). Colonoscopic imaging should ectal cancer (1) identi* ed 14 studies evaluating 19,189 patients be performed in IBS patients with alarm features to rule out or- with lower GI symptoms and reported on the accuracy of rectal ganic diseases and in those over the age of 50 years for the pur- bleeding in this regard. Rectal bleeding had a pooled sensitivity pose of colorectal cancer screening (Grade 1C). When colonoscopy of 64% (95% CI = 55 – 73 % ) and pooled speci* city of 52 % (95 % is performed in patients with IBS-D, obtaining random biopsies CI = 42 – 63 % ) for diagnosing colorectal cancer. Seven studies should be considered to rule out microscopic colitis (Grade 2C). involving 4,404 patients evaluated the diagnostic utility of anemia and found a pooled sensitivity of 19% (95% CI = 5.5 – 33 % ) IBS is a disorder of heterogeneous pathophysiology for which and pooled speci* city of 90% (95% CI = 87 – 92 % ) for diag- speci* c biomarkers are not yet available. Diagnostic tests are nosing colorectal cancer in patients with lower GI symptoms. therefore performed to exclude organic diseases that may mas- ere were * ve studies that assessed the accuracy of weight querade as IBS and, in so doing, reassure both the clinician and loss in 7,418 patients with lower GI symptoms. Weight loss had the patient that the diagnosis of IBS is correct. Historically, IBD, a pooled sensitivity of 22 % (95 % CI = 14 – 31 % ) and pooled colorectal cancer, diseases associated with malabsorption, sys- speci* city of 89 % (95 % CI = 81 – 95 % ). temic hormonal disturbances, and enteric infections are of the It also has been suggested that the presence of nocturnal greatest concern to clinicians caring for patients with IBS symp- symptoms may identify a group of patients more likely to har- toms. e broad di7 erential diagnosis of IBS symptoms as well bor organic disease. Studies suggest, however, that nocturnal as medicolegal concerns related to making an incorrect diagno- abdominal pain is no more likely in patients with organic sis of IBS drives most clinicians to view IBS as a “ diagnosis of diseases than it is in in patients with IBS (61,62) . exclusion” . is practice has tangible consequences for patients, ere is evidence to suggest that individuals with a family payors, and society at large. Physicians who feel that IBS is a history of colorectal cancer, IBD, and celiac sprue are at higher diagnosis of exclusion order more diagnostic tests and spend risk of having these organic diseases. e increased risk of more money to evaluate their patients than do experts who feel colorectal cancer among individuals with an a7 ected * rst- more con* dent about diagnosing IBS (66) . Given this informa- degree relative under 60 years of age is well documented (63) . tion, it is important to review the value of commonly ordered ere is epidemiologic evidence of a 4- to 20-fold increased diagnostic tests in patients with suspected IBS, including com- risk of IBD disease in * rst-degree relatives of an a7 ected patient plete blood count, serum chemistries, thyroid function studies, (64) . Recent evidence also has shown that between 4 and 5 % of markers of inK ammation, testing for celiac sprue, breath testing individuals with an a7 ected * rst-degree relative will have celiac for lactose maldigestion and bacterial overgrowth, and colonic sprue (65) . imaging. Overall, the accuracy of alarm features is disappointing. When deciding on the necessity of a diagnostic test in a Rectal bleeding and nocturnal pain o7 er little discriminative patient with IBS symptoms, one should consider * rst the pretest value in separating patients with IBS from those with organic probability of the disease in question. If the pretest probability diseases. Whereas anemia and weight loss have poor sensitivity of a particular disease is su+ ciently small, diagnostic testing for organic diseases, they o7 er very good speci* city. As such, directed at uncovering that improbable disease is unlikely to in patients who ful* ll symptom-based criteria, the absence of be either clinically useful or cost e7 ective. Clinicians also selected alarm features, including anemia, weight loss, and a should consider the performance characteristics (e.g., sensitivity,

S14 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

of stool ova and parasite examination in IBS patients (69,70) . Table 3 . Prevalence of organic diseases in patients meeting One study of 170 patients with IBS found no abnormal stool symptom-based criteria for IBS ova and parasite examination results, whereas a second study in Organic GI disease IBS patients (%) General population (%) 1,154 patients reported an abnormal result in 1.6% . Symptom response following treatment of identi* ed pathogens was not Colitis/IBD a 0.51 – 0.98 0.3 – 1.2 reported, so causality could not be established. Based on these Colorectal cancera 0 – 0.51 0 – 6 (varies with age) results, the Task Force does not recommend the routine use of Thyroid dysfunctionb 4.2 5 – 9 stool ova and parasite examination in patients with IBS. Gastrointestinal 0 – 1.5 NA ere are very limited data on the utility of abdominal imag- infection b ing tests in patients with IBS. One study evaluated the role of abdominal ultrasound to identify serious abdominal or pelvic Celiac sprueb 3.6 0.7 pathology in 125 patients diagnosed with IBS by the Rome I b Lactose maldigestion 38 26 criteria (73) . Of these patients, 22 (18 %) had abnormal ultra- a Data from Cash et al . (67) . b Data courtesy of Moayyedi, et al . (personal com- sound results, the most common explanation of which was munication, unpublished). gallstones in 6 (5 % ) patients. In no patient did the ultrasound results lead to a revision of the diagnosis of IBS. e Task Force recommends against the routine use of abdominal imaging in speci* city, positive and negative predictive values) of the diag- patients with IBS symptoms and no alarm features. nostic test under consideration when deciding on its relative ere is emerging evidence, however, to suggest that the value. Data from systematic reviews that address the pretest prevalence of celiac sprue is higher among patients with probability of organic diseases in patients with IBS symptoms IBS than in controls. e systematic review performed for are presented in Table 3 . this monograph (74) identi* ed seven case-control studies Application of the available data to routine clinical prac- (68,75 – 80) of 2,978 individuals (1,052 with IBS), which used tice may be limited by a number of factors including the rela- anti-endomysial or tissue-transglutaminase antibodies to tively small size of the study populations, the variable quality screen for celiac sprue. ree percent of the IBS cohorts, of study methodologies, and selected nature of study popula- compared with 0.7 % of controls, were found to have a posi- tions that typically derive from secondary or tertiary care facili- tive anti-endomysial or transglutaminase antibody, or both ties. Accepting these limitations, the prevalences of Crohn’ s (OR = 2.94, 95 % CI = 1.36 – 6.35). In a separate analysis of * ve disease, ulcerative colitis, colon cancer, and thyroid disease studies (68,76,78 – 80) , 34 of 952 IBS patients compared wih 12 do not appear to be signi* cantly di7 erent in patients with IBS of 1,798 controls were found to have serologic (anti-gliadin, symptoms compared with healthy controls. ere is emerg- anti-endomysial, transglutaminase antibodies) and small bowel ing evidence, however, to suggest that celiac sprue and lactose biopsy evidence of celiac sprue (3.6 vs. 0.7 %; OR = 4.34, 95 % intolerance may be more prevalent in patients with IBS symp- CI = 1.78 – 10.6). Two decision analytic models have evaluated toms than in controls. Small Intestinal Bacterial Overgrowth the cost e7 ectiveness of serologic screening for celiac sprue in (SIBO) continues to generate considerable interest as a possi- IBS patients and found that screening was cost e7 ective as long ble cause of IBS symptoms, but this association remains highly as the prevalence of celiac sprue exceeded 1% (81,82) . Based controversial. on the totality of evidence, the Task Force recommends routine Patients who ful* ll the symptom-based diagnostic criteria for serological screening for celiac sprue in patients with IBS-D IBS and who have no alarm features, require little formal test- and IBS-M. ing before con* dently arriving at the diagnosis of IBS. e like- Based on data from seven studies (83– 89) of 2,149 IBS lihood of uncovering important organic diseases by complete patients, the systematic review performed for this mono- blood count and serum chemistries is low and no greater in IBS graph reported the prevalence of lactose maldigestion by patients than in healthy controls (67) . Five studies have evalu- lactose breath testing to be 35 % (95% CI = 17 – 56 % ). In a sepa- ated the utility of checking thyroid function in 2,160 IBS patients rate analysis of data from three case-control studies including (68 – 72) . e prevalence of abnormal thyroid function tests was 425 individuals (251 with IBS), lactose intolerance was found 4.2 % (range = 0 – 5.5 % ), a value very similar to that expected in to be more prevalent in IBS patients than in controls (38 vs. the general population. Furthermore, in the infrequent cases 26% ; OR = 2.57, 95 % CI = 1.27 – 5.22). Unfortunately, these data, in which abnormal test results have been identi* ed, causality which suggest an association, do not prove causation between between the laboratory abnormality and the IBS symptoms has lactose maldigestion and IBS symptoms. It is worth noting not been established. For these reasons, the routine application that a substantial proportion of IBS patients have underlying of thyroid function tests in IBS patients without alarm features abnormalities in intestinal and/ or colonic motility and vis- is not recommended. ceral sensation. erefore, it is reasonable to speculate that the Similarly, stool for ova and parasite examination appears to clinical consequences of lactose maldigestion may be greater in o7 er little value in the evaluation of patients with IBS symp- IBS patients than in controls (85) . For these reasons, the Task toms and no alarm features. Two trials have evaluated the yield Force suggests that providers question patients about a link

between lactose ingestion and their IBS symptoms. A food diary patients and 416 healthy controls undergoing colorectal can- sometimes can help to identify such an association. If, aM er a cer screening, found no di7 erence in the prevalence of colo- careful history and review of a food diary, questions remain rectal cancer (IBS = 0 % , Controls = 0.2 % ) or IBD (IBS = 0.46 % , regarding the presence of lactose maldigestion, performance of C o n t r o l s = 0 % ) a m o n g g r o u p s . e prevalence of adenoma- a lactose hydrogen breath test can be considered (86) . Whether tous polyps (14 vs. 26 %, p = 0.0004) and diverticulosis (13 vs. other carbohydrates such as fructose and sucrose can cause or 21 % , p = 0.01) was lower in the IBS cohort than in the healthy exacerbate IBS symptoms remains poorly de* ned. controls. ese results may be confounded by the younger age A great deal of attention has been focused on the poten- (51 vs. 55 years, p < 0.0001) and greater proportion of women tial role of SIBO in the pathogenesis of IBS symptoms. Stud- (69 vs. 42 % , p < 0.0001) in the IBS cohort (95) . Based on these ies utilizing lactulose and glucose breath testing have yielded results, the Task Force recommends that any patient older than conK icting results. In the systematic review performed for this 50 years of age with IBS symptoms should undergo colonic monograph, which included three studies (87– 89) and 432 IBS imaging for the purpose of colorectal cancer screening. Patients patients, the prevalence of a positive lactulose breath test was younger than 50 years of age who do not have alarm features 65 % (95 % CI = 47 – 81 % ). e corresponding prevalence using need not undergo routine colonic imaging. Patients with IBS glucose breath testing (two studies with 208 patients) was 36 % symptoms, who also present with alarm features, such as ane- (95 % CI = 29 – 43 % ) (90,91) . e strikingly di7 erent results mia or weight loss, should undergo colonic imaging to exclude yielded by lactulose and glucose breath testing highlight the organic disease, however, the clinician may feel that such an absence of a widely available gold standard to diagnose SIBO investigation can be deferred in some young patients with mild (86) . In the only study performed to date that utilized lactu- symptoms, e.g., a young woman with IBS symptoms and mild lose breath testing, glucose breath testing, and jejunal aspira- anemia where heavy periods may be the explanation. tion for quantitative culture, no di7 erences in the likelihood In patients with symptoms consistent with IBS who also have of abnormal test results were identi* ed between IBS patients alarm features, the nature and severity of symptoms as well as and controls. Quantitative increases in small bowel bacteria the patient’ s expectations and concerns inK uence the choice of that did not meet the traditional diagnostic threshold for SIBO diagnostic testing. Most patients will undergo routine blood (>105 CFU / ml aspirate), however, were identi* ed in the IBS and stool tests depending on their predominant symptoms. cohort compared with controls (92) . Although there seems lit- With regard to colonic imaging, it is appealing to suggest that tle doubt that the intestinal and colonic microK ora play a role in patients with diarrhea-predominant symptoms undergo colon- the pathogenesis of a subset of IBS patients, the Task Force feels oscopy with inspection of the distal terminal ileum to exclude that, currently, there is insu+ cient evidence to recommend colon cancer and IBD respectively. e necessity of random breath testing for SIBO in patients with IBS. colonic mucosal biopsies in patients with diarrhea-predomi- Other diagnostic tests have been evaluated in IBS patients. nant symptoms remains controversial. As part of a recent pro- One study utilized erythrocyte sedimentation rate and spective trial, random colonic mucosal biopsies were obtained C-reactive protein to screen for evidence of systemic inK am- at the time of colonoscopy in patients with IBS-D and IBS-M. mation in 300 IBS patients (68) . ree patients (1% ) had an Histologic evidence of microscopic colitis and nonspeci* c abnormal test and were subsequently diagnosed with organic inK ammation was found in 2.3 and 1.4% of IBS patients respec- disease. Fecal serine protease has been associated with activa- tively; all of the patients with microscopic colitis had IBS-D tion of proteinase-activated receptors. Proteinase-activated (95) . Combining the preceding evidence with a retrospective receptors have been implicated in the development of visceral analysis that also identi* ed an association between IBS-D and hypersensitivity in IBS patients. In a recent study, fecal serine- microscopic colitis (96) led the Task Force to recommend that protease activity was assessed in 38 IBS patients, 15 patients if a patient with IBS-D has a colonoscopy, random colonic with ulcerative colitis, and 15 healthy controls (93) . Fecal serine mucosal biopsies to exclude microscopic colitis should be con- protease activity was threefold higher in IBS-D patients than in sidered. Clinical features suggestive of a secretory process, such controls or patients with nondiarrheal IBS; fecal serine protease as nocturnal diarrhea, large volume diarrhea that is una7 ected levels also were elevated in the ulcerative colitis patients. More by fasting, or a low fecal osmotic gap ( < 50 Osm / kg), go against work is eagerly awaited to understand the role of these tests in a diagnosis of IBS and strengthen the rational for obtaining patients with IBS symptoms. random colonic biopsies to exclude microscopic colitis (97) . If Colonic imaging in an IBS patient with no alarm features laboratory and / or stool testing suggest the presence of malab- is unlikely to reveal structural disease that might explain the sorption, upper endoscopy with small bowel biopsies to further patient’ s symptoms. In a recent systematic review, which evaluate for celiac sprue or testing for SIBO may be warranted. included three studies and a total of 636 IBS patients, colonic In patients with IBS-C who have alarm features, the major imaging with colonoscopy or barium enema with or without objective of colonic imaging is to exclude the presence of dis- K exible sigmoidoscopy, uncovered organic/ structural disease ease causing mechanical obstruction. Colonoscopy, virtual in 1.3% (95% CI = 0.06 – 2.3 % ) (54,69,94) . An interim ana- colonography, or barium enema can be used for this purpose. lysis from a prospective, controlled, multicenter U.S. trial that Once the diagnosis of IBS has been established, clinicians compared the yield of colonoscopy in 216 IBS-D or IBS-M should be reassured by the durability of the diagnosis. In two

S16 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

studies with follow-ups ranging from three to more than 20 accuracy of the diagnostic test or that food allergy is not a cause years, less than 1% of patients were given an alternative diag- of IBS symptoms. e gold standard method of addressing this nosis felt to be responsible for their gastrointestinal symptoms issue is the randomized controlled trial (RCT) and there are (98,99) . two such trials that gave patients with adverse food reactions a double-blind trial of the o7 ending agent. In one study (104) , Section 2.6 Diet and irritable bowel syndrome patients correctly identi* ed the o7 ending food in 10 of 12 cases Patients oC en believe certain foods exacerbate their IBS symp- (83% ), whereas the other study (105) essentially was negative. toms. B ere is, however, insu? cient evidence that food allergy An additional trial (108) evaluated the e+ cacy of food elimi- testing or exclusion diets are e? cacious in IBS and their routine nation based on IgG antibody levels to a panel of 29 di7 erent use outside of a clinical trial is not recommended (Grade 2C). food antigens. All IBS patients had food allergies according to this test and patients were randomized to an exclusion diet IBS patients oM en report that food intake exacerbates their eliminating foods identi* ed by allergy testing or a sham diet. symptoms. Surveys (100,101) suggest that 60 – 70 % of IBS suf- e study reported that the food elimination diet was e+ ca- ferers feel that their symptoms are related to food sensitivity cious, however, analysis of the intention-to-treat or all-evalu- and most exclude such o7 ending foods from their diet (101). able patient groups revealed that the impact was only modest: Clinicians also have explored whether dietary intervention can 18 of 65 (28 %) patients responded in the elimination diet group help alleviate symptoms in patients with IBS. Exclusion diets compared with 11 of 66 (17 %) in the sham diet group, (p = 0.19, involve having the patient complete a food diary and then not signi* cant). excluding those foods that seem to exacerbate symptoms. In Even if exclusion diets are shown to have modest e+ cacy addition, some researchers have excluded all dairy products, in ameliorating symptoms in patients with IBS, it will be dif- cereals, citrus fruits, potatoes, ca7 eine drinks, alcohol, addi- * cult to determine whether such bene* t resulted from a change tives, and preservatives (102) , although mechanistic data in intraluminal end products of bacterial metabolism, an supporting the omission of all these foods in the diet are meager. alteration in immunologically mediated food allergy or that A systematic review (103) of the literature on food allergy in IBS the change in diet acted as a prebiotic to vary the intestinal identi* ed eight studies (102,104 – 110) evaluating the response microbiota (112,113) . Currently there is little evidence to sup- of 540 IBS patients to exclusion diets. Most studies were uncon- port exclusion diets for the treatment of IBS, although a mod- trolled and the response rates to various exclusion diets ranged est e7 ect cannot be excluded from these data. More RCTs are from 12.5 to 67 % (Figure 1). Most of these papers claimed that needed aM er carefully excluding patients with celiac sprue and such responses demonstrate the e+ cacy of exclusion diets in lactose intolerance. IBS, a conclusion that is di+ cult to interpret given the high placebo response that can be seen in this condition; more objective Section 2.7 Effectiveness of dietary ﬁ ber, bulking agents, and evidence is required before this conclusion can be accepted. laxatives in the management of irritable bowel syndrome ere is no gold standard test to diagnose food allergy (103). Psyllium hydrophilic mucilloid (ispaghula husk) is moder- Skin prick tests and serum IgE or IgG levels to speci* c food ately e; ective and can be given a conditional recommendation antigens have been advocated, but all have uncertain sensitivity (Grade 2C). A single study reported improvement with calcium and speci* city. Studies that have evaluated adverse food reac- polycarbophil. Wheat bran or corn bran is no more e; ective than tions in IBS patients have found no correlations between the placebo in the relief of global symptoms of IBS and cannot be types of food causing symptoms and the results of food allergy recommended for routine use (Grade 2C). PEG laxative was tests (100,111) . is lack of correlation supports either a lack of shown to improve stool frequency —but not abdominal pain— in one small sequential study in adolescents with IBS-C (Grade 2C). 80 n =21 n =21 70 Most physicians recommend the use of dietary * ber or bulk- n =171 60 ing agents to regularize bowel function and to reduce pain in n =189 50 patients with IBS. e quality of the evidence supporting this n =9 recommendation, however, is poor. Our systematic review 40 n =65 (114) found 12 RCTs with global endpoints dealing with this 30 n=40 n=24 issue (115 – 126) . All but one were conducted outside of North 20 America, most were over 15 years old and, therefore, tended % responding to diet 10 to be small (in aggregate involving 591 subjects), had subopti- 0 mal experimental design, and utilized a variety of experimental 104 105 106 102 107 108 109 110 agents and conditions. IBS-C was di7 erentiated from IBS-D in Study (reference) only three studies; two of these recruited only IBS-C patients Figure 1 . Proportion of irritable bowel syndrome (IBS) patients responding and in the other, almost half of the participants had IBS-C. e to exclusion diets. other nine studies did not specify which IBS subtypes were

