Combination of oral antibiotics may be effective in severe pediatric ulcerative colitis: A preliminary report

Transcription

1 Journal of Crohn's and Colitis (2014) 8, Available online at ScienceDirect Combination of oral antibiotics may be effective in severe pediatric ulcerative colitis: A preliminary report Dan Turner a,, Arie Levine b, Kaija-Leena Kolho c, Ron Shaoul d, Oren Ledder a a The Juliet Keidan Institute of Pediatric Gastroenterology and Nutrition Shaare Zedek Medical Center, The Hebrew University of Jerusalem, Israel b Wolfson Medical Center, Holon, Israel c Pediatric Gastroenterology, Hospital for Children and Adolescents, Helsinki University Central Hospital, Helsinki, Finland d Pediatric Gastroenterology Unit, Rambam Medical Center, Haifa, Israel Received 5 March 2014; received in revised form 25 April 2014; accepted 31 May 2014 KEYWORDS Ulcerative colitis; Children; Antibiotics; Severe UC; Vancomycin; Doxycyline Abstract Background: The results of previous studies on the effectiveness of antibiotics in ulcerative colitis (UC) seem more effective when used orally. In this retrospective, multicenter study, we aimed to report our experience of using a combination of 3 4 antibiotics in children with moderate-severe refractory UC and IBD-unclassified including metronidazole, amoxicillin, doxycycline, and if during hospital admission, also vancomycin (MADoV). Methods: All children treated during 2013 with the antibiotic cocktail for 2 3 weeks in an attempt to alleviate inflammation in refractory colitis were included. Doxycycline was substituted with oral gentamycin or ciprofloxacin in children younger than 8 years or when an allergy was known to one of the drugs. Children were assessed using the PUCAI and CRP weekly for 3 weeks. Results: All 15 included children had moderate to severe disease with refractory disease course to multiple immunosuppressants (mean age 13.6 ± 5.1 years, median disease duration 2 (IQR ) years, 11 females (73%), and 13 (87%) extensive disease; 14 (93%) were corticosteroid-dependent or resistant, and 12 (80%) refractory to anti-tnf therapy). The cocktail was definitely effective in 7 of the 15 included children (47%) who entered complete clinical remission (PUCAI b 10) without additional interventions. Questionable or partial short-term response was noted in another 3 (20%), totaling 67% of patients. Conclusion: The use of oral wide-spectrum antibiotic cocktail in pediatric UC seems promising in half of patients, refractory to other salvage therapy. A pediatric randomized controlled trial to assess this intervention is underway European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. Corresponding author at: Pediatric Gastroenterology and Nutrition Unit, Shaare Zedek Medical Center, P.O. Box 3235, Jerusalem 91031, Israel. Tel.: ; fax: address: turnerd@szmc.org.il (D. Turner) / 2014 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved.

2 Antibiotics in Pediatric UC Introduction The pathogenesis of inflammatory bowel diseases (IBDs) is multifactorial and relates to dysbiosis between an altered immune response and the enteric microbiome. The latter is a term reflecting a fascinating complex biological network that interacts with the immune system, while both are influenced by the environment and the genetic background of the host. Increased permeability of the gut may be associated with translocation of the altered micro-organisms through the mucosa thereby worsening the chronic inflammation. There are now ample of data to implicate the microbiome as a main factor in the occurrence of IBD 1. Our group has shown that children with ASC who respond to steroid therapy have a more diverse microbiome component than those who do not respond and requiring salvage medical therapy or colectomy 2. Antibiotics have long been used in IBD with conflicting results. A recent meta-analysis has concluded that antibiotic therapy is effective in both CD and UC with OR higher in the UC studies (OR 2.17 (95%CI ) as compared with 1.35 ( ) in CD) 3. It is noteworthy that 5 of the 6 placebo-controlled RCT's in UC which used oral antibiotics showed treatment benefit compared with none of three other RCTs applying intravenous antibiotics (Fig. 1). Specifically, two RCTs from Japan suggested the effectiveness of oral anti-fusobacterium varium (F. varium) antibiotic protocol in adult UC including a 14 day triple therapy with amoxicillin 500 mg, metronidazole 250 mg and tetracycline 500 mg all three times daily. A total of 210 mild-moderate ambulatory UC adult patients were randomized to receive the cocktail or placebo with more treated patients achieving a Mayo-score defined response at 3 months (45% vs. 23%, respectively; P = 0.001). 4 Remission rate was similar at 3 months (19% vs 16%, respectively) but steroid-free remission was higher at 12 months (35% vs 14%, respectively; P = 0.02); notably the lost to follow-up rate was as high as 55% a b Figure 1 Previously reported randomized placebo-controlled trials of antibiotic treatment in ulcerative colitis (1a orally administered antibiotics; 1b intravenously administered antibiotics). Amox, amoxicillin; Tetra, tetracycline; Metro, metronidazole; Cipro, ciprofloxacin; ASC, acute severe colitis; IVCS, intravenous corticosteroids; DC, discharge. Footnote: All p-values in Fig. 1b were insignificant; In Gionchetti 1999 we included only the ASC arm; In Dickinson 1985 we included only the UC subgroup of the 40 included patients; Two randomized studies (Gilat T et al and Ohkusa T et al. 2005) were not included in this chart since they did not include a placebo arm.

