the role of antibiotics vs. conventional

Review article:the role of antibiotics vs.conventional pharmacotherapy in treating symptoms of irritable bowel syndrome C.L.FRISSORA*&B.D.CASH

*Division of Gastroenterology and Hepatology,Weill Cornell Medical College of Cornell University,New York,NY,USA; Gastroenterology Department and Colon Health Initiat-ive,National Naval Medical Center, Bethesda,MD,USA

Correspondence to:

Dr C.L.Frissora,Division of Gastroenterology and Hepatology, Weill Cornell Medical College of Cornell University,520E.70th Street, Suite J314,New York,NY10028, USA.

E-mail:cfrissor@https://www.wendangku.net/doc/7e2070225.html,

Publication data

Submitted30November2006

First decision11January2007 Resubmitted6March2007 Accepted6March2007SUMMARY

Background

The concept of augmenting the management of irritable bowel syn-drome with antibiotics is evolving,and many questions remain regard-ing this therapy relative to known and hypothesized irritable bowel syndrome pathophysiology.The clinical evidence of small intestinal bacterial overgrowth as an important aetiology of irritable bowel syn-drome continues to accumulate.Clinical symptoms of bacterial over-growth and irritable bowel syndrome are similar;however,a de?nitive cause-and-effect relationship remains unproven.It is unclear whether motility dysfunction causes bacterial overgrowth or gas products of enteric bacteria affect intestinal motility in irritable bowel syndrome. Aim

To discusses the ef?cacy and tolerability of current symptom-directed pharmacotherapies and of antibiotics in the treatment of irritable bowel syndrome.

Methods

A computerized search of PubMed was performed with search terms ‘IBS’,‘pharmacotherapy’and‘antibiotics’.Relevant articles were selec-ted,and the reference list of selected articles was reviewed to identify additional references.

Results

Antibiotic treatment bene?ts a subset of irritable bowel syndrome patients.The non-absorbed antibiotic rifaximin has a favourable safety and tolerability pro?le compared with systemic antibiotics and demon-strates a therapeutic ef?cacy comparable with symptom-based irritable bowel syndrome pharmacotherapies.

Conclusion

Rifaximin is the only antibiotic with demonstrated sustained bene?t beyond therapy cessation in irritable bowel syndrome patients in a pla-cebo-controlled trial.Whether antibiotics can improve quality of life in patients with irritable bowel syndrome warrants further research.

Aliment Pharmacol Ther25,1271–1281 Alimentary Pharmacology&Therapeutics

a2007The Authors1271 Journal compilationa2007Blackwell Publishing Ltd

doi:10.1111/j.1365-2036.2007.03313.x

INTRODUCTION

Irritable bowel syndrome(IBS)is a functional bowel disorder characterized by the hallmark symptoms of recurrent abdominal pain associated with abnormal defecation(constipation,diarrhoea or both)and mul-tiple ancillary symptoms in the absence of structural or biochemical abnormalities.1The proposed pathophysi-ology of IBS is complex and includes visceral hyper-sensitivity,abnormal colonic motility,dysfunctional brain–gut interactions and immune activation.2–4 Symptoms can cause substantial discomfort in patients and can disrupt their daily lives.5,6Although the severity of IBS symptoms can often wax and wane,IBS is a chronic disorder in most patients,as illustrated by a population-based study in Denmark in which95%of individuals with a diagnosis of IBS still had symptoms 5years after the diagnosis.7

Current IBS treatments include dietary restrictions (e.g.avoidance of arti?cial sweeteners,fructose,carbon-ated beverages,fatty foods and sometimes lactose),diet-ary?bre supplementation,pharmacotherapy (antidiarrhoeals,smooth muscle relaxants,bulking agents,prokinetics,psychotropics and serotonin recep-tor agonists and antagonists),and psychological therapy (e.g.relaxation,stress management).8–11Response to these therapies,many of which are directed at symptom relief rather than modi?cation of any heretofore unidenti?ed physiological derangements,is often incomplete.Additionally,for several classes of IBS pharmacotherapy,information about therapeutic bene-?t is either lacking or has not been observed in placebo-controlled clinical trials.12Certain treatments,such as antispasmotics and tricyclic antidepressants,are limited by their anticholinergic side effects–dry mouth, blurred vision and urinary retention.Pharmacotherapies with demonstrable bene?t in the treatment of IBS gen-erally improve targeted symptoms over a short period of time,and symptoms often recur when treatment is with-drawn.12–14Due to these current treatment limitations, many patients with IBS remain dissatis?ed with cur-rently available options.15,16In a2002study of504 patients with constipation-predominant IBS(IBS-C),a median percentage of35%of patients(across therapeu-tic categories)reported that they were not very satis?ed or not at all satis?ed with their medication.15On aver-age,these patients had tried at least four IBS medica-tions.

While symptom-directed treatment generally has not proven useful for long-term management of IBS,recent approaches appear more promising.Long-term trials demonstrating the safety and ef?cacy of seroton-ergic agents have been reported.17,18More recently, the potential utility of antibiotics,one of several emer-ging treatments for IBS,10has been supported by a growing body of evidence demonstrating the import-ant role of bacteria in IBS pathogenesis.19This review discusses the ef?cacy and tolerability of current symp-tom-directed pharmacotherapies and of antibiotics in the treatment of IBS.

EFFICACY AND TOLERABILITY OF SYMPTOM-DIRECTED IBS PHARMACOTHERAPY Symptom-directed IBS pharmacotherapy is generally prescribed according to the patient’s most troublesome symptom,typically abdominal pain or discomfort, bloating,constipation or diarrhoea(Table1).8,9Anti-spasmodics and antidepressants are often prescribed to treat pain or spasms.Bulking agents,osmotic laxatives and the5-HT4agonist tegaserod are used to treat IBS-C. Antidiarrhoeals,tricyclic antidepressants,and,in severe cases,the5-HT3antagonist alosetron are used to treat IBS with diarrhoea(IBS-D).

