Treatment and Prevention of Pouchitis After Ileal Pouch-Anal Anastomosis for Chronic Ulcerative Colitis (Review)

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review)

Holubar SD, Cima RR, Sandborn WJ, Pardi DS

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2010, Issue 6 http://www.thecochranelibrary.com

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. TABLE OF CONTENTS HEADER...... 1 ABSTRACT ...... 1 PLAINLANGUAGESUMMARY ...... 2 BACKGROUND ...... 2 OBJECTIVES ...... 3 METHODS ...... 3 RESULTS...... 4 Figure1...... 6 Figure2...... 7 Figure3...... 7 Figure4...... 8 Figure5...... 9 Figure6...... 9 Figure7...... 10 Figure8...... 10 Figure9...... 11 Figure10...... 11 Figure11...... 12 DISCUSSION ...... 12 AUTHORS’CONCLUSIONS ...... 13 ACKNOWLEDGEMENTS ...... 14 REFERENCES ...... 15 CHARACTERISTICSOFSTUDIES ...... 17 DATAANDANALYSES...... 27 Analysis 1.1. Comparison 1 Acute Pouchitis, Outcome 1 Ciproﬂoxacin vs. Metronidazole: Achieved Remission (PDAI <7)...... 28 Analysis 1.2. Comparison 1 Acute Pouchitis, Outcome 2 Budesonide enema vs. Metronidazole: Achieved Remission (PDAI <7)...... 28 Analysis 1.3. Comparison 1 Acute Pouchitis, Outcome 3 Rifaximin vs. Placebo: Achieved Remission (PDAI <7). . . 29 Analysis 1.4. Comparison 1 Acute Pouchitis, Outcome 4 Lactobacillus GG vs. Placebo: Improved (PDAI reduction ≥ 3). 29 Analysis 2.1. Comparison 2 Chronic Pouchitis, Outcome 1 Glutamine suppositories vs. Butyrate suppositories: Achieved Remission (no recurrence of symptoms)...... 30 Analysis 2.2. Comparison 2 Chronic Pouchitis, Outcome 2 Bismuth Carbomer Foam Enemas vs. Placebo: Improved (PDAI reduction ≥ 3)...... 30 Analysis 2.3. Comparison 2 Chronic Pouchitis, Outcome 3 VSL#3 vs. Placebo: Relapse of Pouchitis (PDAI increase ≥ 2)...... 31 Analysis 3.1. Comparison 3 Prevention of Pouchitis, Outcome 1 VSL#3 vs. Placebo: No Episodes of Acute Pouchitis (PDAI ≥ 7)...... 31 Analysis 3.2. Comparison 3 Prevention of Pouchitis, Outcome 2 VSL#3 vs. No Treatment: No Episodes of Acute Pouchitis (PDAI ≥ 7)...... 32 Analysis 3.3. Comparison 3 Prevention of Pouchitis, Outcome 3 Allopurinol vs. Placebo: No Episodes of Pouchitis (clinical diagnosis)...... 32 WHAT’SNEW...... 32 HISTORY...... 33 CONTRIBUTIONSOFAUTHORS ...... 33 DECLARATIONSOFINTEREST ...... 33 SOURCESOFSUPPORT ...... 33 INDEXTERMS ...... 34

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) i Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. [Intervention Review] Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Stefan D Holubar2, Robert R Cima2, William J Sandborn1, Darrell S Pardi1

1Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN, USA. 2 Division of Colon and Rectal Surgery, Department of Surgery, Mayo Clinic, Rochester, MN, USA

Contact address: Darrell S Pardi, Division of Gastroenterology and Hepatology, Mayo Clinic, 200 First Street SW, Rochester, MN, 55905, USA. [email protected].

Editorial group: Cochrane Inﬂammatory Bowel Disease and Functional Bowel Disorders Group. Publication status and date: New search for studies and content updated (conclusions changed), published in Issue 6, 2010. Review content assessed as up-to-date: 4 October 2009.

Citation: Holubar SD, Cima RR, Sandborn WJ, Pardi DS. Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis. Cochrane Database of Systematic Reviews 2010, Issue 6. Art. No.: CD001176. DOI: 10.1002/14651858.CD001176.pub2.

ABSTRACT Background Pouchitis may occur following ileal pouch-anal anastomosis for chronic ulcerative colitis in approximately 30% of patients. Objectives The primary objective was to determine the efficacy of medical therapies for pouchitis (including antibiotic, probiotic, and other agents) as substantiated by data from randomized controlled trials (RCTs). Search strategy A search for RCTs from 1966 to October 2009 was performed using the MEDLINE, Cochrane Library, EMBASE, Web of Science, and Scopus databases. Selection criteria Randomized controlled treatment or prevention trials of adult patients who underwent ileal pouch-anal anastomosis for ulcerative colitis who subsequently developed pouchitis or were at risk for pouchitis were considered for inclusion. Data collection and analysis Extracted data were converted to 2X2 tables and then synthesized in to a summary statistic using the Peto odds ratio (OR) and [95% confidence intervals], or weighted mean difference (WMD), using RevMan-5 for Mac OS 10.6. Main results Eleven RCTs fulfilled the inclusion criteria and were included in the review. The efficacy of 10 different pharmacologic agents was assessed. For the treatment of acute pouchitis (4 RCTS, 5 agents), ciprofloxacin was more effective at inducing remission than metronidazole. Neither rifaximin nor lactobacillus GG were more effective than placebo, while budesonide enemas and metronidazole were similarly effective, for inducing remission of acute pouchitis. For the treatment and maintenance of remission of chronic pouchitis (4 RCTs, 4 agents), glutamine suppositories were not more effective than butyrate suppositories, and bismuth carbomer foam enemas were not more effective than placebo, while VSL#3 was more effective than placebo in maintaining remission of chronic pouchitis in

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 1 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. patients with chronic pouchitis who achieved remission with antibiotics. For the prevention of pouchitis (3 RCTs, 2 agents), in one study VSL#3 was more effective than placebo while in another study VSL#3 was not more effective than no treatment. Allopurinol was not more effective than placebo, while inulin was more effective than placebo but the results were not clinically signiﬁcant.

Authors’ conclusions

For acute pouchitis, ciproﬂoxacin was more effective than metronidazole, while budesonide enemas and metronidazole were similarly effective. For chronic pouchitis, VSL#3 was more effective than placebo. For the prevention of pouchitis, VSL#3 was more effective than placebo. Larger RCTs are needed to determine the optimal agent(s) for the treatment and prevention of pouchitis.

PLAIN LANGUAGE SUMMARY

Several antibiotic agents, including metronidazole, ciproﬂoxacin, and rifaximin, as well as oral probiotics, may be effective treatments for pouchitis.

Some patients with ulcerative colitis have their colon and rectum removed with construction of a reservoir or pouch (made from a loop of small intestine) to serve in place of the rectum. This is known as an ileal pouch-anal anastomosis (IPAA) surgery. Pouchitis is acute inflammation of the surgically constructed pouch which may cause diarrhea and other problems. The exact cause of pouchitis is not known, but it may be caused by an imbalance in bacteria (similar to an infection) and can be treated by antibiotics, probiotics (bacteria important for the health of the bowel), or other agents that may reduce or prevent inflammation. Metronidazole and Ciprofloxacin (two antibiotics), budesonide enemas (a topical steroid that may decrease inflammation), and oral probiotic therapy with VSL#3 all appear to be effective therapies for acute and/or chronic pouchitis. Current evidence does not support the use of lactobacillus GG (a different probiotic), bismuth (a metal that may be useful in some diarrheal disorders), butyrate and glutamine (two nutrients required by the bowel), allopurinol (a gout medication which may decrease inflammation), or inulin (a non-absorbable sugar which may decrease inflammation). So far the research performed has generally consisted of small studies that were not reproduced, so more research is needed to determine which of these different medications are best for treatment of pouchitis.

BACKGROUND ing villous atrophy, crypt hyperplasia, and chronic inflammatory Pouchitis is an idiopathic chronic inflammatory disease that may cell infiltration (Moskowitz 1986; Shepherd 1987). Patients with occur in the ileal pouch after restorative proctocolectomy with ileal pouchitis can be classified according to disease activity and symp- pouch-anal anastomosis (IPAA) or Kock pouch (Sandborn 1994a). tom duration (Sandborn 1997; Pardi 2009). Disease activity can Although pouchitis can occur after construction of pouches for be classified as: remission (no active pouchitis); mild-to-moder- either chronic ulcerative colitis or familial adenomatous polypo- ately active (increased stool frequency, urgency, infrequent incon- sis, pouchitis occurs much less frequently in the latter, evidence tinence); or severely active (hospitalization for dehydration, fre- that pouchitis is less related to the structure of the pouch, but quent incontinence). Symptom duration can be classified as: acute is a function of the patients’ underlying immune dysregulation (≤ 4 weeks); or chronic (> 4 weeks). The Pouchitis Disease Ac- interacting with the pouch. The diagnosis of pouchitis is sug- tivity Index (PDAI) is a 19 point index of pouchitis activity based gested by variable clinical symptoms including increased stool fre- on both clinical symptoms and endoscopic and histologic findings quency, rectal bleeding, abdominal cramping, fecal urgency and (Sandborn 1994b). Active pouchitis is defined as a PDAI ≥ 7 and tenesmus, incontinence, and fever. A clinical diagnosis should remission is defined as a PDAI < 7. Clinical response to treatment ideally be confirmed by endoscopy and mucosal biopsy of the can also be quantified by reduction in the PDAI. pouch. Endoscopic examination shows inflammatory changes that may include mucosal edema, granularity, contact bleeding, loss At the time of the original Cochrane review of this topic (Sandborn of vascular pattern, hemorrhage, and ulceration (Di Febo 1990; 1998), pouchitis was a relatively new disease; criteria for diagno- Moskowitz 1986). Histologic examination shows acute inflam- sis, classification, and measurement of disease activity had only mation, including neutrophil infiltration and mucosal ulceration recently been proposed and adopted. At that time, the previous superimposed on a background of chronic inflammation includ- lack of consensus on these issues had hampered the design and