included. Most studies did not use criteria-based diagnosis, Notwithstanding the possibility that their actions may concealed allocation, adequate blinding, or other methods now reside elsewhere, these agents generally have been referred to recommended in modern study design. Nine trials were double- as “ antispasmodics ” and marketed as such. blind, two were single-blind, and one was unblinded. Few were Our systematic review (114) suggested that there is evidence at least eight weeks in duration and none followed patients for the e+ cacy of antispasmodics as a class in IBS, however, beyond the period of treatment. there are signi* cant variations in the availability of these Most studies we reviewed examined the e7 ect of wheat bran agents in di7 erent countries. For example, of the 22 separate or psyllium hydrophilic mucilloid (ispaghula husk). Taken as a studies identi* ed (117,120,121,132 – 150) , all but four (three trials group, treatment with wheat bran provided no global bene* t using hyoscine (117,120,132) , and one with dicyclomine (143)) in patients with IBS. Only one study (which did not include a involved drugs that are not available in the United States: oti- placebo-controlled group) demonstrated improvement in pain lonium (138,145,147,148) , cimetropium (133,135), pinaverium frequency, severity and stool frequency with wheat bran (116) , (144,146,149) , trimebutine (137,142) , alverine (140) , mebev- while the others showed no signi* cant improvement with treat- erine (121) , pirenzipine (139) , pri* nium (141) , propinox ment. Overall, the relative risk of IBS symptoms not improv- (150) , and a combination of trimebutine and rociverine (136) . ing with wheat bran was 1.02 (95 % CI = 0.82 – 1.27) (114) . In Furthermore, the preparation of hysocine used in reported trials contrast, global IBS symptoms were improved in four of the di7 ers from that currently available in the United States. Very six studies with psyllium hydrophilic mucilloid. e relative few of the trials are recent (only three since 2000 (138,140,150) ) risk of IBS symptoms not improving with psyllium hydrophilic and earlier trials vary considerably in terms of diagnostic mucilloid was 0.78 (95 % CI = 0.63 – 0.96) (114) . e NNT with criteria (only two (138,140) featured a standardized methodo- psyllium hydrophilic mucilloid was six (95 % CI = 3 – 50). logy, e.g., one of the Rome iterations), inclusion criteria, A single study of the e7 ectiveness of corn * ber in patients dosing schedule, duration of therapy, and study endpoints. with IBS showed no substantial bene* t over placebo (127) . IBS Many are of poor quality with only three studies (132,138,140) patients preferred calcium polycarbophil to placebo in another including more than 100 subjects and only one utilizing a vali- controlled trial (128) . dated outcome measure to de* ne improvement in IBS symp- Safety issues and adverse events were not addressed formally toms following therapy (140). e available data also do not in these studies of bulking agents. Clinical studies and expert permit ready identi* cation of a likely responder to this class opinion suggest that increased * ber intake may cause bloating, of drugs or a particular agent; for example, only six studies abdominal distention, and K atulence, especially if increased reported on subtype of IBS according to predominant stool suddenly (129,130) . Gradual titration is advised if these agents pattern (133,134,136,137,141,146) and the vast majority of are used. studies used a global endpoint rather than including results Laxatives have not been studied in randomized, placebo-con- of individual symptoms, such as pain, which might be expected trolled trials in adults with IBS and have mostly been studied to respond to this drug class. in patients with chronic constipation. A single small sequen- e 22 trials included 1,778 IBS patients and the relative risk tial study compared symptoms before and with PEG laxative of symptoms persisting with antispasmodics compared with treatment in adolescents with IBS-C (131). Stool frequency placebo was 0.68 (95 % CI = 0.57 – 0.81). e NNT to prevent IBS improved from an average of 2.07± 0.62 bowel movements symptoms persisting in one patient was * ve (95 % CI = 4 – 9). per week to 5.04 ± 1.51 bowel movements per week (p < 0.05), Of all drugs studied, the most data were available for otilo- but there was no e7 ect on pain intensity. nium (138,145,147,148) , trimebutine (136,137,142) , cimetropium (133 – 135) , hyoscine (117,120,132) , and pinaverium Section 2.8 Effectiveness of antispasmodic agents, including (144,146,149) . Trimebutine appeared to have no bene* t over peppermint oil, in the management of irritable bowel syndrome placebo in IBS, whereas the other four drugs all signi* cantly Certain antispasmodics (hyoscine, cimetropium, and pinaveri- reduced the risk of IBS patients remaining symptomatic with um) may provide short-term relief of abdominal pain / discomfort therapy. ere was considerable heterogeneity, however, among in IBS (Grade 2C). Evidence for long-term e? cacy is not avail- individual trials, with each study only including a small number able (Grade 2B). Evidence for safety and tolerability are limited of patients. e best evidence for e+ cacy appears to exist for (Grade 2C). Although peppermint oil appears superior to pla- the use of hyoscine, the e+ cacy of which was studied in more cebo in IBS, this conclusion is based on a small number of studies than 400 patients with no statistically signi* cant heterogene- (Grade 2B). ity detected, and four (95% CI = 2 – 25) patients needing to be treated to prevent one patient ’ s symptoms from persisting aM er Based on clinical observations as well as some experimen- completion of therapy. tal evidence, it has long been postulated that IBS symptoms Furthermore, the adverse event pro* le of these agents has not including pain, in particular, emanate from colonic smooth been de* ned adequately. muscle spasm. A variety of agents, some acting directly on irteen studies reported on the total number of adverse smooth muscle and others on cholinergic receptors, therefore, events in 1,379 patients (121,132 – 138,140,142,143,147,149) . have been developed and tested in IBS over the decades. e commonest adverse events were dry mouth, dizziness,

S18 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

and blurred vision, and there were no serious adverse events one study (157) , and four adverse events in each arm of the reported in either treatment arm in any of the trials. e rela- other trial (158) . tive risk of experiencing adverse events with antispasmodics compared with placebo was 1.62 (95 % CI = 1.05 – 2.50), with Section 2.10 Effectiveness of antibiotics in the management of statistically signi* cant heterogeneity detected among studies irritable bowel syndrome (I 2 = 3 8 % , p = 0.07). e NNH with antispasmodic drugs was A short-term course of a nonabsorbable antibiotic is more 18 (95 % CI = 7 – 217). e; ective than placebo for global improvement of IBS and for A variety of preparations containing various formulations of bloating (Grade IB). B ere are no data available to support the peppermint oil are available through conventional and comple- long-term safety and e; ectiveness of nonabsorbable antibiotics mentary routes and have been used for some time on a largely for the management of IBS symptoms. empiric basis for the treatment of IBS-like symptoms. Limited experimental data suggest the ability of peppermint oil to relax Rifaximin, a nonabsorbable antibiotic, has demonstrated e+ - smooth muscle, thus its inclusion in the same category as antispas- cacy in three RCTs evaluating 545 IBS patients (159 – 162) . modics. Only four studies (151 – 154) were identi* ed in a system- All of these RCTs were well designed, meeting all criteria for atic review (114) comparing peppermint oil with placebo in 392 appropriately designed RCTs (i.e., truly randomized studies patients; all but one (154) were short-term and only one reported with concealment of treatment allocation, implementation on the type of IBS patient according to stool pattern (153) . of masking, completeness of follow-up and intention-to-treat e relative risk of IBS symptoms persisting with peppermint analysis) and meeting most criteria of the Rome committee oil compared with placebo was 0.43 (95 % CI = 0.32 – 0.59), with for design of treatment trials of functional GI disorders (e.g., statistically signi* cant heterogeneity detected between studies patients met Rome criteria for IBS, no placebo run-in, base- (I 2 = 3 1 % , p = 0.23) (114) . e NNT with peppermint oil to pre- line observation of patients to assess IBS symptoms, and pri- vent one patient with IBS remaining symptomatic was 2.5 (95 % mary study outcome is improvement in global IBS symptoms) CI = 2 – 3) (114) . Only three studies reported adverse events data (163) . All of these RCTs demonstrated statistically signi* cant (152 – 154) , and these were few in number. improvement in symptoms with rifaximin, and rifaximin- treated patients were 8 – 23 % more likely to experience Section 2.9 Effectiveness of antidiarrheals in the management global improvement in IBS symptoms, bloating symptoms, of irritable bowel syndrome or both compared with placebo-treated patients. Rifaximin is B e antidiarrheal agent loperamide is not more e; ective than not FDA-approved for treatment of IBS, although it is FDA- placebo at reducing abdominal pain or global symptoms of IBS, approved for treatment of traveler ’ s diarrhea at the dose of but is an e; ective agent for treatment of diarrhea, improving 200 mg twice daily for three days. However, IBS trials utilized stool frequency and stool consistency (Grade 2C). RCTs with higher doses of rifaximin for longer periods: 400 mg three other antidiarrheal agents have not been performed. Safety and times daily for 10 days (162,164) , 400 mg twice daily for 10 tolerability data on loperamide are lacking. days (161) , and 550 mg twice daily for 14 days (159,160) . e largest RCT (n = 388 patients) only examined IBS-D patients, Patients with IBS who have diarrhea display faster colonic and in this trial, rifaximin-treated patients demonstrated signi- transit than healthy subjects (155,156) ; therefore, agents that * cant improvement in their diarrhea compared with placebo- slow colonic transit may be bene* cial in reducing symptoms. treated patients (164) . Based on these results, rifaximin is most Loperamide is the only antidiarrheal agent su+ ciently evaluated likely to be e+ cacious in IBS-D patients or IBS patients with a in RCTs for the treatment of diarrhea-predominant IBS. predominant symptom of bloating and the appropriate dosage ere have been two RCTs involving 42 patients that evalu- is approximately 1,100 –1,200 mg / day for 10 –14 days. ated the e7 ectiveness of loperamide in the treatment of IBS In the largest trial, 388 IBS-D patients were randomized to with diarrhea-predominant symptoms (157,158) . ere were rifaximin 550 mg twice daily for two weeks followed by placebo no statistically signi* cant e7 ects of loperamide on overall for another two weeks or, alternatively, they took placebo for symptoms compared with placebo (relative risk of IBS symp- four weeks. In this trial, patients had to experience adequate toms not improving = 0.44; 95 % CI = 0.14 – 1.42). Both trials relief of IBS symptoms in two of the three * nal weeks to be were double-blinded, but neither reported adequate methods de* ned as a responder. Rifaximin-treated patients were signi* - of randomization nor adequate concealment of allocation. cantly more likely to be responders (52.4 vs. 44.2 %, p = 0.03). e proportion of women in each trial was unclear. Both Notably, most of the improvement was not noted until aM er trials used a clinical diagnosis of IBS supplemented by nega- completion of the course of treatment. In a well-publicized RCT tive investigations to de* ne the condition. Both trials reported (162,164) , 87 IBS patients were randomized to rifaximin 400 mg that 100 % of the loperamide-treated group had improved stool three times daily for 10 days or placebo with a 10-week follow-up consistency compared with 20 – 45 % of controls (p = 0.006). period. In this study, severity of global IBS symptoms was based e pooled analysis of stool frequency suggested that the on a composite symptom score, and patients had to experience a relative risk of stool frequency not improving with loperamide 50% improvement in global IBS symptoms from baseline to one was 0.2. (95 % CI = 0.05 – 0.9). ere were no adverse events in week aM er completion of antibiotics to be de* ned as a responder

(37.2 vs. 15.9 % , p < 0.05). Based on assessment of the entire individual adverse events were noted between rifaximin-treated 10-week follow-up period, rifaximin-treated patients were and placebo-treated patients. signi* cantly more likely than placebo-treated patients to Overall, rifaximin consistently demonstrates improvement experience 50% improvement in bloating (49.2 vs. 22.6 %), in global IBS symptoms and bloating in well-designed trials. diarrhea (50.6 vs. 35.3 % ), abdominal pain (39.7 vs. 28.9 % ), e majority of patients in rifaximin trials had IBS-D. ere- and constipation (35.1 vs. 28.1 %) (164) , although a sepa- fore, rifaximin is most likely to be bene* cial in IBS-D patients rate mixed-model statistical analysis of the same data did not or IBS patients with bloating as their primary symptom. e demonstrate signi* cant improvement for the individual symp- most appropriate dose of rifaximin for IBS is unclear. Based toms of diarrhea, abdominal pain, or constipation (162) . Finally, on currently available data, 400 mg three times a day for 10 – 14 another RCT examined 103 patients with a primary complaint days is e+ cacious. IBS symptom relief appears to last for 10– 12 of bloating, 70 of whom met Rome II criteria for IBS. Among weeks, but symptoms may recur over three to nine months. IBS patients, rifaximin-treated patients were signi* cantly more Neomycin also demonstrated e+ cacy in a single, small RCT of likely than placebo-treated patients to state that “ symptoms IBS patients. Adverse events were not more common in anti- have improved since starting the drug ” aM er completion of biotic-treated than placebo-treated patients. However, given the study treatment (41 vs. 18 % ), but the percentage of patients oM en chronic and recurrent nature of IBS symptoms and the who continued to state that “ symptoms have improved since theoretical risks related to long-term treatment with any anti- starting the drug ” decreased 10 days aM er completion of study biotic, a recommendation regarding continuous or intermittent treatment in both groups (27 vs. 9 %). is * nding suggests that use of this agent in IBS must await further, long-term studies. relief of IBS symptoms may not be durable aM er completion of It must also be stressed that available data on rifaximin is based antibiotics, although other RCTs (162,164) have demonstrated on phase II studies; phase III studies have yet to be reported. that IBS symptom improvement lasts for at least 10 weeks. Furthermore, an Italian study (165) examined 61 consecutive Section 2.11 Effectiveness of probiotics in the management of patients with positive lactulose hydrogen breath tests, who were irritable bowel syndrome treated with rifaximin 400 mg three times daily for seven days. In single organism studies, lactobacilli do not appear e; ective; ese patients had repeat breath tests at three, six and nine bi dobacteria and certain combinations of probiotics demon- months. Breath tests gradually became positive in a substan- strate some e? cacy (Grade 2C). tial proportion of patients at three months (13 % ), six months (28% ), and nine months (46% ), and recurrences of positive Probiotics have been used on an empiric basis for many years in breath tests were associated with increases in abdominal pain, the treatment of IBS, although recent interest in the science of bloating, K atulence, and diarrhea, based on mean visual analog the intestinal K ora (microbiota) and probiotics, and our increas- scale scores. Based on one open-label retrospective study, IBS ing awareness of putative factors in IBS pathophysiology, such patients with recurrent symptoms respond to repeated courses as exposure to enteric pathogens, qualitative and quantitative of rifaximin (166) . changes in the enteric K ora, and subtle levels of colonic inK am- Among other antibiotics, a single RCT (167) of 111 patients mation or immune activation, have stimulated more extensive demonstrated that neomycin-treated patients were more likely studies of the use of these preparation in IBS. to experience 50 % improvement in global IBS symptoms com- Our systematic review (170) identi* ed 19 studies (171 – 189) pared with placebo-treated patients (43 vs. 23 %, p < 0.05). One including a total of 1,668 participants that were deemed eligible. trial of clarithromycin (168) did not assess e+ cacy of antibiot- e quality of studies was reasonable with nine (173,174,178, ics for IBS as a primary outcome. In this trial, a cohort of 40- to 180,181,183,187– 189) reporting an adequate method of 49-year-old individuals was screened for Helicobacter pylori . I f randomization and six (173,174,181,183,187,189) describ- an individual was positive for H. pylori , t h e n h e / s h e r e c e i v e d ing appropriate methods of concealment of allocation. All but clarithromycin, omeprazole and tinidazole, or placebos for one three (175,176,184) recruited patients according to Rome or week. As part of this study, patients also completed gastroin- Manning criteria. testinal symptom questionnaires at baseline, six months and Eleven trials (173,175 – 177,180 – 182,186 – 189) evaluated 936 two years, and IBS was de* ned as presence of three or more participants and reported IBS symptoms as a dichotomous out- Manning criteria. Among 274 participants with IBS at baseline, come. Taken as a group, probiotics had a statistically signi* cant 42% of the antibiotic group and 42% of the placebo group had e7 ect to reduce IBS symptoms (RR symptoms persisting in pro- IBS two years aM er their one-week course of treatment. Finally, biotic group = 0.71; 95 % CI = 0.57 – 0.87) with an NNT of four one trial (169) reported that metronidazole was more e7 ective (95 % CI = 3 – 12.5). ese data probably overestimate the e7 ects than placebo at improving global IBS symptoms, but this study of probiotics, however, as there was heterogeneity and evidence did not present data that were extractable. of funnel asymmetry, suggesting there may be publication bias No study reported on overall adverse events, but all stated that with an overrepresentation of small positive studies in the pub- antibiotics were well tolerated with no severe adverse events. lished literature. Furthermore, higher quality studies reported Two trials assessing rifaximin (159,160,162,164) provided data a more modest treatment e7 ect compared with lower quality on individual adverse events, and no signi* cant di7 erences in trials. ere was no di7 erence among the di7 erent types of

S20 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

probiotics used, with Lactobacillus (175 – 177,189) , Bi dobacte- the GI tract, the largest proportion of which is present in the rium (186,187) , Streptococcus (188) , and combinations of pro- enterochroma+ n cells. When released, serotonin can interact biotics (173,180 – 182) all showing a trend toward bene* t. with a number of di7 erent 5-HT receptors, the 5-HT1 , 5-HT3 ,

Fourteen publications (171–174,177–185,187) with 1,351 and 5-HT4 receptor subtypes playing major roles in GI motor/ participants reported IBS symptoms as a continuous variable. secretory function and visceral sensation (190) . 5-HT 3 recep- Probiotics had a statistically signi* cant e7 ect to improve IBS tor antagonists delay GI transit, reduce colonic tone, blunt the symptoms compared with placebo (standardized mean di7 er- gastrocolic reK ex and decrease visceral sensation (190 – 193) , ence = − 0.34; 95 % CI = − 0.60 to − 0.07). Four trials (172,177 – making members of this drug class potentially attractive as a 179) evaluated Lactobacillus in 200 patients and found no e7 ect treatment for patients with IBS-D. on IBS symptoms. Nine trials (171,173,174,180 – 185) evaluated Alosetron originally was approved for the treatment of combinations of probiotics in 772 patients with a signi* cant women with IBS-D in the United States in February 2000. I n e7 ect in improving IBS symptoms, whereas two trials (178,187) the systematic review commissioned to assist the develop- evaluated Bi dobacterium in 379 patients with a trend toward ment of this guideline (194) , eight placebo-controlled trials were improving IBS symptoms. found that evaluated alosetron use in 4,987 patients (195 – 202) . e main limitation of this review is that there were a variety Considering the primary therapeutic endpoint as “ adequate relief ” of species, strains, and doses of probiotics used and, therefore, of abdominal pain and discomfort or urgency, the relative risk it was di+ cult to reach a conclusion about the optimal probi- of IBS persisting with alosetron treatment was 0.79 (95 % CI = otic strategy to use in patients with IBS. Data from this review 0.69 – 0.90 with NNT = 8; 95 % CI = 5 – 17). In a comparative trial, are conK icting. e dichotomous data suggest that all probiotic alosetron also proved superior to the antispasmodic medication therapies show a trend for being e+ cacious in IBS. In contrast, mebeverine in female patients with nonconstipating IBS (203) . the continuous data suggest (1) Lactobacilli have no impact Studies have consistently shown bene* ts of alosetron for on symptoms; (2) probiotic combinations improve symptoms global and individual symptoms in female patients with in IBS patients; and (3) there was a trend for Bi dobacteria t o IBS-D. One randomised, double-blind, placebo-controlled improve IBS symptoms. e review was conservative as we study of women with IBS-D established that patient satisfaction decided a priori to include all doses of probiotics. One trial (187) was signi* cantly greater with alosetron compared with placebo of Bi dobacterium infantis 35624 was a dose-ranging study in for overall symptom relief (69 vs. 46 %, p < 0.001) as well as for which the authors found on post hoc evaluation that the prepara- the relief of urgency, speed of relief, time to return of normal tion methods had resulted in the higher dose of organisms being activities, relief of abdominal pain, and prevention of return of clumped together and inactivated. is dose was still included urgency (204). In another placebo-controlled study, alosetron in the analysis, but had it been excluded, the Bi dobacteria d a t a demonstrated sustained relief of abdominal pain and discom- would have reached statistical signi* cance. Almost all probiotic fort as well as urgency in IBS-D patients for up to 48 weeks combinations contained both Bi dobacteria a n d Lactobacilli with a safety pro* le comparable to that of placebo (200) . A ran- and the latter did not have an e7 ect in the continuous data meta- domized, placebo-controlled trial also found alosetron to be analysis. It is therefore possible that Bi dobacteria are the active e7 ective in men with IBS-D. (201) In this phase II dose-ranging agent in probiotic combinations. Alternatively, it is possible that study, 1 mg of alosetron taken twice daily resulted in adequate di7 erent species of probiotics are synergistic in promoting a relief of abdominal pain and discomfort in 53 % of men with therapeutic e7 ect on IBS. IBS-D compared with 40 % of men given placebo ( P < 0.001). e three trials of cilansetron evaluating a total of 2,229 IBS

Section 2.12 Effectiveness of the 5HT3 receptor antagonists in patients (205 – 207) were also identi* ed in the systematic review the management of irritable bowel syndrome (194) . Cilansetron 3 mg twice daily was statistically signi* cantly