3 1466 D. Turner et al. (Fig. 1). The treatment led to a decrease in the titer of antibodies to F. varium, especially in the responders, and improved endoscopic score. Similarly, a smaller RCT showed superiority of this antibiotic regimen to induce remission in 10 UC patients compared to 10 controls 5. The antibiotic cocktail was also used in an open label study of 27 steroid refractory and 61 steroid dependent UC patients yielding short term responses of 71% and 82%, and 3-month remission of 70% and 75%, respectively (UEGW Amsterdam 2012; oral presentation). Another open label study has shown the effectiveness of this intervention in steroid-dependent UC patients 6. The role of F. varium in UC has been suggested in an experimental model, wherein mice developed colitis after being exposed to the bacteria isolated from UC patients 7. In ASC, oral vancomycin led to reduced need for colectomy by hospital discharge (2 of 18 (11%) in the antibiotic arm vs. 7 of 15 (47%) in the placebo arm) 8 (Fig. 1). Based on the above observations, we have used a combination of oral antibiotics in patients with refractory UC or IBD-unclassified (IBD-U) with encouraging results. We aimed to retrospectively report all consecutively treated UC children with combinations of antibiotics in our institutions in the year of Methods This is a retrospective cohort report of children with UC and IBD-U treated with oral antibiotics during the year IBD was confirmed in all patients according to accepted pediatric criteria, including upper and lower endoscopy with biopsies and small bowel imaging 9. As our experience with this intervention accumulated, we have tuned the protocol as follows (all antibiotics were prescribed orally): Children over the age of 7 years were prescribed triple therapy with amoxicillin 50 mg/kg divided by 3 (up to 500 mg X3/d), metronidazole 5 mg/kg X3/d (up to 250 mgx 3/d), and doxycycline 2 mg/kg X2/d (up to 100 mg X2/d). Doxycycline was substituted to ciprofloxacin (10 mg/kg X2/d -up to 250 mg X2/d) in children 2 7 years and to gentamycin (2.5 mg/kg X3/d) in infants younger than 2 years. Patients with known allergy to one of the drugs have been treated with oral gentamycin (2.5 mg/kg X3/d) instead of the allergenic drug. If the child has been admitted to the hospital, vancomycin was added as the fourth regimen (250 mg X4/d or 125 mg X4/d in those younger than 8 years). We initially prescribed the antibiotic cocktail for 2 weeks, but we revised our practice to 3 weeks as we noticed that complete response may be delayed. All children were repeatedly tested negative for fecal Clostridium difficile and bacterial cultures and only one was positive for CMV colitis as described below. To avoid selection bias, this report includes all children treated with the cocktail in our institutions during the year However, children who were commenced on steroid therapy concurrently with the antibiotics were excluded. Disease activity was assessed prospectively by the PUCAI at every ambulatory visit, and daily in hospitalized patients. The PUCAI has been validated in pediatric UC 10,11 and proved to perform well also in the acute colitis setting 12,13. Other recorded data included demographics and standard laboratory variables. Predefined time points for analyses were at introduction of antibiotics (i.e. baseline), and 7, 14 and 21 days thereafter. Endoscopic evaluation, when performed, was scored using the 4-point Mayo-endoscopic subscore (0 remission, 1 mild, 2 moderate and 3 severe). We did not aim to report the long-term response as we considered this cocktail for induction of remission only; however the longest follow-up available is described in the text of each case description (Table 1 and Supplementary Table 1). The primary analysis was based on the intention-to-treat principle, in which all patients were included, regardless of the response. A further secondary per-protocol analysis was added including only those completing the antibiotic protocol without salvage therapy. Missing laboratory data were imputed using the last observation carried forward. The last PUCAI and laboratory values recorded before treatment escalation in children not responding to the antibiotic cocktail were carried forward through the last visit, to avoid attributing the response to the failed antibiotic treatment. Statistical analyses included description summary presentation of mean ± SD, medians (interquartile range) and rates (95% confidence intervals). Values before and after the intervention were compared by paired Student's t-test or the sign test, as appropriate for the data distribution. All analyses were performed using IBM SPSS V21, taking a P b 0.05 as the threshold for significance. 