Systematic review with meta‐analysis: the efficacy of probiotics in inflammatory bowel disease

<jats:title>Summary</jats:title><jats:sec><jats:title>Background</jats:title><jats:p>Ulcerative colitis (<jats:styled-content style="fixed-case">UC</jats:styled-content>) and Crohn's disease (<jats:styled-content style="fixed-case">CD</jats:styled-content>) are inflammatory bowel diseases (<jats:styled-content style="fixed-case">IBD</jats:styled-content>). Evidence implicates disturbances of the gastrointestinal microbiota in their pathogenesis.</jats:p></jats:sec><jats:sec><jats:title>Aim</jats:title><jats:p>To perform a systematic review and meta‐analysis to examine the efficacy of probiotics in <jats:styled-content style="fixed-case">IBD</jats:styled-content>.</jats:p></jats:sec><jats:sec><jats:title>Methods</jats:title><jats:p><jats:styled-content style="fixed-case">MEDLINE</jats:styled-content>,<jats:styled-content style="fixed-case"> EMBASE</jats:styled-content>, and the Cochrane Controlled Trials Register were searched (until November 2016). Eligible randomised controlled trials (<jats:styled-content style="fixed-case">RCT</jats:styled-content>s) recruited adults with <jats:styled-content style="fixed-case">UC</jats:styled-content> or <jats:styled-content style="fixed-case">CD</jats:styled-content>, and compared probiotics with 5‐aminosalicylates (5‐<jats:styled-content style="fixed-case">ASA</jats:styled-content>s) or placebo. Dichotomous symptom data were pooled to obtain a relative risk (<jats:styled-content style="fixed-case">RR</jats:styled-content>) of failure to achieve remission in active <jats:styled-content style="fixed-case">IBD</jats:styled-content>, or <jats:styled-content style="fixed-case">RR</jats:styled-content> of relapse of disease activity in quiescent <jats:styled-content style="fixed-case">IBD</jats:styled-content>, with 95% confidence intervals (<jats:styled-content style="fixed-case">CI</jats:styled-content>s).</jats:p></jats:sec><jats:sec><jats:title>Results</jats:title><jats:p>The search identified 12 253 citations. Twenty‐two <jats:styled-content style="fixed-case">RCT</jats:styled-content>s were eligible. There was no benefit of probiotics over placebo in inducing remission in active <jats:styled-content style="fixed-case">UC</jats:styled-content> (<jats:styled-content style="fixed-case">RR</jats:styled-content> of failure to achieve remission=0.86; 95% <jats:styled-content style="fixed-case">CI</jats:styled-content>=0.68‐1.08). However, when only trials of <jats:styled-content style="fixed-case">VSL</jats:styled-content>#3 were considered there appeared to be a benefit (<jats:styled-content style="fixed-case">RR</jats:styled-content>=0.74; 95% <jats:styled-content style="fixed-case">CI</jats:styled-content>=0.63‐0.87). Probiotics appeared equivalent to 5‐<jats:styled-content style="fixed-case">ASA</jats:styled-content>s in preventing <jats:styled-content style="fixed-case">UC</jats:styled-content> relapse (<jats:styled-content style="fixed-case">RR</jats:styled-content>=1.02; 95% <jats:styled-content style="fixed-case">CI</jats:styled-content>=0.85‐1.23). There was no benefit of probiotics in inducing remission of active <jats:styled-content style="fixed-case">CD</jats:styled-content>, in preventing relapse of quiescent <jats:styled-content style="fixed-case">CD</jats:styled-content>, or in preventing relapse of <jats:styled-content style="fixed-case">CD</jats:styled-content> after surgically induced remission.</jats:p></jats:sec><jats:sec><jats:title>Conclusions</jats:title><jats:p><jats:styled-content style="fixed-case">VSL</jats:styled-content>#3 may be effective in inducing remission in active <jats:styled-content style="fixed-case">UC</jats:styled-content>. Probiotics may be as effective as 5‐<jats:styled-content style="fixed-case">ASA</jats:styled-content>s in preventing relapse of quiescent <jats:styled-content style="fixed-case">UC</jats:styled-content>. The efficacy of probiotics in <jats:styled-content style="fixed-case">CD</jats:styled-content> remains uncertain, and more evidence from <jats:styled-content style="fixed-case">RCT</jats:styled-content>s is required before their utility is known.</jats:p></jats:sec>