B e 5-HT3 receptor antagonist alosetron is more e; ective than superior to placebo (RR of symptoms persisting = 0.75; 95 % placebo at relieving global IBS symptoms in male (Grade 2B) CI = 0.69 – 0.82), with an NNT of six (95 % CI = 5 – 8). Cilanset- and female (Grade 2A) IBS patients with diarrhea. Potentially ron is not available in any country and is unlikely to be mar- serious side e; ects including constipation and colon ischemia keted in view of the adverse event data seen with alosetron. occur more commonly in patients treated with alosetron Unfortunately, alosetron has been linked to the development compared with placebo (Grade 2A). B e bene ts and harms of severe constipation and colon ischemia in a small percentage balance for alosetron is most favorable in women who have not of patients. In the systematic review conducted for this guide- responded to conventional therapies (Grade 1B). B e quality of line, seven studies (196 – 202) were identi* ed that presented evidence for e? cacy of 5-HT3 antagonists in IBS is high. overall adverse event data in 4,609 patients. Patients randomized to receive alosetron were statistically signi* cantly more A number of drugs targeting serotonin receptors have demon- likely to report an adverse event than were those randomized strated e+ cacy for improving global symptoms in IBS patients. to placebo (RR of adverse event = 1.18; 95 % CI = 1.08 – 1.29). Serotonin has been found to play key roles in the physiology e NNH with alosetron was 10 (95 % CI = 7 – 16). Constipa- of the GI tract in health and disease. Approximately 95% of tion was the most commonly reported adverse event with alos- the body ’ s serotonin (5-hydroxytryptamine; 5-HT) is found in etron, and its development appeared to be dose-dependent (1 mg

twice daily = 29 % ; 0.5 mg twice daily = 11 % ). e risk of ischemic to 20 weeks. Almost all RCTs, however, assessed tegaserod at a colitis was independent of dose. A systematic review of the dose of 6 mg twice daily for 12 weeks. All trials recruited only clinical and postmarketing surveillance data from IBS patients and women or mostly women. erefore, tegaserod was approved the general population con* rmed a greater incidence of severe only for the treatment of IBS-C in women, and the recom- complicated constipation and ischemic colitis in patients taking mendations in this guideline should only be applied to female alosetron (208) , however, the incidence of these events was low, IBS patients. Review of material submitted to the FDA reveals with a rate of 1.1 cases of ischemic colitis and 0.66 cases of compli- that these RCTs (210–219) met criteria for an appropriately cated constipation per 1,000 patients-years of alosetron use. designed RCT (i.e., truly randomized studies with conceal- Because of these rare but serious side e7 ects, alosetron was ment of treatment allocation, implementation of masking, voluntarily withdrawn by GlaxoSmithKline from the U.S. completeness of follow-up and intention-to-treat analysis) and marketplace in November 2000. In June 2002, the FDA met almost all criteria of the Rome committee for design of approved the re-release of alosetron for use in female patients treatment trials of functional GI disorders (e.g., patients met with chronic, severe IBS-D who had failed to respond to con- Rome criteria for IBS, no placebo run-in, treatment duration ventional therapy. Current use of alosetron is regulated by a pre- of eight to 12 weeks, baseline observation of patients to assess scribing program set forth by the FDA and administered by the IBS symptoms, primary study outcome is improvement in manufacturer (Prometheus Laboratories, San Diego, CA). e global IBS symptoms, sample size calculation is provided and recommended starting dose of alosetron is 0.5 mg twice daily. adequate sample size is enrolled, etc.). is lower dose of alosetron is supported by results from a ran- In our meta-analysis, tegaserod 6 mg twice daily was signi* - domized, double-blind, placebo-controlled trial involving 705 cantly more e7 ective than placebo for satisfactory improvement women with severe IBS-D who were randomized to alosetron of global IBS symptoms (relative risk of IBS not improving = 0.85; or placebo (202) . Alosetron proved more e7 ective for global IBS 95% CI = 0.80 to 0.90), however, there was signi* cant heteroge- symptoms at doses of 0.5 mg / day (50.8 % ), 1 mg / day (48 % ), and neity in results, suggesting that treatment populations (IBS-C 1 mg twice daily (42.9 % ) than placebo (30.7 % , p < 0.02 for all vs. IBS-M), study endpoints or study design were too di7 erent comparisons). Constipation was dose dependent, reported by 9, to justify combining the results. In individual RCTs, tegaserod- 16, and 19 % of patients randomized to 0.5 mg / day, 1 mg / day, or treated patients were 5–19% more likely than placebo-treated 1 mg twice daily, respectively. One case of ischemic colitis was patients to achieve satisfactory relief of global IBS symptoms, reported in the 0.5 mg / day group and one case of fecal impac- and tegaserod-treated patients were signi* cantly more likely tion was reported in the 1 mg twice daily group. If aM er four to experience improvement in abdominal discomfort, satis- weeks, the drug is well tolerated but the patient’ s IBS-D symp- faction with bowel habits, and bloating in most RCTs. Also, toms are not adequately controlled, the dose can be increased to tegaserod is the only 5HT4 agonist that has been evaluated 1 mg twice daily. Alosetron should be discontinued if a patient and fully reported in an IBS-M population. In a well-designed develops symptoms or signs suggestive of severe constipation RCT (218), of IBS-M and IBS-C patients, those treated with or ischemic colitis or if there is no clinical response to the 1 mg tegaserod were 15% more likely to demonstrate improvement in twice daily dose aM er four weeks. Alosetron is contraindicated global IBS symptoms compared with placebo-treated patients. in patients with signi* cant liver disease and has been desig- Neither renzapride nor cisapride are marketed for use in nated as a pregnancy category B drug. North America, although cisapride may be obtained through a complicated compassionate use drug protocol. Indeed these

Section 2.13 Effectiveness of 5HT 4 (serotonin) receptor agonists drugs are not available in most developed countries; we still in the management of irritable bowel syndrome conducted a systematic review of their e+ cacy to establish

B e 5-HT 4 receptor agonist tegaserod is more e; ective than pla- whether other drugs in this class had a role in IBS. ere were cebo at relieving global IBS symptoms in female IBS-C (Grade four RCTs randomizing 317 IBS-C patients to either cisapride 1A) and IBS-M patients (Grade 1B). B e most common side or placebo (220 – 223) . All studies used a dose of 5 mg three e; ect of tegaserod is diarrhea (Grade 1A). A small number times daily for 12 weeks, titrating up to 10 mg three times (0.11 % ) of cardiovascular events (myocardial infarction, unstable daily at four weeks if there was no response to therapy and angina, or stroke) were reported among patients who had received there was no signi* cant bene* t compared with placebo (RR tegaserod in clinical trials. of symptoms persisting = 0.91; 95 % CI = 0.58 – 1.43, I 2 = 7 0 % ) . We identi* ed three trials of renzapride in 726 Rome II IBS

Tegaserod was the only 5-HT4 agonist that had been approved patients (224 – 226) . e trials used 1 –4 mg of renzapride once for the treatment of IBS, but it was withdrawn from the market daily for a duration of up to 12 weeks and there was no signi- in March 2007 because of a low rate (0.11 %) of cardiovascular * cant bene* t compared to placebo (RR of symptoms per- events in tegaserod-treated patients. sisting = 0.99; 95 % CI = 0.79 – 1.23, I 2= 48 % ). In April 2008, A systematic review (209) identi* ed multiple RCTs evaluat- Alizyme Pharmaceuticals (Cambridge, UK), manufacturer of ing the e+ cacy of tegaserod in over 9,000 patients with IBS-C or renzapride, announced that they had discontinued develop- IBS-M (210 – 219) . e dose of tegaserod used ranged from ment of renzapride for IBS-C because of disappointing Phase 1 mg to 12 mg twice daily, and duration of therapy from two III trial results that showed only limited clinical improvement

S22 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

compared with the placebo for the primary study endpoint Lubiprostone is the only selective C-2 chloride channel (ClC-2) (~ 0.45 months of IBS symptom relief vs. 0.55 –0.60 months of activator available worldwide. e drug works from the symptom relief in a three-month trial) and that the e+ cacy was luminal surface to promote chloride secretion into the intestine. not su+ cient to justify further development. Chloride channels are proteins inserted into cell membranes to Total numbers of patients experiencing adverse events permit chloride ions to cross the otherwise impermeable cell were reported in only three tegaserod trials, containing 2,827 membrane (228,229) . Because intracellular chloride concentra- patients (212,215,218) . ere was no statistically signi* cantly tion is higher than that in the lumen due to the e7 ect of a baso- increased risk of overall adverse events detected with tegaserod lateral Na-K-2Cl pump, activation of an apical chloride channel (RR = 1.07; 95 % CI = 0.99 – 1.15, I 2 = 0 % ) and 48% of the tegas- in the intestinal epithelium results in chloride secretion (230). erod arm and 45% of the placebo arm reported at least one In the small intestine, sodium enters the lumen through the adverse event. Diarrhea occurred signi* cantly more oM en in the paracellular pathway in response to the negative charge of the tegaserod-treated patients than in the placebo-treated patients secreted chloride ion and water follows passively. us the net with most individual RCTs reporting diarrhea in approximately e7 ect of activation of a chloride channel is secretion of salt 10% of the former group and 5% of the latter group. Approxi- water into the lumen of the intestine. mately 1–2% of tegaserod-treated patients discontinued Activation of the cystic * brosis transmembrane regulator tegaserod because of severe diarrhea. (CFTR), a high-capacity chloride channel inserted into the api- Tegaserod was withdrawn from the market in March 2007 cal membrane of enterocytes, is responsible for many secretory aM er data from the entire clinical trial database of 29 RCTs diarrheas, such as cholera (231) . e C-2 chloride channel is were presented to the FDA (227) . ere were 11,614 patients a lower capacity chloride channel that is thought to be more treated with tegaserod and 7,031 treated with placebo; the involved with the physiologic regulation of paracellular perme- average age of study subjects was 43 years, and 88 % were ability and intracellular volume (229) . No disease states have women. Cardiovascular events occurred in 0.11 % of tegaserod- yet been associated with activation of this channel in humans. treated patients vs. 0.01 % of placebo-treated patients. irteen Although lubiprostone is derived from prostaglandin, it does tegaserod-treated patients had myocardial infarction ( n = 4 ) , not work exclusively via prostaglandin receptors (232,233) . It unstable angina ( n = 6), or stroke (n = 3 ) w h e r e a s o n e p l a c e b o - is poorly absorbed into the systemic circulation and appears to treated patient had a transient ischemic attack. Currently, work topically in the small intestine. Lubiprostone is thought tegaserod is not available under any treatment investigational also to stimulate colonic motility by increasing intraluminal drug protocol, but it is available through the FDA under an volume or by some additional as yet unknown mechanisms. emergency investigational drug protocol. Lubiprostone has shown e+ cacy in RCTs in patients with Serious cardiac arrhythmias including ventricular tachycardia, chronic idiopathic constipation at a dose of 24 g twice daily ventricular * brillation, torsades de pointes, and QT prolongation (234 – 236) . Subgroup analysis of patients entered into those have been reported in patients taking cisapride, especially those trials who had severe abdominal discomfort suggested some using medications that increase cisapride blood levels by inhibit- improvement in abdominal pain and prompted further study ing the cytochrome P450 3A4 enzymes that metabolize cisapride, of lubiprostone in patients with IBS-C (237) . e.g., clarithromycin, erythromycin, troleandomycin, nefazodone, Dose-ranging studies showed e7 ectiveness in reducing K uconazole, itraconazole, ketoconazole, indinavir, and ritonavir. abdominal discomfort from IBS-C in doses ranging from As a result of these adverse events reports, cisapride was with- eight to 24 g twice daily (238) . Side e7 ects were greater at the drawn from the US market in July 2000, but is still available higher doses so the 8 g twice daily dose was selected for fur- under a compassionate-use protocol from the FDA. ther testing in large RCTs lasting 12 weeks (239) . ese studies Overall, tegaserod consistently demonstrates e+ cacy for glo- used a complicated end point designed to minimize placebo bal IBS symptom improvement and individual IBS symptom response rates. To be counted as an overall responder, subjects improvement in women with IBS-C based on well-designed were asked to rate their responses each week on a seven-point trials. However, cisapride is only available under an emergency balanced Likert scale ranging from “ signi* cantly worse ” to investigational drug protocol through the FDA. Cisapride has “ signi* cantly relieved” . Only those responding with “ signi* - not demonstrated improvement compared with placebo. e cantly relieved ” for at least two of four weeks or “ moderately development of renzapride was discontinued because of disap- relieved” for four of four weeks and who did not increase their pointing Phase III trial results about the magnitude of improve- use of relief medications and who did not have any weekly ment with this treatment. erefore, e7 ective 5-HT4 a g o n i s t s ratings of “ moderately worse” or “ signi* cantly worse” were for the management of IBS are not readily available. counted as monthly responders. Only those who were monthly responders for two of three months were counted as overall Section 2.14 Effectiveness of the selective C-2 chloride channel responders. activators in the management of irritable bowel syndrome Placebo response for the pooled Phase III studies was only Lubiprostone in a dose of 8 g twice daily is more e; ective than 10 % . Subjects treated with lubiprostone 8 g twice daily had placebo in relieving global IBS symptoms in women with IBS-C a response rate of 18% ( p < 0.001) (239) . As most participants (Grade 1B). in these studies were women, FDA approval was granted only

for women with IBS-C. Factor analysis was applied to under- with symptoms suggesting obstruction should be evaluated standing whether improvement in one symptom drove the before starting treatment (248) . overall response rate to lubiprostone. Improvement in no individual symptom (e.g., constipation severity) was responsible Section 2.15 The effectiveness of antidepressants in the for the overall response, suggesting that improvement in symp- management of irritable bowel syndrome toms across the board was associated with global response (240) . TCAs and SSRIs are more e; ective than placebo at relieving Quality of life also was investigated in these subjects. Lubipros- global IBS symptoms, and appear to reduce abdominal pain. tone treatment was associated with improvement in domains of B ere are limited data on the safety and tolerability of these health worry (p < 0.025) and body image ( p < 0.015) (241) . agents in patients with IBS (Grade 1B). Two continuation studies were done as part of the Phase III investigations in IBS-C. In the * rst, those who had received Patients with IBS that fails to respond to peripherally acting lubiprostone in the initial double-blinded 12-week study agents oM en are considered for treatment with antidepressants, improved their response rate from 15 to 37 % during the exten- especially if abdominal pain is a prominent symptom; the data sion study (242) . Patients initially receiving placebo increased on e+ cacy of antidepressants in IBS, however, has been ques- their response rate from 8 to 31 %. In the second continuation tioned (249). In the largest, high-quality RCT, desipramine study, subjects initially treated with lubiprostone either were was tested against placebo in 216 patients with moderate-to- continued on therapy or therapy was withdrawn and subjects severe IBS (250) ; 90% of patients included had IBS according were followed for an additional four weeks (243) . ere was no to a physician diagnosis and 80 % ful* lled the Rome I criteria di7 erence in response rates between lubiprostone- and placebo- for IBS. Desipramine was begun at a starting dose of 50 mg, treated subjects at the end of this extension study. is study increased to 150 mg daily (an antidepressant dose) over a three- shows that there is no rebound of symptoms and there may week interval, and then continued for a total of 12 weeks. By be positive “carry over” effect after treatment. 12 weeks, 60 % of patients responded to desipramine compared No electrocardiographic changes were found during initial with 47 % of those on placebo; this di7 erence failed to reach dose-ranging studies and with acute doses of up to 144 g signi* cance in the intention-to-treat analysis. e de* nition of (244,245) . Pooled analysis of studies using 24 g twice daily a responder was based on a measurement of patient satisfac- showed no change in serum electrolytes (246) . Analysis of all tion with the treatment rather than on a symptom evaluation; phase II and III studies using 24 g twice daily dose in patients when individually analyzed, global well being and average daily with chronic constipation for up to 48 weeks showed that the abdominal pain scores were not signi* cantly di7 erent between most common side e7 ects were nausea (31% ), diarrhea (13% ), the desipramine and placebo groups. Overall, 28 % of subjects and headache (13% ) (247) . Abdominal pain, abdominal disten- treated with desipramine dropped out of the trial, most oM en tion, and K atulence also were seen in >4 % of subjects treated because of side e7 ects (250) . Additional analyses from this trial for chronic constipation. Nausea was less common in men suggest that a TCA, speci* cally desipramine, may be particu- (8% ) and in the elderly (19% ). Side e7 ects were less frequent in larly useful in patients with IBS-D, likely because of the anti- Phase III studies of patients with IBS-C given 8 mg twice daily cholinergic e7 ect that characterizes this class of agents; the and included nausea (8 % ), diarrhea (6 % ), and abdominal pain other trials evaluated did not prespecify IBS subgroup analyses (5 % ) (239) . Postmarketing reports include allergic reactions and (251) . e presence of comorbid depression did not predict troubling dyspnea occurring within an hour of the * rst dose response to therapy (251) . and generally resolving within three hours; this may recur with Physicians oM en prefer to use a SSRI rather than a TCA repeat dosing. Dyspnea was noted in 2.5 % of chronic constipa- because of the lower side-e7 ect pro* le. e use of SSRIs in IBS tion patients treated with 24 mg twice daily and in 0.4 % of IBS-C is more controversial, however, because convincing evidence patients treated with 8 g twice daily in the clinical trials (248) . of e+ cacy from individual trials has been lacking (249) . A sys- Lubiprostone has been given a pregnancy category C rating. tematic review on antidepressants in functional gastrointesti- Animal studies showed no teratogenicity with even large doses nal disorders concluded that antidepressants were e+ cacious and three of four animal species had no excess fetal loss when in IBS, but data on SSRIs were not included (252) . dosed during gestation (248) . Guinea pigs had an excess rate of Antidepressants could theoretically provide a bene* t in IBS fetal resorption when dosed during pregnancy and this led to by both central and peripheral mechanisms (253,254) . SSRIs the recommendation that women who can have children have a have e7 ects on the gastrointestinal tract that di7 er from those pregnancy test before starting therapy and practice contracep- of TCAs. For example, K uoxetine has been shown to decrease tion while taking lubiprostone. Six women became pregnant orocecal and whole gut transit times in both constipation- during clinical trials with lubiprostone; four delivered healthy predominant IBS and controls (255) . In contrast, the TCA children, one was lost to follow-up, and one pregnancy was ter- imipramine has been shown to prolong orocecal and whole minated electively (248) . ere is no information about use of gut transit times in controls and in patients with IBS-D (255). lubiprostone in nursing mothers. Venlafaxine (an inhibitor of serotonin and norepinephrine e FDA lists mechanical gastrointestinal obstruction as a reuptake) has been shown to reduce colonic compliance and contraindication to use of lubiprostone and advises that patients relax the colon in healthy volunteers (256) , whereas K uoxetine

S24 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

and citalopram did not change colonic compliance or visceral opinion supports the e+ cacy of psychological therapies hypersensitivity (257) . Antidepressants are oM en prescribed although their bene* ts in IBS remain poorly quanti* ed (270) . when abdominal pain is a prominent feature, and it has been A systematic review evaluating psychological therapies in IBS presumed any bene* t is from a central antinociceptive e7 ect. identi* ed 17 studies, 10 of which had extractable data; 9 of the 10 We conducted a systematic review of the literature (258) studies, however, emanated from a single center (271) . Two and identi* ed 13 RCTs that evaluated either TCAs or SSRIs in other reviews on the subject concluded that the quality of 789 patients (120,250,257,259 – 268) . Global symptoms were the available evidence was low and that these approaches were signi* cantly more likely to improve with an antidepressant, e+ cacious for individual IBS symptoms, but a meta-analysis regardless of type (RR of IBS symptoms not improving = 0.66, was not undertaken (1,2) . A Cochrane Collaboration system- 95 % CI = 0.57 – 0.78), and there was only marginal statistically atic review of the e+ cacy of hypnotherapy identi* ed four trials signi* cant heterogeneity, so pooling of these data appears rea- but the data were not combined (272) . sonable. TCAs were superior to placebo in pooled data from e Task Force (273) identi* ed 20 RCTs, making 21 di7 er- nine trials involving 575 IBS patients, with a NNT of 4 (95 % ent comparisons (250,274 – 292) , including 1,278 IBS patients. CI = 3 – 8; RR of IBS not improving = 0.68, 95 % CI = 0.56 – 0.83) ere was a bene* t of psychological therapy over usual care (258) . Overall, there were * ve trials evaluating SSRI therapy (RR of IBS not improving = 0.67, 95 % CI = 0.57 – 0.79; NNT = 4; in 230 IBS patients and data suggested that this class of drugs 95 % CI = 3 – 5), however, there was signi* cant heterogeneity so also is e+ cacious in IBS with a NNT of 3.5 (95 % CI = 2 – 14; pooling these studies needs to be interpreted very cautiously. RR of IBS not improving = 0.62, 95 % CI = 0.45 – 0.87) (258) . Nine of these studies came from the same US research group ese drugs also have the advantage of being potentially (274,275,277,282,285 – 287,289,292) and overall study quality better tolerated than TCAs (249) , and because the SSRIs have was judged to be low. Relaxation therapy alone (282 – 285,291) a prokinetic e7 ect (255) , this drug class may work better in had no signi* cant bene* t. Cognitive behavioral therapy IBS-C than in those with IBS-D, although the studies per- (250,274 – 278,291) , dynamic psychotherapy (280,281) , and formed did not actually evaluate this issue to con* rm this multicomponent psychological therapy (279,286,287) were all clinical impression . Nevertheless, the data indicate that both similarly e+ cacious when pooled separately. Two additional TCAs and SSRIs appear able to improve global IBS symptoms. studies evaluated the global e+ cacy of hypnotherapy in IBS and It was not possible to show a pooled bene* t for individual overall reported a signi* cant bene* t with no signi* cant hetero- symptoms because few trials reported them in detail, and not geneity (RR of IBS not improving = 0.48, 95 % CI = 0.26 – 0.87; all trial participants had all of the key symptoms at study entry. NNT = 2) (289,290) . Other clinical trial evidence that could not In two of the trials, abdominal pain was the primary endpoint be included in the pooled analyses because global e+ cacy was and a bene* t was observed (259,265) . not assessed also favored hypnotherapy (293) . Data on safety of antidepressants was reported in six IBS Overall, the data suggest that regardless of the type of psycho- trials involving 301 patients (257,259 – 262,267). e results logical therapy applied, it was superior to usual care in terms of suggested an increased risk in overall adverse events in those global symptom improvement (aside from relaxation therapy). taking antidepressants but this did not reach statistical sig- None of the trials reported any adverse events with psycho- ni* cance (RR adverse event with antidepressants = 1.63, 95 % logical therapy although, theoretically, this absence may reK ect CI = 0.94 – 2.80). Given the limited data available on the safety under-reporting bias. Adequate blinding is virtually impossi- and tolerability of antidepressants in IBS, we evaluated other ble with psychological therapy, and this is a major methodo- diseases in which these drugs are used and found that this has logical problem with all studies in this area. Whether there is a been assessed in a systematic review of neuropathic pain (269) . speci* c biological mechanism by which psychological therapy e NNH for major adverse e7 ects, de* ned as an event leading may work in IBS has not been shown. Any bene* t may derive to withdrawal, was 28 (95 % CI = 17.6 – 68.9) for amitriptyline from an empathic attitude of the health provider, reduction of and 16.2 (95 % CI = 8 – 436) for venlafaxine (269) . life stresses because of attention from or discussion with the Head-to-head trials of a low-dose TCAs with an SSRI in health provider, transference of enthusiasm by the provider IBS are also not available, and the long-term outcome of such about the potential e7 ectiveness of therapy, and the quality and therapies is relatively poorly documented, representing major quantity of contact time with the provider. gaps in the literature that remain to be * lled. Section 2.17 Effectiveness of herbal therapies and acupuncture Section 2.16 The effectiveness of psychological therapies in in the management of irritable bowel syndrome the management of irritable bowel syndrome A systematic review of herbal therapy in IBS has been pub- Cognitive behavioral therapy, dynamic psychotherapy, and lished (294) . e Task Force reviewed the available RCTs when hypnotherapy but not relaxation therapy are more e; ective than evaluating the evidence for bene* t in this report (295 – 298) . usual care in relieving global symptoms of IBS (Grade 1C). ese trials mostly tested unique Chinese herbal mixtures, and they appeared to show a bene* t (296 – 298) . It is not possible to Psychological therapies include cognitive behavioral therapy, combine these studies into a meaningful meta-analysis, how- relaxation therapy, hypnosis, and psychotherapy. Expert ever, and overall, any bene* t of Chinese herbal therapy in IBS