3. Results Twenty patients were treated in the four centers with the oral antibiotic cocktail for refractory colitis, five of whom were excluded since concomitant steroid therapy was initiated with the antibiotics, serving as a possible confounding factor. All 15 children (11 females (73%)) had moderate to severe disease at antibiotic initiation with frequently active disease course, refractory to multiple immunosuppressants (Table 1 and Supplementary Table 1). Fourteen (93%) of the children were corticosteroid dependent or resistant, 12 (80%) were refractory to anti-tnf therapy (4 of whom to both infliximab and adalimumab), 11 (73%) failed extended trials of thiopurines with adequate levels, and 1 (7%) showed only partial transient response to tacrolimus. Ten patients were treated as in-patients for ASC. Further reflecting the severity of this cohort, 13 (87%) had extensive or pancolitis and only 2 (13%) had left sided colitis; the mean PUCAI score at antibiotic initiation was 65 ± 13 points. Five patients underwent colonoscopic assessment prior to starting the antibiotic cocktail, four demonstrating a Mayo endoscopic score of 3/3 and one with a score of 2/3. Mean age was 13.6 ± 5.1 years, median disease duration 2 (IQR ) years, 11 (73%) were diagnosed with UC and the others with IBD-U. Of the 15 included children, 9 (60%) children entered complete remission (PUCAI b10 points, 2 after one week, 6 after two weeks and one after 3 weeks) and one (7%) showed a response (PUCAI decrease from 80 to 45 points within 2 weeks of treatment) but is considered in the failure group since she relapsed shortly thereafter (Fig. 2). Of the 9 children entering remission we excluded two questionable cases, totaling seven children (47%) who had definite complete short term remission to the antibiotic cocktails.

4 Antibiotics in Pediatric UC 1467 Table 1 Patients' description. Case# Age IBD type a Disease extent Disease duration (years) Previously used medications b Antibiotic cocktail c Hospital admission Length of treatment (days) Longer-term outcome Responders entered complete clinical remission (PUCAI b 10) 1 16 IBD-U Pancolitis 4.5 CS, AZA, IFX, ADA MADoV Y 16 Remission (7 months) d IBD-U Pancolitis 0.5 CS, IFX VG Y 18 Remission (8 months) d IBD-U Pancolitis 0.2 CS MADo N 14 Remission (6 months) d UC Extensive 1.6 CS, AZA, IFX, ADA MGVC Y 21 Colectomy (at 3 months) e 5 17 UC Pancolitis 3.4 CS, AZA, IFX, ADA, FK506 MADo N 14 Relapsed; awaiting colectomy (5 months) UC Pancolitis 6.3 CS, AZA, IFX, ADA MADo N 14 Remission (10 months) f 7 19 UC Pancolitis 3 CS, AZA, IFX MADo N 14 Remission (2 months) UC Pancolitis 0.5 CS MADo Y 14 Remission (2 months) 9 17 UC Pancolitis 2 CS, AZA, IFX MADo N 21 Remission (3 months) Non-responders UC Left-sided 0.95 CS, AZA, IFX MADoV Y 21 Awaiting colectomy (5 months) IBD-U Pancolitis 2 CS, AZA, IFX MADo Y 14 Remission on MTX (7 months) UC Left-sided 5 CS, AZA, IFX MADoV Y 6 Colectomy (at 2 weeks) UC Pancolitis 3 CS, AZA MADo Y 6 Chronically active (3 months) UC Extensive 1.1 CS, AZA, IFX MADoV Y 14 Remission on FK-506 (3 months) UC Pancolitis 0.7 CS, IFX MADoV Y 5 Colectomy (at 2 weeks) a All patients with moderate to severe disease. b All previously used 5-ASA; CS (corticosteroids), AZA (azathioprine), IFX (infliximab), ADA (adalimumab), and MTX (methotrexate). c A (Amoxicillin), Do (Doxycycline), M (Metronidazole), V (Vancomycin), G (Gentamycin), and C (Ciprofloxacin). d Subsequent flare treated successfully with repeated similar antibiotic cocktail. e Patient stopped all maintenance