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 2 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. conduct of randomized, double-blind, placebo-controlled trials, active agents. Studies which reported duplicate results were ex- and as a result the medical therapy for pouchitis was largely em- cluded, as was a single patient “n-of-1” placebo-controlled trial. piric. The medical therapies that had been reported to be of ben- In addition a cross-over trial of oral metronidazole 400 mg three efit in uncontrolled trials include metronidazole, ciprofloxacin times daily (Madden 1994) and a cross-over trial of oral inulin and other quinolones, amoxicillin/clavulanic acid, erythromycin, 24 g daily (Welters 2002) were excluded because data were not tetracycline, mesalamine enemas, sulfasalazine, oral mesalamine, available before the first cross-over. conventional corticosteroid enemas, budesonide enemas, oral corticosteroids, cyclosporine enemas, azathioprine, short chain fatty acid (SCFA) enemas or suppositories, allopurinol, bismuth carbomer enemas, and bismuth subsalicylate (Sandborn 1994a). At Types of participants the time of the previous review, only four small placebo-controlled Adult patients (age ≥ 18 years) with pouchitis variably defined by trials have been performed, evaluating treatment with metron- 1) solely clinical criteria; 2) clinical criteria in combination with idazole [(Madden 1994) this study has been excluded from the endoscopic and histologic criteria; or 3) the PDAI. Pouchitis was present review as described below), oral probiotic bacteria (VSL#3) categorized as active (defined clinically as the presence of mild-to- (Gionchetti 2000), bismuth carbomer enemas (Tremaine 1997), severe symptoms or by a PDAI ≥ 7) or in remission (absence of and glutamine versus butyrate suppositories (Wischmeyer 1993). symptoms or by a PDAI < 7). Pouchitis was further categorized Since then, an additional 9 randomized controlled trials have as acute (symptom duration ≤ 4 weeks) or chronic (symptom been performed, including 4 studies of treatments for acute pou- duration > 4 weeks). chitis: ciprofloxacin versus metronidazole (Shen 2001), budesonide versus metronidazole (Sambuelli 2002), rifaximin versus placebo (Isaacs 2007), Lactobacillus GG versus placebo (Kuisma 2003); 1 new study of VSL#3 versus placebo for chronic pouchitis Types of interventions (Mimura 2004); and 3 studies of pouchitis prevention: VSL#3 The following interventions were considered for inclusion: versus placebo and VSL#3 versus no treatment (Gionchetti 2003; 1. Oral metronidazole 20 mg/kg/day (Shen 2001), or 500 mg Pronio 2008), allopurinol versus placebo (Joelsson 2001); a study twice daily (Sambuelli 2002); of inulin versus placebo (Welters 2002) was excluded as described 2. Oral VSL-3 probiotic bacterial formulation containing 300 bil- below. This updated Cochrane review will examine the results lion bacteria per gram of viable lyophilized bacteria with 4 strains from a total of 11 randomized controlled trials to determine which of lactobacilli (L. acidophilus, L. delbrueckii subspecies Bulgaricus, of the currently utilized empiric medical therapies for pouchitis L. plantarum, L. casei), 3 strains of Bifidobacterium (B. infantis, B. can be substantiated with valid data from controlled trials. longum, B. breve) and 1 strain of Streptococcus salivarius subspecies Thermophilus; 6 grams per day (Gionchetti 2000), 3 grams per day (Gionchetti 2003), 3 grams twice daily (Mimura 2004), 3 OBJECTIVES grams once per day (Pronio 2008); 3. Bismuth carbomer foam enemas containing 513 mg bismuth The primary objective was to determine the efficacy of various oral citrate (270 mg metallic bismuth) complexed with carbomer (a and topical (per rectum) medical therapies (including antibiotics, synthetic high-molecular weight polymer of acrylic acid cross probiotics, and modulators of inflammation) for the treatment of linked with poly alkenyl polyether) administered once nightly acute pouchitis, chronic pouchitis, and the prevention of pouchitis (Tremaine 1997); as assessed by randomized controlled clinical trials. 4. Glutamine suppositories containing 1 gram of L-glutamine in a polyethylene glycol base administered twice daily (Wischmeyer 1993); METHODS 5. Butyrate suppositories containing 40 mmol sodium butyrate in a polyethylene glycol base administered twice daily (Wischmeyer 1993); Criteria for considering studies for this review 6. Ciprofloxacin 1000 mg/day (Shen 2001); 7. Rifaximin 400 mg orally three times daily (Isaacs 2007); 8. Lactobacillus GG in two gelatine capsules orally twice daily versus microcrystalline cellulose-only gelatine placebo capsules ( Types of studies Kuisma 2003); Randomized, controlled trials were considered for inclusion. 9. Budesonide enema 2 mg/100 mL at bedtime plus oral placebo Study designs could include parallel arm placebo-controlled tri- tablets (Sambuelli 2002); and als, crossover placebo-controlled trials, and trials comparing two 10. Allopurinol 100 mg twice daily (Joelsson 2001).

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 3 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Types of outcome measures Eligible articles were reviewed in triplicate (SDH, RRC, and WJS) The primary outcome was the proportion of patients with clinical and the results of the primary research trials were abstracted into improvement or remission of pouchitis. The exact definition of specially designed data extraction forms. The proportion of pa- improvement and remission varied from study to study limiting tients who had clinical improvement in the active treatment and the ability to make comparisons across studies. However, the def- control groups of each study was derived from life tables, survival inition of improvement or remission used in each study was used curves, or where possible, by calculating life tables from the data for extraction of data from the individual studies for the purposes provided. of this systematic review. Risk of Bias: Three authors (SDH, RRC, WJS) independently assessed methodological quality, using the Cochrane risk of bias tool as described in the Cochrane Handbook for Systematic Reviews of Interventions Search methods for identification of studies (Higgins 2008). Briefly, an assessment was made of the method of Original review: A computer-assisted search of the on-line bib- allocation generation (i.e. was the allocation sequence adequately liographic database MEDLINE was carried out so as to identify generated?), allocation concealment (i.e. was allocation adequately potentially relevant studies published between 1966 and Decem- concealed?), blinding (i.e. was knowledge of the allocated interven- ber 1997. The Medical Subject Heading (MeSH) terms “Coli- tion adequately prevented during the study?), incomplete outcome tis, Ulcerative/su [surgery]” or Ileitis/et [etiology]“ or ”Ileitis/th data (i.e. were incomplete outcome data adequately addressed?); [therapy]“ or ”Proctocolectomy, Restorative/ae [adverse events]“, and selective outcome reporting (i.e. are reports of the study free of and key words: ”Antibiotics“ or ”Metronidazole“ or ”Ciproflo- suggestion of selective outcome reporting?). A judgement of ’Yes’ xacin“ or ”Amoxicillin“ or ”Erythromycin“ or ”Tetracycline“ or indicates low risk of bias, ’No’ indicates high risk of bias, and ’Un- ”Sulfasalazine“ or ”Aminosalicylic acids“ or ”Glucocorticoids“ or clear’ indicates unclear or unknown risk of bias. Disagreements ”Azathioprine“ or ”6-Mercaptopurine“ or ”Cyclosporine“ or ”Al- were resolved by consensus. lopurinol“ or “Fatty Acids, Volatile”, or “Butyric Acids” or “Glu- Statistical Analysis: tamine” or “Probiotic” or “Bismuth” were used to perform searches Data extracted from the original research articles were converted of the database. Manual searches of the reference lists from the into individual 2X2 tables (response versus no response X med- potentially relevant studies were performed in order to identify ical therapy versus placebo, or medical therapy versus medical additional studies that may have been missed using the computer- therapy). All results were tabulated on an intention-to-treat basis assisted search strategy. Proceedings from the American Gastroen- if available; otherwise modified-intention-to-treat or per-protocol terological Association, American Society of Colon and Rectal results are reported. Where available, individual 2X2 tables for Surgery, and the British Society of Gastroenterology were also strata within studies were also used. The 2X2 tables were synthe- manually searched from 1965-1997 in order to identify unpub- sized into a summary test statistic using an un-pooled (or pooled lished studies. where appropriate) Peto odds ratio and 95% confidence intervals The updated review: A computer-assisted search for randomized as described by Cochran and Mantel and Haenszel. A fixed effects controlled trials from 1966 to October 2009 was performed using model was used for the pooling of data. the MEDLINE, Cochrane Library, EMBASE, Web of Science, and Scopus databases. Search terms included “pouchitis”, “remission induction”, “proctocolectomy, restorative”, “drug therapy”, “colitis, ulcerative” and “postoperative complication”. This search RESULTS was limited to clinical trials (phase 1, 2, 3, or 4) or comparative studies or meta-analyses or multicenter studies or practice guidelines or randomized controlled trial. A text word search was also Description of studies employed. See: Characteristics of included studies; Characteristics of excluded studies. Two hundred and ninety-seven abstracts were reviewed and a total Data collection and analysis of 19 RCTs were identified that were considered for inclusion. A Study Selection: total of 11 trials satisfied the inclusion criteria (Wischmeyer 1993; Potentially relevant articles were reviewed in an independent un- Tremaine 1997; Gionchetti 2000; Joelsson 2001; Shen 2001; blinded fashion by two authors (SDH, DSP) to determine if they Sambuelli 2002; Gionchetti 2003; Kuisma 2003; Mimura 2004; met the criteria specified above. The studies were then rated as Isaacs 2007; Pronio 2008). Eight studied were excluded (see table being eligible, ineligible, or as having insufficient information to of excluded studies), including a single patient “n-of-1” trial for make a judgment as to eligibility and were then excluded. active chronic pouchitis (metronidazole vs. placebo), two random- Data Collection: ized trials of alicaforsen enema vs. mesalazine enema for ulcerative

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 4 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. colitis (not pouchitis), a non-randomized sub-study (VSL#3), and two non-randomized follow-up studies of Cultura (a fermented milk probiotic), a randomized cross-over trial of metronidazole and a randomized cross-over trial of inulin both without outcome data from before the ﬁrst cross-over.

Risk of bias in included studies The Cochrane risk of bias tool indicates that the risk of bias was low for 3 of the 11 included studies (See Figure 1). The risk of bias in the other studies was high for blinding in 2 studies (open studies) and allocation concealment in 1 study. Some quality items were unclear in 10 studies (due to inadequate descriptions of methods used for sequence generation, allocation concealment and blinding and possible selective reporting).

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 5 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 1. Methodological quality summary: review authors’ judgements about each methodological quality item for each included study.

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 6 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Effects of interventions The eleven included studies had a variety of designs (although all were randomized controlled trials) and categories of outcomes. For analysis, these eleven studies were categorized according to the type of study 1) treatment of acute pouchitis, 2) treatment, or maintenance of remission, for chronic pouchitis, and 3) prevention of pouchitis. 1. Treatment of Acute Pouchitis - 4 Studies, 5 agents Ciproﬂoxacin versus metronidazole (Figure 2)

Figure 2. Forest plot of comparison: 1 Acute Pouchitis, outcome: 1.1 Ciproﬂoxacin vs. Metronidazole: Achieved Remission (PDAI < 7).