continues to be potentially confounded by the variable compo- ion transport and K uid secretion. Crofelemer is an extract from nents used and their purity. Publication bias may also explain the Croton lechleri tree in South America that inhibits CFTR the lack of more negative trials. Furthermore, concerns about and also has anti-inK ammatory and analgesic properties, mak- toxicity, especially liver failure, and also other serious side ing it an attractive agent for the treatment of IBS-D. A 12-week e7 ects remain regarding use of any Chinese herbal mixture. randomized, double-blind, placebo-controlled, phase 2 dose- A Cochrane systematic review (299) of acupuncture identi* ed ranging study involving 246 adults with IBS-D demonstrated six poor quality trials that compared acupuncture with sham safety and signi* cant improvement in pain as well as trends to- acupuncture. e outcomes assessed were heterogeneous and ward improvement in urgency, stool frequency, and adequate the review reported that the e+ cacy of this intervention is relief of overall symptoms (301) . A phase 2b trial assessing uncertain. Further work is needed before any recommenda- crofelemer ’ s safety and e+ cacy in adult women with IBS-D is tions on acupuncture or herbal therapy can be made. underway. Guanylate cyclase-C is another intestinal transmembrane Section 2.18 Emerging therapies for the irritable bowel receptor responsible for chloride, bicarbonate, and K uid secre- syndrome tion into the intestinal lumen via production of cyclic guanos- Our expanding knowledge of the pathogenesis of IBS has led ine monophosphate and consequent activation of CFTR (302). to the identi* cation of a wide variety of novel agents, now in Linaclotide is a guanylate cyclase C agonist being developed as various stages of development. is discussion will focus on a treatment for IBS-C and chronic constipation. Linaclotide has drugs that have progressed beyond the proof of concept stage been reported to increase colonic transit, improve stool con- of development and will consider agents with predominantly sistency, stool frequency, and ease of stool passage in a rand- peripheral e7 ects, as well as those with both peripheral and omized, double-blind, placebo-controlled trial of 36 women central e7 ects. with IBS-C (303). Preliminary data from a recently completed To prepare for this discussion, it is helpful to understand phase 2b study, which randomized 420 patients with IBS- the steps involved in the FDAs’ drug development process C to placebo, 75, 150, 300, and 600 g of linaclotide daily for (300) . Preclinical development consists of animal studies, 12 weeks demonstrated bene* ts for stool frequency as well as which address questions involving mechanism of action and global and other individual IBS-C symptoms. Phase III studies drug toxicity. AM er submission of an Investigational New Drug are expected to begin in the near future (304) . Application to the FDA, clinical development can commence Calcium channel blockers. Arverapamil (AGI-003) is the and consists of phases 1, 2, and 3 trials. Phase 1 trials typically r-isomer of the calcium channel blocker verapamil and are conducted in small numbers of healthy volunteers and eval- is reported to selectively inhibit intestinal calcium channels. uate drug toxicity and pharmacokinetics, i.e., the absorption, Averapamil recently demonstrated e+ cacy compared with distribution, metabolism, and excretion of the drug being stud- placebo in a study of 129 adults with IBS-D (305) . Phase 3 trials ied. Phase 2a trials evaluate the drug ’ s pharmacodynamics, i.e., in adults with IBS-D are expected to begin in 2008. biochemical and physiologic e7 ects, mechanisms of action, and Opioid receptor ligands. -Opioid agonists and -opioid anta- the relationship between drug concentration and e7 ects of the gonists are capable of modulating visceral sensation through drug in healthy volunteers or patients. Phase 2b trials are rand- e7 ects on peripheral visceral a7 erent nerves. Compared with omized, placebo-controlled trials that involve larger numbers of placebo, asimadoline, a peripheral -opioid agonist, decreased patients and typically evaluate the e+ cacy and safety of a range pain perception from colonic distention in female IBS patients of drug doses. Ideally, phase 2b study results inform the selec- (306) . In a phase 2b dose-ranging study in 596 patients, asima- tion of the drug dose o7 ering the best combination of e+ cacy doline (0.15 mg, 0.5 mg, or 1.0 mg twice daily) was shown to and safety for phase 3 trials. Phase 3 trials are large, randomized, improve pain, urgency, stool frequency, and bloating in patients controlled registration trials that assess the e+ cacy and safety of with IBS-D and, to a lesser extent, patients with IBS-M. No the investigational drug vs. placebo in patients with the disease bene* ts were observed in patients with IBS-C (307) . of interest. Typically, positive results from two phase 3 trials e peripheral -opioid antagonist, methylnaltrexone, are necessary for drug approval. e speci* cs of the phase 3 has proven e7 ective for inducing “ laxation ” (passage of a trials regarding patient population, study methodology, bowel movement) in terminally ill patients taking opioids and main study results determine the eventual product label and recently has been FDA approved for the treatment of contents should a drug gain FDA approval. opioid-induced constipation (308) . is drug is typically administered every other day to daily as a subcutaneous Agents with predominantly peripheral e; e c t s injection. e role of this drug in the treatment of IBS-C remains to be established. Drugs which a; ect chloride secretion. Multiple types of chloride Motilin receptor ligands. Mitemcinal is a motilin receptor channels are present in nearly all cells, and are responsible agonist that has demonstrated prokinetic properties in the for many cellular functions including modulation of cellular lower gastrointestinal tracts of several animal models (309,310) . volume and K uid transport. CFTR, which is located on the A phase 2 clinical trial assessing mitemcinal’ s safety and apical membrane of intestinal cells, plays a major role in chloride e+ cacy in IBS is expected in the near future.

S26 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

Agents with peripheral and central e; e c t s Several novel 5-HT receptor agents in early stages of clinical development may have future applications in the

Emerging serotonergic agents. Several serotonergic agents are in treatment of IBS-C. TD-5108 is a highly selective full 5-HT4 development for IBS. A recent randomized, placebo-control- agonist associated with increased stool frequency and decreased led study found the 5-HT4 receptor agonist prucalopride to stool consistency in preclinical trials. A recent four-week be more e7 ective at increasing stool frequency than placebo in multi-center, randomized, double-blinded, placebo-controlled patients with chronic constipation (311) . Studies evaluating the phase 2 trial involving 400 patients with chronic constipation e+ cacy of prucalopride in patients with IBS-C are anticipated. demonstrated TD-5108’ s safety, tolerability, and superiority

Ramosetron, a 5-HT3 antagonist, currently is being evalu- over placebo in increasing weekly spontaneous bowel move- ated as a treatment for patients with IBS-D. In a 12-week ments (313) . In a recent press release by the manufacturer, randomized, double-blind, placebo-controlled phase 3 trial of three di7 erent daily doses of TD-5108 (15 mg, 30 mg, and 539 patients from Japan, ramosetron was signi* cantly more 50 mg) each were superior to placebo in achieving the primary likely than placebo to achieve the primary endpoint of relief endpoint of increased spontaneous bowel movements and of global IBS symptoms (312) ; there were no serious drug- in key secondary endpoints including time to * rst spontane- associated adverse events reported during this study. A new ous bowel movement and percentage of patients achieving a drug application for ramosetron treatment of IBS-D has been spontaneous bowel movement in the initial 24 h. is drug * led in Japan. In a 12-week phase 2 trial, which enrolled 691 is also being considered as a potential treatment for patients IBS-D patients from Europe, all four ramosetron groups (2.5, 5, with IBS-C. 10, 20 g once daily) had a numerically higher responder rate e 5-HT3 agonist DDP-733 demonstrated a statistically for relief of global IBS symptoms and abdominal pain compared signi* cant bene* t for the subjective global assessment of with placebo (312) . Ischemic colitis has not been reported with IBS compared with placebo (54 vs. 15 % ) in a phase 2a trial ramosetron, though a relatively small number of patients have in IBS-C patients (314) . A randomized, blinded, placebo- thus far been exposed to this drug. ere are plans for phase controlled study is underway at multiple centers in Canada to 3 trials of this drug in the United States and Europe. assess the safety and e+ cacy of this drug in IBS-C.

Table 4 . Emerging therapies for IBS

Agent Mechanism of action Targeted disorder Clinical status

Peripheral acting agents Crofelemer (301) CFTR inhibitor IBS-D Phase 2b complete Linaclotide (MD-1100) (303) Guanylate cyclase-c agonist IBS-C Phase 3 Arverapamil (AGI-003) (305) Calcium channel blocker IBS-D Phase 3 Asimadoline (306) Kappa opioid agonist IBS Phase 2b complete Mitemcinal (326) Motilin receptor agonist IBS-C Phase 2 Peripheral and central acting agents

Ramosetron (312) 5-HT3 antagonist IBS-D Phase 3

TD-5108 (313) 5-HT 4 agonist IBS-C Phase 2

DDP-773 (314) 5-HT 3 agonist IBS-C Phase 2

DDP-225 (315) 5-HT 3 antagonist and NE reuptake inhibition IBS-D Phase 2 BMS-562086 (318) Corticotropin-releasing hormone antagonist IBS-D Phase 2 GW876008 (319) Corticotropin-releasing hormone antagonist IBS Phase 2 GTP-010 (327) Glucagon-like peptide IBS pain Phase 2 AGN-203818 (322) Alpha receptor agonist IBS pain Phase 2 Solabegron (323) Beta-3 receptor agonist IBS Phase 2 Espindolol (AGI-011) (324) Beta receptor antagonist IBS (all subtypes) Phase 2 Dextoﬁ sopam (325) 2,3 benzodiazepinereceptors IBS-D and IBS-M Phase 3

IBS-C, irritable bowel syndrome with constipation; IBS-D, irritable bowel syndrome with diarrhea, IBS-M, mixed irritable bowel syndrome; CFTR, cystic ﬁ brosis transmembrane conductance regulator.

DDP-225 is a novel partial 5-HT 3 antagonist and norepine- data extraction with Dr Moayyedi in the systematic reviews phrine reuptake inhibitor. Preclinical studies with DDP-225 that contributed to this monograph. Christine Young also have reported decreases in GI motility and visceral hyper- supported the systematic reviews for this monograph. We sensitivity (315). A phase 2 clinical trial in patients with IBS-D thank Dr Premysl Bercik, Dr Peter Bytzer, Cathy Yuan and is currently underway in Canada. Heidi Krall for assisting us with the translation of foreign Corticotropin-releasing hormone antagonists. Corticotropin- language articles. We are indebted to the following investi- releasing hormone (CRH) is one of the primary mediators gators for answering our data queries and, where applica- of the hypothalamic pituitary adrenal axis. Exogenous CRH ble, providing us with their original datasets for analysis: led to exaggerated colonic motility and associated abdominal Vanessa Ameen, Philip Boyce, Kevin Cain, Francis Creed, discomfort in IBS patients compared with controls (316) . A Douglas Drossman, David Earnest, Alessio Fasano, Margaret nonselective CRH antagonist reduced the anxiety, sensation, Heitkemper, Roger Jones, Dr Marotta, David Sanders, and motility evoked by electrical stimulation of the colon in a Kathryn Sanders, Paul Seed, Jan Tack, Barbara Tomenson, small cohort of IBS patients (317) . A multi-center, randomized, a n d Al a n Z i n s m e i s t e r . placebo-controlled phase 2 trial of the CRH antagonist, BMS- 562086 was recently completed in a group of women with REFERENCES IBS-D; results from this trial have not yet been announced 1 . B r a n d t L J , B j o r k m a n D , F e n n e r t y M B et al. S y s t e m a t ic review on the (318). Another CRH antagonist, GW876008, is currently being management of irritable bowel syndrome in North America. Am J G a s t r o e n t e r o l 2 0 0 2 ; 9 7 ( 1 1 S u p p l ) : S 7 – 2 6 . evaluated in a multi-center, randomized, placebo-controlled 2 . T a c k J , F r i e d Met ,al. SH y o s ut e g m h ta o t ic n review: L A the e+ cacy of phase 2 trial in IBS patients (319). treatments for irritable bowel syndrome — a European perspective . Autonomic modulators. ere is growing evidence that speci* c Aliment Pharmacol e r 2 0 0 6 ; 2 4 : 1 8 3 – 2 0 5 . 3 . An o n y m o u s . Sic y s treview e m a t on the management of irritable bowel syndrome forms of autonomic dysfunction can be identi* ed in di7 erent in the European Union. Eur J Gastroenterol Hepatol 2007 ; 19 (suppl 1): subgroups of IBS patients (320,321) . ese discoveries have S 1 1 – 3 7 . generated interest in evaluating autonomic receptor ligands as 4 . Q u a r t e r o AO , Mhmidt e i n e c h V e , - SM c u r i s J et al. B u l k i n g a g e n ts, antispasmodic and antidepressant medication for the treatment of irritable potential treatments for IBS. bowel syndrome . Cochrane Database Syst Rev 2005 (2): CD003460 . AGN-203818, an 2-receptor agonist, currently is being eval- 5 . L e s b r os-PantoK i c k o v a D , M i c h e t t i P , F r i e d M et al. M e t a - a n a l y s i s: e uated as a treatment for abdominal pain in a phase 2 clinical treatment of irritable bowel syndrome . Aliment Pharmacol er 2004 ; 20 : 1253 – 6 9 . trial in patients with IBS (322) . Solabegron is a 3 - a d r e n e r g i c 6 . M a nning AP , o m p s o n W G , H e a t o n K W et al. To w a r d s p o s i t i v e d i a g n o s i s receptor agonist being evaluated as a treatment for the global of the irritable bowel . Br Med J 1978 ; 277 : 653 – 4 . symptoms of IBS in a multinational phase 2 clinical trial 7 . K r u i s W , i e m e C H , W e inzierl M et al. A d i a g n o s t i c s c o r e f o r t h e irritable bowel syndrome. Its value in the exclusion of organic disease. involving sites in Europe and Australia (323). e - a d r e n e r g i c G a s t r o e n t e r o l o g y 1 9 8 4 ; 8 7 : 1 – 7 . receptor antagonist, espindolol (AGI-001), recently has been 8 . D r o s s m a n o m D p A s o n, W G , T a l l e y N J . I d e n t i * cation of sub-groups of evaluated as a treatment for IBS in a randomized, double-blind, functional gastrointestinal disorders . Gastroenterol Intl 1990 ; 3 : 159 – 7 2 . 9 . L o n g s t r e t h G F , o m p s o n W G , C h e y W D et al. F u n c t i o n a l b o w e l placebo-controlled trial employing a forced dose escalation disorders . Gastroenterology 2006 ; 130 : 1480 – 9 1 . protocol (324) Preliminary data did not demonstrate a di7 er- 1 0 . o m p s o n W G , L o n g s t r e t h G F , Det r oal. s s m F u a n n c t i o D n a A l b o w e l ence in e+ cacy compared with placebo when all doses were disorders and functional abdominal pain . Gut 1999 ; 45 (suppl II) : II43 – 7 . 1 1 . J a d a d AR , M o o r e et R al. A , AsC a s re r s o s i l n l g tD h e q u a l i t y o f r e p o r t s taken into account, however, there was a trend toward signi* - of randomized clinical trials: is blinding necessary? Control Clin Trials cant improvement at the highest dose compared with placebo. 1 9 9 6 ; 1 7 : 1 – 1 2 . Dexto* sopam is a nonsedating homophthalazine compound 1 2 . D e r S i m o n i a n R , L a i r d N . M e t a - a n a l y s i s in clinical trials. Control Clin Trials 1 9 8 6 ; 7 : 1 7 7 – 8 8 . that is structurally distinct from traditional benzodiazepines and 1 3 . H i g g i n s J P T , o m p s o n S G , D e e k s J J et al. M e a s u r i n g inconsistency in binds to 2,3 benzodiazepine receptors concentrated in the sub- meta-analyses. Br Med J 2003 ; 327 : 557 – 6 0 . cortical and hypothalamic regions of the brain. Such receptors 1 4 . E g g e r M , D a v e y - S m i t h G , S c h n eider M et al. B i a s in meta-analysis detected by a simple, graphical test. Br Med J 1997 ; 315 : 629 – 3 4 . are known to have modulatory e7 ects on autonomic function 1 5 . G u y a t t G H , C o o k D J , J a e s c h k e R et al. G r ades of recommendation for and consequently, gastrointestinal motility and sensation. In a antithrombotic agents: American College of Chest Physicians Evidence- double-blind, placebo-controlled trial involving 140 patients Based Clinical Practice Guidelines (8th Edition). Chest 2008 ; 133 (6 Suppl): 1 2 3 S – 3 1 S . (66 IBS and 74 placebo) with IBS-D and IBS-M, dexto* sopam 1 6 . E l - S e r a g H B , O l d e n K , B j o r k a m a n D . H e a l t h - r e l a t e d q u a l i t y o f life among was well tolerated and proved superior to placebo in providing persons with irritable bowel syndrome: a systematic review . Aliment adequate relief of overall IBS symptoms as well as reducing stool Pharmacol er 2002 ; 16 : 1171 – 8 5 . 1 7 . G r a l n e k I M , H a y s R Det al. , K e i limpact b o u r n e of irritable A bowel frequency and improving stool consistency (325) . An 18-month syndrome on health-related quality of life. Gastroenterology 2000 ; 119 : phase 2b trial is currently underway to further assess the e+ cacy 6 5 4 – 6 0 . of this drug in 480 women with IBS-D and IBS-M. 1 8 . S a i t o YA , S c h o e n f e l d I I I .P , e epidemiology L o c k e G R of irritable bowel syndrome in North America: a systematic review. Am J Gastroenterol A summary of emerging therapies is given in Table 4 . 2002 ; 97 : 1910 – 5 . 1 9 . S y s t e m a t ic review on the management of irritable bowel syndrome in the ACKNOWLEDGMENTS European Union . Eur J Gastroenterol Hepatol 2007 ; 19 (Suppl 1) : S11 – 3 7 . 2 0 . S a i t o YA , T a l l e y et Nal. J , e e7M eect l t o nof new L diagnostic criteria We are grateful to Dr Alexander Ford for conducting all for irritable bowel syndrome on community prevalence estimates . literature searches, assessing eligibility and performing all N e u r o g a s t r o e n t e r o l M o t i l 2 0 0 3 ; 1 5 : 6 8 7 – 9 4 .