medications after induction. f Mild distal colitis was noted after 9 months during a minor flare responding to local therapy. The first questionable case (Case 5) had ASC and failed intravenous steroids and infliximab (imminent colectomy). Significant response followed 5 days of antibiotic treatment (PUCAI decreased from 85 to 40 points), but then ganciclovir was added due to finding of CMV in a previous rectal biopsy. He was discharged in complete remission (PUCAI = 0) after two weeks. All other children did not have any additional treatments commenced besides the antibiotic cocktail. The second questionable case (Case 6) entered complete remission within 2 weeks (PUCAI decreased from 50 to 0 points) for the first time in 3 years (despite multiple immunosuppressants and biologics; the family refused colectomy). However, the disease relapsed 1 week after stopping the antibiotics. One of the five children who we defined as non-responder, Case 15, was admitted for ASC and failed intravenous steroids and infliximab. She showed a significant improvement to the antibiotic cocktail within 5 days (PUCAI decreased from 85 to 20) but she then elected to have fecal microbiome transplantation with subsequent prompt worsening of diarrhea and severe abdominal pain; colectomy followed within 2 days. In an intention to treat analysis of the total cohort (responders and non-responders), the mean PUCAI score dropped from 65 ± 13 points pretreatment to 23 ± 28 points thereafter (P b ) (Fig. 2). CRP dropped from 2.9 ± 2.6 to 2.3 ± 2.9 mg/dl (P = 0.09). There was no change in hemoglobin (P = 0.32) and albumin (P = 0.41) possibly given the fact that these tests are less responsive to change in the short term. In a per-protocol analysis of the 9 responders who completed antibiotic treatment, there was a clinical (PUCAI decrease from 64 ± 19 to 5 ± 13; P b ) and biochemical (CRP dropped from 1.73 ± 1.5 to 0.82 ± 1.1 mg/dl; P = 0.007) improvement within three weeks of treatment (Fig. 2). Hemoglobin and albumin increased from 10 ± 1.9 to 11.2 ± 1.4 (P = 0.035) and 3.6 ± 0.6 to 4 ± 0.4 (P = 0.063), respectively. Since some of these patients received blood transfusions and total parenteral nutrition (TPN) the results of the hemoglobin and albumin must be interpreted with caution. Three of the five patients who had baseline endoscopic evaluation repeated sigmoidoscopy within the treatment period. The period is too short to judge mucosal healing, but nonetheless, one had an endoscopic improvement from Mayo score of 3 to 2 points after 10 days (Case 10), a second (Case 6) from 3 to 2 and the third (Case 9) from 2 to 1, both after 21 days. Serial fecal calprotectin levels were measured on Case 9 with initial measure of 155 μg/g and normalization after treatment (26 μg/g upon treatment completion and 38 μg/g 6 weeks thereafter). The mean follow-up period of the 9 children who entered remission was 5.1 ± 2.8 months (Table 1, Supplementary

5 1468 D. Turner et al. PUCAI P< in ITT (black) P< in PP (gray) CRP (mg/dl) P=0.09 in ITT (black) P=0.009 in PP (gray) Baseline 7 days 14 days 21 days Baseline 7 days 14 days 21 days Figure 2 Change in PUCAI and C-reactive protein (CRP) in the entire cohort (bars represent means ± standard deviation; black intention to treat and gray per protocol analysis (PP)). Footnote: The five non-responders are indicated by dashed lines and have their data carried forward from the time of initiation of salvage therapy; one of the responders initiated salazopyrin and ganciclovir after response has been achieved by antibiotics only. No therapies except for the antibiotic cocktail have been changed or initiated in all other 9 responders. Table 1), and at the last follow-up visit, 7/9 maintained their remission (6 of whom without steroids). 