Several factors should be considered when interpre-ting clinical studies,reviews and meta-analyses on the ef?cacy and tolerability of symptom-based IBS therap-ies.First,many studies have small sample sizes and lack standard criteria for accurately identifying patients with IBS.Secondly,long-term ef?cacy and safety of some therapies have not been systematically assessed.The degree to which the results of short-term clinical trials can be extrapolated to long-term therapy is largely unknown,constituting an important gap in knowledge given the chronic course of IBS.Thirdly,a variable and often high placebo response rate compli-cates the interpretation of controlled clinical trials of IBS pharmacotherapies.20,21The placebo measurement in any study is complicated but has been discussed previously.22The placebo response rate in IBS clinical trials averaged44%and ranged from16%to71%in a recent systematic review.20Therapeutic gain(i.e.dif-ference in treatment response between placebo and active therapy)may be a more meaningful measure-ment of clinical ef?cacy in these studies,allowing derivation of a number needed to treat(NNT)in order to achieve the outcome of interest.For instance,the therapeutic yield of tegaserod in one study was15% (69%tegaserod vs.54%placebo),yielding an NNT of 6.6[95%con?dence interval(CI): 4.0–18.6]for the

1272 C.L.FRISSORA AND B.D.CASH

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

outcome of satisfactory relief of IBS symptoms.23It should be remembered,however,that the NNT derived from a study only applies to the speci?c study circum-stances,and real-world clinical results may be quite different.

Antispasmodics

Antispasmodics are most commonly used to alleviate pain,ostensibly from gastrointestinal(GI)smooth muscle spasms,associated with IBS.Antispasmodics available in the United States include anticholiner-gic?antimuscarinic agents(e.g.dicyclomine,hyoscy-amine)that block activation of smooth muscle spasms by the parasympathetic nervous system.It is suggested that antispasmodics reduce abdominal pain in IBS by decreasing intestinal smooth muscle spasms thought to cause abdominal discomfort.12

The ef?cacy of anticholinergic antispasmodics was evaluated in four placebo-controlled clinical trials:two involving dicyclomine and two involving hyoscyam-ine.24–27In three of the four trials,the antispasmodic was not signi?cantly more effective than placebo at improving individual or global IBS symptoms. However,results should be interpreted cautiously,as the studies investigated these treatments over short periods of time(￡4weeks)in small patient populations and did not employ standard diagnostic criteria for identifying patients with IBS.

Although data do not support the administration of anticholinergic antispasmodics available in the United States,other classes of antispasmodics available else-where do show evidence of ef?cacy.12,28In particular,the direct smooth muscle relaxants mebeverine and pinaverine,which are not available in the United States,can improve some IBS symptoms;however,the poor quality of clinical studies,in general,limits the ability to draw?rm conclusions.12Anticholinergic antispasmodics are associated with a dose-related increase in the incidence of adverse effects secondary to muscarinic blockade,including disturbances in vision,urinary retention,dry mouth and constipa-tion.12Antispasmodics should be used with caution in patients with constipation.29In a comprehensive review conducted by the American College of Gastro-enterology(ACG)Functional GI Disorders Task Force, antispasmodics for the management of IBS received a Grade B recommendation,indicating that clinical stud-ies supporting their use were of intermediate quality.29 Antidepressants

The visceral hypersensitivity observed in some patients with IBS is thought to arise from a dysfunctional interaction between the central and enteric nervous systems.Tricyclic antidepressants are hypothesized to modify central–enteric nervous system interaction and have been used in the treatment of multiple visceral pain syndromes,including IBS.Because these agents can cause constipation,they are most often adminis-tered for IBS-D.30–33

The ef?cacy of tricyclic antidepressants for the treatment of IBS symptoms was evaluated in three meta-analyses of double-blind,placebo-controlled clinical trials.28,31,32These meta-analyses differed in the studies selected for inclusion(although several

Table1.Symptom-directed

pharmacotherapy for irritable bowel syndrome8,9Predominant symptom Pharmacotherapy

ACG grade

recommendation*

Pain Antispasmodics B

Tricyclic antidepressants B

Constipation Bulking agents B

Osmotic laxatives A

5-HT4agonist tegaserod A

Diarrhoea Antidiarrhoeals(e.g.loperamide)B

Tricyclic antidepressants–

5-HT3antagonist alosetron A

*Grade recommendations given by ACG Functional Gastrointestinal Disorders Task Force29or Chronic Constipation Task Force.35

A,supported by strongest level evidence;B,supported by intermediate quality evidence; C,supported by observational studies;ACG,American College of Gastroenterology.

REVIEW:SYMPTOM-BASED TREATMENTS AND ANTIBIOTICS IN IBS1273

a2007The Authors,Aliment Pharmacol Ther25,1271–1281 Journal compilationa2007Blackwell Publishing Ltd

were included in multiple analyses)and the patient populations included.The?rst meta-analysis involved 11clinical studies,including eight that enrolled only patients with IBS,two that enrolled only patients with non-ulcer dyspepsia and one that enrolled patients with IBS or non-ulcer dyspepsia.31Results were not separated by functional GI disorder.The summary odds ratio for improvement in GI symptoms with act-ive treatment vs.placebo was4.2(95%CI:2.3–7.9).

A meta-analysis published in2004involved12clin-ical trials,including10with tricyclic antidepressants, one with the tricyclic antidepressant mianserin and one with a selective serotonin reuptake inhibitor.32The summary odds ratio for global improvement in IBS symptoms with active treatment vs.placebo was2.6 (95%CI:1.9–3.5).Tricyclic antidepressants were con-sidered signi?cantly effective even when low-quality studies were excluded[odds ratio of global improve-ment:1.9(95%CI:1.3–2.7)].The authors emphasized the poor quality of tricyclic antidepressant clinical tri-als,which generally had small sample sizes,did not employ Rome diagnostic criteria,and were of short duration.32

A meta-analysis involving six clinical trials of tricy-clic antidepressants was published in2005.28This ana-lysis employed more stringent criteria for the inclusion of clinical trials than the meta-analyses described above and showed that tricyclic antidepressants were no more effective than placebo.For relief vs.no relief of abdominal pain,the pooled relative risk was0.83 (95%CI:0.33–2.12,two studies;pooled,n=81).For relief of abdominal pain as a continuous variable,the standardized mean difference between active treatment and placebo was)0.53(95%CI:)2.29to1.23;two studies;pooled,n=101).For global improvement vs. no improvement,the pooled relative risk was 1.2 (95%CI:0.78–1.73,four studies;pooled,n=241).