S28 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

2 1 . An d r e w s E B , E a t o n et S al. C P, r e vH a l o e l n l ice s and K Ademographics of 4 7 . S p i e g e l B M R , K a n w a l F , N a l i b o 7 B et al. e impact of somatization on irritable bowel syndrome: resuls from a large web-based survey . Aliment gastrointestinal health resource use in irritable bowel syndrome. Am J Pharmacol e r 2 0 0 5 ; 2 2 : 9 3 5 – 4 2 . Gastroenterol 2005 ; 100 : 2262 – 7 3 . 2 2 . H u n g i n AP , C h a n g et al. L I , r r i t L a b o l c e k b e o w eG l sR y n d r o m e in the 4 8 . P a r e P , G r a y J , L a m S et al. H e a l t h - r e l a t e d q u a l i t y o f life, work productivity, United States: prevalence, symptom patterns and impact . Aliment and health care resource utilization of subjects with irritable bowel Pharamacol er 2005 ; 21 : 1365 – 7 5 . syndrome: baseline results from LOGIC (Longitudinal Outcomes Study 2 3 . M i n o c h a A , J o h n s o n et W al. D P , r e v aAb l e e n l cl e , s T o c L i o d e m o g r a p h y , of Gastrointestinal Symptoms in Canada), a naturalistic study. Clin e r and quality of life of older versus younger patients with irritable bowel 2006 ; 28 : 1726 – 3 5 . syndrome: a population-based study . Dig Dis Sci 2006 ; 51 : 446 – 5 3 . 4 9 . D e a n B B , Aq u i l a r Det ,al. IB m a r p g a h i r o m u t e n t Vin work productivity 2 4 . Am e r i c a n C of l lGastroenterology e g e o Functional Gastrointestinal and health-related quality of life in patients with IBS. Am J Manag Care D i s o r d e r s T a s k F o r c e . E v idence-based position statement on the manage- 2 0 0 5 ; 1 1 : S 1 7 – 2 6 . ment of irritable bowel syndrome in North America. Am J Gastroenterol 5 0 . S p i e g e l B M R , H a r r i s L , L u c a k S et al. P r e d i c t o r s o f w o r k p r o d u c t i v i t y in 2 0 0 2 ; 9 7 : S 1 – S 2 . irritable bowel syndrome (IBS): results from the PROOF cohort. 2 5 . D u n p h y R C , B r i d g w a t e r L , P r i c e D D et al. V i s c e r a l a n d c u t a n e o u s h y p e r - Gastroenterology 2008 ; 134 : AB157 . sensitivity in Persian Gulf war veterans with chronic gastrointestinal 5 1 . F o r d AC , T a l l e y N J , et al.v a n W Z i l a l n t h t ee n history S V and physical s y m p t o m s . P a i n 2 0 0 3 ; 1 0 2 : 7 9 – 8 5 . examination help establish that irritable bowel syndrome is causing my 2 6 . G r a y G C , R e e d R J , K a i s e r K S et al. S e lf-reported symptoms and medical patient’s lower gastrointestinal symptoms? JAMA 2008 . conditions among 11,868 Gulf War-era veterans: the Seabee Health Study . 5 2 . M a nning AP , o m p s o n W G , H e a t o n K W et al. To w a r d s p o s i t i v e d i a g n o s i s Am J Epidemiol 2002 ; 155 : 1033 – 4 4 . of the irritable bowel . Br Med J 1978 ; 277 : 653 – 4 . 2 7 . H u n t S C , R i c h a r d s o n R D . C h r o n ic multisystem illness among Gulf War 5 3 . K r u i s W , i e m e C H , W e inzierl M et al. A d i a g n o s t i c s c o r e f o r t h e veterans . JAMA 1999 ; 282 : 327 – 8 . irritable bowel syndrome. Its value in the exclusion of organic disease. 2 8 . M i l l e r V , H o p k i n s L , W h o r w e l l P . S u icidal ideation in patients with irritable G a s t r o e n t e r o l o g y 1 9 8 4 ; 8 7 : 1 – 7 . bowel syndrome. Clin Gastroenterol and Hepatol 2004 ; 2 : 1064 – 8 . 5 4 . B e l l e n t a n i S , B a l d o n i P , P e t r e l l a S et al. A simple score for the identi* cation 2 9 . S p i e g e l B M R , S c h o e n f e l d P , N a l i b o 7 B . P r e v a l e n ce of suicidal behavior in of patients at high risk of organic diseases of the colon in the family doctor patients with chronic abdominal pain and irritable bowel syndrome: c o n s u l t i n g r o o m . F a m P r a c t 1 9 9 0 ; 7 : 3 0 7 – 1 2 . a systematic review . Aliment Pharmacol e r 2 0 0 7 ; 2 6 : 1 8 3 – 9 3 . 5 5 . F r i g e r i o G , B e r e t t a et A al. , I r rO i t r a s b e l n e i gb o o w e lG syndrome. Still far 3 0 . F o r d AC , F o r m a n D , B a i l e y AG , Ax o n bowelAT , M o a y y e d i P from . I ra r positive i t a b l e diagnosis. Dig Dis Sci 1992 ; 37 : 164 – 7 . syndrome: a 10 year natural history of symptoms and factors that inK uence 5 6 . J e o n g H , L e e H R , Yo o B C et al. M a nning criteria in irritable bowel consultation behaviour. Am J Gastroenterol 2008 ; 103 : 1229 – 3 9 . syndrome: its diagnostic signi* cance. Korean J Intern Med 1993 ; 8 : 34 – 9 . 3 1 . P a l s s o n O S , J o n e s K R , T u r n e r M J et al. I m p a c t o f s o m a t i z a t i o n a n d 5 7 . R a o K P , G u p tet a al. S E v a, l u aJ t a i io n n o f AK M a nning’s criteria in the diag- comorbid medical conditions on health care utilization, disability, and nosis of irritable bowel syndrome. J Assoc Physicians India 1993 ; 41 : 357 – 6 3 . quality of life in irritable bowel syndrome (IBS) . Gastroenterology 2002 ; 122 5 8 . D o g a n U B , U n a l S . K r u i s s c o r i n g system and Manning’s criteria in (Suppl 1) : A501 – 2 . diagnosis of irritable bowel syndrome: is it better to use combined? 3 2 . H a h n B , K i r c h d o e r f e r L , F u l l e r t o n S et al. P a t i e n t - p e r c e i v e d s e v e r i t y o f Acta Gastroenterol Belg 1996 ; 59 : 225 – 8 . irritable bowel syndrome in relation to symptoms, health resource 5 9 . T i b b l e J A , S i g t h o r s s o net al. G U s, e o fF s o u s rt e r or g a t eR m a r k e r s o f i n K am- utilization, and quality of life . Aliment Pharmacol e r 1 9 9 7 ; 1 1 : 5 5 3 – 9 . mation and Rome criteria to distinguish organic from nonorganic intestinal 3 3 . N a l i b o 7 B D , B a lice G , M a y e r E A . P s y c h oderators s o c i a l m of o quality of life d i s e a s e . G a s t r o e n t e r o l o g y 2 0 0 2 ; 1 2 3 : 4 5 0 – 6 0 . in irritable bowel syndrome . Eur J Surg Suppl 1998 ; 583 : 57 – 9 . 6 0 . F o r d ACyzen , Vvan e l d hZanten u SJO , R o d g e r s C C et al. D i a g n o s t ic utility 3 4 . C r e e d F , R a t c l i 7 e J , F e r n a n d e z L et al. H e a l t h - r e l a t e d q u a l i t y o f life and of alarm features for colorectal cancer: systematic review and meta-analysis. health care costs in severe, refractory irritable bowel syndrome. Ann Intern Gut 2008 ; online publication doi:10.1136/gut.2008.159723 . M e d 2 0 0 1 ; 1 3 4 : 8 6 0 – 8 . 6 1 . H a m m e r J , E s l i c k G D , H o w e l l S C et al. D i a g n o s t ic yield of alarm features in 3 5 . S p i e g e l B M , G r a l n e k I M , B o l u s R et al. C l inical determinants of health- irritable bowel syndrome and functional dyspepsia. Gut 2004 ; 53 : 666 – 7 2 . related quality of life in patients with irritable bowel syndrome. Arch Intern 6 2 . W h i t e h e a d W E , P a l s s o n et al.O S U t, i l i t yF eo l f d r e d AD K ag symptom Med 2004 ; 164 : 1773 – 8 0 . exclusions in the diagnosis of irritable bowel syndrome . Aliment Pharmacol 3 6 . v a n der Veek PP , v a n R o o d Y R , M a sinical c l e e trial: AA short-. C l and e r 2 0 0 6 ; 2 4 : 1 3 7 – 4 6 . long-term bene* t of relaxation training for irritable bowel syndrome . 6 3 . W i n a w e r S , Fletcher R , R e x D et al. C o l o r e c t a l c a n cer screening and Aliment Pharmacol e r 2 0 0 7 ; 2 6 : 9 4 3 – 5 2 . surveillance: clinical guidelines and rationale-update based on new 3 7 . S h a w G , S r i v a s t a v a E D , S a d lier M et al. S t r e s s m a n a g e m e n t f o r irritable e v i d e n c e . G a s t r o e n t e r o l 2 0 0 3 ; 1 2 4 : 5 4 4 – 6 0 . bowel syndrome: a controlled trial. Digestion 1991 ; 50 : 36 – 4 2 . 6 4 . P o d o l s k y D . I n K ammatory bowel disease. New Engl J Med 2002 ; 347 : 417 – 2 9 . 3 8 . S p i e g e l B M R , N a l i b o 7 B , M a y e r E et al. e e7 ectiveness of a model 6 5 . F a s a n o A , B e r t i etI al., P G r e e v r a rl e d n uce z z iof celiac T disease in physician-patient relationship versus usual care in irritable bowel at-risk and not-at-risk groups in the United States. Arch Intern Med syndrome: a randomized controlled trial. Gastroenterology 2006 ; 130 : A773 . 2 0 0 3 ; 1 6 3 : 2 8 6 – 9 2 . 3 9 . Am e r i c a n G a s t r o e n t e r o l o g i c a l As s on c iPublication a t i o . e Burden of 6 6 . S p i e g e l B M . D o p h ysicians follow evidence-based guidelines in the diagnostic Gastrointestinal Diseases. American Gastroenterological Association Press: work-up of IBS? Nat Clin Pract Gastroenterol Hepatol 2007 ; 4 : 296 – 7 . Bethesda, MD, 2001 . 6 7 . C a s h B D , S c h o e n f e l d P , C h e y W D . e utility of diagnostic tests in 4 0 . T a l l e y N J , G a b r i e l S E , H a r m s e n W S et al. M e d i c a l costs in community sub- irritable bowel syndrome patients: a systematic review. Am J Gastroenterol jects with irritable bowel syndrome. Gastroenterology 1995 ; 109 : 1736 – 4 1 . 2002 ; 97 : 2812 – 9 . 4 1 . L o n g s t r e t h G F , W i l s o n et al. A I r, r i t aK b n l e i g b h o t w e lK s y n d r o m e , h e a l t h 6 8 . S a n d e r s D S , C a r t e r M J , H u r l s t o n e D P et al. As s o c i a t i o nf adult o coeliac care use, and costs: a US managed care perspective. Am J Gastroenterol disease with irritable bowel syndrome: a case-control study in patients 2 0 0 3 ; 9 8 : 6 0 0 – 7 . ful* lling ROME II criteria referred to secondary care. Lancet 2001 ; 358 : 4 2 . L o n g s t r e t h G F , Ya o J F . I r r i t a b l e b o w e l s y n d r o m e a n d s u r g e r y : a multivari- 1 5 0 4 – 8 . able analysis . Gastroenterology 2004 ; 126 : 1665 – 7 3 . 6 9 . To l l i v e r eB r A a , J LH e , r r D i P a l m a J A . E v a l u a tts i o who n o f meet p a t i e n 4 3 . L i e b e r m a n D A , H o l uet b al. UJ t i,lization E i s e n of colonoscopy G in the clinical criteria for irritable bowel syndrome. Am J Gastroenterol United States: results from a national consortium . Gastrointest Endosc 1 9 9 4 ; 8 9 : 1 7 6 – 8 . 2 0 0 5 ; 6 2 : 8 7 5 – 8 3 . 7 0 . H a m m L R , S o r r e l l s S C , H a r d i n g J P et al. Ad d i t i ol ninvestigations a fail to 4 4 . S p i e g e l B M R , G r a l n e k I M , B o l u s R et al. I s a negative colonoscopy alter the diagnosis of irritable bowel syndrome in subjects ful* lling the associated with improved health-related quality of life or reassurance in Rome criteria. Am J Gastroenterol 1999 ; 94 : 1279 – 8 2 . irritable bowel syndrome? Gastrointest Endosc 2005 ; 62 : 892 – 9 . 7 1 . B a n e r j e e R , C h o u n g O W , G u p t a R et al. R o m e I criteria are more sensitive 4 5 . W h i t e h e a d W E , P a l s s o n O , J o n e s K R . S y s t e m a t ic review of the comorbidity than Rome II for diagnosis of irritable bowel syndrome in Indian patients . of irritable bowel syndrome with other disorders: what are the causes and Indian J Gastroenterol 2005 ; 24 : 164 – 6 . implications? Gastroenterology 2002 ; 122 : 1140 – 5 6 . 7 2 . C a s h B D , K i m C H , L e e D H et al. Y i e l d o f diagnostic testing in patients with 4 6 . M i l l e r AR , N o r t h C S , et Cal. l o u s e association R E of irritable bowel suspected irritable bowel syndrome: a prospective, US multi-center trial. syndrome and somatization disorder. Ann Clin Psychiatry 2001 ; 13 : 25 – 3 0 . Gastroenterology 2007 ; 132 (suppl 1) : A678 .

7 3 . F r a n c i s C Y , D u 7 y J N , W h o r w e l l P J et al. D o es routine ultrasound enhance 9 6 . L i m s u i D , P a r d i D , L o M u s E et al. S y m p t o m a t ic overlap between irritable diagnostic accuracy in irritable bowel syndrome? Am J Gastroenterol bowel syndrome and microscopic colitis . InK amm Bowel Dis 2007 ; 13 : 1996 ; 91 : 1348 – 5 0 . 1 7 5 – 8 1 . 7 4 . F o r d AC , C h e y Wet D al. , U t T i l a i t l y l e o y f d iN a g J n o s t ic tests for celiac 9 7 . F i n e K D , S c h i l l e r L R . AG A t e c h n i c a l r e v i e w o n t h e e v a l u a t i o n a n d disease in irritable bowel syndrome: systematic review and meta-analysis management of chronic diarrhea. Gastroenterology 1999 ; 116 : 1464 – 8 6 . (abstract). Am J Gastroenterol 2008 ; 103 (suppl 1): S463 . 9 8 . Ya w n B P , L y d i c k E , L o c k e G R et al. D o p u b l i s h e d guidelines for evalu- 7 5 . L o c k e G R I I I , M u r r a y J A , Z i n s met e i sal. t e r C e l i ARac disease serology in ation of irritable bowel syndrome reK ect practice? BMC Gastroenterol irritable bowel syndrome and dyspepsia: a population-based case-control 2 0 0 1 ; 1 : 1 1 . s t u d y . M a y o C l i n P r o c 2 0 0 4 ; 7 9 : 4 7 6 – 8 2 . 9 9 . O w e n s D M , N e l s o n D K , T a l l e y N J . e irritable bowel syndrome: long- 7 6 . S h a h b a z k h a n i B , F o r o o t a n M , M e r a t S et al. C o e l i ac disease presenting term prognosis and the physician-patient interaction. Ann Intern Med with symptoms of irritable bowel syndrome . Aliment Pharmacol e r 1 9 9 5 ; 1 2 2 : 1 0 7 – 1 2 . 2 0 0 3 ; 1 8 : 2 3 1 – 5 . 1 0 0 . M o n s b a k k e n K W , Va n d v i k P O , F a r u p P G . P e r c e i v e d f o o d i n t o l e r a n c e 7 7 . Ag r e u s Lu d , d K S, v T a ri bd bs l i n G et al. E n d o m y s i um antibodies are in subjects with irritable bowel syndrome— etiology, prevalence and superior to gliadin antibodies in screening for coeliac disease in patients consequences . Eur J Clin Nutr 2006 ; 60 : 667 – 7 2 . presenting supposed functional gastrointestinal symptoms . Scand J 1 0 1 . S i m r e n M , M a n s sl do ne AMA et , al. L F a o n o g d k - i r e l a t e d gastrointestinal G a s t r o e n t e r o l 2 0 0 0 ; 1 8 : 1 0 5 – 1 1 0 . symptoms in the irritable bowel syndrome . Digestion 2001 ; 63 : 108 – 1 1 5 . 7 8 . S a n d e r s D S , P a t e l D , S t e p h e n s o n T J et al. A p r i m a r ye ccross-sectional a r 1 0 2 . N a n d a R , J a m e s R , S m i t h H et al. F o o d i n t o l e r a n ce and the irritable bowel study of undiagnosed coeliac disease. Eur J Gastroenterol Hepatol syndrome. Gut 1989 ; 30 : 1099 – 1 0 4 . 2 0 0 3 ; 1 5 : 4 0 7 – 1 3 . 1 0 3 . P a r k M I , C a m i l l e r i M . I s t h e r e a role of food allergy in irritable bowel syn- 7 9 . C h e y W D , N o j k o v B , S a a d R J et al. S c r e e n i n g f o r celiac sprue in patients drome and functional dyspepsia? A systematic review . Neurogastroenterol with suspected irritable bowel syndrome: results from a prospective US M o t i l 2 0 0 6 ; 1 8 : 5 9 5 – 6 0 7 . multi-center trial. Gastroenterology 2007 ; 132 (suppl 1): A147 . 1 0 4 . J o n e s VA , M c L a u g h l a n Pet , al. S F h o o o r d t h i n o t u o slerance: e M a 8 0 . O z d i l K , S o k m e n M , E r s o y O et al. As s o c i a t i o n o f gn l uenteropathy t e major factor in the pathogenesis of irritable bowel syndrome . Lancet and irritable bowel syndrome in adult Turkish population . Dig Dis Sci 1982 ; 2 : 1115 – 7 . 2008 ; 53 : 1852 – 5 ( I n p r e s s ) . 1 0 5 . B e n t l e y S J , P e a r s o n D J , R i x K J . F o o d h y p e r s e n s i t i v i t y in irritable bowel 8 1 . M e i n S M , L a d a b a u m U . S e r o l o g i c a l t e s t i n g f o r coeliac disease in patients s y n d r o m e . L a n c e t 1 9 8 3 ; 2 : 2 9 5 – 7 . with symptoms of irritable bowel syndrome: a cost e7 ectiveness analysis . 1 0 6 . S t e f a nini GF , S a g g i o r o A , et Al al. v i sO i r a l Vcromolyn sodium in Aliment Pharmacol er 2004 ; 19 : 1199 – 2 1 0 . comparison with elimination diet in the irritable bowel syndrome, 8 2 . S p i e g e l B M R , D e R o s a V P , G r a l n e k I M et al. Te s t i n g f o r c e l i ac sprue in diarrheic type. Multicenter study of 428 patients. Scand J Gastroenterol irritable bowel syndrome with predominant diarrhea: a cost-e7 ectiveness 1 9 9 5 ; 3 0 : 5 3 5 – 4 1 . analysis . Gastroenterology 2004 ; 126 : 1721 – 3 2 . 1 0 7 . Z w e t c h k e n b a u m J , B u r a k o 7 R . e irritable bowel syndrome and food 8 3 . S c i a r r e t t a G , G i a c o b azzi G , V e r ret i al. A H y d r o g en breath test quanti* ca- h y p e r s e n s i t i v i t y . An n Al l e r g y 1 9 8 8 ; 6 1 : 4 7 – 9 . tion and clinical correlation of lactose malabsorption in adult irritable 1 0 8 . At k i n s o n W , S h e l d o n Tet A al. , F S o h o a d a t e hlimination N based on IgG bowel syndrome and ulcerative colitis. Dig Dis Sci 1984 ; 29 : 1098 – 1 0 4 . antibodies in irritable bowel syndrome: a randomised controlled trial. 8 4 . F a r u p P G , M o n s b a k k e n K W , Va n d v i k P O . L a c t o s e m a l a b s o r p t i o n in a Gut 2004 ; 53 : 1459 – 6 4 . population with irritable bowel syndrome: prevalence and symptoms. 1 0 9 . M c K e e AM , P r i o r A , W hn o rdiets w e l l in irritableP J . E bowel x c l u s i o A case-control study . Scand J Gastroenterol 2004 ; 39 : 645 – 9 . syndrome: are they worthwhile? J Clin Gastroenterol 1987 ; 9 : 526 – 8 . 8 5 . d i S t e f a n o M , M i c e l i E , M a z z o c c h i P et al. V i s c e r a l h y p e r s e n s i t i v i t y a n d 1 1 0 . P e t i t p i e r r e M , G u m o w s k i P , G i r a r d J P . I r r i t a b l e b o w e l syndrome and intolerance symptoms in lactose malabsorption. Neurogastroenterol Motil h y p e r s e n s i t i v i t y t o f o o d . An n Al l e r g y 1 9 8 5 ; 5 4 : 5 3 8 – 4 0 . 2 0 0 7 ; 1 9 : 8 8 7 – 9 5 . 1 1 1 . D a i n e s e R , G a l l iazzari a n i FE Aet al., D D i s e c r L e p a n c i es between 8 6 . S a a d R , C h e y W D . e role of breath tests in clinical GI practice . reported food intolerance and sensitization test * ndings in irritable bowel Gastroenterol 2007 ; 133 : 1763 – 6 . syndrome patients. Am J Gastroenterol 1999 ; 94 : 1892 – 9 7 . 8 7 . P i m e n t e l M , C h o w E J , L i n H C . E r adication of small intestinal bacterial 1 1 2 . D r i s k o J , B i s c h o 7 B , H a l l M et al. T r e a t i n g irritable bowel syndrome with a overgrowth reduces symptoms of irritable bowel syndrome. Am J food elimination diet followed by food challenge and probiotics. J Am Coll Gastroenterol 2000 ; 95 : 3503 – 6 . N u t r 2 0 0 6 ; 2 5 : 5 1 4 – 2 2 . 8 8 . N u cera G , G a b r i e l l i M , Let u pal. a s c u Ab nA o r m a l b r e a t hts t eto s lactose, 1 1 3 . C o s t a b i l e der A A , , F aK v l a i n F et al. W h o l e - g r a in wheat breakfast cereal fructose and sorbitol in irritable bowel syndrome may be explained has a prebiotic e7 ect on the human gut microbiota: a double-blind, by small intestinal bacterial overgrowth . Aliment Pharmacol e r placebo-controlled, crossover study. Br J Nutr 2008 ; 99 : 110 – 2 0 . 2005 ; 21 : 1391 – 5 . 1 1 4 . F o r d AC , T a l l e y Net Jal. , E +S pcacy i e g e lof * Bbre, M R antispasmodics, 8 9 . P a r o d i A , G r e c o et A al. , M S a ya v b a r r e i an t o h t e sE t b e u s e f u l in diagnosis and peppermint oil in irritable bowel syndrome: systematic review and of IBS patients? An Italian study. Gastroenterology 2007 ; 132 (suppl 1): m e t a - a n a l y s i s . B M J 2 0 0 8 ( i n p r e s s ) . A1 9 2 . 1 1 5 . S o l t o M J , G u d m a n d - H o y e r E , K r a g B et al. A d o u b l e - b l i n d t r i a l o f t h e 9 0 . L u p a s c u A , G a b r i e l l i Met ,al. L H a y u d r r i o t a g nen o glucose E C breath e7 ect of wheat bran on symptoms of irritable bowel syndrome . Lancet test to detect small intestinal bacterial overgrowth: a prevalence 1 9 7 6 ; I : 2 7 0 – 2 . case-control study in irritable bowel syndrome . Aliment Pharmacol e r 1 1 6 . M a nning AP , H e a t o n K W , H a r v e y R F et al. W h e a t * bre and irritable bowel 2005 ; 22 : 1157 – 6 0 . syndrome: a controlled trial . Lancet 1977 ; II : 417 – 8 . 9 1 . M c C a l l u m R , S c h u l t z C , S o s t a r i c h S . E v a l u a t i n g t h e r o le of small intestinal 1 1 7 . R i t c h i e J A , T r u e l o v e S C . f Tirritable r e a t m en bowel t o syndrome with bacterial overgrowth in diarrhea predominant irritable bowel syndrome lorazepam, hyoscine butylbromide, and ispaghula husk . Br Med J patients utilizing the glucose breath test. Gastroenterology 2005 ; 128 1 9 7 9 ; 1 : 3 7 6 – 8 . (suppl 2) : A460 . 1 1 8 . L o n g s t r e t h G F , F o x D D , Yo u k e l e s L et al. P s y l l i um therapy in the irritable 9 2 . P o s s e r u d I , S t o t z e r P , B j ö r n s s o n E S et al. S m a l l i n t e s t i n a l b a c t e r i a l o v e r - bowel syndrome. Ann Intern Med 1981 ; 95 : 53 – 6 . growth in patients with irritable bowel syndrome. Gut 2007 ; 56 : 802 – 8 . 1 1 9 . Ar t h u r s Y , F i e l d i n g J F . D o u b l e b l i n d t r i a l o f iin s p athe g h u l a / p o l o x a m e r 9 3 . R o k a R , R o s z t i o c z y A , L e v e qet u eal. M A p i l o t s t u d y o f f e c a l s e r i n e - irritable bowel syndrome. Ir Med J 1983 ; 76 : 253 . protease activity: a pathophysiologic factor in diarrhea-predominant 1 2 0 . N i g a m P , K a p o o r K K , R a s t o g C K et al. D i 7 erent therapeutic regimens in irritable bowel syndrome . Clin Gastroenerol Hepatol 2007 ; 5 : 550 – 5 . irritable bowel syndrome. J Assoc Physicians India 1984 ; 32 : 1041 – 4 . 9 4 . Am e e n V Z , P a t t e r s o n M Het al., C C o o n l o* prmation y M W of presump- 1 2 1 . K r u i s W , W einzierl P , S c h u s s l e r P et al. C o m p a r i s o n o f t h e t h e r a p e u t i c tive diagnosis of irritable bowel syndrome utilizing Rome II criteria e7 ects of wheat bran, mebeverine and placebo in patients with the irritable and simple laboratory screening tests with diagnostic GI evaluation . bowel syndrome . Digestion 1986 ; 34 : 196 – 2 0 1 . Gastroenterology 2001 ; 120 (suppl 1) : A635 . 1 2 2 . L u c e y M R , C l a r k M L , L o w n d e s J O et al. I s b r a n e+ cacious in irritable 9 5 . N o j k o v B , R u b e n s t e i n J H , C a s h B D et al. e yield of colonoscopy in bowel syndrome? A double-blind, placebo controlled crossover study . patients with non-constipated irritable bowel syndrome (IBS): results G u t 1 9 8 7 ; 2 8 : 2 2 1 – 5 . from a prospective, controlled US trial. Gastroenterology 2008 ; 134 1 2 3 . P r i o r A , W h o r wle e lblind l P Jstudy . Dof o ispaghula u b in irritable bowel (Suppl 1) : A30 . syndrome. Gut 1987 ; 28 : 1510 – 3 .