Shen 2001 was a randomized, non-blinded, parallel-arm trial of ciprofloxacin compared to metronidazole for the induction of remission of acute pouchitis. All patients had a PDAI score > 7 at entry, and there was no statistically significant difference in mean baseline PDAI scores between groups. A total of 16 patients were enrolled; 7 patients received ciprofloxacin and 9 received metronidazole for two-weeks. After two weeks of therapy, both groups had significantly lower PDAI scores compared to baseline. Overall 7/7 (100%) of the ciprofloxacin group achieved remission (defined as PDAI score < 7), compared with 6/9 (67%) of the metronidazole group (P = 0.008). The Peto odds ratio for induction of remission by ciprofloxacin compared to metronidazole was 14.39 (95% CI 2.00 to 103.76). Budesonide enema versus metronidazole (Figure 3)

Figure 3. Forest plot of comparison: 1 Acute Pouchitis, outcome: 1.2 Budesonide enema vs. Metronidazole: Achieved Remission (PDAI < 7).

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 7 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Sambuelli 2002 was a randomized, double-blind, double dummy, placebo-controlled, parallel-arm trial of budesonide enemas plus age (metronidazole group older, P = 0.044). Overall, 7/12 (58%) placebo tablets compared to placebo enemas and metronidazole of the budesonide group and 7/14 (50%) of the metronidazole tablets for the induction of remission of acute pouchitis (Sambuelli group had a reduction in the PDAI score ≥ 3 (58% versus 50%, 2002). At entry patients had a PDAI score ≥ 7 and no treatment P = 0.67). In addition, 6/12 (50%) in the budesonide group and during the previous month. A total of 26 patients were enrolled; 12 6/14 (43%) in the metronidazole achieved remission as deﬁned received budesonide enemas plus placebo tablets and 14 received by a PDAI < 7 (50% versus 43%, P = 0.72). The Peto odds ratio placebo enema plus metronidazole tablets for 6 weeks. Of note, for induction of remission of pouchitis by budesonide compared the randomization resulted in un-balanced groups with respect to to metronidazole was 1.32 (95% CI 0.29 to 6.01). Rifaximin versus placebo (Figure 4)

Figure 4. Forest plot of comparison: 1 Acute Pouchitis, outcome: 1.3 Rifaximin vs. Placebo: Achieved Remission (PDAI < 7).

Isaacs 2007 was a randomized, double-blind, placebo-controlled, parallel-arm, multi-center trial of rifaximin tablets compared to placebo for the induction of remission of acute pouchitis. A total of 18 patients were enrolled; 8 patients received rifaximin and 10 patients received placebo for 4 weeks. One patient in the placebo group did not undergo post-treatment evaluation and was therefore excluded from the modified intention-to-treat analysis. Clin- ical remission (defined as a PDAI score < 7) was achieved in 2/8 (25%) patients in the rifaximin group and 0/9 (0%) in the placebo group (25% vs 0%, P = 0.1). The modified intention-to-treat Peto odds ratio for induction of remission of pouchitis by rifaximin compared to placebo was 9.65 (95% CI 0.55 to 169.75); the un- modified intention-to-treat Peto odds ratio for induction of remission by rifaximin compared to placebo was 10.92 (95% CI 0.62 to 193.55). Lactobacillus GG versus placebo (Figure 5)

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 8 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 5. Forest plot of comparison: 1 Acute Pouchitis, outcome: 1.4 Lactobacillus GG vs. Placebo: Improved (PDAI reduction ≥ 3).

Kuisma 2003 was a randomized, double-blind, placebo-controlled, parallel arm of Lactobacillus GG compared to placebo for was not provided; hence only per protocol analysis results were inducing clinical response of acute pouchitis. All patients had a pre- available. A total of 10 patients received Lactobacillus GG and 10 vious history of pouchitis and endoscopic inﬂammation. Patients patients received placebo for 3 months. Clinical response (deﬁned who had received antibiotics within one month were excluded, as as a PDAI score reduction of ≥ 3) occurred in 1/10 (10%) patients were active chronic pouchitis (intermittent or continuous medical in the Lactobacillus group and 0/10 (0%) patients in the placebo treatment) or symptoms [fever, bleeding, and increased number of group (10% vs. 0%, P = 0.32). The Peto odds ratio for induction stools (> 2/day within 1 week)] indicative of an acute exacerbation of remission by Lactobacillus GG compared to placebo was 7.39 of pouchitis. A total of 22 patients were enrolled; 2 patients were (95% CI 0.15 to 372.38). excluded after enrollment for severe endoscopic pouchitis requir- 2. Treatment of Chronic Pouchitis - 4 studies, 4 agents ing immediate antibiotics. The assigned group of these 2 patients Glutamine versus Butyrate Suppositories (Figure 6)

Figure 6. Forest plot of comparison: 2 Chronic Pouchitis, outcome: 2.1 Glutamine suppositories vs. Butyrate suppositories: Achieved Remission (no recurrence of symptoms).

Wischmeyer 1993 was a randomized, double-blind trial of glutamine compared to butyrate suppositories for maintenance of medical therapy. A total of 19 patients were enrolled into a single remission in patients with chronic pouchitis after withdrawal of study; 10 received glutamine suppositories and 9 received butyrate other suppressive medical therapy. All patients had chronic pouch- suppositories. The relapse rate was 4/10 (40%) for glutamine and itis (deﬁned as at least 4 episodes of pouchitis within one year and 6/9 (67%) for butyrate. Therefore, the remission rate was 6/10 the need for suppressive medications for at least 6 of the prior 12 (60%) for glutamine and 3/9 (33%) for butyrate (60% vs. 33%, months) and all discontinued suppressive medical therapy 7 days P = 0.26). The Peto odds ratio for maintenance of remission with prior to study entry. At entry, patients could have either active or glutamine compared to butyrate at 3 weeks was 2.75 (95% CI remitted pouchitis following the recent withdrawal of suppressive 0.48 to 15.94). Bismuth Carbomer Foam Enemas versus placebo (Figure 7)

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 9 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Figure 7. Forest plot of comparison: 2 Chronic Pouchitis, outcome: 2.3 Bismuth Carbomer Foam Enemas vs. Placebo: Improved (PDAI reduction ≥ 3).

Tremaine 1997 was a randomized, double-blind, placebo-controlled, parallel-arm trial of bismuth carbomer enemas for induction of clinical response in active chronic pouchitis. All patients had active, chronic pouchitis (deﬁned as a PDAI score ≥ 7 points and failure or intolerance to treatment with antibiotics and other medical therapies). A total 40 patients were enrolled; 20 received bismuth carbomer foam enemas and 20 received placebo. The response rate, as measured by a decrease in the PDAI score of ≥ 3 points, was 9/20 (45%) for bismuth carbomer foam enemas and 9/20 (45%) for placebo (P = 1.0). The Peto odds ratio for response to bismuth carbomer enemas compared to placebo was 1.00 (95% CI 0.29 to 3.42). Oral Probiotic Bacteria (2 studies, Figure 8):

Figure 8. Forest plot of comparison: 2 Chronic Pouchitis, outcome: 2.4 VSL#3 vs. Placebo: Relapse of Pouchitis (PDAI increase ≥ 2).

Gionchetti 2000 was a randomized, double-blind, placebo-controlled, parallel-arm trial of probiotic bacteria (VSL#3) to main- therapy compared to 20/20 (100%) for placebo. Therefore, the tain remission in chronic pouchitis. All patients were in remission rates of maintenance of remission was 17/20 (85%) for probiotic (deﬁned as a PDAI score of 0 following open label induction with therapy and 0/20 (0%) for placebo (P < 0.0001). antibiotics) and had chronic pouchitis (history of requiring con- Mimura 2004 was a randomized, placebo-controlled, multicen- tinuing medical suppressive therapy and recurrence within a few ter, parallel-arm trial of probiotic bacteria (VSL#3) to maintain weeks of discontinuing suppression). In total, 40 patients were remission in chronic pouchitis. All patients had pouchitis at least enrolled; 20 received probiotic bacteria and 20 received placebo. twice in the previous year or requiring continuous antibiotics, as- The relapse rate, as measured by an increase ≥ 2 points in the 6 sociated with a PDAI ≥ 7, in whom remission was induced by point clinical portion of the PDAI, was 3/20 (15%) for probiotic four weeks of combined metronidazole and ciproﬂoxacin. Active

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 10 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. refractory or recurrent pouchitis was deﬁned as a total PDAI score of >7 and a history of relapsing or persistent pouchitis. A total of 36 patients were enrolled for 1 year of treatment; 20 patients received VSL#3 and 16 patients received placebo. Remission at one year, as deﬁned by a clinical PDAI score ≤ 2 and endoscopic PDAI ≤ 1, was maintained in 17/20 (85%) in the VSL#3 group and 1/16 (6%) in the placebo group (P < 0.0001). The pooled Peto odds ratio for these two studies for the combined rate of maintenance of remission with probiotic bacteria compared to placebo (97% versus 3%, P < 0.0001) was 25.39 (95% CI 10.37 to 62.17). The number needed to treat with oral probiotic therapy to prevent one additional relapse was 2. 3. Prevention of Pouchitis - 3 studies, 2 agents Oral Probiotic Bacteria (2 studies, Figure 9, Figure 10):

Figure 9. Forest plot of comparison: 3 Prevention of Pouchitis, outcome: 3.1 VSL#3 vs. Placebo: No Episodes of Acute Pouchitis (PDAI ≥ 7).

Figure 10. Forest plot of comparison: 3 Prevention of Pouchitis, outcome: 3.2 VSL#3 vs. No Treatment: No Episodes of Acute Pouchitis (PDAI ≥ 7).

Gionchetti 2003 was a randomized, double-blind, placebo-controlled, parallel-arm trial which assessed the efﬁcacy of oral pro- the placebo group had an episode of pouchitis (no episodes 80% biotic therapy with VSL#3 for the prevention of pouchitis. All versus 60%, P = 0.03). The Peto odds ratio for prevention of pou- patients were randomized within one week of ileostomy closure chitis by VSL#3 compared with placebo was 4.76 (95% CI 1.16 after IPAA. A total of 40 patients were enrolled for 12 months of to 19.56). treatment; 2/10 (20%) of the VSL#3 group and 8/20 (40%) of Pronio 2008 was a randomized, open-label, parallel-arm trial

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 11 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. which assessed the efﬁcacy of oral probiotic therapy with VSL#3 compared with no treatment for the prevention of pouchitis. All patients were taking no medication at entry. Exclusion criteria included patients with active pouchitis (PDAI ≥ 7), perianal disease (including abscess, ﬁssure, stricture, or anal sphincter weakness), and pregnant or lactating women. A total of 31 patients were enrolled for 12 months. A total of 0/16 patients (0%) in the VSL#3 group and 1/12 (8.3%) in the control group developed pouchitis (no pouchitis 100% versus 92%, P = 0.24). The Peto odds ratio for prevention of pouchitis by VSL#3 compared with no treatment was 10.31 (95% CI 0.20 to 541.25). Allopurinol versus placebo (Figure 11)

Figure 11. Forest plot of comparison: 3 Prevention of Pouchitis, outcome: 3.3 Allopurinol vs. Placebo: No Episodes of Pouchitis (clinical diagnosis).