In the highest-quality randomized trial to date,com-paring the ef?cacy of the tricyclic antidepressant desipramine to placebo,the intent-to-treat analysis did not show a statistically signi?cant improvement in composite symptom scale between desipramine and placebo groups(60%vs.47%,P=0.13).31This?nding was largely due to the28%of patients who did not complete the trial.The most commonly cited reason for patient dropout was adverse effects.There were also a small number of patients who had undetectable serum levels of desipramine despite reporting medical compliance.When these patients were excluded from the per-protocol analysis,desipramine resulted in a statistically signi?cant bene?t compared with placebo (73%vs.49%;P=0.006,NNT=4).These?ndings indicate that patients who tolerate tricyclic antidepres-sants are likely to experience symptomatic bene?t but that many patients will suffer unacceptable adverse effects.Furthermore,this study also con?rms that the risk:bene?t ratio associated with tricyclic antidepres-sants is most favourable in individuals with persistent, moderate-to-severe IBS symptoms.

Bulking agents

Bulking agents(i.e.natural or synthetic?bres that add bulk to stool)have also been evaluated for the treat-ment of IBS-C to increase stool frequency and facili-tate stool passage.12Common bulking agents include psyllium,wheat bran,corn?bre,calcium polycarbo-phil and ispaghula husk.The ef?cacy of bulking agents for IBS symptoms was evaluated in three meta-analyses.28,33,34Similar to the meta-analyses of anti-depressants,these meta-analyses differed in the studies selected for inclusion and the patient populations studied.None provided evidence that bulking agents improve global IBS symptoms.

In a2004meta-analysis of13randomized,double-blind clinical trials,?bre did not signi?cantly improve global IBS symptoms after exclusion of low-quality trials(odds ratio, 1.4;95%CI: 1.0–2.0;P=0.06).33 Another2004meta-analysis included17randomized-controlled clinical trials of?bre in the treatment of IBS,and tests of heterogeneity were signi?cant for all analyses,34suggesting that the individual studies were too variable to be pooled for a quantitative review. This meta-analysis found that?bre had a signi?cant effect on global IBS symptoms(relative risk,1.3;95% CI:1.19–1.50;P<0.05).While soluble?bre(e.g.psy-llium,ispaghula,calcium polycarbophil)signi?cantly improved global IBS symptoms,insoluble?bre showed an insigni?cant trend towards worsening them.The authors concluded that?bre provides only marginal bene?ts in the treatment of IBS and that insoluble ?bre might actually worsen clinical outcome.34 Additionally,a2005meta-analysis of11clinical trials of bulking agents reported no evidence of ef?c-acy of bulking agents vs.placebo in the treatment of IBS.28For relief vs.no relief of abdominal pain,the pooled relative risk was1.2(95%CI:0.86–1.73,three studies;pooled,n=159).For relief of abdominal pain as a continuous variable,the standardized mean difference between active treatment and placebo was

1274 C.L.FRISSORA AND B.D.CASH

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

0.68(95%CI:)0.86to 2.33;three studies;pooled, n=128).For global improvement of symptoms,the pooled relative risk was1.1(95%CI:0.78–1.50,nine studies;pooled,n=482).Tests of heterogeneity among studies were statistically signi?cant for all analyses,again suggesting that the individual studies were too divergent to be pooled for a meta-analysis. The most common adverse effect found with bulking agents was increased gas caused by their metabolism by enteric bacteria.12Increased gas production can exacerbate bloating and abdominal discomfort in patients with IBS.Rare adverse effects from bulking agents include anaphylactic reactions and oesophageal and bowel obstruction.

Osmotic laxatives

Delayed intestinal transit time and impaired water secretion may contribute to symptoms of https://www.wendangku.net/doc/7e2070225.html,ctu-lose and polyethylene glycol3350are both osmotic laxatives that cause accumulation of water and elec-trolytes in the colonic lumen,thereby increasing intes-tinal transit.Both agents are approved therapies for constipation and received Grade A recommendations from the ACG Chronic Constipation Task Force,35 based on the strength of evidence demonstrating ef?c-acy for relieving chronic constipation.While osmotic laxatives have been prescribed to treat constipation in IBS,they have not been examined in randomized-controlled trials of patients with IBS,and evidence-based conclusions about their ef?cacy and tolerability cannot be drawn.

Antidiarrhoeals

Accelerated intestinal transit time,a common?nding in IBS-D,may contribute to symptoms of frequent, watery bowel movements,abdominal pain and cramp-ing.At doses of2–12mg daily for3–5weeks,the antidiarrhoeal loperamide signi?cantly reduced stool frequency and improved stool consistency for patients in three placebo-controlled clinical trials,36–38but was not more effective than placebo at improving global IBS symptoms,abdominal distention or abdominal pain.Results should be interpreted cautiously,given the short treatment periods(￡5weeks),small sample sizes and lack of standard criteria(e.g.Rome I or II) for identifying patients with IBS.Antidiarrhoeals are generally considered safe;however,safety data were not systematically reported in these studies.Patients taking these medications should be monitored for the development of constipation or worsening abdominal pain.Additionally,antidiarrhoeals should be adminis-tered with caution in IBS patients with alternating bowel habits(IBS-A)or a history of constipation.

5-HT4receptor agonist:tegaserod

The5-HT4receptor agonist tegaserod stimulates bowel motility,reduces small-bowel and colonic transit time, and has the potential to modulate visceral hypersensi-tivity through an effect on colonic afferent nerves.12 Tegaserod is indicated in the United States for the treatment of IBS-C in women and chronic idiopathic constipation in men and women<65years of age. Clinical trials evaluating tegaserod for IBS are metho-dologically rigorous and meet most or all Rome cri-teria for appropriately designed IBS clinical trials.1The US Food and Drug Administration women-only IBS indication for tegaserod re?ects lack of information about tegaserod in men rather than a demonstrated lack of ef?cacy.33

In a2004Cochrane meta-analysis of eight placebo-controlled clinical trials,tegaserod4mg or12mg daily for up to12weeks was effective at improving global IBS symptoms and increasing the frequency of bowel movements in patients with IBS-C.39The relat-ive increase for achieving global relief of GI symptoms with tegaserod vs.placebo was 1.2(95%CI: 1.09–1.29)at the12-mg dose and1.2(95%CI:1.02–1.31)at the4-mg dose.With respect to individual symptoms of IBS-C,tegaserod increased the frequency of bowel movements but did not lessen abdominal pain and discomfort.