S30 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

1 2 4 . J a l i h a l A , K u r i a n G .y in I s irritable p a g h u l a bowel t h e r a psyndrome: 1 5 2 . L i u J - H , C h e n G - H , Ye h H - Z et al. E n t e r ic-coated peppermint-oil capsules Improvement in overall well-being is related to reduction in bowel in the treatment of irritable bowel syndrome: a prospective, randomized dissatisfaction. J Gastroenterol Hepatol 1990 ; 5 : 507 – 1 3 . trial. J Gastroenterol 1997 ; 32 : 765 – 8 . 1 2 5 . F o w l i e S , E a s t w o o d M A , P r e s c o t t Rl syndrome: . I r r i t a b Assessment l e b o w e 1 5 3 . C a p p e l l o G , S p e z z a f e r r o M , G r o s s i L et al. P e p p e r m i n t o i l (Mintoil) in of psychological disturbance and its inK uence on the response to * ber the treatment of irritable bowel syndrome: a prospective double blind s u p p l e m e n t a t i o n . J P s y c h o s o m R e s 1 9 9 2 ; 3 6 : 1 7 5 – 8 0 . placebo-controlled randomized trial. Dig Liver Dis 2007 ; 39 : 530 – 6 . 1 2 6 . R e e s G , D a v i e s J , o m p s o n R et al. R a n d o m i s e d - c o n t r o l l e d t r i a l o f a * ber 1 5 4 . C a p a n n i M , S u r r e n t i E , B i a g i n i M et al. E + cacy of peppermint oil in the supplement on the symptoms of irritable bowel syndrome. JR Soc Health treatment of irritable bowel syndrome: a randomized, controlled trial . 2 0 0 5 ; 1 2 5 : 3 0 – 4 . Gazz Med Ital 2005 ; 164 : 119 – 2 6 . 1 2 7 . C o o k I J , I r v i n e E J , C a m p b e l l D et al. E 7 ect of dietary * ber on symptoms 1 5 5 . Va s s a l l o M J , C a m i l l e r i M , P h i l l i p s S F et al. C o l o nic tone and motility in and rectosigmoid motility in patoients with irritable bowel syndrome. patients with IBS . Mayo Clin Proc 1992 ; 67 : 725 – 3 1 . G a s t r o e n t e r o l o g y 1 9 9 0 ; 9 8 : 6 6 – 7 2 . 1 5 6 . C h e y W Y , J i n H O , S u n S W et al. C o l o nic motility abnormality in patients 1 2 8 . To s k e s P P , C o n n e r y K L , R i t c h e y T W . C a l c i um polycarbophil compared with irritable bowel syndrome exhibiting abdominal pain and diarrhea. with placebo in irritable bowel syndrome . Aliment Pharmacol e r Am J Gastroenterol 2001 ; 96 : 1499 – 5 0 6 . 1 9 9 3 ; 7 : 8 7 – 9 2 . 1 5 7 . H o v d e n a k N . L o p e r a mide treatment of the irritable bowel syndrome. 1 2 9 . m Ar a n 7 n S , An d e r s e n J R , et al.H e g n e h oe7 j ect J of coarse wheat bran Scand J Gastroenterol 1987 ; 130 : 81 – 4 . in the irritable bowel syndrome. A double-blind cross-over study . Scand J 1 5 8 . L a v o B , S t e n s t a m M , N i e l s e nde in A- treatment L . L o p eof r airritable m i Gastroenterol 1985 ; 20 : 295 – 8 . bowel syndrome— a double blind placebo controlled study . Scand J 1 3 0 . S n o o k J , S h e p h e r d H A . B rn a in n sthe u p ptreatment l e m e n t a t iof o irritable G a s t r o e n t e r o l 1 9 8 7 ; 1 3 0 : 7 7 – 8 0 . bowel syndrome . Aliment Pharmacol e r 1 9 9 4 ; 8 : 5 1 1 – 4 . 1 5 9 . R i n g e l Y , P a l s s o n O S , Z a k k o S F et al. P r e d i c t o r s o f c l i n i c a l r e s p o n s e f r o m 1 3 1 . K h o s h o o V , Ar m s t e a d Cect ,of aL laxative a n d r y with L .and E 7without a phase 2 multi-center e+ cacy trial using rifaximin, a gut-selective, non- tegaserod in adolescents with constipation predominant irritable bowel absorbed antibiotic for the treatment of diarrhea associated irritable bowel syndrome . Aliment Pharmacol e r 2 0 0 6 ; 2 3 : 1 9 1 – 6 . sydnrome . Gastroenterology 2008 ; 134 (suppl 1) : A550 (T1141) . 1 3 2 . S c h a f e r V E , E w e K . e treatment of irritable colon. E+ cacy and tolerance 1 6 0 . L e m b o A , Z aeira k k o NL S et F al. , T1390 F e r r Rifaximin for the treatment of buscopan plus, buscopan, paracetamol and placebo in ambulatory of diarrhea associated irritable bowel syndrome: short term treatment patients with irritable colon . Fortschr Med 1990 ; 108 : 488 – 9 2 . leading to long term sustained response . Gastroenterology 2008 ; 134 1 3 3 . C e n t o n z e V , I m b i b o B P , C a m p a n o z z i F et al. O r a l cimetropium bromide, (suppl 1) : A 545 (T1390) . a new antimuscarinic drug, for long-term treatment of irritable bowel 1 6 1 . S h a r a r a AI , Ao u n E et , al. Ab d u A l - B r a amized n k di o H double-blind syndrome. Am J Gastroenterol 1988 ; 83 : 1262 – 6 . placebo-controlled trial of rifaximin in patients with abdominal bloating 1 3 4 . D o b r i l l a G , I m b i b o B P , P i azzi L et al. L o n g t e r m t r e a t m e n t o f irritable and K atulence. Am J Gastroenterol 2006 ; 101 : 326 – 3 3 . bowel syndrome with cimetropium bromide: a double blind placebo 1 6 2 . P i m e n t e l M , P a r k S , M i r o c h a J et al. e e7 ect of a nonabsorbed oral c o n t r o l l e d c l i n i c a l t r i a l . G u t 1 9 9 0 ; 3 1 : 3 5 5 – 8 . antibiotic (rifaximin) on the symptoms of the irritable bowel syndrome: 1 3 5 . P a s s a r e t t i S , G u s l a n d i M , I m b i b o B P et al. E 7 ects of cimetropium bromide a randomized trial. Ann Intern Med 2006 ; 145 : 557 – 6 3 . on gastrointestinal transit time in patients with irritable bowel syndrome. 1 6 3 . D r o s s m a n razziari D A , E , C D o e l v a u x M et al. ( e d s ) R o m e III: e Aliment Pharmacol e r 1 9 8 9 ; 3 : 2 7 6 . Functional Gastrointestinal Disorders, 3rd edn Degnon Associates: 1 3 6 . G h idini O , S a p o n a t i G , I n t r i e r i L . S i n g le drug treatment for irritable McLean, VA , 2006 . colon: rociverine versus trimebutine maleate . Curr er Res Clin Exp 1 6 4 . P i m e n t e l M , P a r k S , K o n g S et al. 10-day A course of rifaximin, a non- 1 9 8 6 ; 3 9 : 5 4 1 – 8 . absorbable antibiotic, produces a durable improvement in all symptoms 1 3 7 . M o s h a l M G , H e r r o n M . A c l i n i c a l t r i a l o f t r i m e b u t i n e ( M e b u t i nof ) irritable i n s p a s t ibowel c syndrome: a double-blind randomized controlled study. c o l o n . J I n t M e d R e s 1 9 7 9 ; 7 : 2 3 1 – 4 . G a s t r o e n t e r o l o g y 2 0 0 6 ; 1 3 0 : A1 3 4 . 1 3 8 . G l e n d e M , M o r s e l l i - L a b a t e AMet al. , E x B t e a n t t d a e g d l i a n a l yG s i s o f a 1 6 5 . L a u r i t a n o E C , G a b r i e l l i M , S c a r p e l l i n i E et al. S m a l l i n t e s t i n a l b a c t e - double blind, placebo-controlled, 15-week study with otilinium bromide rial overgrowth recurrence aM er antibiotic therapy. Am J Gastroeterol in irritable bowel syndrome . Eur J Gastroenterol Hepatol 2002 ; 14 : 1331– 8 . 2008 ; 103 : 2031 – 5 . 1 3 9 . G i l v a r r y J , lding K e n n JF y . A e non-e7, F i e ect of pirenzipine in dietary 1 6 6 . Ya n g J , L e e H R , L o w K et al. R i faximin versus other antibiotics in the resistant irritable bowel syndrome. Ir J Med Sci 1989 ; 158 : 262 . primary treatment and retreatment of bacterial overgrowth in IBS. 1 4 0 . M i t c h e l l S A , M e e ASet al. , S m Al i t vh ee r icitrate n G D fails to relieve the Dig Dis Sci 2008 ; 53 : 169 – 7 4 . symptoms of irritable bowel syndrome: results of a double-blind, rando- 1 6 7 . P i m e n t e l M , C h o w E J , L i n H C . N o r m a lization of lactulose breath testing mized, placebo-controlled trial . Aliment Pharmacol er 2002 ; 16 : 1187 – 9 5 . correlates with symptom improvement in irritable bowel syndrome. a 1 4 1 . P i a i G , M a zzacca G . P r i * nium bromide in the treatment of the irritable double-blind, randomized, placebo-controlled study. Am J Gastroenterol colon syndrome . Gastroenterology 1979 ; 77 : 500 – 2 . 2 0 0 3 ; 9 8 : 4 1 2 – 9 . 1 4 2 . F i e lding JF . D o u b l e b l i n d t r i a l o f t r i m e b u t i n e in the irritable bowel 1 6 8 . M o a y y e d i P , D u 7 e t t S , M a s o n S et al. e inK uence of antibiotics on ir- syndrome . Ir Med J 1980 ; 73 : 377 – 9 . ritable bowel syndrome: a randomised controlled trial. Gastroenterology 1 4 3 . P a g e J G , D i r n b e r g e r G M . T r e a t m e n t o f t h e i r r i t a b le bowel syndrome 2002 ; 122 (suppl 4) : A465 . with Bentyl (dicyclomine hydrochloride). J Clin Gastroenterol 1 6 9 . N a y a k AK , K a r n a d D R , Ab r a h a m P , le relievesM i s t r y F P . M e t r o n i d a z o 1 9 8 1 ; 3 : 1 5 3 – 6 . symptoms in irritable bowel syndrome: the confusion with so-called 1 4 4 . L e v y C , C h a r b o n n i e r A , Cum a c hbromide i n M and . functionalP i n a v e r i ‘ chronic amebiasis’ . I n d i a n J Gastroenterol 1997 ; 16 : 137 – 9 . colonic disease (double-blind study) . Sem Hop e r 1 9 7 7 ; 5 3 : 3 7 2 – 4 . 1 7 0 . M o a y y e d i P , F o r d et AC al. , eB re+ a n cacy d t ofL probiotics in the 1 4 5 . ienzo D ’ Ar A , D ’ Ag o s t i n o L nio . bromuroL ’ o t t i l o nel trattamento della therapy of irritable bowel syndrome (IBS): a systematic review (abstract) . sí ndrome del colon irritabile. Rass Int Clin Ter 1980 ; 60 : 649 – 5 6 . Am J Gastroenterol 2008 ; 103 (suppl 1): S481 (1230) . 1 4 6 . V i r a t J , H u e b e r D . C o l o p a t h y p a in and dicetel. Prat Med 1987 ; 43 : 32 – 4 . 1 7 1 . G u y o n n e t D , C h a s s a n y O , D u c r o t t e P et al. E 7 ect of a fermented milk con- 1 4 7 . B a l d i F , C o r i n a l d e s i R , F e r r a r i n i F et al. C l inical and functional evaluation taining Bi dobacterium animalis DN-173 010 on the health-related quality of octilonium bromide in the treatment of irritable bowel syndrome: a of life and symptoms in irritable bowel syndrome in adults in primary double-blind controlled trial . Clin Trials J 1983 ; 20 : 77 – 8 8 . care: a multicentre, randomized, double blind, controlled trial . Aliment 1 4 8 . C a s t i g l i o n e F , D a n i e l e B , M a zzacca G . erapeutic strategy for the irritable Pharmacol e r 2 0 0 7 ; 2 6 : 4 7 5 – 8 6 . bowel syndrome . Ital J Gastroenterol 1991 ; 23 (suppl 1) : 53 – 5 . 1 7 2 . N i v E , N a M a l i T , H a l l a k R et al. e e+ cacy of Lactobacillus reuteri AT C C 1 4 9 . D e l m o n t J . I n t e r et de l’ adjonction d ’ un antispasmodique musculotrope au 55730 in the treatment of patients with irritable bowel syndrome— a double traitement des constipations douloureuses des colopathies fonctionnelles blind, placebo-controlled, randomized study. Clin Nutr 2005 ; 24 : 925 – 3 1. par le son. Med Chir Dig 1981 ; 10 : 365 – 7 0 . 1 7 3 . K i m H J , C a m i l l e r i H , M c K inzie S et al. A r amized n d o controlled trial of a 1 5 0 . P u l p e i r o A , M aos r t iSantos M L AR , et D al. e PL r o p i n o x en sindrome de probiotic, VSL#3, on gut transit and symptoms in diarrhea-predominant intestino irritable. Prensa Med Argent 2000 ; 87 : 299 – 3 0 7 . irritable bowel syndrome . Aliment Pharmacol e r 2 0 0 3 ; 1 7 : 8 9 5 – 9 0 4 . 1 5 1 . L e c h Y , O l e s e n K M , H e y H et al. T r e a t m e n t o f irritable bowel syndrome 1 7 4 . K i m H J , Va z q u e z R o q u e M I , C a m i l l e r i M et al. A r amized n d o controlled with peppermint oil. A double-blind investigation with a placebo. trial of a probiotic combination VSL #3 and placebo in irritable bowel Ugeskr Laeger 1988 ; 150 : 2388 – 9 . s y n d r o m e w i t h b l o a t i n g . N e u r o g a s t r o e n t e r o l M o t i l 2 0 0 5 ; 1 7 : 6 8 7 – 9 6 .

1 7 5 . H a l p e r n G , P r i n d i v i l l e T , B l a n k e n b u r g M et al. T r e a t m e n t o f irritable 1 9 7 . B a r d h a n K D , B o d e m a r G , G e l d o f H et al. A d o u b l e - b, l irandomized, n d bowel syndrome with Lacteol Fort: a randomized, double-blind, cross over placebo-controlled dose-ranging study to evaluate the e+ cacy of alosetron trial. Am J Gastroenterol 1996 ; 91 : 1579 – 8 4 . in the treatment of irritable bowel syndrome . Aliment Pharmacol e r 1 7 6 . N i e dzielin K , K o r d e c k i H , B i r k e n f e l d B . A c o n t r o l l e d , d o u b l e b l i n d 2, 0 0 0 ; 1 4 : 2 3 – 3 4 . randomized study on the e+ cacy of Lactobacillus plantarum 299V in 1 9 8 . C a m i l l e r i M , C h e y etW al. Y , MA a r yamized ne r d o E controlled A patients with irritable bowel syndrome. Eur J Gastroenterol Hepatol clinical trial of the serotonin type 3 receptor antagonist alosetron in 2001 ; 13 : 1143 – 7 . women with diarrhea-predominant irritable bowel syndrome . 1 7 7 . N o b a e k S , J o h a n s s o n M L , M o l i n G et al. Al t e r a t i of n intestinal o microK ora Arch Intern Med 2001 ; 161 : 1733 – 4 0 . is associated with reduction in abdominal bloating and pain in patients 1 9 9 . L e m b o T , W r i g h t et al.R A , Al B o a s g e bt r y o n cB o n t r o l s b o w e l u r g e n c y a n d with irritable bowel syndrome. Am J Gastroenterol 2000 ; 95 : 1231 – 8 . provides global symptom improvement in women with diarrhea-predomi- 1 7 8 . O ’ M a h o n y L , M c C a r t h y J , K e l l y P et al. Lactobacillus and Bi dobacterium nant irritable bowel syndrome. Am J Gastroenterol 2001 ; 96 : 2662 – 7 0 . in irritable bowel syndrome: symptom responses and relationship to 2 0 0 . C h e y W D , C h e y W Yet , al. H L e o a n t h g - t e AT r m s a f e t y a n d e+ cacy of cytokine pro* l e s . G a s t r o e n t e r o l o g y 2 0 0 5 ; 1 2 8 : 5 4 1 – 5 1 . alosetron in women with severe diarrhea-predominant irritable bowel 1 7 9 . S i m r e n M , S y r o u set al. A E 7, ects L i nof d Lactobacillus h A Plantarum syndrome. Am J Gastroenterol 2004 ; 99 : 2195 – 2 0 3 . 299V on symptoms and rectal sensitivity in patients with irritable bowel 2 0 1 . C h a n g L , Am e e n etV Zal. , DA ue-ranging, d k o e s s G E phase II study of syndrome (IBS)— a randomized double blind controlled trial. the e+ cacy and safety of alosetron in men with diarrhea-predominant Gastroenterology 2006 ; 130 (suppl 1) : T2043 . IBS. Am J Gastroenterol 2005 ; 100 : 115 – 2 3 . 1 8 0 . K a j a n d e r K , H a t a k k a K , P o u s s a T et al. A p r o b i omixture t i c alleviates 2 0 2 . K r a u s e R , Am e e n et V al. , G A o r dr a omized, n n d o S Hdouble-blind, symptoms in irritable bowel syndrome patients: a controlled 6 month placebo-controlled study to assess e+ cacy and safety of 0.5 mg and intervention . Aliment Pharmacol e r 2 0 0 5 ; 2 2 : 3 8 7 – 9 4 . 1 mg alosetron in women with severe diarrhea-predominant IBS. 1 8 1 . T s u c h i y a J , B a r r e t o R , O k u r a R et al. S i n g l e - b l i n d f o l l o w u p s t u d y o n t h e Am J Gastroenterol 2007 ; 102 : 1709 – 1 9 . e7 ectiveness of a symbiotic preparation in irritable bowel syndrome. 2 0 3 . J o n e s R , H o l t m a n n G , R o d r i g i L et al. Al o s e t r o n r ees l i epain v and Chin J Dig Dis 2004 ; 5 : 169 – 7 4 . improves bowel function compared with mebeverine in female non- 1 8 2 . D r o u a u l t - H o l o wacz S , B i e u v e l e t S , etB ual. r c k e l A A d o u b l e d b lrando- i n constipated irritable bowel syndrome patients . Aliment Pharmacol e r mized controlled trial of a probiotic combination in 100 patients with 1999 ; 13 : 1419 – 2 7 . irritable bowel syndrome . Gastroenterolo Clin Biol 2008 ; 32 : 147 – 5 2 . 2 0 4 . O l d e n K , D e G a r m o R G , J h i n g r a n P et al. P a t i e n t s a t i s f a c t i o n w i t h 1 8 3 . K a j a n d e r K , M y l l y l u o m a E , R a j i lic-Stojanovics M et al. C l inical trial: alosetron for the treatment of women with diarrhea-predominant multispecies probiotic supplementation alleviates the symptoms of irritable bowel syndrome. Am J Gastroenterol 2002 ; 97 : 3139 – 4 6 . irritable bowel syndrome and stabilizes intestinal microbiota . Aliment 2 0 5 . B r adette M , Moennikes H , C a r t e r F et al. C i l a n s e t r o n in irritable bowel Pharmacol e r 2 0 0 8 ; 2 7 : 4 8 – 5 7 . syndrome with diarrhea predominance (IBS-D): e+ cacy and safety in a 1 8 4 . K i m Y G , M o o n J T , L e e K M et al. e e7 ects of probiotics on symptoms of 6 month global study . Gastroenterology 2004 ; 126 (suppl 2) : A42 . irritable bowel syndrome. Korean J Gastroenterol 2006 ; 47 : 413 – 9 . 2 0 6 . F r a n c i s c o n i C F , D r o s s m a n etD Aal. , I n tM e r r a u y pe r t i o n E o A f d a i l y 1 8 5 . S i m r e n M , L i n d h A et , al. S E a 7 mect u e of l s s yoghurt o n L containing activities in cilansetron-treated patients with irritable bowel syndrome with three probiotic bacteria in patients with irritable bowel syndrome (IBS)— diarrhea-predominance (IBS-D): results from a 16-week, placebo-controlled, a randomized, double-blind, controlled trial. Gastroenterology 2007 ; 132 rerandomization trial. Gastroenterology 2006 ; 130 (suppl 2): A600 . (suppl 1) : S1269 . 2 0 7 . M i n e r P , S t a n t o n D , C a r t e r F et al. C i l a n s e t r o n in irritable bowel syndrome 1 8 6 . L o n g Z R , Yu C H , Ya n g Y et al. C l inical observation on acupuncture with diarrhea predominance (IBS-D): e+ cacy and safety in a 3 month combined with microorganism pharmaceutical preparations for treatment US study. Am J Gastroenterol 2008 ; 99 (suppl) : S277 . of irritable bowel syndrome of constipation type. Zhongguo Zhen Jiu 2 0 8 . C h a n g L , C h e y W D , H a r r i s L et al. I n cidence of ischemic colitis and 2 0 0 6 ; 2 6 : 4 0 3 – 5 . serious complications of constipation among patients using alosetron: 1 8 7 . W h o r w e l l P J , Al t r i n g eet r al EL + cacy, M of o ran e l encapsulated J systematic review of clinical trials and post-marketing surveillance data. probiotic Bi dobacterium infantis 35624 in women with irritable bowel Am J Gastroenterol 2006 ; 101 : 1069 – 7 9 . syndrome. Am J Gastroenterol 2006 ; 101 : 1581 – 9 0 . 2 0 9 . F o r d AC , Bxx-Orenstein r a n d t L , A F et o al. E + cacy of 5HT4-agonists in 1 8 8 . G a d e J , o r n P . P a r a g h u r t f o r p a t i e n ts with irritable bowel syndrome. non-diarrhea predominant irritable bowel syndrome: systematic review Scand J Prim Health Care 1989 ; 7 : 23 – 6 . and meta-analysis. Am J Gastroenterol 2008 ; 103 (suppl 1): S478 (1222) .

