Search

Background: Alterations in the intestinal microbiota are thought to be involved in the pathogenesis of inflammatory bowel diseases (IBD). Klebsiella oxytoca is an intestinal pathobiont that can produce a cytotoxin (tillivaline).Aim: We aimed to elucidate the pathogenetic relevance of toxin-producing K. oxytoca in patients with IBD flares and investigated the clonal relationship of K. oxytoca isolates from IBD patients using multilocus sequence typing (MLST).Methods: Fecal samples of 235 adult IBD patients were collected from January 2008 to May 2009 and were tested for K. oxytoca, C. difficile toxin, and other pathogens by standard microbiological methods. Clinical data and disease activity scores were collected. K. oxytoca isolates were tested for toxin production using cell culture assays. A total of 45 K. oxytoca isolates from IBD patients, healthy, asymptomatic carriers and from patients with antibiotic-associated hemorrhagic colitis in part from our strain collection were tested for their clonal relationship using MLST.Results: The prevalence of K. oxytoca in IBD overall was 4.7%. Eleven K. oxytoca isolates were detected. Two of 11 isolates were tested positive for toxin production. There was no significant difference in the distribution of K. oxytoca isolates between the groups (active vs. remission in UC and CD). MLST yielded 33 sequence types. K. oxytoca isolates from IBD did not cluster separately from isolates from asymptomatic carriers.Conclusions: Our data demonstrate that toxin (tilivalline)-producing K. oxytoca is not associated with IBD flares. [ABSTRACT FROM AUTHOR]

Background and Aim: Many patients with quiescent Crohn's disease are maintained on long-term treatment with azathioprine (AZA), but controlled data are limited. We aimed to evaluate the efficacy of AZA therapy for more than 4 years to maintain clinical remission. Methods: We performed a randomized double-blind placebo-controlled AZA withdrawal trial with a follow-up period of 24 months. Patients had to have continuous AZA therapy ≥4 years without exacerbation of disease during the 12 months before enrollment, and a Crohn's disease activity index <150 at baseline. Patients were randomized to continue on AZA or switch to placebo. The primary endpoint was time to clinical relapse during follow-up. Results: After inclusion of 52 patients, the trial was stopped prematurely due to slow recruitment. During the 2-year follow-up, clinical relapse occurred in 4 of 26 (15 %) patients on continued AZA and in 8 of 26 (31 %) patients on placebo. Time to clinical relapse averaged 22.3 months (95 % CI 20.6-24.0) on AZA and 19.2 months (95 % CI 16.4-22.1) on placebo ( p = 0.20). According to life-table analysis, the proportion of patients in remission after 12 and 24 months was 96 ± 4 and 86 ± 7 % in patients receiving AZA versus 76 ± 8 and 68 ± 9 % in patients receiving placebo (month 12, p = 0.035; month 24, p = 0.30). A higher AZA dose at enrollment was an independent predictor for relapse ( p < 0.05). Conclusions: AZA withdrawal resulted in a significantly increased relapse risk after 1 year and a nonstatistically significant trend for relapse after 2 years. Our results are in line with previous observations. [ABSTRACT FROM AUTHOR]