Mounting evidence suggests an important role for the intestinal microbiota in the chronic mucosal inflammation that occurs in inflammatory bowel disease (IBD), and novel molecular approaches have further identified a dysbiosis in these patients. Several mechanisms of action of probiotic products that may interfere with possible aetiological factors in IBD have been postulated. Our objective was to discuss the rationale for probiotics in IBD and to systematically review clinical intervention studies with probiotics in the management of IBD in adults. A systematic search was performed in PubMed up to 1 October 2011, using defined keywords. Only full-text papers in the English language addressing clinical outcomes in adult patients were included. The 41 eligible studies were categorized on disease type (ulcerative colitis [UC] with/without an ileo-anal pouch and Crohn's disease [CD]) and disease activity. Pooled odds ratios were only calculated per probiotic for a specific patient group when more than one randomized controlled trial was available. Well designed randomized controlled trials supporting the application of probiotics in the management of IBD are still limited. Meta-analyses could only be performed for a limited number of studies revealing overall risk ratios of 2.70 (95% CI 0.47, 15.33) for inducing remission in active UC with Bifido-fermented milk versus placebo or no additive treatment (n = 2); 1.88 (95% CI 0.96, 3.67) for inducing remission in active UC with VSL#3 versus placebo (n = 2); 1.08 (95% CI 0.86, 1.37) for preventing relapses in inactive UC with Escherichia coli Nissle 1917 versus standard treatment (n = 3); 0.17 (95% CI 0.09, 0.33) for preventing relapses in inactive UC/ileo-anal pouch anastomosis (IPAA) patients with VSL#3 versus placebo; 1.21 (95% CI 0.57, 2.57) for preventing endoscopic recurrences in inactive CD with Lactobacillus rhamnosus GG versus placebo (n = 2); and 0.93 (95% CI 0.63, 1.38) for preventing endoscopic recurrences in inactive CD with Lactobacillus johnsonii versus placebo (n = 2). Further well designed studies based on intention-to-treat analyses by several independent research groups are still warranted to support the promising results for E. coli Nissle in inactive UC and the multispecies product VSL#3 in active UC and inactive pouch patients. So far, no evidence is available to support the use of probiotics in CD. Future studies should focus on specific disease subtypes and disease location. Further insight into the aetiology of IBD and the mechanisms of probiotic strains will aid in selecting probiotic strains for specific disease entities and disease locations.