1 8 9 . S i n n D H , S o n g J H , K i m H J et al. erapeutic e7 ect of Lactobacillus 2 1 0 . M u l l e r - L i s s n e r S A , F u m a g a l l i etI al., Te B ag r a d s eh r a o n d , a K5-HT D 4 acidophilus -SDC 2012, 2013 in patients with irritable bowel syndrome . receptor partial agonist, relieves symptoms in irritable bowel syndrome Dig Dis Sci 2008 ; 53 : 2714 – 2718 Internet * rst publication DOI 10.1007/ patients with abdominal pain, bloating and constipation . Aliment s10620-007-0196-4 . Pharmacol er 2001 ; 15 : 1655 – 6 6 . 1 9 0 . G e r s h o n M D , T a c k J . e serotonin signaling system: from basic 2 1 1 . T a c k J , M u l l e r - L i s s n e r S , B y tzer P et al. A r a n d o m i s e d c o n t r o l l e d t r i a l understanding to drug development for functional GI disorders . assessing the e+ cacy and safety of repeated tegaserod therapy in women G a s t r o e n t e r o l o g y 2 0 0 7 ; 1 3 2 : 3 9 7 – 4 1 4 . with irritable bowel syndrome with constipation . Gut 2005 ; 54 : 1707 – 1 3 . 1 9 1 . D e l v a u x M , L o u v e l D , M a m e t J P et al. E 7 ect of alosetron on responses 2 1 2 . N o v i c k J , M i n e r P , K r a u s e R et al. A r amized, n d o double-blind, to colonic distention in patients with irritable bowel syndrome . Aliment placebo-controlled trial of tegaserod in female patients su7 ering from Pharmacol e r 1 9 9 8 ; 1 2 : 8 4 9 – 5 5 . irritable bowel syndrome with constipation . Aliment Pharmacol e r 1 9 2 . H o u g h t o n L A , F o r s t e r J M , W hn, o ra w 5HT3 e l l Preceptor J . Al o s e t r o 2002 ; 16 : 1877 – 8 8 . antagonist, delays colonic transit in patients with irritable bowel 2 1 3 . K e l l o w J , L e e O Y , C h a n g F Y et al. An As iaci* a - P c, double-blind, pla- syndrome and healthy volunteers . Aliment Pharmacol e r 2 0 0 0 ; 1 4 : cebo-controlled, randomised study to evaluate the e+ cacy, safety, and 7 7 5 – 8 2 . tolerability of tegaserod in patients with irritable bowel syndrome . Gut 1 9 3 . M a y e r E A , B e r m a n S , et Dal. e r b y es h e7 i r eect of S Wthe G 2 0 0 3 ; 5 2 : 6 7 1 – 6 . 5-HT3 receptor antagonist, alosetron, on brain responses to visceral 2 1 4 . H a r i s h K , H a z e e n a K , Va r g h e s e T et al. E 7 ect of tegaserod on colonic tran- stimulation in irritable bowel syndrome patients . Aliment Pharmacol e r sit time in male patients with constipation-predominant irritable bowel 2002 ; 16 : 1357 – 6 6 . syndrome. J Gastroenterol Hepatol 2007 ; 22 : 1183 – 9 . 1 9 4 . F o r d AC , Bxx-Orenstein r a n d t L , A F et o al. E + cacy of 5HT3-antagonists 2 1 5 . N y h l i n H , B a n g C , E l s b o r g L et al. A d o u b l e - b l i n d , p lo-control- a c e b in non-constipation predominant irritable bowel syndrome: systematic led, randomized study to evaluate the e+ cacy, safety and tolerability of review and meta-analysis (abstract). Am J Gastroenterol 2008 ; 103 tegaserod in patients with irritable bowel syndrome . Scand J Gastroenterol (suppl 1): S477 (1220) . 2 0 0 4 ; 3 9 : 1 1 9 – 2 6 . 1 9 5 . C a m i l l e r i M , M a y e r E A et , al. D I mr o p s s r mo v a e n m e n Dt in A pain and 2 1 6 . H a m l i n g J , B a n g C J , T a r p i l a S et al. T i t r a t i o n r e g i m en indicates partial

bowel function in female irritable bowel patients with alosetron, a 5-HT3 5-HT 4 agonist HTF 919 improves symptoms of constipation predominant receptor antagonist . Aliment Pharmacol er 1999 ; 13 : 1149 – 5 9 . irritable bowel syndrome (C-IBS) . Digestion 1998 ; 59 (suppl 3) : 735 . 1 9 6 . C a m i l l e r i M , N o r t h c u t tet al. AR e, e+ K cacy o n g and S safety of 2 1 7 . N o v a r t i s P h a r m aceuticals Corporation . Zelmac TM (tegaserod) Advisory alosetron in female patients with irritable bowel syndrome: a randomised, Committee Brie* ng Document . www.fda.gov/ohrms/dockets/ac/00/ placebo controlled study . Lancet 2000 ; 355 : 1035 – 4 0 . backgrd/3627b1a.pdf A ccessed 20 July 2008 2000 .

S32 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

2 1 8 . C h e y W D , P a r eet al. P Te , g a sV e r i eo g d a fs o r f eA m a l e p a t i e n ts su7 ering 2 4 1 . D r o s s m a n D A , C h e y et Wal. D H e , a l t hS - c r o e t l ta t e d C q u a l i t y o f life in adults from IBS with mixed bowel habits or constipation: a randomized control- with irritable bowel syndrome with constipation: results of a combined analysis led trial. Am J Gastroenterol 2008 ; 103 : 1217 – 2 5 . of two Phase 3 studies with lubiprostone. Gastroenterology 2008 ; 1 3 4 : A4 69 .

2 1 9 . L a n g a k e r K J , M o r r i s D , P r u i t t R et al. e partial 5-HT 4 agonist (HTF 2 4 2 . C h e y W D , D r o s s m a net al. D L A u biprostone , S c o t t is C e7 ective and well 919) improves symptoms in constipation-predominant irritable bowel tolerated through 48 weeks of treatment in adults with irritable bowel syndrome (C-IBS) . Digestion 1998 ; 59 (suppl 3) : 20 . syndrome and constipation. Gastroenterology 2008 ; 134 : A215 . 2 2 0 . S c h u t z e K , B r a n d s t a t t e r G , D r a g o sics B et al. D o u b le-blind study of the ef- 2 4 3 . C h e y W D , Saad R , P a n a s R et al. D i s c o n t i n u a t i o n o f l u b i p r o s t o n e t r e a t - fect of cisapride on constipation and abdominal discomfort as components ment for irritable bowel syndrome with constipation is not associated with of the irritable bowel syndrome . Aliment Pharmacol e r 1 9 9 7 ; 1 1 : 3 8 7 – 9 4 . symptom increase or recurrence: results from a randomized withdrawal 2 2 1 . Va n O u t r y v e M , M i l o R , To u s s a i n t J et al. ‘ P r o k i n e t i c ’ t r e a t m e n t o f consti- s t u d y . G a s t r o e n t e r o l o g y 2 0 0 8 ; 1 3 4 : A4 0 1 . pation-predominant irritable bowel syndrome: a placebo-controlled study 2 4 4 . U e n o R . M u l t i p l e , e s c a l a t i n g , o r a l-dose study to assess the safety, tolerance of cisapride . J Clin Gastroenterol 1991 ; 13 : 49 – 5 7 . and pharmacodynamic pro* le of lubiprostone in normal healthy 2 2 2 . Z i e g e n h a g e n D J , K r u i s W . C i s a p r i d e t r e a t m e n t o f constipation-predomi- v o l u n t e e r s . N e u r o g a s t r o e n t e r o l M o t i l 2 0 0 5 ; 1 7 : 6 2 6 . nant irritable bowel syndrome is not superior to placebo . J Gastroenterol 2 4 5 . S p r e n g e r C , C o p a Aet , al. M E 7 oect r g a of n r olubiprostone, t h J a unique H e p a t o l 2 0 0 4 ; 1 9 : 7 4 4 – 9 . agent for the treatment of chronic idiopathic constipation, on clinical 2 2 3 . F a r u p P G , H o v d e n a k N , W e t t e r h u s S et al. e symptomatic e7 ect of cis- e l e c t r o c a r d i o g r a p h i c r e s u l t s . G a s t r o e n t e r o l o g y 2 0 0 7 ; 1 3 2 : A3 2 5 . apride in patients with irritable bowel syndrome and constipation . Scand J 2 4 6 . R i v e r a E , Wa h l e et A al. , L u biprostone,J o s w i c k T a R novel type-2 chloride Gastroenterol 1998 ; 33 : 128 – 3 1 . channel (ClC-2) activator, does not a7 ect serum electrolyte balance in 2 2 4 . S p i l l e r R C , M e y e r s N L , H i c k l i n g R I . I d e n t i * cation of patients with nonelderly and nonelderly patients with chronic idiopathic constipation. D, non-C irritable bowel syndrome and treatment with renzapride: an G a s t r o e n t e r o l o g y 2 0 0 7 ; 1 3 2 : A1 9 1 . exploratory, multicenter, randomized, double-blind, placebo-controlled 2 4 7 . U e n o R , eWa r a hE l. e P o o A l e d , a n R a l iy v s i s o f t h e m o s t f r e q u e n t ad- clinical trial . Dig Dis Sci 2008 ; e-pub ahead of print . verse events associated with the use of lubiprostone. Am J Gastroenterol 2 2 5 . C a m i l l e r i M , M c K inzie S , F o x J et al. E 7 ect of renzapride on transit in 2006 ; 101 : S489 . constipation-predominant irritable bowel syndrome. Clin Gastroenterol 2 4 8 . U S F o o d a n d Drug Administration. MedWatch: Detailed View: Safety H e p a t o l 2 0 0 4 ; 2 : 8 9 5 – 9 0 4 . and Labeling Changes Approved by FDA Center for Drug Evaluation 2 2 6 . G e o r g e AM , M e y e r s N L inical , H trial: i c k l i n Renzapride g R I . therapyC l and Research (CDER)— April 2008. http://www.fda.gov/medwatch/ for constipation-predominant irritable bowel syndrome— multicentre, SAFETY/2008/apr08.htm, A ccessed 27 July 2008 . randomized, placebo-controlled, double-blind study in primary healthcare 2 4 9 . T a l l e y N J . S S R I s in IBS: sensing a dash of disappointment. setting . Aliment Pharmacol e r 2 0 0 8 ; 2 7 : 8 3 0 – 7 . Clin Gastroenterol Hepatol 2003 ; 1 : 155 – 9 . 2 2 7 . h t t p://www.fda.gov/cder/drug/advisory/tegaserod.htm , Ac c e s sd e 8 January 2 5 0 . D r o s s m a n D , To n e r B B , W h i t e h e a d W E et al. C o g n i t i v e - b e h a v i o r a l 2008, http://clinicaltrials.gov/ct2/show/NCT00401258 . therapy versus education and desipramine versus placebo for moderate to 2 2 8 . S u z u k i M , M o r i t a T , I w a m o t o T . D i v e r s i t y o f C l - channels. Cell Mol Life s e v e r e f u n c t i o n a l b o w e l d i s o r d e r s . G a s t r o e n t e r o l o g y 2 0 0 3 ; 1 2 5 : 1 9 – 3 1 . S c i 2 0 0 6 ; 6 3 : 1 2 – 2 4 . 2 5 1 . H a l p e r t A , D a l t o n C B et , al. D Ci l ainical m a n tresponse N E to tricyclic 2 2 9 . Z i f a r e l l i G , P u s c h M . C l C c h l o r ide channels and transporters: a bio- antidepressants in functional bowel disorders is not related to dosage. physical and physiological perspective . Rev Physiol Biochem Pharmacol Am J Gastroenterol 2005 ; 100 : 664 – 7 1 . 2 0 0 7 ; 1 5 8 : 2 3 – 7 6 . 2 5 2 . J a c k s o n J L , O ’ M a l l e y P G , To m k i n s G et al. T r e a t m e n t o f f u n c t i o n a l gastro- 2 3 0 . C u p p o l e t t i J , M a l i n o w s k a D H , Te w a r i K P et al. S P I-0211 activates T84 cell intestinal disorders with antidepressant medications: a meta-analysis . chloride transport and recombinant human ClC-2 chloride currents. Am J Am J M e d 2 0 0 0 ; 1 0 8 : 6 5 – 7 2 . Physiol Cell Physiol 2004 ; 287 : C1173 – 8 3 . 2 5 3 . T a l l e y N , S p i l l e r R C . I r r i t a b l e b o w e l s y n d r o m e: a little understood organic 2 3 1 . V e r k m a n kacs AS GL , , G L a ulietta LJ . C F TR chloride channel drug bowel disease? Lancet 2002 ; 360 : 555 – 6 4 . discovery— inhibitors as antidiarrheals and activators as therapy of cystic 2 5 4 . C l o u s e R E . M a n a g i n g f u n c t i o n a l b o w e l disorders from the top down: * brosis. Curr Pharm Des 2006 ; 12 : 2235 – 4 7 . lessons from a well-designed treatment trial. Gastroenterology 2 3 2 . B a s s i l AK , B o r m a n etR Aal. , J Ac a r vt i i v e a t i o E n M o f p r o s t a gdin l a n EP 2 0 0 3 ; 1 2 5 : 2 4 9 – 5 3 . receptors by lubiprostone in rat and human stomach and colon . 2 5 5 . G o r a r d D A , L i b b y G W , Fuence a r t h i n of g antidepressants M J G . I n K on B r J P h a r m a c o l 2 0 0 8 ; 1 5 4 : 1 2 6 – 3 5 . whole gut orocaecal transit times in health and irritable bowel syndrome. 2 3 3 . C u p p o l e t t i J , M a l i n o w s k a D H , C h a k r a b a r t i J et al. E 7 ects of lubiprostone Aliment Pharmacol e r 1 9 9 4 ; 8 : 1 5 9 – 6 6 . on human uterine smooth muscle cells . Prostaglandins Other Lipid 2 5 6 . C h i a l H J , C a m i l l e r i M , F e r b e r I et al. E 7 ects of venlafaxine, buspirone, M e d i a t 2 0 0 8 ; 8 6 : 5 6 – 6 0 . and placebo on colonic sensorimotor functions in healthy humans . 2 3 4 . J o h a n s o n J F , M o r t o n D , G e e n a n J et al. M u l t icenter, 4-week, double-blind, Clin Gastroenterol Hepatol 2003 ; 1 : 211 – 8 . randomized, placebo-controlled trial of lubiprostone, a locally-acting 2 5 7 . K u i k e n S D , T y t g a t G N , B o e c k x s t a e n s G E . e selective serotonin reuptake type-2 chloride channel activator, in patients with chronic constipation . inhibitor K uoxetine does not change rectal sensitivity and symptoms Am J Gastroenterol 2008 ; 103 : 170 – 7 . in patients with irritable bowel syndrome: a double blind, randomized, 2 3 5 . J o h a n s o n J F , G a r g a n o Met A al. , I n H i t o i a l l a a n n d d s u sP t C a i n e d e7 ects placebo-controlled study . Clin Gastoenterol Hepatol 2003 ; 1 : 219 – 2 8 . of lubiprostone, a chloride channel-2 (ClC-2) activator for the treatment 2 5 8 . F o r d AC , T a l l e y N J , et al.S ch E o +e ncacy f e l d of antidepressantsP in of constipation: data from a 4-week Phase III study. Am J Gastroenterol irritable bowel syndrome: systematic review and meta-analysis. Am J 2005 ; 100 : S324 – 5 . Gastroenterol 2008 ; 103 (suppl 1): S476 (1218) . 2 3 6 . J o h a n s o n J F , G a r g a n o Met A al. , P H h a o s le l aIII n d study P C of lu biprostone, 2 5 9 . H e e f n e r J D , W i lder RM , W i l s o n I D . I r r i t a b l e c o l o n a n d depression . a chloride channel-2 (ClC-2) activator for the treatment of constipation: P s y c h o s o m a t i c s 1 9 7 8 ; 1 9 : 5 4 0 – 7 . safety and primary e+ cacy. Am J Gastroenterol 2005 ; 100 : S328 – 9 . 2 6 0 . B o e r n e r D , E b e r h a r d t R , M e t z K et al. W i r k s a m k e i t u n d v e r t raqlichkeit 2 3 7 . J o h a n s o n J F , Wa h l e cacy A , and U safety e n o of R lubiprostone . E + in a eines antidepressivuns beim colon irritabile. e r a p i e w o c h e 1 9 8 8 ; 3 8 : 2 0 1 – 8 . subgroup of constipation patients diagnosed with irritable bowel 2 6 1 . V i j J C , J i l o h a R C , K u m a r N et al. E 7 ect of antidepressant drug (doxepin) syndrome with constipation (IBS-C). Am J Gastroenterol 2006 ; 101 : S491 . on irritable bowel syndrome patients . Indian J Psychiatry 1991 ; 33 : 243 – 6 . 2 3 8 . J o h a n s o n J F , D r o s s m a n et al. D A C l inical, P a trial: n a s phase R 2 study of 2 6 2 . M y r e n J , G r o t h H , L a r s s e n S E et al. e e7 ect of trimipramine in patients lubiprostone for irritable bowel syndrome with constipation. with the irritable bowel syndrome. A double-blind study . Scand J Aliment Pharmacol e r 2 0 0 8 ; 2 7 : 6 8 5 – 9 6 . G a s t r o e n t e r o l 1 9 8 2 ; 1 7 : 8 7 1 – 5 . 2 3 9 . D r o s s m a n D A , C h e y etW al. D L , u biproston P a n a s e Rsigni* cantly 2 6 3 . Va h e d i H , M e r a t S , M o m t a h e n S et al. C linical trial: e e7 ect of improves symptom relief rates in adults with irritable bowel syndrome and amitriptyline in patients with diarrhea-predominant irritable bowel constipation (IBS-C): data from two twelve-week, randomized, placebo- syndrome . Aliment Pharmacol e r 2 0 0 8 ; 2 7 : 6 7 8 – 8 4 . controlled, double-blind trials. Gastroenterology 2007 ; 132 : 2586 – 7 . 2 6 4 . B e r g m a n n M , H e ddergott A , S c h l o s s e r T . Die therapie des colon irritabile 2 4 0 . C h e y W D , D r o s s m a net al. D W A h , a t sS y c m o pt t t o m Cs drive global mit trimaprimin (Herphonal)— Eine kontrollierte studie. Z Klin Med symptom improvement with lubiprostone in patients with irritable bowel 1991 ; 46 : 1621 – 8 . syndrome and constipation: data from two multicenter, randomized, 2 6 5 . Va h e d i H , M e r a t Set ,al. R a e s he7 i d ect i o o of n K uoxetine A in patients placebo-controlled trials. Gastroenterology 2008 ; 134 : A28 . with pain and constipation-predominant irritable bowel syndrome: a