4. Discussion We have shown that patients with severe refractory colitis may benefit from a course of combination antibiotic therapy. The cocktail was effective in inducing complete remission in 7 of the 15 included children (47%) without additional interventions. Questionable or partial short-term response was noted in another 3 (20%), totaling 67% of patients. We included in this report all eligible children treated with the antibiotic cocktail in our institutions during 2013, to avoid selection bias. Given the novelty of the intervention, we have used it in disease refractory to multiple treatments and thus our cohort included children at the most severe end of the spectrum, some of whom had no options left besides colectomy. The fact that this is a real life case series with a heterogeneous case load reflects better generalization but also reduced power to draw firm conclusions and to ascertain the best candidates for this intervention. There are four main possible explanations why our antimicrobial intervention might be effective in some cases. First, the antibiotics might be targeting a specific unknown UC associated pathogen underlying some of the exacerbations. Indeed, this has been described with CMV 14, cryptosporidium 15 and C. difficile 16 that are more commonly seen with acute UC compared to the general population and tend to be more severe. Bacteroides adherent to the mucosal surface have beenshowntobemoreabundantinchildrenwithibd 17.In these circumstances, effective therapy directed towards the identified pathogens may alleviate the severe attack. It is plausible to assume that more potentially pathogenic organisms can lead to the same result and our antibiotic cocktail might be effective against these yet unidentified bacteria (F. Varium, for instance, as suggested by the Japanese group 4 ). Secondly, the mucosal inflammation in severe UC further impairs the epithelial barrier which might enable translocating bacteria to stimulate the adaptive immune response to a greater extent than in patients with an intact epithelium, thus increasing inflammation and aggravating the disease; antibiotics may also alleviate this second hit effect. Third, antibiotics may alter the composition of the luminal microbiome while changing the immunological balance with the immune response. Finally, some antibiotics may posses anti-inflammatory properties, as shown for azythromycin. It may very well be that other antibiotic regimens may be more effective or that fewer than the four proposed antibiotics can suffice. However, we first attempted to prove the concept by maximizing treatment. The choice of the four antibiotics was based on the published success of the Japanese cocktail (while substituting Tetracycline for Doxycyline) and adding oral Vancomycin based on a positive small RCT and since orally administered Vancomycin is not associated with adverse events, except, of course, for the risk for emerging resistant strains. Three clinical trials showed no benefit for intravenous antibiotics in ASC including IV metronidazole 18, IV ciprofloxacin 19, and IV metronidazole with tobramycin 20, all in combination with IV steroids (Fig. 1). In contrast, almost all UC trials to date that used oral preparations showed some response as compared with placebo (Fig. 1). In addition to the Japanese studies of amoxicillin, metronidazole and tetracycline 4 6 and the oral vancomycin trial 8 described in the introduction, three other placebo-controlled trials showed a significant benefit to oral antibiotics, including tobramycin 21, long term ciprofloxacin 22 and rifaximin 23, although the latter was underpowered to show statistical significance. One study of oral ciprofloxacin was negative 24. Endoscopic evaluation was available in three of our patients, showing improvement within a short period (severe to moderate in two patients and moderate to mild in one patient within 21 days). Although we did not have endoscopic evaluation for all patients, a strong PUCAI-endoscopy concordance has been well verified in several cohorts in children 10,25,26. In previous oral antibiotic trials, endoscopic improvement followed the clinical benefit when included as an outcome. In Burke et al., oral tobramycin led to complete histological remission in 17% of patients within days of treatment as compared with 0% in those receiving placebo