Joelsson 2001 was a randomized, double-blind, multi-center, placebo-controlled, parallel-arm study assessing the efficacy of al- odds ratio. As an example, in the comparison of oral ciprofloxacin lopurinol for the prevention of pouchitis. All patients were ran- versus metronidazole, we are 95% confident that the true pooled domized 1 week after IPAA surgery. A total of 184 patients were odds ratio is between 2.00 and 103.76. This interval describes a enrolled for 24 months; 94 patients received allopurinol and 90 statistically significant benefit in favour of oral ciprofloxacin over placebo. Of these patients, 116 (36%) completed the study. At one metronidazole, however, the estimate of the therapeutic advantage year, pouchitis, as defined by a functional score, occurred in 19/ of oral ciprofloxacin over metronidazole is imprecise. 62 (31%) patients in the allopurinol group, and in 15/54 (28%) Additionally for each comparison, with the exception of VSL#3 of patients in the placebo group (no pouchitis 69% versus 72%, P for chronic pouchitis, only one trial was eligible. Therefore, the = 0.74). A per protocol analysis showed the Peto odds ratio for the generalizability and external validity of these results must be ques- prevention of pouchitis by allopurinol compared to placebo was tioned. The reader must exercise clinical judgement, and ensure 0.87 (95% CI 0.39 to 1.94); on an intention-to-treat basis the Peto the patient populations (those being generalized to, and those in odds ratio for the prevention of pouchitis by allopurinol compared the trials) are similar before these findings can be implemented to placebo was 1.10 (95% CI 0.62 to 1.97). into clinical practice. Two cross-over studies which were excluded (because of lack of data prior to cross-over) merit further discussion. Madden 1994 was DISCUSSION a randomized, double-blind, placebo-controlled, crossover trial of metronidazole for induction of clinical response in active chronic The results presented in this review must be interpreted with ex- pouchitis. All patients had active pouchitis (defined as chronic, treme caution given the small numbers of trials and patients evalu- recurrent clinical symptoms, > 6 stools/24 hours, and endoscopic ated for any one comparison. For each comparison, the confidence inflammation) at entry. In total 13 patients were enrolled, and 11 intervals about the pooled odds ratio were wide. This indicates patients completed both arms of the crossover trial. The overall a moderate amount of uncertainty in the estimate of the pooled response, as measured by a reduction in stool frequency of at least 3

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 12 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. stools/24 hours, was 8/11 (73%) for metronidazole compared to 1/ may indirectly set the stage for pouchitis to the extent that “colon- 11 (9%) for placebo (P = 0.003). The Peto odds ratio for response like” ileal mucosa may be more susceptible to a recurrence of UC- to metronidazole compared to placebo was 12.34 (95% CI 2.34 to like inflammation. Strategies directed towards reducing fecal con- 64.94). The number needed to treat with metronidazole to prevent centrations of anaerobic bacteria through the use of antibiotics, or one additional relapse was 2. Welters 2002 was a randomized, altering the relative balance of anaerobes and other bacteria using double-blind, placebo-controlled, cross-over study assessing the probiotic bacteria therapy, may be useful in treating pouchitis. efficacy of inulin for the prevention of pouchitis. Patients with IPAA for CUC or FAP were included. Patients with a history of Using this rationale, clinicians have empirically treated patients pouchitis were not excluded, none of the included patients had with antibiotics and have observed that most patients with pou- overt clinical pouchitis at the time of the study, and none were chitis experience clinical improvement. In the absence of ade- taking pouchitis-related medications. A total of 20 patients were quate and well controlled trials, antibiotics have become the de enrolled to receive inulin followed by placebo, or placebo followed facto “standard medical therapy” for pouchitis. In the past, the by insulin, for 3 weeks each. One patient who started in the placebo most commonly used antibiotic for pouchitis was metronidazole group dropped out due to lactose intolerance. Overall, there was (Sandborn 1994a; Moskowitz 1986; Zuccaro 1989; Lohmuller no difference in pouchitis prevention, as measured by the final 1990; Svaninger 1993; Kelly 1983; Boerr 1995; Boerr 1996; mean PDAI score, which was 1.42 for inulin and 1.66 for the Kmiot 1993; Hurst 1996; Tytgat 1988; Scott 1989; Bonello 1981; placebo. However, the modified intention-to-treat weighted mean Klein 1983; Knobler 1986). More recently, ciprofloxacin has be- difference (MD) in PDAI for the prevention of pouchitis by inulin come the more commonly used antibiotic (Hurst 1996). Amox- compared to placebo was 1.34 (P < 0.001, 95% CI 0.85 to 1.83). icillin/clavulanic acid, erythromycin, and tetracycline have also Although this was statistically significant, this difference was not been reported to be of benefit (Scott 1989). Most patients with clinically significant. pouchitis initially respond to ciprofloxacin at a dose of 500 to 1000 mg/day or metronidazole at doses of 750 to 1500 mg/day. Based on the results of this analysis the following conclusions may Symptomatic improvement usually occurs within 1 to 2 days. Pa- be drawn. For the treatment of acute pouchitis ciprofloxacin ap- tients with relapsing or continuous pouchitis may require chronic pears to be more effective than metronidazole, metronidazole and maintenance ciprofloxacin therapy with doses ranging from 250 budesonide enemas were equally effective, while neither Lacto- mg every 3rd day up to 1000 mg/day. Based on the Sambuelli bacillus GG nor Rifaximin were more effective than placebo. For 2002 study, budesonide enemas and metronidazole appear to have the treatment of chronic pouchitis VSL#3 was more effective than similar efficacy (58% and 50%). Although the authors of that placebo in maintaining remission; glutamine suppositories were study suggested that the enema preparation may be more tolerable not more effective than butyrate suppositories for maintenance of than oral metronidazole, which is known to produce a metallic remission, and bismuth carbomer foam enemas were not more ef- taste when it comes in contact with the tongue, many patients pre- fective than placebo for induction of remission. For the prevention fer oral administration to administration per anus. Nonetheless, of pouchitis VSL#3 was more effective than placebo, but not more the results of that study suggest that additional study of budes- effective than no treatment hence the efficacy of VSL#3 for pre- onide, whether administered orally or as a suppository, is war- vention is questionable. Allopurinol was not more effective than ranted. There are two uncontrolled reports supporting this con- placebo for prevention of pouchitis. clusion (Gionchetti 2007; Chopra 2006).

To a limited extent, this empiric approach to the manipulation of pouch bacterial flora has been validated by controlled trials AUTHORS’ CONCLUSIONS of metronidazole and the oral probiotic formulation VSL#3 in Implications for practice patients with chronic pouchitis. Madden 1994 reported that the After IPAA, the primary function of the terminal ileum changes odds ratio for a decrease in stool frequency following metronida- from absorption to storage, and bacterial overgrowth occurs with zole therapy compared with placebo in patients with active chronic bacterial concentrations increasing to levels which are interme- pouchitis was 12.34 (95% CI 2.34 to 64.94). The single patient diate between an end ileostomy and the colon (Sandborn 1995; “n-of-1” study of metronidazole for active chronic pouchitis also Shepherd 1989). However, there is no correlation between fecal suggested benefit (McLeod 1986). However, despite the signifi- bacterial concentrations and histologic changes of acute inflamma- cant decrease in stool frequency, there was no difference between tion (Sandborn 1995; Shepherd 1989), demonstrating that pou- metronidazole and placebo for changes of endoscopic inflamma- chitis is not directly related to the quantity of bacterial overgrowth. tion, general well-being, stool blood, and histologic inflammation However, anaerobic bacterial overgrowth of the pouch is associ- (Madden 1994). The pooled odds ratio for maintaining remission ated with transformation of the ileal mucosa to a “colon-like” mor- in patients with chronic pouchitis during administration of the phology (villous atrophy, chronic inflammatory cell infiltration) oral probiotic bacteria formulation VSL#3 was 25.39 (95% CI (Sandborn 1995; Natori 1992). Thus, pouch bacterial overgrowth 10.37 - 62.17) (Gionchetti 2000; Mimura 2004). However, when

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 13 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. examining the efficacy of VSL#3 for prevention of pouchitis, con- mesalamine for the treatment of left-sided UC (Pullan 1993). flicting results were obtained, with VSL#3 being more efficacious Given the proven benefit of bismuth in traveller’s diarrhea, and its than placebo (Gionchetti 2003) but not more efficacious than potential benefit in UC and collagenous colitis, therapeutic trials no treatment (Pronio 2008). The reason for this discrepancy is of bismuth in patients with acute pouchitis seemed reasonable. An not clear. Thus, while antibiotic and probiotic bacterial therapies uncontrolled study of bismuth complexed to carbomer suggested appear to be efficacious, antibiotic selection, dose, and duration beneficial effects for both inducing and maintaining remission in of therapy remain empiric. Ingestion of other lactobacillus-based patients with chronic pouchitis (Gionchetti 1997) while another formulations (such as Lactobacillus GG) do not appear to be as uncontrolled study suggested that bismuth subsalicylate may also effective as VSL#3 probiotic therapy. At the present time, based on be of benefit in chronic pouchitis (Tremaine 1998). In contrast, the limited controlled clinical trial data available, it is reasonable a placebo-controlled trial with the bismuth carbomer formula- to treat pouchitis with antibiotics and perhaps probiotic bacterial tion in patients with active chronic pouchitis showed no benefit therapy (at least for chronic pouchitis), but additional controlled (Tremaine 1997). It is possible that the xanthan gum placebo in trials are needed to optimize therapeutic recommendations. this study was actually a positive control (Parihar 1997), and another uncontrolled study suggested that bismuth subsalicylate may In the well-functioning ileal pouch, the bacterial flora produce be of benefit in chronic pouchitis (Tremaine 1998). Additional SCFAs including acetate, propionate, and butyrate at concentra- controlled trials using an inactive placebo are needed to determine tions similar to those in the colon of healthy controls, and in- with certainty whether or not bismuth is effective for pouchitis. creased compared to stomal SCFA concentrations in ileostomy pa- However, based on the currently available controlled data, bismuth tients (Sandborn 1995; Nasmyth 1989). In the colon, SCFAs are cannot be advocated as a therapy for pouchitis. Finally, allopuri- the preferred fuel for colonic epithelial cell metabolism, whereas nol (Joelsson 2001) did not appear to be clinically effective for in the small bowel, glutamine is the preferred fuel for enterocyte the prevention of pouchitis. An initial uncontrolled prospective metabolism. Some (Clausen 1992; Wischmeyer 1993) but not all trial (Levin 1992) indicated that allopurinol, a oxygen-related free (Sandborn 1995) studies have reported that patients with pouch- radical scavenger, may be efficacious in decreasing inflammation itis have significantly lower fecal concentrations of SCFAs than hypothesized to be related to chronic ischemia and subsequent free patients with well-functioning IPAAs, perhaps as a result of dilu- radical damage; however the results of the Joelsson 2001 trial, the tion (Clausen 1992). Strategies directed at replacing “nutritional largest randomized trial of any treatment for any form of pouchitis deficits” of SCFAs or glutamine in the ileal pouch by adminis- (184 patients randomized) have refuted this hypothesis. tering SCFA or glutamine enemas or suppositories may be useful in treating pouchitis. However, small uncontrolled studies of Implications for research SCFA enemas have, for the most part, shown low overall clinical response rates, suggesting that SCFA enemas are not bene- The medical therapy of pouchitis remains largely empiric, and ficial for pouchitis (De Silva 1989; Tremaine 1994; den Hoed additional multicenter, randomized, double-blind, placebo-con- 1996). This lack of efficacy of SCFA-based therapies is supported trolled, dose-ranging trials of antibiotics, probiotic bacteria, and by the Welters 2002 study, which examined the use of inulin, a other therapies known to be beneficial in UC such as mesalamine sugar which is fermented to SCFA, which was not clinically ef- and corticosteroids are needed. Trials of VSL#3 compared to ci- fective for the prevention of pouchitis; although it did produce a profloxacin or metronidazole for the treatment and maintenance statistically significant reduction in the mean PDAI scores (1.34 of remission of chronic pouchitis are warranted, and perhaps ad- WMD). Wischmeyer 1993 reported that the pooled odds ratio ditional trials of rifaximin and inulin. In addition trials of newer for maintenance of remission in patients with chronic pouchitis agents, such as biologic therapies, need to be performed in order by glutamine suppositories compared with butyrate suppositories to establish the true efficacy of these agents prior to widespread was 2.75 (95% CI 0.48 to 15.94). It is not clear from this single clinical application. In future trials, treatment indications such as controlled trial whether glutamine and butyrate suppositories are active acute or chronic pouchitis and maintenance of remission equally effective or equally ineffective, but considering the low for acute or chronic pouchitis should be clearly defined using the overall response rates observed during open therapy with SCFA gold standard PDAI definitions. enemas, it appears that “nutritional replacement” with SCFA, butyrate, or glutamine is probably not beneficial for pouchitis and cannot be advocated as standard therapy. ACKNOWLEDGEMENTS