double-blind randomized-controlled study . Aliment Pharmacol e r 2 9 3 . W h o r w e l l P J , P r i o r A , F a r a g h e r E B . C o n t r o l l e d t r i a l o f h y p n o t h e r a p y 2 0 0 5 ; 2 2 : 3 8 1 – 5 . in the treatment of severe refractory irritable bowel syndrome . Lancet 2 6 6 . T a b a s G , B e a v e s M , Wa n g J et al. P a r o x e t i n e t o t r e a t irritable bowel 1984 ; 2 : 1232 – 3 . syndrome not responding to high-* ber diet: a double-blind, 2 9 4 . L i u J P , Ya n g M , L i u Y X et al. H e r b a l m e dicines for treatment of irritable placebo-controlled trial. Am J Gastroenterol 2004 ; 99 : 914 – 2 0 . bowel syndrome . Cochrane Database Syst Rev 2006 (1) CD004116 . 2 6 7 . T a c k J , B r o e k a e r t D , F i s c h l e r B et al. A c o n t rd o lcrossover l e study of 2 9 5 . Ya d a v S K , J a i n AKet al. , I r rT i r t ia p b a l te h b i o w eSl N syndrome: thera- the selective serotonin reuptake inhibitor citalopram in irritable bowel peutic evaluation of indigenous drugs. Indian J Med Res 1989 ; 90 : 496 – 5 0 3 . syndrome. Gut 2006 ; 55 : 1095 – 1 0 3 . 2 9 6 . B e n s o u s s a n A , T a l l e y et al. N J T r , e a t mH e i n n gt o f irritable M bowel 2 6 8 . T a l l e y N J , K e l l o w J E , B o y c e P et al. idepressantAn t therapy (imipramine syndrome with Chinese herbal medicine: a randomized controlled trial . and citalopram) for irritable bowel syndrome: a double-blind, JAMA 1998 ; 280 : 1585 – 9 . randomized, placebo-controlled trial . Dig Dis Sci 2008 ; 53 : 108 – 1 5 . 2 9 7 . S a l l o n S , B e n - Ar y e etE al. , D aA v i dn s o o v n e l t r R e a t m e n t f o r c o n s t i p a - 2 6 9 . S a a r t o T , W i 7 e n P J . An t i d e p r e s s a n t s f o r n e u r o p a t h i c p a i n . C o c h r a n e tion-predominant irritable bowel syndrome using Padma Lax, a Tibetan Database Syst Rev 2007 (4); CD005454 . h e r b a l f o r m u l a . D i g e s t i o n 2 0 0 2 ; 6 5 : 1 6 1 – 7 1 . 2 7 0 . R a i n e R , H a i n e s et al. A S , y s t eS m e n a s t kic y review T of mental health inter- 2 9 8 . M a d i s c h A , H o l te m i n a Kn n et al. G Tr , e a t mP l e n t o f irritable bowel syndrome ventions for patients with common somatic symptoms: can research evidence with herbal preparations: results of a double-blind, randomized, placebo- from secondary care be extrapolated to primary care? BMJ 2002 ; 325 : 1082 . controlled, multi-centre trial . Aliment Pharmcol er 2004 ; 19 : 271 – 9 . 2 7 1 . L a c k n e r J M , M e s m e r C , M o r l e y S et al. P s y c h o l o g i c a l t r e a t m e n ts for 2 9 9 . L i m B , M a n h e i m e r E , L a o L et al. Ac u p u n c t u r e f o r t r e a t m e nf tirritable o bowel irritable bowel syndrome: a systematic review and meta-analysis. syndrome. Cochrane Database Syst Rev 2006 Issue 4. Art. No.: CD005111. J Consult Clin Psychol 2004 ; 72 : 1100 – 1 3 . 3 0 0 . U S F o o d a n d Drug Administration (homepage on the internet) e FDA’s 2 7 2 . We b b AN , K u k u r u z o v i c R H , et C al. a t tH o y- S p m n o i t t h h e r a AG p y f o r t r e a t m e n t Drug Review Process: Ensuring Drugs Are Safe and E7 ective (updated of irritable bowel syndrome. Cochrane Database Syst Rev 2007 (4) CD005110. 2005 Sep, cited 2008 May 5); available from http://www.fda.gov/fdac/ 2 7 3 . F o r d A , T a l l e y Net al., E S + c hcacy o e n fof e l dpsychological P therapies features/2002/402_drug.html . in irritable bowel syndrome: systematic review and meta-analysis. 3 0 1 . L e m b o AJ , R o s e n b a u m Det P al. , S C a fh e e t y a n dW e+ D cacy of Am J Gastroenterol 2008 ; 103 (suppl 1): S477 (1219) . crofelemer in patients with diarrhea predominant irritable bowel syn- 2 7 4 . G r e e n e B , B l a n c h a r d E B . C o g n i t i v e t h e r a p y f o r irritable bowel syndrome. J drome (IBS-D). Gastroenterology 2007 (abstract) 132 (4 Suppl2): A-141 . Consult Clin Psychol 1994 ; 62 : 576 – 8 2 . 3 0 2 . F o r t e L R . G u a n y lin regulatory peptides: structures, biological activities 2 7 5 . P a y n e A , B l a n c h a r d dE Bcomparison . A c o n tof r o cognitive l l e therapy and mediated by cyclic GMP and pathobiology. Regul Pept 1999 ; 81 : 25 – 3 9 . self-help support groups in the treatment of irritable bowel syndrome . 3 0 3 . An d e r s e n V l i, o B I u s c, ig G r uet d eal. l l E 7A ects of a novel, * rst-in-class J C o n s u l t C l i n P s y c h o l 1 9 9 5 ; 6 3 : 7 7 9 – 8 6 . guanylate cyclase-c activator, linaclotide acetate (Md-1100), on gastro- 2 7 6 . T k a c h u k G A ,L A G , r a 7M a r t i n Gmized L . controlledR a n d o trial of intestinal and colonic transit and bowel habits in patients with consti- cognitive-behavioral group therapy for irritable bowel syndrome in a pation-predominant irritable bowel syndrome (c-IBS). Gastroenterology medical setting . J Clin Psychol Med Settings 2003 ; 10 : 57 – 6 9 . 2007 (Abstract) 132 (4 Suppl 2): A-82 . 2 7 7 . V o l i m e r A , B l a n c h a r d E B . C o n t rf o individual l l e d c o m p aversus r i s o n o 3 0 4 . I r o n w o o d P h a r m a c e u t i c a l s ( h o m e p a g e o n t h e i n t e r n e t ) . M i c r o b i a a n d group cognitive therapy for irritable bowel syndrome. Behav e r Forest Laboratories Announce Preliminary Results of Linaclotide Phase 2B 1 9 9 8 ; 2 9 : 1 9 – 3 3 . Studies. (updated 2008 Mar 4, cited 2008 Apr 29); Available from http:// 2 7 8 . K e n n e d y T M , J o n e s R , D a r n l e y S et al. C o g n i t i v e b e h a v i o r a l t h e r a p y in ad- ironwoodpharma.com/newsPDF/Linaclotide.Phase.2b.03.04.08.FINAL.pdf . dition to antispasmodic treatment for irritable bowel syndrome in primary 3 0 5 . Q u i g l e y E M , D e v a n e J , Yo u n g D et al. A r amized, n d o double-blind, care: Randomized, controlled trial . BMJ 2005 ; 331 : 435 – 7 . placebo-controlled study of r-verapamil in non-constipated irritable bowel 2 7 9 . H e i t k e m p e r M , J a r r e t t M E , L e v y R et al. S e lf-management for women with syndrome. Am J Gastroenterol 2007 (Abstract) 102 (S2) : S502 . irritable bowel syndrome. Clin Gastoenterol Hepatol 2004 ; 2 : 585 – 9 6 . 3 0 6 . D e l v a u x M , B e c ket al. A E 7, ect J aof c oasimadoline, b J a kappa opioid 2 8 0 . G u t h r i e E , C r e e d F , D a w s o n D et al. A c o n t r o l l e d t rf i apsychological l o treat- agonist, on pain induced by colonic distension in patients with irritable ment for the irritable bowel syndrome . Gastroenterology 1991 ; 100 : 450 – 7 . bowel syndrome . Aliment Pharmacol e r 2 0 0 4 ; 2 0 : 2 3 7 – 4 6 . 2 8 1 . C r e e d F , F e r n a n d e s L , G u t h r i e E et al. e cost-e7 ectiveness of 3 0 7 . M a n g e l AW , Be i o n r Jn , s H t a m m L et al. C l inical trial: asimadoline in the psychotherapy and paroxetine for severe irritable bowel syndrome . treatment of patients with irritable bowel syndrome . Aliment Pharmacol G a s t r o e n t e r o l o g y 2 0 0 3 ; 1 2 4 : 3 0 3 – 1 7 . e r a p e u t 2 0 0 8 ; 2 8 : 2 3 9 – 4 9 . 2 8 2 . B l a n c h a r d E B , G r e e n e B , S c h a r 7 L et al. R e l a x a t i o n t r a ining as a treatment 3 0 8 . o m a s J , K a r v e r S , et C al. o o M n e e y t h y l nG a A l t r e x o n e f o r o p i o i d - for irritable bowel syndrome . Biofeedback Self Regul 1993 ; 18 : 125 – 3 2 . induced constipation in advanced illness . N Engl J Med 2008 ; 358 : 2332 – 4 3 . 2 8 3 . L y n c h P M , Z a m b l e E . A c o n t r o l l e d b e h a v i o rf a irritable l t r e a t m e n t s t u 3 d0 9y . oO z a k i K , S u d o H , M u r a m a t s u H et al. M i t e m c i n a l (GM-611), an orally bowel syndrome. Behav e r 1 9 8 9 ; 2 0 : 5 0 9 – 2 3 . active motilin receptor agonist, accelerates colonic motility and bowel 2 8 4 . v a n der Veek PP , v a n R o o d Y R , M a slinical c l e e trial: AA short-. C and movement in conscious dogs. InK a m m o p h a r m a c o l o g y 2 0 0 7 ; 1 5 : 3 6 – 4 2 . long-term bene* t of relaxation training for irritable bowel syndrome . 3 1 0 . S u d o H , O z a k i K , M u r a m a t s u H et al. M i t e m c i n a l (GM-611), an orally Aliment Pharmacol e r 2 0 0 7 ; 2 6 : 9 4 3 – 5 2 . active motilin agonist, facilitates defecation in rabbits and dogs without 2 8 5 . K e e f e r L , B l a n c h a r d E B . e e7 ects of relaxation response meditation on causing loose stools. Neurogastroenterol Motil 2007 ; 19 : 318 – 3 2 6 . the symptoms of irritable bowel syndrome: results of a controlled treat- 3 1 1 . C a m i l l e r i M , K e r s t e net s al. R ,A pR l a yo-controlled c k e x b A trial of prucalo- ment study . Behav Res e r 2 0 0 1 ; 3 9 : 8 0 1 – 1 1 . pride for severe chronic constipation . N Engl J Med 2008 ; 358 : 2344 – 5 4 . 2 8 6 . B l a n c h a r d E B , S c h w a r z S P , S u l s J M et al. T w o controlled evaluations of 3 1 2 . As t e l l a s P h a r mc. a (homepage I n on the internet) As t es l lR a & D Meeting multicomponent psychological treatment of irritable bowel syndrome 2006 [updated 2006 Jul 3, cited 2008 Apr 29]; Available from http://www. (study 1). Behav Res e r 1 9 9 2 ; 3 0 : 1 7 5 – 8 9 . astellas.com/global/ir/library/pdf/rd2006_1_eg.pdf . 2 8 7 . N e 7 D F , B l a n c h a r d Ei-component B . A m u l ttreatment for irritable bowel 3 1 3 . eravance Inc. (homepage on the internet) eravance announces syndrome . Behav e r 1 9 8 7 ; 1 8 : 7 0 – 8 3 . positive results from Phase 2 clinical study in chronic constipation with 2 8 8 . S h a w G , S r i v a s t a v a E D , S a d lier M et al. S t r e s s m a n a g e m e n t f o r irritable investigational compound TD-5108: All three doses studied achieved bowel syndrome: a controlled trial . Digestion 1991 ; 50 : 36 – 4 2 . primary endpoint (updated 2007 jun 25, cited 2008 Apr 29); Available 2 8 9 . G a l o v s k i T E , B l a n c h a r d E B . e treatment of irritable bowel syndrome from http://ir.theravance.com/ReleaseDetail.cfm?ReleaseID=250925 . with hypnotherapy. Appl Psychophysiol Biofeedback 1998 ; 23 : 219 – 3 2 . 3 1 4 . At l a s V e ne t u(homepage r on the internet) D y n o g en to receive two key 2 9 0 . S i m r e n M , R i n g s t r o m G , B j o r n s s o n E S et al. T r e a t m e n t w i t h h y p n o t h e r a p y European patents for DDP733: Claims cover IBS-c and GERD (updated reduces the sensory and motor component of the gastrocolonic response 2007 Apr 17, cited 2008 Apr 29); Available from http://www.atlasventure. in irritable bowel syndrome . Psychosom Med 2004 ; 66 : 233 – 8 . com/newsandevents/news.cfm?id=1523 . 2 9 1 . B o y c e P M , T a l l e y N J , B a l a a m B et al. A r amized n d o controlled trial of 3 1 5 . P R N e w s w i r e (homepage on the internet) D y n o g en initiates phase II cognitive behavior therapy, relaxation training, and routine clinical care trial of DDP225 for treatment of patients with diarrhea-predominant for the irritable bowel syndrome. Am J Gastroenterol 2003 ; 98 : 2209 – 1 8 . irritable bowel syndrome: trial marks the second compound to 2 9 2 . S a n d e r s K A , B l a n c h a r d E B , S y k e s M Af a .self- P r e l i m i n a r y s t u d y oenter proof-of-concept studies within the last month (cited 2008 Apr administered treatment for irritable bowel syndrome: Comparison to a 30); Available from http://www.faxmarketing.co.uk/cgi-bin/stories. wait list control group. Appl Psychophysiol Biofeedback 2007 ; 32 : 111 – 9 . pl?ACCT=104& STORY=/www/story/10-17-2005/0004170360 & EDATE= .

S34 VOLUME 104 | SUPPLEMENT 1 | JANUARY 2009 www.amjgastro.com Evidence-Based Systematic Review on the Management of IBS

3 1 6 . F u k u d o S , N o m u r a T , H o n g o M . I m p a c t o f corticotropin-releasing 3 2 2 . Al l e r g a nc. (homepage I n on the internet) Al l e r g a n R e s e a r c h a n d D e v e l o p - hormone on gastrointestinal motility and adrenocorticotropic hormone ment Clinical Trials: Ongoing Trials: NCT00441766— Safety and E+ cacy in normal controls and patients with irritable bowel syndrome . Gut of Oral AGN 203818 for the Relief of Irritable Bowel Syndrome Pain 1 9 9 8 ; 4 2 : 8 4 5 – 9 . [cited 2008 Apr 29]; Available from http://www.allerganclinicaltrials. 3 1 7 . S a g a m i Y , S h i m a d a Y , T a y a m a J et al. E 7 ect of a corticotropin releasing com/ongoing/stomach_intestine_conditions.htm . hormone receptor antagonist on colonic sensory and motor function in 3 2 3 . C l inical Trials.gov (homepage on the internet) An en-label o p patients with irritable bowel syndrome. Gut 2004 ; 53 : 958 – 6 4 . trial of duloxetine for the treatment of irritable bowel syndrome. 3 1 8 . B r i s t o l - M y e r s S q u i b b (homepage on the internet) C l inical trial http://clinicaltrials.gov/ct2/show/NCT00401258. registry: Trial details for Trial CN148-013 ST (cited 2008 Apr 30); 3 2 4 . Q u i g l e y E M , D e v a n e J , Yo u n g D et al. A r amized, n d o controlled, Available from http://ctr.bms.com/ctd/InitTrialDetailAction.do?pnum= double-blind trial of s-pindolol in irritable bowel syndrome (IBS) . CN148-013%20ST . Am J Gastroenterol 2007 (Abstract) 102 (S2) : S501 . 3 1 9 . G l a x o S m i t h K l i n e (homepage on the internet) P r o d u c t p i p e l i n e (cited 3 2 5 . L e v e n t e r S , R a u d i b a u g h K , F r i s s o r a C et al. D e x t o * sopam in the treatment 2008 Apr 30); Available from http://www.gsk.com/investors/pp_pipeline_ of patients with diarrhoea-predominant or alternating irritable bowel s t a n d a r d . h t m . syndrome . Aliment Pharmacol e r 2 0 0 8 ; 2 8 : 1 9 7 – 2 0 6 . 3 2 0 . To u s i g n a n t - L a K a m m e Y , G o 7 a u x P , B o u r g a u l t P et al. D i 7 erent autonomic 3 2 6 . C h u g a i P h a r m a (homepage on the internet) D e v e l o p m e n t p i p e l i n e responses to experimental pain in IBS patients and healthy controls. J Clin (as of July 31, 2007) (cited 2008 Apr 30); Available from http://www. Gastroenterol 2006 ; 40 : 814 – 8 2 0 . chugai-pharm.co.jp/pdf/pipeline/english/070731ePipeline.pdf . 3 2 1 . N g C , M a l c o l m et al.A F, e e dH i na n g s a e n n d c o R l o nic distension 3 2 7 . G a s t r o t e c h P h a r m a (homepage on the internet) A p i p e l i n e o f p r o j e c t s provoke altered autonomic responses in irritable bowel syndrome . in clinical development (cited 2008 Apr 30); Available from http://www. Scand J Gastroenterol 2007 ; 42 : 441 – 6 . g a s t r o t e c h p h a r m a . c o m / P r o d u c t _ P i p e L i n e . h t m .

Lawrence J. Brandt declared no financial interests.

William D. Chey has received consulting fees from AGI, Novartis, Procter & Gamble, Salix, Takeda, and Prometheus, and lecture fees from Novartis, Procter & Gamble, Salix, Takeda, and Prometheus.

Jason Connor has received consulting fees from Tranzyme and lecture fees from Janssen.

Amy E. Foxx-Orenstein has received consulting fees from Novartis, GlaxoSmithKline, Salix, Easton Associates, MGI Pharma, AstraZeneca, Salix, TAP, Prometheus, and Strategic Consultants. She has also received grant support from Novartis and Salix.

Paul Moayyedi has received consulting fees from AstraZeneca and lecture fees from Abbot, Procter & Gamble, AstraZeneca, Nycomed, and Johnson & Johnson. He has also received grant support from AstraZeneca and Axcan.

Eamonn M.M. Quigley has received consulting fees from Boehringer Ingelheim, Nycomed, Reckitts Benckiser, Salix, AGI Therapeutics, Procter & Gamble, and Ironside, and lecture fees from Procter & Gamble, GlaxoSmithKline, Pfizer, Janssen-Cilag, Novartis, Norgine, Danone, and Yakult. He has received grant support from Procter & Gamble, Pfizer, Alimentary Health, and AGI Therapeutics. Dr. Quigley has equity ownership/ stock options in Alimentary Health.

Lawrence R. Schiller has received consulting fees from Takeda, Prometheus, Novartis, Napo, UCB, McNeil, Procter & Gamble, Santarus, Adolor, Salix, TAP, and Movetis, and lecture fees from Takeda, IMPACT, Abbott, Santarus, Scientific Frontiers, Facilitate, Fission, Sucampo, Prometheus, Primary Care Network, UCB, AstraZeneca, Procter & Gamble, Pri-Med, EBMed, and Novartis. He has also equity ownership/stock options in Salix.

Philip S. Schoenfeld has received consulting fees from Salix, Shire, Tioga, AGI, Epigenomics, Vertex, Altus, and Takeda, and lecture fees from Salix and Shire. He also has equity ownership/stock options in Wyeth, Merck, and GlaxoSmithKline and is a partner with MD Evidence, LLC.

Brennan Spiegel has received consulting fees from Novartis, AstraZeneca, Phynova, and Johnson & Johnson, and lecture fees from Takeda, Sucampo, AstraZeneca, and Prometheus. He has also received grant support from Amgen and Novartis.

Nicholas J. Talley has received consulting fees from AccreditEd, Addex Pharmaceuticals, SA, the Annenberg Center, Astellas Pharma US, AstraZeneca R&D Lund, Axcan Pharma, Conexus, Dyogen, the F Network, Medscape from WebMD, Metabolic Pharma, MGI Pharma, Microbia, Novartis, Oakstone Publishing, Optum HC, Procter & Gamble, Salix, and SK Life Science. He also holds a patent for Talley BDQ. official publication of the american college of gastroenterology