6 Antibiotics in Pediatric UC 1469 (P = 0.025). 21 Six months of oral ciprofloxacin led to improved endoscopic findings at 6 months but not at 3 months 22 and the Japanese 2-week cocktail was associated with better endoscopic scores both at 3 and 12 months follow-up visits 4. In contrast, the negative oral ciprofloxacin trial showed no apparent endoscopic benefit 24. We have considered our antibiotic intervention as an induction of remission regimen which mandates a strategy for long-term maintenance of remission. Indeed, most studies that followed the patients receiving oral antibiotic treatment did not show a sustained benefit over 6 12 months 8,27,22, except for the Japanese cocktail that showed superior remission rate at 12 month compared with placebo 4. Our findings are intriguing and give some short-term hope to half of children otherwise refractory to standard and salvage therapy. Our cohort was enrolled consecutively from several centers which increases our confidence in the results and lessens the likelihood for selection bias. However, our study is merely an uncontrolled, retrospective report. Moreover, our longer term follow-up has been anecdotal, since we focused on induction-of-remission effect. Therefore, this preliminary report should not be used to change clinical practice and guidelines of treating children with UC 28,29. However, it formed the basis for further research and a pediatric RCT in ASC is underway, evaluating the 3-week MADoV protocol described here. Supplementary data to this article can be found online at Conflict of interest The authors declare no conflict of interest in relation to the content of this manuscript. References 1. Bakhtiar SM, LeBlanc JG, Salvucci E, Ali A, Martin R, Langella P, et al. Implications of the human microbiome in inflammatory bowel diseases. FEMS Microbiol Lett May 2013;342(1):10 7 [PubMed PMID: ]. 2. Michail S, Durbin M, Turner D, Griffiths AM, Mack DR, Hyams J, et al. Alterations in the gut microbiome of children with severe ulcerative colitis. Inflamm Bowel Dis Oct 2012;18(10): [PubMed PMID: Pubmed Central PMCID: ]. 3. Wang SL, Wang ZR, Yang CQ. Meta-analysis of broad-spectrum antibiotic therapy in patients with active inflammatory bowel disease. Exp Ther Med Dec 2012;4(6): [PubMed PMID: , Pubmed Central PMCID: ]. 4. Ohkusa T, Kato K, Terao S, Chiba T, Mabe K, Murakami K, et al. Newly developed antibiotic combination therapy for ulcerative colitis: a double-blind placebo-controlled multicenter trial. Am J Gastroenterol 2010;105(8): Ohkusa T, Nomura T, Terai T, Miwa H, Kobayashi O, Hojo M, et al. Effectiveness of antibiotic combination therapy in patients with active ulcerative colitis: a randomized, controlled pilot trial with long-term follow-up. Scand J Gastroenterol 2005;40(11): Terao S, Yamashiro K, Tamura I, Hirano T, Ohkusa T, Kato K. Antibiotic combination therapy for steroid withdrawal in steroid-dependent ulcerative colitis. Digestion 2011;83(3): [PubMed PMID: ]. 7. Ohkusa T, Okayasu I, Ogihara T, Morita K, Ogawa M, Sato N. Induction of experimental ulcerative colitis by Fusobacterium varium isolated from colonic mucosa of patients with ulcerative colitis. Gut Jan 2003;52(1):79 83 [PubMed PMID: , Pubmed Central PMCID: ]. 8. Dickinson RJ, O'Connor HJ, Pinder I, Hamilton I, Johnston D, Axon AT. Double blind controlled trial of oral vancomycin as adjunctive treatment in acute exacerbations of idiopathic colitis. Gut 1985;26(12): Inflammatory bowel disease in children and adolescents: recommendations for diagnosis the Porto criteria. J Pediatr Gastroenterol Nutr Jul 2005;41(1):1 7 [PubMed PMID: ]. 10. Turner D, Otley AR, Mack D, Hyams J, de Bruijne J, Uusoue K, et al. Development, validation, and evaluation of a pediatric ulcerative colitis activity index: a prospective multicenter study. Gastroenterology 2007;133(2): [PubMed PMID: ]. 11. Turner D, Hyams J, Markowitz J, Lerer T, Mack DR, Evans J, et al. Appraisal of the pediatric ulcerative colitis activity index (PUCAI). Inflamm Bowel Dis Aug 2009;15(8): [PubMed PMID: , Epub 2009/01/24. Eng]. 12. Turner D, Walsh CM, Benchimol EI, Mann EH, Thomas KE, Chow C, et al. Severe paediatric ulcerative colitis: incidence, outcomes and optimal timing for second-line therapy. Gut Mar 2008;57(3): [PubMed PMID: , Epub 2007/11/06. eng]. 