Bismuth has both anti-microbial and anti-diarrheal properties, Funding for the IBD/FBD Review Group (October 1, 2005 - and has been useful for the treatment of traveller’s diarrhea and September 30, 2010) has been provided by the Canadian Insti- collagenous colitis. A randomized, double-blind, controlled trial tutes of Health Research (CIHR) Knowledge Translation Branch; suggested that bismuth citrate may have efﬁcacy comparable to the Canadian Agency for Drugs and Technologies in Health

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 14 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. (CADTH); and the CIHR Institutes of Health Services and Pol- icy Research; Musculoskeletal Health and Arthritis; Gender and Health; Human Development, Child and Youth Health; Nutri- tion, Metabolism and Diabetes; and Infection and Immunity. Miss Ila Stewart has provided support for the IBD/FBD Review Group through the Olive Stewart Fund. Miss Patricia J. Erwin has provided technical support in creating the electronic search pattern and is an employee of Mayo Clinic, Rochester MN.

REFERENCES

References to studies included in this review dummy, controlled trial. Aliment Pharmacol Ther 2002;16(1): 27–34. Gionchetti 2000 {published data only} Shen 2001 {published data only} Gionchetti P, Rizzello F, Venturi A, Brigidi P, Matteuzzi D, Shen B, Achkar JP, Lashner BA, Ormsby A, Remzi F, Bevins CL, et Bazzocchi G, et al.Oral bacteriotherapy as maintenance treatment al.Ciprofloxacin is more effective in treating acute pouchitis than in patients with chronic pouchitis: a double-blind, placebo- metronidazole: A randomized clinical trial. Gastroenterology 2001; controlled trial. Gastroenterology 2000;119(2):305–9. 120(5 Suppl 1):A453. Gionchetti 2003 {published data only} Tremaine 1997 {published data only} Gionchetti P, Rizzello F, Helwig U, Venturi A, Lammers KM, Tremaine WJ, Sandborn WJ, Wolff BG, Carpenter HA, Brigidi P, et al.Prophylaxis of pouchitis onset with probiotic Zinsmeister AR, Metzger PP. Bismuth carbomer foam enemas for therapy: a double-blind, placebo-controlled trial. Gastroenterology active chronic pouchitis: a randomized, double-blind, placebo- 2003;124(5):1202–9. controlled trial. Aliment Pharmacol Ther 1997;11:1041–6. Isaacs 2007 {published data only} Wischmeyer 1993 {published data only} Isaacs KL, Sandler RS, Abreu M, Picco MF, Hanauer SB, Bickston Wischmeyer P, Pemberton JH, Phillips SF. Chronic pouchitis after SJ, et al.Rifaximin for the treatment of active pouchitis: a ileal pouch-anal anastomosis: responses to butyrate and glutamine randomized, double-blind, placebo-controlled pilot study. Inflamm suppositories in a pilot study. Mayo Clin Proc 1993;68:978–81. Bowel Dis 2007;13(10):1250–5. [: 1078–0998] References to studies excluded from this review Joelsson 2001 {published data only} Joelsson M, Andersson M, Bark T, Gullberg K, Hallgren T, Jiborn Kuhbacher 2006 {published data only} H, et al.Allopurinol as prophylaxis against pouchitis following ileal Kuhbacher T, Ott SJ, Helwig U, Mimura T, Rizzello F, Kleessen B, pouch-anal anastomosis for ulcerative colitis. A randomized et al.Bacterial and fungal microbiota in relation to probiotic therapy placebo-controlled double-blind study. Scand J Gastroenterol 2001; (VSL#3) in pouchitis. Gut 2006;55(6):833–41. 36(11):1179–84. Laake 2005 {published data only} Kuisma 2003 {published data only} Laake KO, Bjørneklett A, Aamodt G, Aabakken L, Jacobsen M, Kuisma J, Mentula S, Jarvinen H, Kahri A, Saxelin M, Farkkila M. Bakka A, et al.Outcome of four weeks’ intervention with probiotics Effect of Lactobacillus rhamnosus GG on ileal pouch inflammation on symptoms and endoscopic appearance after surgical and microbial flora. Aliment Pharmacol Ther 2003;17(4):509–15. reconstruction with a J-configurated ileal-pouch-anal-anastomosis Mimura 2004 {published data only} in ulcerative colitis. Scand J Gastroenterol 2005;40(1):43–51. Mimura T, Rizzello F, Helwig U, Poggioli G, Schreiber S, Talbot Madden 1994 {published data only} IC, et al.Once daily high dose probiotic therapy (VSL#3) for Madden MV, McIntyre AS, Nicholls RJ. Double-blind crossover maintaining remission in recurrent or refractory pouchitis. Gut trial of metronidazole versus placebo in chronic unremitting 2004;53(1):108–14. pouchitis. Dig Dis Sci 1994;39:1193–6. Pronio 2008 {published data only} McLeod 1986 {published data only} Pronio A, Montesani C, Butteroni C, Vecchione S, Mumolo G, McLeod RS, Taylor DW, Cohen Z, Cullen JB. Single-patient Vestri A, et al.Probiotic administration in patients with ileal pouch- randomised clinical trial. Use in determining optimum treatment anal anastomosis for ulcerative colitis is associated with expansion for patients with inflammation of Kock continent ileostomy of mucosal regulatory cells. Inflamm Bowel Dis 2008;14(5):662–8. reservoir. Lancet 1986;1(8483):726–8. Sambuelli 2002 {published data only} Meijer 2000 {published data only} Sambuelli A, Boerr L, Negreira S, Gil A, Camartino G, Huernos S, Meijer HP, Welters CF, Heineman E, Salomons GS, Buller HA, et al.Budesonide enema in pouchitis - A double-blind, double- Dekker J, et al.Enteral inulin doles not affect epithelial gene

