The role of probiotics in pouchitis Georgios Nalmpantidis, Theofanis Maris.

In their invited narrative review article, Gionchetti et al (April-June issue) [1] report 6 studies in which probiotics were used for maintenance of remission in pouchitis or induction of remission in acute pouchitis. In all but one studies VSL#3 probiotic was used. In 4 studies treatment group (VSL#3 regi-men was used in 3 studies) was compared to a control group [2-5]. [Q1. references are missing]. The studies of VSL#3 may form a coherent group, although the duration of treatment varied between 9 and 12 months. A recent meta-analysis of probiotic efficacy for gastrointestinal diseases synthesized data of 4 studies of pouchitis treatment [6]. In this systematic review the relative risk ratio for the probiotic group was 0.17 [95% Confidence Interval (CI) 0.10-0.30]. However, most of the studies cited by the article of Gionchetti et al were not included in this meta-analysis. The heterogeneity among the 3 probiotic-control studies referred by Gionchetti et al is not statistically significant (Cochrane’s Q for Odds Ratio, Risk Ratio and Risk Difference is 3.435, 0.040 and 4.730, with 2 degrees of freedom, and P value of 0.178, 0.980 and 0.094, respectively). Using a fixed-effect meta-analysis model (NCSS 2007 software) the combined Odds Ratio for the disease is

0.036 (95% CI 0.011-0.113), the Risk Ratio for the disease is 0.182 (95% CI 0.100-0.328) and Risk Difference is -0.718 (95% CI -0.835 - -0.602) in favor of the probiotic group (Fig. 1). These results are in accordance with the findings of the systematic review [6]. Statistically significant heterogeneity will be detected if the study of Kuisma (Lactobacillus rhamnosus GG was used for 3 months) is included in the meta-analysis model.

Figure 1

Forest plot of Risk Ratio of pouchitis in patients receiving VSL#3

It is of interest that a case report of 2 patients suggested that another type of probiotic, Escherichia coli Nissle 917, might be beneficial for the treatment of active pouchitis and for maintenance therapy [7].

The results of small clinical trials and anecdotal reports indicate the urgent need for large-scale, randomized, placebo-control trials.