13. Turner D, Mack D, Leleiko N, Walters TD, Uusoue K, Leach ST, et al. Severe pediatric ulcerative colitis: a prospective multicenter study of outcomes and predictors of response. Gastroenterology Jun 2010;138(7): [PubMed PMID: , Epub 2010/ 03/03. Eng]. 14. Inokuchi T, Kato J, Hiraoka S, Suzuki H, Nakarai A, Hirakawa T, et al. Long-term follow-up of ulcerative colitis patients treated on the basis of their cytomegalovirus antigen status. World J Gastroenterol Jan ;20(2): [PubMed PMID: ]. 15. Chen W, Chadwick V, Tie A, Harp J. Cryptosporidium parvum in intestinal mucosal biopsies from patients with inflammatory bowel disease. Am J Gastroenterol Dec 2001;96(12): [PubMed PMID: ]. 16. Berg AM, Kelly CP, Farraye FA. Clostridium difficile infection in the inflammatory bowel disease patient. Inflamm Bowel Dis Jan 2013;19(1): [PubMed PMID: ]. 17. Zitomersky NL, Atkinson BJ, Franklin SW, Mitchell PD, Snapper SB, Comstock LE, et al. Characterization of adherent bacteroidales from intestinal biopsies of children and young adults with inflammatory bowel disease. PLoS ONE 2013;8(6):e63686 [PubMed PMID: , Pubmed Central PMCID: ]. 18. Chapman RW, Selby WS, Jewell DP. Controlled trial of intravenous metronidazole as an adjunct to corticosteroids in severe ulcerative colitis. Gut Oct 1986;27(10): [PubMed PMID: , Epub 1986/10/01. eng]. 19. Mantzaris GJ, Petraki K, Archavlis E, Amberiadis P, Kourtessas D, Christidou A, et al. A prospective randomized controlled trial of intravenous ciprofloxacin as an adjunct to corticosteroids in acute, severe ulcerative colitis. Scand J Gastroenterol Sep 2001;36(9):971 4 [PubMed PMID: ]. 20. Mantzaris GJ, Hatzis A, Kontogiannis P, Triadaphyllou G. Intravenous tobramycin and metronidazole as an adjunct to corticosteroids in acute, severe ulcerative colitis. Am J Gastroenterol Jan 1994;89(1):43 6 [PubMed PMID: ]. 21. Burke DA, Axon AT, Clayden SA, Dixon MF, Johnston D, Lacey RW. The efficacy of tobramycin in the treatment of ulcerative colitis. Aliment Pharmacol Ther 1990;4(2): Turunen UM, Farkkila MA, Hakala K, Seppala K, Sivonen A, Ogren M, et al. Long-term treatment of ulcerative colitis with ciprofloxacin: a prospective, double-blind, placebo-controlled study. Gastroenterology 1998;115(5): Gionchetti P, Rizzello F, Ferrieri A, Venturi A, Brignola C, Ferretti M, et al. Rifaximin in patients with moderate or severe ulcerative colitis refractory to steroid-treatment: a double-blind, placebo-

7 1470 D. Turner et al. controlled trial. Dig Dis Sci Jun 1999;44(6): [PubMed PMID: , Epub 1999/07/02. eng]. 24. Mantzaris GJ, Archavlis E, Christoforidis P, Kourtessas D, Amberiadis P, Florakis N, et al. A prospective randomized controlled trial of oral ciprofloxacin in acute ulcerative colitis. Am J Gastroenterol Mar 1997;92(3):454 6 [PubMed PMID: , Epub 1997/03/01. eng]. 25. Turner D, Griffiths AM, Veerman G, Johanns J, Damaraju L, Blank M, et al. Endoscopic and clinical variables that predict sustained remission in children with ulcerative colitis treated with infliximab. Clin Gastroenterol Hepatol Nov 2013;11(11): [PubMed PMID: ]. 26. Turner D, Seow CH, Greenberg GR, Griffiths AM, Silverberg MS, Steinhart AH. A systematic prospective comparison of noninvasive disease activity indices in ulcerative colitis. Clin Gastroenterol Hepatol Oct 2009;7(10): [PubMed PMID: , Epub 2009/07/07. eng]. 27. Lobo AJ, Burke DA, Sobala GM, Axon AT. Oral tobramycin in ulcerative colitis: effect on maintenance of remission. Aliment Pharmacol Ther Apr 1993;7(2):155 8 [PubMed PMID: , Epub 1993/04/01. eng]. 28. Turner D, Levine A, Escher JC, Griffiths AM, Russell RK, Dignass A, et al. Management of pediatric ulcerative colitis: joint ECCO and ESPGHAN evidence-based consensus guidelines. J Pediatr Gastroenterol Nutr Sep 2012;55(3): [PubMed PMID: ]. 29. Turner D, Travis SP, Griffiths AM, Ruemmele FM, Levine A, Benchimol EI, et al. Consensus for managing acute severe ulcerative colitis in children: a systematic review and joint statement from ECCO, ESPGHAN, and the Porto IBD Working Group of ESPGHAN. Am J Gastroenterol Apr 2011;106(4): [PubMed PMID: eng.].