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 15 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. expression and cell turnover within the ileoanal pouch - reservoir after restorative ileoanal anastomosis. Gastrointest Endosc Randomized, placebo-controlled, crossover study. Dis Colon 1990;36:6–9. Rectum 2000;43(10):1427–34. Gionchetti 1997 Miner 2006 {published data only} Gionchetti P, Rizzello F, Venturi A, Ferretti M, Brignola C, Peruzzo Miner Jr PB, Wedel MK, Xia S, Baker BF. Safety and efficacy of S, et al.Long-term efficacy of bismuth carbomer enemas in patients two dose formulations of alicaforsen enema compared with with treatment-resistant chronic pouchitis. Aliment Pharmacol Ther mesalazine enema for treatment of mild to moderate left-sided 1997;11:673–8. ulcerative colitis: A randomized, double-blind, active-controlled trial. Aliment Pharmacol Ther 2006;23(10):1403–13. Gionchetti 2007 Gionchetti P,Rizzello F, Poggioli G, Pierangeli F, Laureti S, Morselli Van 2006 {published data only} C, Tambasco R, Calabrese C, Campieri M. Oral budesonide in the Van Deventer SJ, Wedel MK, Baker BF, Xia S, Chuang E, Miner Jr treatment of chronic refractory pouchitis.. Aliment Pharmacol Ther PB. A Phase II dose ranging, double-blind, placebo-controlled 2000 May 15;25(10):1231–6. study of alicaforsen enema in subjects with acute exacerbation of mild to moderate left-sided ulcerative colitis. Aliment Pharmacol Higgins 2008 Ther 2006;23(10):1415–25. Higgins JPT, Altman DG (editors). Chapter 8: Assessing risk of Welters 2002 {published data only} bias in included studies. In: Higgins JPT, Green S editor(s). Welters CF, Heineman E, Thunnissen FB, van den Bogaard AE, Cochrane Handbook for Systematic Reviews of Interventions Version Soeters PB, Baeten CG. Effect of dietary inulin supplementation on 5.0.0 [updated February 2008]. The Cochrane Collaboration, inflammation of pouch mucosa in patients with an ileal pouch-anal Available from www.cochrane–handbook.org. 2008. anastomosis. Dis Colon Rectum 2002;45(5):621–7. Hurst 1996 Additional references Hurst RD, Molinari M, Chung TP, Rubin M, Michelassi F. Prospective study of incidence, timing, and treatment of pouchitis Boerr 1995 in 104 consecutative patients after restorative proctocolectomy. Boerr LA, Sambuelli AM, Sugai E, Graziano A, Valero J, Kogan Z, Arch Surg 1996;131:497–502. et al.Faecal alpha-1-antitrypsin concentration in the diagnosis and management of patients with pouchitis. Eur J Gastroenterol Hepatol Kelly 1983 1995;7:129–33. Kelly DG, Phillips SF, Kelly KA, Weinstein WM, Gilchrist MJR. Boerr 1996 Dysfunction of the continent ileostomy: clinical features and Boerr LR, Sambuelli AM, Filinger E, Peredo A, Graziano A, Valero bacteriology. Gut 1983;24:193–201. J, et al.Increased mucosal levels of leukotriene B4 in pouchitis: Klein 1983 evidence for a persistent inflammatory state. Eur J Gastroenterol Klein K, Stenzel P, Katon RM. Pouch ileitis: report of a case with Hepatol 1996;8:57–61. severe systemic manifestations. J Clin Gastroenterol 1983;5:149–53. Bonello 1981 Kmiot 1993 Bonello JC, Thow GB, Manson RR. Mucosal enteritis: a Kmiot WA, Hesslewood SR, Smith N, Thompson H, Harding LK, complication of the continent ileostomy. Dis Colon Rectum 1981; Keighley MR. Evaluation of the inflammatory infiltrate in pouchitis 24:37–41. with 111-In-labeled granulocytes. Gastroenterology 1993;104: Chopra 2006 981–8. Chopra A, Pardi DS, Loftus EV Jr, Tremaine WJ, Egan LJ, Faubion WA, Hanson KA, Johnson TA, Sandborn WJ. Budesonide in the Knobler 1986 treatment of inflammatory bowel disease: the first year of experience Knobler H, Ligumsky M, Okon E, Ayalon A, Nesher R, in clinical practice.. Inflamm Bowel Dis 2006 Jan;12(1):29–32. Rachmilewitz D. Pouch ileitis - recurrence of the inflammatory bowel disease in the ileal reservoir. Am J Gastroenterol 1986;81: Clausen 1992 199–201. Clausen MR, Tvede M, Mortensen PB. Short-chain fatty acids in pouch contents with and without pouchitis after ileal pouch-anal Levin 1992 anastomosis. Gastroenterology 1992;103:1144–53. Levin KE, Pemberto n JH, Phillip s SE, Zinsmeister AR, Pezim ME. Role of oxyge n free radical s in aetiolog y of pouchitis.. Dis De Silva 1989 Colon Rectum 1992;35:452–6. De Silva HJ, Ireland A, Kettlewell M, Mortensen N, Jewell DP. Short-chain fatty acid irrigation in severe pouchitis. N Engl J Med Lohmuller 1990 1989;321:416–7. Lohmuller JL, Pemberton JH, Dozois RR, Ilstrup D, Van Heerden den Hoed 1996 J. Pouchitis and extraintestinal manifestations of inflammatory den Hoed PT, van Goch JJ, Veen HF, Ouwendijk RJT. Severe bowel disease after ileal pouch-anal anastomosis. Ann Surg 1990; pouchitis sucessfully treated with short-chain fatty acids. Canadian 211:622–9. J Surg 1996;39:168–9. Moskowitz 1986 Di Febo 1990 Moskowitz RL, Shepherd NA, Nicholls RJ. An assessment of Di Febo G, Miglioli M, Lauri A, Biasco G, Paganelli GM, Poggioli inflammation in the reservoir after restorative proctocolectomy with G, et al.Endoscopic assessment of acute inflammation of the ileoanal ileal reservoir. Int J Colorectal Dis 1986;1:167–74.

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 16 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Nasmyth 1989 Kluwer Academic Publishers 1997;-:Kluwer Academic Publishers Nasmyth DG, Godwin PG, Dixon MF, Williams NS, Johnston D. 1997;51-63. Ileal ecology after pouch-anal anastomosis or ileostomy. Sandborn 1998 Gastroenterology 1989;96:817–24. Sandborn W, McLeod R, Jewell D. Pharmacotherapy for induction Natori 1992 and maintenance of remission in pouchitis. Cochrane Database of Natori H, Utsunomiya J, Yamamaura T, Benno Y, Uchida K. Fecal Systematic Reviews 1998, Issue 2. [Art. No.: CD001176. DOI: and stomal bile acid composition after ileostomy or ileoanal 10.1002/14651858.CD001176] anastomosis in patients with chronic ulcerative colitis and Scott 1989 adenomatous coli. Gastroenterology 1992;102:1278–88. Scott AD, Phillips RK. Ileitis and pouchitis after colectomy for Pardi 2009 ulcerative colitis. Br J Surg 1989;76:668–9. Pardi DS, D’Haens G, Shen B, Campbell S, Gionchetti P. Shepherd 1987 Guidelines for the Clinical Manangement of Pouchitis. Inflamm Shepherd NA, Jass JR, Duval I, Moskowitz RL, Nicholls RJ, Bowel Dis 2009 Sep;15(9):1423–31. Morson BC. Restorative proctocolectomy with ileal reservoir: Parihar 1997 pathological and histochemical study of mucosal biopsy specimens. Parihar GA, Malhotra P,Singh GB, Ludtke R, Safayhi H, Ammon J Clin Pathol 1987;40:601–7. HP. Effects of boswella serrata gum resin in patients with ulcerative Shepherd 1989 colitis. Eur J Med Res 1997;2:37–43. Shepherd NA, Hulten L, Tytgat GN, Nicholls RJ, Nasmyth DG, Pullan 1993 Hill MJ, et al.Workshop: pouchitis. Int J Colorectal Dis 1989;4: Pullan RD, Ganesh S, Mani V, Morris J, Evans BK, Williams GT, et 205–29. al.Comparison of bismuth citrate and 5-aminosalicylic acid enemas Svaninger 1993 in distal ulcerative colitis: a controlled trial. Gut 1993;34:676–9. Svaninger G, Nordgren S, Oresland T, Hulten L. Incidence and Sandborn 1994a characteristics of pouchitis in the Kock continent ilesotomy and the Sandborn WJ. Pouchitis following ileal pouch-anal anastomosis: pelvic pouch. Scand J Gastroenterol 1993;28:695–700. definition, pathogenesis, and treatment. Gastroenterology 1994;107: Tremaine 1994 1856–60. Tremaine WJ, Sandborn WJ, Phillips SF, Pemberton JH, Carpenter Sandborn 1994b HA. Short chain fatty acid (SCFA) enema therapy for treatment- Sandborn WJ, Tremiane WJ, Batts KP, Pemberton JH, Phillips SF. resistant pouchitis following ileal pouch-anal anastomosis (IPAA) Pouchitis following ileal pouch-anal anastomosis: a pouchitis for ulcerative colitis (UC). Gastroenterology 1994;106:A784. disease activity index. Mayo Clin Proc 1994;69:409–15. Tremaine 1998 Sandborn 1995 Tremaine WJ, Sandborn WJ, Kenan ML. Bismuth subsalicylate Sandborn WJ, Tremaine WJ, Batts KP, Pemberton JH, Rossi SS, tablets for chronic antibiotic-resistant pouchitis. Gastroenterology Hofmann AF, et al.Fecal bile acids, short chain fatty acids, and 1998;114(1):A1101. bacteria after ileal pouch-anal anastomosis do not differ in patients Tytgat 1988 with pouchitis. Dig Dis Sci 1995;40:1474–83. Tytgat GN, Van Deventer SJ. Pouchitis. Int J Colorectal Dis 1988;3: Sandborn 1997 226–8. Sandborn WJ. Pouchitis: definition, risk factors, frequency, natural Zuccaro 1989 history, classification, and public health perspectives. In: Trends In Zuccaro G, Fazio VW, Church JW, Lavery IC, Ruderman WB, Inflammatory Bowel Disease 1996. McLeod RS, Martin F, Farmer RG. Pouch ileitis. Dig Dis Sci 1989;34:1505–10. Sutherland LR, Wallace JL, Williams CN, eds. Lancaster, UK: ∗ Indicates the major publication for the study

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 17 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. CHARACTERISTICSOFSTUDIES

Characteristics of included studies [ordered by study ID]

Gionchetti 2000

Methods Randomized, double-blind, placebo-controlled. Parallel-arm design. Intention to treat. 36 weeks. N=40

Participants Patients with pouchitis in remission (deﬁned as a PDAI = 0) and chronic (require con- tinuing medical suppression therapy and recur within a few weeks of discontinuing suppression). Prior to study entry patients were treated with open antibiotics (types not speciﬁed) to induce remission. Antibiotic therapy discontinued at study entry

Interventions VSL#3, 6 grams per day versus placebo

Outcomes Maintenence of remission. Relapse from remission to active pouchitis (deﬁned as an increase by 2 points in the 6 point clinical portion of the PDAI, from a baseline score of 0)

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Computer generated

Allocation concealment? Yes Adequate

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes No drop outs All outcomes

Free of selective reporting? Yes The published report includes all expected outcomes

Gionchetti 2003

Methods Randomized, double-blind, placebo-controlled. Parallel-arm design. Intention to treat. 12 months. N=40

Participants Patients were randomized within one week of ileostomy closure after ileal pouch-anal anastomosis for ulcerative colitis

Interventions VSL#3, 1 packet per day versus 3 g/day placebo (maize)

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 18 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Gionchetti 2003 (Continued)

Outcomes Prophylaxis. Primary endpoint was pouchitis, deﬁned as PDAI score ≥ 7, with symptoms, histologic, and endoscopic evidence of inﬂammation. Secondary endpoints included IBDQ score, fecal bacterial concentrations, and stool frequency

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Computer generated

Allocation concealment? Yes Adequate

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes No drop outs All outcomes

Free of selective reporting? Yes The published report includes all expected outcomes

Isaacs 2007

Methods Randomized, double-blind, multi-center, placebo-controlled. Parallel-arm design. In- tention-to-treat analysis. 4 weeks. N=18

Participants Patients with active pouchitis deﬁned as a PDAI score of at least 7

Interventions Oral rifaximin 400 mg or placebo 3 times per day for 4 weeks

Outcomes Acute treatment. Clinical remission deﬁned as a PDAI score < 7 points and a decrease in the baseline PDAI score of 3 points

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Computer generated with permuted blocks

Allocation concealment? Unclear Not described

Blinding? Yes Double blind All outcomes

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 19 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Isaacs 2007 (Continued)

Incomplete outcome data addressed? Yes Two patients withdrew from placebo group All outcomes due to lack of efﬁcacy. These patients were classiﬁed as treatment failures.