An additional2004meta-analysis of four random-ized,placebo-controlled clinical trials of tegaserod and a2005review of four randomized,placebo-controlled clinical trials assessing the ef?cacy of the US-recom-mended tegaserod dose of6mg twice daily(b.d.) support similar conclusions.12,33Across studies,app-roximately15%more tegaserod-treated patients than placebo-treated patients reported global symptom relief.The latter two assessments,unlike the Cochrane Collaboration meta-analysis,39cited signi?cant improvement of abdominal discomfort with tegaserod relative to placebo.

Cessation of tegaserod therapy is associated with rapid recurrence of symptoms,14,40and repeated cour-ses of therapy or continuous daily therapy is necessary in most patients to control symptoms.In an open-label

REVIEW:SYMPTOM-BASED TREATMENTS AND ANTIBIOTICS IN IBS1275

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

study of tegaserod6mg b.d.,85%of513patients with IBS who had constipation responded to tegaserod dur-ing a12-week course of therapy.14Of403responders who were assessed after cessation of treatment,symp-toms recurred in84%over an average of38days. Among the307patients(a subset of the403)with recurrent symptoms who were retreated with tegaserod 6mg b.d.for up to4weeks,89%responded to retreat-ment.A similar pattern of results was observed in another placebo-controlled trial(n=2660for?rst treatment;n=1191for retreatment).40Response rates in the placebo-controlled trial,however,were lower than in the open-label trial.In the placebo-controlled trial,the percentage of patients who were responders for relief of IBS symptoms was34%with tegaserod vs. 24%with placebo after initial treatment and45%with tegaserod vs.29%with placebo after retreatment.

The incidence of adverse effects was generally com-parable between tegaserod and placebo,with the exception of diarrhoea and headache,reported in9% and15%,respectively,of patients treated with tegas-erod in IBS clinical trials.12Based on the strength of clinical trial?ndings,the ACG Functional GI Disorders Task Force gave tegaserod a Grade A recommendation for the treatment of IBS-C.29

5-HT3receptor antagonist:alosetron

The5-HT3receptor antagonist alosetron slows small-bowel and colonic transit and also appears to modu-late visceral hypersensitivity through an effect on colonic afferent nerves.12Alosetron is indicated in the United States for the treatment of severe IBS-D in women whose IBS has failed to respond to conven-tional treatments.The indication for alosetron is nar-row because of its association with rare but serious side effects,including severe constipation and ischae-mic colitis.In clinical trials of alosetron,the incidence of ischaemic colitis was higher in alosetron-treated patients(0.15%)than in placebo-treated patients (0%;P=0.03).41Alosetron is available with restric-tions and is subject to an extensive postmarketing sur-veillance programme.During this surveillance,the incidence of complications per1000patient-years of alosetron use is1.1for ischaemic colitis and0.66for serious complications of constipation,values similar to the incidence of these conditions in IBS patients not taking alosetron.41

Studies of alosetron in IBS have been well designed.12In a2003meta-analysis of six randomized,placebo-controlled clinical trials of alosetron in IBS (alosetron,n=1762;placebo,n=1356),the pooled odds ratio for adequate relief of pain or global symp-tom improvement was1.9(95%CI:1.57–2.10).42The average NNT with alosetron over placebo was7 (95%CI:5.74–9.43).These trials had12-week treat-ment periods and included mostly women with IBS-D. Authors of a2004meta-analysis of?ve randomized, placebo-controlled clinical trials of alosetron derived conclusions similar to those of the2003meta-analy-sis.33The odds ratio for overall improvement of GI symptoms with alosetron vs.placebo in well-con-trolled clinical trials was2.2(95%CI:1.9–2.6). Cessation of alosetron,like tegaserod,is associated with rapid recurrence of IBS symptoms.18,43–45There-fore,repeated courses of alosetron therapy or continu-ous daily therapy are often necessary to control symptoms.In a study of daily treatment with alose-tron,1mg b.d.for up to48weeks was more effective than placebo at conferring adequate relief(P=0.01) and control of urgency(P<0.001)throughout the treatment period.However,alosetron was poorly toler-ated compared with placebo,18having higher inci-dences of adverse effects(85%vs.72%,P<0.001), drug-related adverse effects(33%vs.16%,P<0.001) and premature withdrawals from the study due to adverse effects(54%vs.40%,P<0.001).Furthermore, symptoms recurred rapidly after cessation of treatment at the end of the48-week treatment period.18 Overall,the most common adverse effect associated with alosetron was constipation,reported in approxi-mately25–39%of alosetron-treated patients in clinical trials of up to12weeks in duration.12,42In the48-week clinical trial described previously,constipation was reported in23%of patients treated with alosetron compared with5%of placebo-treated patients.18The ACG Functional GI Disorders Task Force on IBS gave alosetron a Grade A recommendation based on the quality and results of the aforementioned studies.29 Conclusions:symptom-directed IBS pharmacotherapies

For the provision of relief from IBS symptoms,current data indicate that bulking agents and antispasmodics do not differ from placebo;tegaserod is more effective than placebo for women with IBS-C;loperamide improves stool consistency and frequency in patients with IBS-D but does not lessen their abdominal pain and alosetron is more effective than placebo for

1276 C.L.FRISSORA AND B.D.CASH

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

women with IBS-D.The evidence of ef?cacy is incon-clusive for tricyclic antidepressants.Both tricyclic antidepressants and alosetron are associated with safety and tolerability issues that limit their wide-spread use.No conclusions about laxatives can be drawn because they have not been evaluated in rand-omized-controlled trials as therapies for IBS. ANTIBIOTICS:AN EMERGING THERAPEUTIC APPROACH IN IBS

