There are few trials for treating mild-moderate pouchitis with small numbers of adult participants. Kuisma et al. [51] recruited 20 patients (10 intervention arm) for a DBRPC trial of L. rhamnosus GG 2 × 10¹⁰ CFU/day for 3 months. Patients with chronic, active pouchitis were excluded. The Pouchitis Disease Activity Index [52] was utilized for evaluation of clinical effect and prior to study entry, the mean PDAI was in the mild range (8.0 ± 0.8). There was no difference following the intervention period with clinical response (defined as a PDAI score reduction of ≥ 3) occurring in 1/10 (10 %) patients in the probiotic group and 0/10 (0 %) patients in the placebo group (10 % vs. 0 %, p = 0.32).

In an open-label trial of 51 UC patients with ileal pouch-anal anastomosis using a fermented milk product with a blend of probiotic strains (L. acidophilus strain La5 + B. lactis strain Bb12) containing 5 × 10¹⁰ CFU/day [53] there was a reported improvement in endoscopic evaluation. In another open-label trial [54], 16 of 23 patients (69 %) with mild pouchitis were in remission after treatment with VSL#3, and the median total Pouchitis Disease Activity Index scores reported before therapy improved following therapy (10 versus 4, p < 0.01).

Thus, there is limited evidence for a role of probiotics as monotherapy for mild-moderate pouchitis at the present time. Limiting access of microbiota to the mucosa of the pouch may be a mechanism, whereby probiotics provide benefit. Alternatively, changing the composition of the pouch microbiota may be important, although it is interesting that no long-term colonization of probiotic strains is achieved [55]. Thus, it may not be surprising that once the deleterious microbiota have colonized the pouch, probiotic monotherapy can do little to alter the situation.

Two trials have studied whether there is an advantage to initiate probiotics immediately following ileal pouch-anal anastomosis, and both found there to be benefit based on delayed onset of pouchitis. In a placebo-controlled trial [56], only 2 of 20 (10 %) subjects in the probiotic arm had developed pouchitis at the end of 1 year as compared to 8 of 20 (40 %, no episodes 80 % vs. 60 %, p = 0.03) in the control arm. The Peto odds ratio for prevention of pouchitis by the probiotic (VSL#3™) compared with placebo was 4.76, (95 % CI 1.16–19.56). Singh et al. [57] concluded for this part of their Cochrane review that low-quality evidence suggests that VSL#3 may be more effective than placebo for prevention of pouchitis.

The initial controlled trial for maintenance of remission of pouchitis was in the year 2000 using the blend probiotic product VSL#3™ and reported an outstanding effect in the prevention of recurrence in patients with antibiotic-dependent pouchitis. Prior to the administration of the blend probiotic, participants in his trial were successfully treated with a combination of antibiotics (ciprofloxacin + rifaximin). At the end of the 9-month study period, only 3 of 20 (15 %) subjects had developed pouchitis in the intervention group, whereas all 20 participants in the control group had a recurrence of pouchitis within 4 months of cessation of antibiotics [58]. A similar result was noted in another European trial of VSL#3™ also evaluating prevention of recurrence in relapsing or chronic pouchitis patients [55]. Remission of the pouchitis was induced in these participants by administering 4 weeks of a combination of antibiotics (metronidazole + ciprofloxacin) followed by either VSL#3™ or a placebo. In the probiotic treatment group, remission was maintained in 17 of 20 (85 %) subjects, but only 1 of 16 (6 %, p < 0.0001) receiving placebo. The pooled Peto odds ratio for these two studies for the combined rate of maintenance of remission with probiotic bacteria compared to placebo (97 % versus 3 %, p < 0.0001) was 25.39 (95 % CI 10.37–62.17). The number needed to treat with oral probiotic therapy to prevent one additional relapse was 2 [59]. However, a GRADE analysis indicated that the quality of the evidence supporting this outcome was low due to very sparse data [57].