Free of selective reporting? Yes The published report includes all expected outcomes

Joelsson 2001

Methods Randomized, double-blind, multi-center, placebo-controlled. Parallel-arm design. Per protocol analysis. 24 months. N=184

Participants Patients with IPAA for UC were randomized 1 week after surgery. Excluded: age < 18 or > 75 years, breastfeeding, pregnancy or planned pregnancy during the coming year, psychiatric disease, known alcohol abuse, other disease of relevance according to the clinician judgement, allergy/intolerance to allopurinol, or current azathioprine, 6- mercaptopurine, cyclophosphamide, cyclosporine or anticoagulation use

Interventions Allopurinol 100 mg twice daily or placebo

Outcomes Prophylaxis. Pouchitis deﬁned by a functional score (similar to PDAI)

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Unclear Not described

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes Missing outcome data balanced in num- All outcomes bers across intervention groups with similar reasons for missing data across groups

Free of selective reporting? Yes The published report includes all expected outcomes

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 20 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Kuisma 2003

Methods Randomized, double-blind, placebo-controlled. Parallel arm design. Per protocol analysis. 3 months. N=20

Participants Patients with a previous history of pouchitis and endoscopic inﬂammation. Patients who had received antibiotics within one month were excluded, as were active chronic pouchitis (intermittent or continuous medical treatment) or symptoms [fever, bleeding, increased number of stools (> 2/day within 1 week)] indicative of an acute exacerbation of pouchitis

Interventions Lactobacillus GG in two gelatine capsules orally twice daily versus microcrystalline cellulose-only gelatine placebo capsules

Outcomes Acute treatment. Total PDAI score and quantitative bacterial culture of fresh faecal samples and biopsies taken from the pouch and afferent limb was performed before and after supplementation

Notes After randomization 2 patients were excluded after the ﬁrst endoscopy because of severe endoscopic pouchitis (large ulcers, stricture) requiring immediate antibiotic therapy

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Unclear Not described

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes Two patients were excluded because of se- All outcomes vere disease before treatment began. No patients dropped out

Free of selective reporting? Yes The published report includes all expected outcomes

Mimura 2004

Methods Randomized, placebo-controlled. Multi-center. Parallel-arm design. Intention to treat. 1-year. N=36.

Participants Patients with pouchitis at least twice in the previous year or requiring continuous antibiotics, associated with a PDAI ≥ 7, in whom remission was induced by four weeks of combined metronidazole and ciproﬂoxacin. Active refractory or recurrent pouchitis was deﬁned as a total PDAI score of > 7 and a history of relapsing or persistent pouchitis

Interventions VSL#3 3 g orally twice per day versus maize placebo

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 21 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Mimura 2004 (Continued)

Outcomes Maintence of remission. Remission was deﬁned as a clinical PDAI ≤ 2 and endoscopic PDAI ≤ 1. Relapse was deﬁned as an increased clinical PDAI score ≥ 2 and increased endoscopic PDAI score ≥ 3

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Yes Computer generated in blocks of four at each institution

Allocation concealment? Yes Adequate

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes One patient dropped out from the VSL#3 All outcomes group due to adverse events

Free of selective reporting? Yes The published report includes all expected outcomes

Pronio 2008

Methods Randomized, open-label study. Parallel-arm design. Modiﬁed intention to treat. 12 months. N=28

Participants Patients had no active disease, no chronic pouchitis, and were taking no medication. Excluded patients with acute pouchitis (PDAI ≥7), patients with perianal disease including abscess, ﬁssure, stricture, or anal sphincter weakness, and pregnant or lactating women

Interventions VSL#3, 2 packets once per day versus no treatment

Outcomes Acute pouchitis deﬁned as a PDAI score ≥ 7. Secondary endpoint included prevalence of regulatory T-cells

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described in sufﬁcient detail

Allocation concealment? No Open study

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 22 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Pronio 2008 (Continued)

Blinding? No Open study All outcomes

Incomplete outcome data addressed? Yes Two patients in the VSL#3 group dropped All outcomes out (refusals) and one control patient developed anal carcinoma

Free of selective reporting? Yes The published report includes all expected outcomes

Sambuelli 2002

Methods Randomized, double-blind, placebo-controlled. Parallel-arm design. Intention to treat. 6-weeks. N=26

Participants Patients with an active episode of pouchitis, deﬁned as a PDAI score ≥ 7 and no treatment during the previous month or during the trial. Patients were deﬁned as having chronic pouchitis, if they presented with recurrence of symptoms at least four times a year or if they needed therapy for more than 6 months in the last year, and required a 1 month treatment-free period prior to enrollment

Interventions Budesonide enema (2 mg/100 mL at bedtime) plus placebo tablets or oral metronidazole (0.5 g twice per day.) plus placebo enema

Outcomes Acute treatment. Decrease in PDAI ≥ 3 and/or remission or non-pouchitis criteria (PDAI < 7 points with a decrease of ≥ 3 points from baseline

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Unclear Not described

Blinding? Yes Double blind, double dummy All outcomes

Incomplete outcome data addressed? Yes Missing outcome data balanced in num- All outcomes bers across intervention groups with similar reasons for missing data across groups

Free of selective reporting? Yes The published report includes all expected outcomes

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 23 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Shen 2001

Methods Randomized, non-blinded, placebo-controlled. Parallel-arm design. Intention to treat. 2-weeks. N=16

Participants Patients with acute pouchitis, deﬁned as a PDAI score of ≥ 7 and symptom duration of 4 weeks or less who had not received any treatment for pouchitis within 2-weeks prior to enrollment. Excluded patients with chronic pouchitis deﬁned as symptoms lasting 4 weeks or longer

Interventions Oral ciproﬂoxacin 1 g/d or metronidazole 20 mg/kg/d

Outcomes Acute treatment. Total PDAI score as well as in the symptom, endoscopy, and histology subscores

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Unclear Not described

Blinding? No Not used (pouch biopsies were blindly All outcomes scored by a GI pathologist)

Incomplete outcome data addressed? Unclear Dropouts were not described All outcomes

Free of selective reporting? Unclear Outcomes were not prespeciﬁed

Tremaine 1997

Methods Randomized, double-blind, placebo-controlled. Parallel-arm design. Intention to treat. 3 weeks. N=40

Participants Patients with pouchitis that was active (deﬁned as a PDAI ≥ 7) and chronic (deﬁned as failure or intolerance to treatment with antibiotics and other medical therapies)

Interventions Bismuth carbomer foam enemas containing bismuth citrate 513 mg (metallic bismuth 270 mg) complexed with carbomer (a synthetic high molecular weight polymer of acrylic acid cross linked with polyalkenyl polyether) versus placebo administered nightly

Outcomes Treatment of chronic pouchitis. Improvement or remission deﬁned as a decrease in the PDAI ≥ 3 or a PDAI of 0, respectively

Notes

Risk of bias

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 24 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Tremaine 1997 (Continued)

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Yes Adequate

Blinding? Yes Double blind All outcomes

Incomplete outcome data addressed? Yes Two patients in the bismuth carbomer All outcomes foam group with missing data at the end of treatment were considered to be treatment failures

Free of selective reporting? Yes The published report includes all expected outcomes

Wischmeyer 1993

Methods Randomized, double-blind, controlled trial comparing two active agents. Parallel-arm design. Intention to treat. 3 weeks. N=19

Participants Patients with pouchitis that was chronic (deﬁned as at least 4 episodes of recurrent pouchitis within one year and the need for suppressive medications for at least 6 of the prior 12 months). Eligible patients with chronic pouchitis discontinued all concomitant medications for 7 days and then began investigational therapy. At the time that the investigational therapy was begun, the patients could have either active or remitted pouchitis following discontinuation of suppressive medical therapy

Interventions L-glutamine 1 g in a polyethylene glycol base suppository versus sodium butyrate 40 mmol in a polyethylene glycol base suppository administered twice per day

Outcomes Treatment of chronic pouchitis. Induction of symptomatic remission (resolution of increased stool frequency, urgency and cramping, fever, and blood in stools), if necessary, and subsequent maintenance of symptomatic remission (asymptomatic) after discontinuing other suppressive medical therapy

Notes

Risk of bias

Item Authors’ judgement Description

Adequate sequence generation? Unclear Not described

Allocation concealment? Unclear Not described

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 25 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Wischmeyer 1993 (Continued)

Blinding? Unclear Not described All outcomes

Incomplete outcome data addressed? Yes One patient withdrew from each group All outcomes

Free of selective reporting? Yes The published report includes all expected outcomes

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Kuhbacher 2006 Non-randomized sub-study

Laake 2005 Non-randomized follow-up study

Madden 1994 Data were not available before the ﬁrst cross-over

McLeod 1986 Randomized, double-blind, placebo-controlled “n-of-1” trial in a single patient with chronic pouchitis in a Kock pouch. Five 14 day treatment courses with oral metronidazole 250 mg/day and five 14 day treatment courses with placebo, administered in random order. Overall, the study demonstrated a beneficial effect for metronidazole as compared to placebo. The study was excluded from the analysis because the purpose of such trials is to determine whether a given intervention is beneficial for a single patient, not to determine whether one intervention is more effective than another in groups of patients with a given disease

Meijer 2000 Follow-up study (actually published before the main results) of the randomized trial by Welters et al in 2002

Miner 2006 Not pouchitis but mild to moderate active left-sided ulcerative colitis

Van 2006 Not pouchitis but mild to moderate left-sided ulcerative colitis

Welters 2002 Data were not available before the ﬁrst cross-over

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 26 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. DATA AND ANALYSES

Comparison 1. Acute Pouchitis

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Ciproﬂoxacin vs. Metronidazole: 1 16 Peto Odds Ratio (Peto, Fixed, 95% CI) 14.39 [2.00, 103.76] Achieved Remission (PDAI <7) 2 Budesonide enema vs. 1 26 Peto Odds Ratio (Peto, Fixed, 95% CI) 1.32 [0.29, 6.01] Metronidazole: Achieved Remission (PDAI <7) 3 Rifaximin vs. Placebo: Achieved 1 18 Peto Odds Ratio (Peto, Fixed, 95% CI) 10.92 [0.62, 193.55] Remission (PDAI <7) 4 Lactobacillus GG vs. Placebo: 1 20 Peto Odds Ratio (Peto, Fixed, 95% CI) 7.39 [0.15, 372.38] Improved (PDAI reduction ≥ 3)

Comparison 2. Chronic Pouchitis

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Glutamine suppositories 1 19 Peto Odds Ratio (Peto, Fixed, 95% CI) 2.75 [0.48, 15.94] vs. Butyrate suppositories: Achieved Remission (no recurrence of symptoms) 2 Bismuth Carbomer Foam 1 40 Peto Odds Ratio (Peto, Fixed, 95% CI) 1.0 [0.29, 3.42] Enemas vs. Placebo: Improved (PDAI reduction ≥ 3) 3 VSL#3 vs. Placebo: Relapse of 2 76 Peto Odds Ratio (Peto, Fixed, 95% CI) 25.39 [10.37, 62.17] Pouchitis (PDAI increase ≥ 2)