Antibiotics,an emerging therapy for IBS,differ from the symptom-based therapies described above by tar-geting a putative pathogenic mechanism of IBS, namely the hypothesis that small intestinal bacterial overgrowth(SIBO)might explain the physiological hallmarks of altered gut motility,visceral hypersensi-tivity,abnormal brain–gut interaction and immune activation seen in IBS.19Multiple lines of evidence support the putative role of bacteria in the develop-ment of IBS symptoms.Gas analysis is abnormal in 10–84%of IBS patients undergoing lactulose breath testing(Table2).46–51The variable rates of abnormal gas production reported in studies evaluating bacter-ial overgrowth are most likely due to the different types of breath tests employed.For instance,using glucose as a substrate provides limited results because glucose is absorbed only from the upper small intes-tine and can only detect increased bacterial growth in this region,unlike lactulose,which can detect bacter-ial overgrowth in the distal GI tract.Furthermore, regardless of the substrate administered,clinicians should interpret premature rises in breath hydrogen carefully because such an increase can be indicative of both bacterial overgrowth and rapid intestinal transit.19

Additionally,the distribution of in?ammatory medi-ators and?or in?ammatory cells have been shown to be disturbed in some patients with IBS.52,53It is thought that SIBO may contribute to many of the clinical manifestations of IBS through bacterial fer-mentation and stimulation of a gut immune response, characterized by release of in?ammatory media-tors,such as interleukins and tumour necrosis factor-a,which may affect motility,secretion and sensation.19Bloating and gas,potentially arising from

Table2.Prevalence of SIBO in

irritable bowel syndrome based on the lactulose breath test46–50Lactulose study

Patients

(n)

Prevalence

(%)Requirements for SIBO diagnosis

Pimentel et al.4620276(1)Two distinct H2peaks represent-

ing small intestinal and colonic

?ora

(2)Increase in H2production

<90min after lactulose ingestion

(3)Absolute change in H2con-

centration>20p.p.m.

Nucera et al.479865(1)Two distinct H2peaks representing

small intestinal and colonic?ora

or

(2)Increase in H2production

<90min after lactulose ingestion Pimentel et al.48

and Pimentel

et al.49

11184Increase in H2production

>20p.p.m.within90–180min of

lactulose ingestion

Walters and Vanner503910

(1)Two distinct H2peaks represent-

ing small intestinal and colonic

?ora

(2)Increase in H2production

<90min after lactulose ingestion

(3)Absolute change in H2con-

centration>20p.p.m.

H2,hydrogen;SIBO,small intestinal bacterial overgrowth.

REVIEW:SYMPTOM-BASED TREATMENTS AND ANTIBIOTICS IN IBS1277

a2007The Authors,Aliment Pharmacol Ther25,1271–1281 Journal compilationa2007Blackwell Publishing Ltd

bacterial fermentation of dietary starch,19are among the primary symptoms of IBS regardless of subtype (IBS-C,IBS-D or IBS-A).Postinfectious IBS,which occurs in4–31%of individuals assessed up to 12months after an episode of acute gastroenteritis,54 also supports an aetiological role of bacteria in IBS. Results of clinical trials demonstrate that the reduc-tion or elimination of SIBO with antibiotics can allevi-ate IBS symptoms(Table3).The systemic antibiotic neomycin has been evaluated in two clinical stud-ies.48,55In a double-blind,randomized,placebo-con-trolled study,neomycin dosed at500mg b.d.for 10days was more effective than placebo at improving symptom scores among patients meeting Rome I cri-teria for IBS.48A subanalysis of a double-blind,rand-omized,placebo-controlled trial demonstrated that treatment with neomycin improved global symptoms in individuals with IBS-C compared with placebo (P<0.001).55

The non-absorbed(<0.4%),oral antibiotic rifaximin is the most thoroughly studied antibiotic for the treat-ment of IBS.56–64Rifaximin appears to be well suited for the treatment of IBS because of its broad-spectrum bactericidal activity in vitro,its ef?cacy for SIBO in vivo,favourable tolerability pro?le and lack of association with clinically relevant resistance or Clos-tridium dif?cile colitis.56,59,63,65,66The ef?cacy of rifaximin in the treatment of IBS was evaluated in a randomized,double-blind,placebo-controlled,paral-lel-group study of87patients who met Rome I criteria for IBS.These patients had not received oral antibiot-ics in the3months prior to the study and were not being treated with tegaserod or antidepressants. Patients received rifaximin at a dose of400mg three times daily(t.d.s.)or identical placebo t.d.s.for 10days.62Importantly,statistically signi?cant improvement in IBS symptoms was observed with rif-aximin over placebo during the10-week follow-up period among80of the87patients who had at least one follow-up assessment(36%vs.21%,P=0.02; Figure1).62Although the effects of rifaximin were not signi?cant(P=0.96)for individual weeks,the sub-stantial improvement of global IBS symptoms during the10-week follow-up period after a?nite course of therapy suggests that chronic antibiotic treatment is not necessary for sustained clinical bene?t.In this regard,this preliminary study is one of the?rst to report the sustained bene?ts of IBS pharmacotherapy after cessation of treatment,a characteristic of rifaxi-min that separates it from other pharmacotherapies, with which symptoms return after treatment discon-tinuation.However,certain limitations of this study deserve comment.First,this study,like other studies of IBS therapies,reported a high placebo response rate.20–22Because of the study design,all patients were aware they were no longer taking medication after 10days;thus,the placebo response rate should have been minimal during the follow-up period.62Addition-ally,although rifaximin improved patients’self-repor-ted global symptoms of IBS in this preliminary,

Table3.Clinical response to antibiotic treatment in patients with IBS and?or SIBO48,55–57,61,62,64

Antibiotic n*Clinical response(%)P-value Measure of clinical response

Neomycin

1g?day,10days485543<0.05 50%improvement in composite scoreà

1g?day,10days551937<0.001 Global improvement of IBS symptoms Chlortetracycline

1g?day,7days561127N.S.Normalized breath test results

Rifaximin

1200mg?day,7days561070<0.01§Normalized breath test results

1200mg?day,7days573060<0.001–Normalized breath test results

800mg?day,10days6163410.03 Improvement in IBS symptoms

1200mg?day,10days6243360.02 Global improvement in IBS symptoms 800–1200mg?day,10days641464–Improvement in bloating and constipation *Number of patients treated with antibiotic; vs.placebo;àBased on symptoms of abdominal pain,diarrhoea and constipa-tion;§vs.chlortetracycline1g?day for7days;–vs.rifaximin600mg?day for7days.