In contrast, an open-label trial by Shen and colleagues [60] reported a poor response for prevention of pouchitis recurrence using the same probiotic. In their trial, 31 subjects were prescribed a 2-week treatment of a single antibiotic (ciprofloxacin) followed by VSL#3™. In contrast to the other studies, the VSL#3™ was bought by patients rather than supplied through the study. Probiotic therapy was stopped by 9 of 31 (29 %) 7 weeks into therapy and 25 of 31 (81 %) by 8 months because of either failure to prevent pouchitis (n = 23) or side effects of the probiotic administration (n = 2). Only 6 of 31 (19 %) subjects did not develop clinical evidence of pouchitis by the end of the 8-month trial period. Even among these six subjects, endoscopy revealed some level of pouch inflammation. In contrast to other studies, in this trial [60], there was a single antibiotic administered, and endoscopy was not performed prior to probiotic administration to ensure pouch inflammation had completely resolved.

For those with poor pouch function, pouch-related dysfunction was not improved in a study [61] of 33 patients randomized to receive probiotics (L. plantarum 299v + B. infantis CURE 21) or placebo as measured by the pouch functional score (p = 0.119), pouch disease activity index (p = 0.786), or any of the fecal biomarkers tested (calprotectin, lactoferrin, myeloperoxidase, and eosinophilic cationic protein).

A recent clinical practice guideline on management of pouchitis [50] did not recommend probiotics as sole therapy for acute treatment of pouchitis but did suggest VSL#3 (or chronic use of antibiotics) for patients with prompt recurrence of pouchitis following antibiotic usage or multiple recurrences of pouchitis despite antibiotics. Table 28.2 lists randomized trials of probiotics in prevention of onset or recurrence of pouchitis.

There is a paucity of studies investigating the use of probiotics to treat active inflammation. In two open-label studies [63, 64], probiotics (using combination of B. breve + L. casei + B. longum + prebiotics or L. rhamnosus GG, respectively) were added as adjuvant therapy to immunomodulators and corticosteroids. In the former study [55], seven of ten patients were reported to respond as determined by Crohn’s Disease Activity Index scores with the most noticeable improvement in diarrhea. However, there was no improvement in inflammation as measured by erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). In the open-label trial using L. rhamnosus GG [64], three of the four children were reported to have improved Pediatric Crohn’s Disease Activity Index (PCDAI) scores or serial determinations over the 6 months of the trial. Specifics with regard to ESR or CRP are not reported although the ESR is a component of the PCDAI [62].

A placebo-controlled trial using L. rhamnosus GG was the sole study included in a Cochrane review of efficacy of probiotic supplementation for the induction of remission in CD [65]. Four of five patients in the probiotic group achieved remission compared to five of six in the placebo group (OR 0.80, 95 % CI 0.04–17.20). Subjects received antibiotics and concurrent therapy of corticosteroids raising some methodological concerns. Thus, although this one small study did not show that probiotics had any effect in treating active CD, at most, one could conclude there is insufficient evidence regarding the effectiveness of probiotics for treatment of active CD. Subgroup analysis by Shen et al. [39] of three studies including patients with CD did not determine a significant benefit in favor of probiotics for inducing a remission or clinical response in CD (p = 0.35, RR=0.89).

Randomized trials (see Table 28.3) studying probiotics as maintenance therapy in CD have been performed, using probiotics as monotherapy or in combination with 5-aminosalicylates [66–69]. Among these studies [70–72], there were no differences in the number of relapses in patients receiving E. coli Nissle compared to placebo (p = 0.11), Saccharomyces boulardii (1 g/day) plus mesalazine (2 g/day) compared to mesalazine alone (3 g/day) (p = 0.08), or patients receiving L. rhamnosus GG (p = 0.77).

In the largest maintenance trial in children, Bousvaros et al. [66] reported no difference in relapse rate for subjects receiving L. rhamnosus strain GG 2 × 10¹⁰ CFU/day (31 %, 12 of 39) or placebo (17 %, 6 of 36, p = 0.18). The time to relapse is shown in Fig. 28.1, and although there was a trend to a shorter time to relapse in those receiving probiotics, the results were not statistically different (p = 0.10) in this study.