Comparison 3. Prevention of Pouchitis

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 VSL#3 vs. Placebo: No Episodes 1 40 Peto Odds Ratio (Peto, Fixed, 95% CI) 4.76 [1.16, 19.56] of Acute Pouchitis (PDAI ≥ 7) 2 VSL#3 vs. No Treatment: No 1 28 Peto Odds Ratio (Peto, Fixed, 95% CI) 10.31 [0.20, 541.25] Episodes of Acute Pouchitis (PDAI ≥ 7)

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 27 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 3 Allopurinol vs. Placebo: No 1 184 Peto Odds Ratio (Peto, Fixed, 95% CI) 1.10 [0.62, 1.97] Episodes of Pouchitis (clinical diagnosis)

Analysis 1.1. Comparison 1 Acute Pouchitis, Outcome 1 Ciproﬂoxacin vs. Metronidazole: Achieved Remission (PDAI <7).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 1 Acute Pouchitis

Outcome: 1 Ciproﬂoxacin vs. Metronidazole: Achieved Remission (PDAI <7)

Studyorsubgroup Cipro Metronizadole PetoOddsRatio Weight PetoOddsRatio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Shen 2001 7/7 3/9 100.0 % 14.39 [ 2.00, 103.76 ] Total (95% CI) 7 9 100.0 % 14.39 [ 2.00, 103.76 ] Total events: 7 (Cipro), 3 (Metronizadole) Heterogeneity: not applicable Test for overall effect: Z = 2.65 (P = 0.0082)

0.01 0.1 1 10 100 Favors Metronizadole Favors Cipro

Analysis 1.2. Comparison 1 Acute Pouchitis, Outcome 2 Budesonide enema vs. Metronidazole: Achieved Remission (PDAI <7).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 1 Acute Pouchitis

Outcome: 2 Budesonide enema vs. Metronidazole: Achieved Remission (PDAI <7)

Studyorsubgroup Budesonide Metronizadole PetoOddsRatio Weight PetoOddsRatio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Sambuelli 2002 6/12 6/14 100.0 % 1.32 [ 0.29, 6.01 ] Total (95% CI) 12 14 100.0 % 1.32 [ 0.29, 6.01 ] Total events: 6 (Budesonide), 6 (Metronizadole) Heterogeneity: not applicable Test for overall effect: Z = 0.36 (P = 0.72)

0.05 0.2 1 5 20 Favors Metronizadole Favours Budesonide

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 28 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.3. Comparison 1 Acute Pouchitis, Outcome 3 Rifaximin vs. Placebo: Achieved Remission (PDAI <7).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 1 Acute Pouchitis

Outcome: 3 Rifaximin vs. Placebo: Achieved Remission (PDAI <7)

Studyorsubgroup Rifaximin Placebo PetoOddsRatio Weight Peto Odds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Isaacs 2007 2/8 0/10 100.0 % 10.92 [ 0.62, 193.55 ] Total (95% CI) 8 10 100.0 % 10.92 [ 0.62, 193.55 ] Total events: 2 (Rifaximin), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.63 (P = 0.10)

0.01 0.1 1 10 100 Favors Placebo Favors Rifaximin

Analysis 1.4. Comparison 1 Acute Pouchitis, Outcome 4 Lactobacillus GG vs. Placebo: Improved (PDAI reduction ≥ 3).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 1 Acute Pouchitis

Outcome: 4 Lactobacillus GG vs. Placebo: Improved (PDAI reduction ≥ 3)

Studyorsubgroup LGG Placebo PetoOddsRatio Weight PetoOdds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Kuisma 2003 1/10 0/10 100.0 % 7.39 [ 0.15, 372.38 ] Total (95% CI) 10 10 100.0 % 7.39 [ 0.15, 372.38 ] Total events: 1 (LGG), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.00 (P = 0.32)

0.005 0.1 1 10 200 Favors Placebo Favors LGG

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 29 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 2.1. Comparison 2 Chronic Pouchitis, Outcome 1 Glutamine suppositories vs. Butyrate suppositories: Achieved Remission (no recurrence of symptoms).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 2 Chronic Pouchitis

Outcome: 1 Glutamine suppositories vs. Butyrate suppositories: Achieved Remission (no recurrence of symptoms)

Studyorsubgroup Glutamine Butyrate PetoOddsRatio Weight Peto Odds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Wischmeyer 1993 6/10 3/9 100.0 % 2.75 [ 0.48, 15.94 ] Total (95% CI) 10 9 100.0 % 2.75 [ 0.48, 15.94 ] Total events: 6 (Glutamine), 3 (Butyrate) Heterogeneity: not applicable Test for overall effect: Z = 1.13 (P = 0.26)

0.01 0.1 1 10 100 Favors Butyrate Favors Glutamine

Analysis 2.2. Comparison 2 Chronic Pouchitis, Outcome 2 Bismuth Carbomer Foam Enemas vs. Placebo: Improved (PDAI reduction ≥ 3).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 2 Chronic Pouchitis

Outcome: 2 Bismuth Carbomer Foam Enemas vs. Placebo: Improved (PDAI reduction ≥ 3)

Studyorsubgroup BismuthEnema Placebo PetoOddsRatio Weight PetoOddsRatio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Tremaine 1997 9/20 9/20 100.0 % 1.00 [ 0.29, 3.42 ] Total (95% CI) 20 20 100.0 % 1.00 [ 0.29, 3.42 ] Total events: 9 (Bismuth Enema), 9 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 0.0 (P = 1.0)

0.1 0.2 0.5 1 2 5 10 Favors Placebo Favors Bismuth

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 30 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 2.3. Comparison 2 Chronic Pouchitis, Outcome 3 VSL#3 vs. Placebo: Relapse of Pouchitis (PDAI increase ≥ 2).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 2 Chronic Pouchitis

Outcome: 3 VSL#3 vs. Placebo: Relapse of Pouchitis (PDAI increase ≥ 2)

Studyorsubgroup Placebo VSL#3 PetoOddsRatio Weight PetoOdds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Gionchetti 2000 20/20 3/20 52.3 % 29.70 [ 8.61, 102.44 ]

Mimura 2004 15/16 3/20 47.7 % 21.38 [ 5.85, 78.17 ] Total (95% CI) 36 40 100.0 % 25.39 [ 10.37, 62.17 ] Total events: 35 (Placebo), 6 (VSL#3) Heterogeneity: Chi2 = 0.13, df = 1 (P = 0.72); I2 =0.0% Test for overall effect: Z = 7.08 (P < 0.00001)

0.01 0.1 1 10 100 Favors Placebo Favors VSL#3

Analysis 3.1. Comparison 3 Prevention of Pouchitis, Outcome 1 VSL#3 vs. Placebo: No Episodes of Acute Pouchitis (PDAI ≥ 7).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 3 Prevention of Pouchitis

Outcome: 1 VSL#3 vs. Placebo: No Episodes of Acute Pouchitis (PDAI ≥ 7)

Studyorsubgroup VSL#3 Placebo PetoOddsRatio Weight PetoOdds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Gionchetti 2003 18/20 12/20 100.0 % 4.76 [ 1.16, 19.56 ] Total (95% CI) 20 20 100.0 % 4.76 [ 1.16, 19.56 ] Total events: 18 (VSL#3), 12 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 2.16 (P = 0.031)

0.01 0.1 1 10 100 Favors pacebo Favors VSL#3

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 31 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 3.2. Comparison 3 Prevention of Pouchitis, Outcome 2 VSL#3 vs. No Treatment: No Episodes of Acute Pouchitis (PDAI ≥ 7).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 3 Prevention of Pouchitis

Outcome: 2 VSL#3 vs. No Treatment: No Episodes of Acute Pouchitis (PDAI ≥ 7)

Studyorsubgroup VSL#3 Notreatment PetoOddsRatio Weight Peto Odds Ratio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Pronio 2008 16/16 11/12 100.0 % 10.31 [ 0.20, 541.25 ] Total (95% CI) 16 12 100.0 % 10.31 [ 0.20, 541.25 ] Total events: 16 (VSL#3), 11 (No treatment) Heterogeneity: not applicable Test for overall effect: Z = 1.15 (P = 0.25)

0.005 0.1 1 10 200 Favors no treatment Favors VSL#3

Analysis 3.3. Comparison 3 Prevention of Pouchitis, Outcome 3 Allopurinol vs. Placebo: No Episodes of Pouchitis (clinical diagnosis).

Review: Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis

Comparison: 3 Prevention of Pouchitis

Outcome: 3 Allopurinol vs. Placebo: No Episodes of Pouchitis (clinical diagnosis)

Studyorsubgroup Allopurinol Placebo PetoOddsRatio Weight PetoOddsRatio n/N n/N Peto,Fixed,95% CI Peto,Fixed,95% CI Joelsson 2001 43/94 39/90 100.0 % 1.10 [ 0.62, 1.97 ] Total (95% CI) 94 90 100.0 % 1.10 [ 0.62, 1.97 ] Total events: 43 (Allopurinol), 39 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 0.33 (P = 0.74)

0.1 0.2 0.5 1 2 5 10 Favors Placebo Favors Allopurinol

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 32 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. WHAT’S NEW Last assessed as up-to-date: 4 October 2009.

Date Event Description

5 October 2009 New search has been performed New search, new studies included.

5 October 2009 New citation required and conclusions have changed Change in authors, additional conclusions due to new data.

HISTORY Protocol ﬁrst published: Issue 3, 1998 Review ﬁrst published: Issue 3, 1998

CONTRIBUTIONSOFAUTHORS Dr. Holubar - data analysis, drafting of review, critical revision. Dr. Cima - data analysis, critical revision. Dr. Sandborn - data analysis, critical revision. Dr. Pardi - data analysis, critical revision.

DECLARATIONSOFINTEREST Disclosures: Dr. Holubar and Dr. Cima - none; Dr. Sandborn - consulting for Salix Pharmaceuticals; Dr. Pardi - research and consulting for Salix Pharmaceuticals.

SOURCES OF SUPPORT Internal sources • none, Not speciﬁed.

Treatment and prevention of pouchitis after ileal pouch-anal anastomosis for chronic ulcerative colitis (Review) 33 Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. External sources • none, Not speciﬁed.

INDEX TERMS

Medical Subject Headings (MeSH) Ciproﬂoxacin [therapeutic use]; Colitis, Ulcerative [∗surgery]; Enema; Gastrointestinal Agents [∗therapeutic use]; Metronidazole [therapeutic use]; Postoperative Complications [∗drug therapy; prevention & control]; Pouchitis [∗drug therapy; prevention & control]; Randomized Controlled Trials as Topic; Remission Induction; Rifamycins [therapeutic use]; Suppositories

MeSH check words Adult; Humans