IBS,irritable bowel syndrome;N.S.,not signi?cant;SIBO,small intestinal bacterial overgrowth.

1278 C.L.FRISSORA AND B.D.CASH

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

single-centre study,these ?ndings warrant additional multicentre trials with a larger patient population,lon-ger study duration and less subjective outcome meas-ures to con?rm the long-term effects of rifaximin on individual symptoms of IBS.

Long-term therapeutic bene?ts of rifaximin were also observed in two studies that assessed the effects of rifaximin on other functional GI symptoms.60,61In the rifaximin in Abdominal Bloating and Flatulence Trial,which involved 104patients with functional GI symptoms (74ful?lled Rome I criteria for IBS),the percentage of patients with global symptom relief was higher with rifaximin 400mg b.d.than with placebo at the end of a 10-day course of therapy (41%vs.23%,P =0.03).61This difference persisted through the 30-day post-treatment observation period,with 29%of rifaximin-treated patients and 12%of placebo-trea-ted patients reporting global symptom relief at the conclusion of this period (P =0.02).

In another study,34patients with functional GI dis-orders received rifaximin 400mg b.d.for 7days or activated charcoal 300mg b.d.for 7days.60Rifaximin,but not activated charcoal,reduced intestinal gas pro-duction (measured by hydrogen excretion)on the ?rst and 10th days after completion of therapy compared with baseline values.Rifaximin also signi?cantly

reduced the number of ?atus episodes,abdominal girth and overall severity of symptoms on the 10th day after therapy completion compared with baseline values (P <0.05).Numeric,but not statistically signi?-cant,reductions from baseline in bloating and abdom-inal pain were also observed on the 10th day after completion of therapy with rifaximin.With activated charcoal,no statistically signi?cant effect was observed for any of these parameters.

CONCLUSIONS

Antibiotics are an emerging therapeutic option for IBS,and many questions surrounding the exact role of enteric bacteria in the diagnosis and management of IBS remain unanswered.Results to date suggest that appropriately targeted antibiotic therapy may offer important bene?ts as an alternative to,or in addition to,symptom-directed pharmacotherapies in the treat-ment of IBS.Clear demonstration of superiority of one treatment approach over another will require a suit-ably designed,randomized,placebo-controlled trial.Given the breadth of the IBS symptom spectrum and current limitations in symptom-directed therapies rela-ted to adverse effects and blinding,it is unlikely that such a trial will be conducted.More likely,continued reports of ef?cacy with directed antibiotic therapy in patients with symptoms consistent with IBS or other functional GI disorders will lead to the complementary use of these medications in a stepwise or combined fashion.The potential bene?t of antibiotics in dimin-ishing the need for chronic daily pharmacotherapy or reducing the frequency of repeated courses of therapy in IBS warrants additional exploration.

ACKNOWLEDGEMENTS

Declaration of personal interests:Christine L.Frissora MD,has served as a consultant for Vela Pharmaceuti-cals and Pharmos Corporation;is on speaker’s bureau for Salix Pharmaceuticals,Novartis,and Wyeth;and is involved in clinical trials with Tioga Pharmaceuticals.Brooks D.Cash MD,has served as a speaker,a consult-ant,and an advisory board member for Novartis,Salix,Takeda,and Sucampo Pharmaceuticals.Declaration of funding interests:

None.

Figure 1.Percentage global improvement in functional gastrointestinal symptoms after cessation of treatment with rifaximin 400mg t.d.s.or placebo for 10days in patients with IBS diagnosed according to Rome I

criteria.62*P ￡0.05,rifaximin vs.placebo.IBS,irritable bowel syndrome;t.d.s.,three times daily.

REVIEW:SYMPTOM-BASED TREATMENTS AND ANTIBIOTICS IN IBS 1279

a2007The Authors,Aliment Pharmacol Ther 25,1271–1281Journal compilation a2007Blackwell Publishing Ltd

REFERENCES

1Drossman DA(senior ed.),Corazziari E, Talley NJ,Thompson WG,Whitehead WE,and the Rome II Multinational Working Teams(eds).Rome II:The Functional Gastrointestinal Disorders, 2nd edn.McLean,VA:Degnon Associ-ates,2000.

2Quigley EM.Disturbances of motility and visceral hypersensitivity in irritable bowel syndrome:biological markers or epiphenomenon.Gastroenterol Clin North Am2005;34:221–33.

3Bercik P,Verdu EF,Collins SM.Is irrit-able bowel syndrome a low-grade in?ammatory bowel disease?Gastroen-terol Clin North Am2005;34:235–45. 4Chang L.Brain responses to visceral and somatic stimuli in irritable bowel syn-drome:a central nervous system disor-der?Gastroenterol Clin North Am2005;

34:271–9.

5Hahn BA,Kirchdoerfer LJ,Fullerton S, Mayer E.Patient-perceived severity of irritable bowel syndrome in relation to symptoms,health resource utilization and quality of life.Aliment Pharmacol Ther1997;11:553–9.

6Hahn BA,Yan S,Strassels S.Impact of irritable bowel syndrome on quality of life and resource use in the United States and United Kingdom.Digestion 1999;60:77–81.

7Kay L,Jorgensen T,Jensen KH.The epidemiology of irritable bowel syn-drome in a random population:preval-ence,incidence,natural history and risk factors.J Intern Med1994;236: 23–30.

8Tillisch K,Chang L.Diagnosis and treat-ment of irritable bowel syndrome:state of the art.Curr Gastroenterol Rep2005;

7:249–56.

9Cash BD,Chey WD.Advances in the management of irritable bowel syn-drome.Curr Gastroenterol Rep2003;5: 468–75.

10Farthing MJ.Treatment options in irrit-able bowel syndrome.Best Pract Res Clin Gastroenterol2004;18:773–86.

11Talley NJ.Pharmacologic therapy for the irritable bowel syndrome.Am J Gastroenterol2003;98:750–8.