In a subgroup meta-analysis [(39)] there was no significant difference in clinical relapse (p = 0.71, RR=1.09; 7 studies) or endoscopic relapse (p = 0.75, RR=1.08) with the use of probiotics.

Another aspect of CD that has been studied is prevention of recurrence of disease following surgical resection. Details of the trials are included in Table 28.4. Three of the trials involved a single probiotic strain and two trials included a blend of probiotics. In the L. rhamnosus GG trial [67], 9 of 15 (60 %) in L. rhamnosus GG group in clinical remission had endoscopic recurrence, and 6 of 17 (35 %) in placebo group in clinical remission had endoscopic recurrence (p = 0.297). In the first L. johnsonii LA1 trials [68], endoscopic recurrence was seen in 21 of 43 (43 %) in the L. johnsonii LA1 group and 30 of 47 (64 %) in the placebo group at 6 months (p = 0.15). The second L. johnsonii LA1 trial [69] showed similar overall endoscopic scores between the probiotic and placebo groups (p = 0.48) and similar numbers of endoscopic recurrence (21 % of those taking L. johnsonii LA1 versus 15 % taking placebo (p = 0.33)). None of the secondary outcomes (clinical recurrence, histological score, C-reactive protein) showed any difference, either. Similarly, the two studies using a blend of probiotics failed to reveal any differences in endoscopic recurrence between treatment and placebo groups [73, 74]. A Cochrane meta-analysis [75] that combined all the studies failed to find efficacy for postoperative prophylaxis. The relative risk of clinical recurrence with any probiotic relative to placebo (n = 213) was 1.41 (95 % CI 0.59–3.36), any endoscopic recurrence (n = 333) was 0.98 (95 % CI 0.74–1.29), and severe endoscopic recurrence (n = 213) was 0.96 (95 % CI 0.58–1.59). In another meta-analysis [76] that reviewed the effect of Lactobacillus alone for maintenance of remission induced by medical or surgical intervention [(67–69, 73, 74)] no difference was found vs. placebo. However, for the trials that compared placebo specifically with L. rhamnosus GG, pooled estimates showed a significant benefit in placebo (RR 1.68; 05 % CI 1.07–2.64) compared to subjects receiving L. rhamnosus GG.

Subsequent to this analysis but consistent with these previous studies, a randomized trial of VSL#3 (9 × 10¹¹ CFU daily) for 90 days versus placebo (given within 30 days of ileocolic resection and reanastomosis) was performed. For those with no or mild endoscopic recurrence at day 90, VSL#3 was continued until day 365 [78]. At 90 days, a similar number of patients in each group had severe endoscopic lesions (9.3 % vs. 15.7 %, p = 0.19). At day 365, a similar number of patients that took the probiotic for all 365 days versus those that only took the probiotic from day 90 through 365 had severe endoscopic lesions (10.0 % vs. 26.7 %, p = 0.09).

In an open-label trial, 16 patients with either Crohn’s disease or ulcerative colitis completed a 3-month course of 9 × 10¹¹ CFU/day of a blend probiotic (VSL^#3™) to assess whether there was clinical improvement in arthralgia [79]. Participants had quiescent IBD at entry and no clinical or laboratory evidence of arthritis and were not taking nonsteroidal anti-inflammatory medications, and other medications were unchanged. An improvement in peripheral but not axial arthralgia was reported using an articular index score. There was no joint pain improvement as reported using a patient-completed visual analog scale. Notably, this study had a dropout rate of 45 %.

With the reported benefit of antibiotics in pilot clinical studies of primary sclerosing cholangitis [80], interest has extended to the potential benefits of probiotics in this condition. Fourteen participants with concurrent IBD were randomized to the treatment with a blend probiotic (L. acidophilus, L. casei, L. salivarius, L. lactis, B. bifidum, and B lactis; total daily dose of 10¹⁰ CFU/day) or placebo during 3 months in a double-blind crossover design that included a 1-month washout period [77]. The subjects remained on ursodeoxycholic acid during the trial. The results of this study showed no evidence of benefit from the probiotics on PSC-related symptoms, serum liver biochemistry, or liver function.