12Schoenfeld P.Ef?cacy of current drug therapies in irritable bowel syndrome: what works and does not work.

Gastroenterol Clin North Am2005;34: 319–35.13Jailwala J,Imperiale TF,Kroenke K.

Pharmacologic treatment of the irritable

bowel syndrome:a systematic review of

randomized,controlled trials.Ann

Intern Med2000;133:136–47.

14Muller-Lissner S,Holtmann G,Rueegg

P,Weidinger G,Lof?er H.Tegaserod is

effective in the initial and retreatment

of irritable bowel syndrome with consti-

pation.Aliment Pharmacol Ther2005;

21:11–20.

15Lembo A.Irritable bowel syndrome

medications side effects survey.J Clin

Gastroenterol2004;38:776–81.

16Frissora CL.Diagnosis,treatment,and

management of irritable bowel syn-

drome with constipation and chronic

constipation.MedGenMed2005;7:71.

17Tougas G,Snape WJ Jr,Otten MH,et al.

Long-term safety of tegaserod in

patients with constipation-predominant

irritable bowel syndrome.Aliment Phar-

macol Ther2002;16:1701–8.

18Chey WD,Chey WY,Heath AT,et al.

Long-term safety and ef?cacy of alose-

tron in women with severe diarrhea-

predominant irritable bowel syndrome.

Am J Gastroenterol2004;99:

2195–203.

19Lin HC.Small intestinal bacterial over-

growth:a framework for understanding

irritable bowel syndrome.JAMA2004;

292:852–8.

20Pitz M,Cheang M,Bernstein CN.De?-

ning the predictors of the placebo

response in irritable bowel syndrome.

Clin Gastroenterol Hepatol2005;3:

237–47.

21Patel SM,Stason WB,Legedza A,et al.

The placebo effect in irritable bowel

syndrome trials:a meta-analysis.Neuro-

gastroenterol Motil2005;17:332–40.

22Mertz HR.Irritable bowel syndrome.

N Engl J Med2003;349:2136–46.

23Kellow J,Lee OY,Chang FY,et al.An

Asia-Paci?c,double blind,placebo con-

trolled,randomised study to evaluate

the ef?cacy,safety,and tolerability of

tegaserod in patients with irritable

bowel syndrome.Gut2003;52:671–6.

24Wheatley D.Irritable colon syndrome

treated with an antispasmodic drug.

Practitioner1976;217:276–80.

25Page JG,Dirnberger GM.Treatment of

the irritable bowel syndrome with Ben-

tyl(dicyclomine hydrochloride).J Clin

Gastroenterol1981;3:153–6.

26Ritchie JA,Truelove SC.Treatment of

irritable bowel syndrome with loraze-

pam,hyoscine butylbromide,and ispag-

hula husk.BMJ1979;1:376–8.

27Schafer E,Ewe K.Behandlung des colon

irritable[The treatment of irritable

colon:ef?cacy and tolerance of busco-

pan plus,buscopan,paracetamol and

placebo in ambulatory patients with

irritable colon].Fortschr Med1990;

108:488–92.

28Quartero AO,Meineche-Schmidt V,

Muris J,Rubin G,de Wit N.Bulking

agents,antispasmodic and antidepres-

sant medication for the treatment of

irritable bowel syndrome.Cochrane

Database Syst Rev2005;2:CD003460.

29Brandt LJ,Bjorkman D,Fennerty MB,

et al.Systematic review on the manage-

ment of irritable bowel syndrome in

North America.Am J Gastroenterol

2002;11(Suppl.):S7–26.

30Crowell MD,Jones MP,Harris LA,Din-

een TN,Schettler VA,Olden KW.Anti-

depressants in the treatment of irritable

bowel syndrome and visceral pain syn-

dromes.Curr Opin Investig Drugs2004;

5:736–42.

31Drossman DA,Toner BB,Whitehead

WE,et al.Cognitive-behavioral therapy

versus education and desipramine ver-

sus placebo for moderate to severe

functional bowel disorders.Gastroenter-

ology2003;125:19–31.

32Jackson JL,O’Malley PG,Tomkins G,

Balden E,Santoro J,Kroenke K.Treat-

ment of functional gastrointestinal dis-

orders with antidepressant medications:

a meta-analysis.Am J Med2000;108:

65–72.

33Lesbros-Panto?ickova D,Michetti P,

Fried M,Beglinger C,Blum AL.Meta-

analysis:the treatment of irritable

bowel syndrome.Aliment Pharmacol

Ther2004;20:1253–69.

34Bijkerk CJ,Muris JW,Knottnerus JA,

Hoes AW,de Wit NJ.Systematic review:

the role of different types of?bre in the

treatment of irritable bowel syndrome.

Aliment Pharmacol Ther2004;19:

245–51.

35American College of Gastroenterology

Chronic Constipation Task Force.An

evidence-based approach to the man-

agement of chronic constipation in

North America.Am J Gastroenterol

2005;100(Suppl.1):S1–4.

36Cann PA,Read NW,Holdsworth CD,

Barends D.Role of loperamide and pla-

cebo in management of irritable bowel

1280 C.L.FRISSORA AND B.D.CASH

a2007The Authors,Aliment Pharmacol Ther25,1271–1281

Journal compilationa2007Blackwell Publishing Ltd

syndrome.Dig Dis Sci1984;29:239–

47.

37Hovdenak N.Loperamide treatment of the irritable bowel syndrome.Scand J Gastroenterol Suppl1987;130:81–4. 38Efskind PS,Bernklev T,Vatn MH.A double-blind,placebo-controlled trial with loperamide in irritable bowel syn-drome.Scand J Gastroenterol1996;31: 463–8.

39Evans BW,Clark WK,Moore DJ,Whor-well PJ.Tegaserod for the treatment of irritable bowel syndrome.Cochrane Database Syst Rev2004;1:CD003960. 40Tack J,Muller-Lissner S,Bytzer P,et al.

A randomised controlled trial assessing

the ef?cacy and safety of repeated tegaserod therapy in women with irrit-able bowel syndrome with constipation.

Gut2005;54:1707–13.

41Chang L,Chey WD,Harris L,Olden K, Surawicz C,Schoenfeld P.Incidence of ischemic colitis and serious complica-tions of constipation among patients using alosetron:systematic review of clinical trials and post-marketing sur-veillance data.Am J Gastroenterol 2006;101:1069–79.

42Cremonini F,Delgado-Aros S,Camilleri M.Ef?cacy of alosetron in irritable bowel syndrome:a meta-analysis of randomized controlled trials.Neurogast-roenterol Motil2003;15:79–86.

43Camilleri M,Northcutt AR,Kong S, Dukes GE,McSorley D,Mangel AW.

Ef?cacy and safety of alosetron in women with irritable bowel syndrome:a randomised,placebo-controlled trial.

Lancet2000;355:1035–40.

44Camilleri M,Chey WY,Mayer EA,et al.

A randomized controlled clinical trial of

the serotonin type3receptor antagonist alosetron in women with diarrhea-pre-dominant irritable bowel syndrome.

Arch Intern Med2001;161:1733–40. 45Lembo T,Wright RA,Bagby B,et al.

and the Lotronex Investigator Team.

Alosetron controls bowel urgency and provides global symptom improvement in women with diarrhea-predominant irritable bowel syndrome.Am J Gast-roenterol2001;96:2662–70.

46Pimentel M,Chow EJ,Lin HC.Eradica-tion of small intestinal bacterial over-growth reduces symptoms of irritable

bowel syndrome.Am J Gastroenterol

2000;95:3503–6.

47Nucera G,Gabrielli M,Lupascu A,et al.

Abnormal breath tests to lactose,fruc-

tose and sorbitol in irritable bowel syn-

drome may be explained by small

intestinal bacterial overgrowth.Aliment

Pharmacol Ther2005;21:1391–5.

48Pimentel M,Chow EJ,Lin HC.Normal-

ization of lactulose breath testing corre-

lates with symptom improvement in

irritable bowel syndrome:a double-

blind,randomized,placebo-controlled

study.Am J Gastroenterol2003;98:

412–9.

49Pimentel M,Wallace D,Hallegua D,

et al.A link between irritable bowel

syndrome and?bromyalgia may be

related to?ndings on lactulose breath

testing.Ann Rheum Dis2004;63:

450–2.

50Walters B,Vanner SJ.Detection of bac-

terial overgrowth in IBS using the lactu-

lose H2breath test:comparison with

14C-D-xylose and healthy controls.Am

J Gastroenterol2005;100:1566–70.

51Noddin L,Callahan M,Lacy BE.Irritable

bowel syndrome and functional dyspep-

sia:different diseases or a single disor-

der with different manifestations?

MedGenMed2005;7:17.

52O’Mahony L,McCarthy J,Kelly P,et al.

Lactobacillus and bi?dobacterium in

irritable bowel syndrome:symptom

responses and relationship to cytokine

pro?le.Gastroenterology2005;128:

541–51.

53Spiller R,Campbell E.Post-infectious

irritable bowel syndrome.Curr Opin

Gastroenterol2006;22:13–7.

54Connor BA.Sequelae of traveler’s diar-

rhea:focus on postinfectious irritable

bowel syndrome.Clin Infect Dis2005;

41(Suppl.8):S577–86.

55Pimentel M,Chatterjee S,Chow EJ,Park

S,Kong Y.Neomycin improves consti-

pation-predominant irritable bowel syn-

drome in a fashion that is dependent on

the presence of methane gas:subanaly-

sis of a double-blind randomized con-

trolled study.Dig Dis Sci2006;51:

1297–301.

56Di Stefano M,Malservisi S,Veneto G,

Ferrieri A,Corazza GR.Rifaximin versus

chlortetracycline in the short-term treat-

ment of small intestinal bacterial over-

growth.Aliment Pharmacol Ther2000;

14:551–6.

57Lauritano EC,Gabrielli M,Lupascu A,

et al.Rifaximin dose-?nding study for

the treatment of small intestinal bacter-

ial overgrowth.Aliment Pharmacol Ther

2005;22:31–5.

58Corazza GR,Ventrucci M,Strocchi A,

et al.Treatment of small intestine bac-

terial overgrowth with rifaximin,a non-

absorbable rifamycin.J Int Med Res

1988;16:312–6.

59Baidoo LR,Kundu R,Berenbaum PL,

et al.Rifaximin is effective for small

intestinal bacterial overgrowth.Gastro-

enterology2006;128(Suppl.2):W1732.

60Di Stefano M,Strocchi A,Malservisi S,

Veneto G,Ferrieri A,Corazza GR.Non-

absorbable antibiotics for managing

intestinal gas production and gas-rela-

ted symptoms.Aliment Pharmacol Ther

2000;14:1001–8.

61Sharara AI,Aoun E,Abdul-Baki H,

Mounzer R,Sidani S,Elhajj I.A rand-

omized double-blind placebo-controlled

trial of rifaximin in patients with

abdominal bloating and?atulence.Am

J Gastroenterol2006;101:326–33.

62Pimentel M,Park S,Kane SV.Rifaxi-

min,a nonabsorbable,gut-selective

antibiotic,improves the symptoms of

irritable bowel syndrome:a randomized,

double-blind,placebo-controlled study.

Ann Intern Med2006;145:557–63.

63Cuoco L,Salvagnini M.Small intestine

bacterial overgrowth in irritable bowel

syndrome:a retrospective study with

rifaximin.Minerva Gastroenterol Dietol

2006;52:89–95.

64George G,Hunt M,Kornbluth A,Leg-

nani P.An open-label pilot trial of

high-dose rifaximin in the treatment of

patients with constipation-predominant

irritable bowel syndrome(abstract

T1287).Gastroenterology2006;130

(Suppl.2):A-515.

65Robins GW,Wellington K.Rifaximin:a

review of its use in the management of

traveller’s diarrhoea.Drugs2005;65:

1697–713.

66Ericsson CD.Safety and tolerability of

the antibacterial rifaximin in the treat-

ment of travellers’diarrhoea.Drug Saf

2006;29:201–7.

REVIEW:SYMPTOM-BASED TREATMENTS AND ANTIBIOTICS IN IBS1281

a2007The Authors,Aliment Pharmacol Ther25,1271–1281 Journal compilationa2007Blackwell Publishing Ltd