Systematic Review With Meta-analysis

The Efficacy of Prebiotics, Probiotics, Synbiotics and Antibiotics in Irritable Bowel Syndrome

Alexander C. Ford; Lucinda A. Harris; Brian E. Lacy; Eamonn M. M. Quigley; Paul Moayyedi


Aliment Pharmacol Ther. 2018;48(10):1044-1060. 

In This Article


This systematic review and meta-analysis has demonstrated that particular combinations of probiotics, or specific species and strains, appear to have beneficial effects in IBS in terms of effect on global IBS symptoms and abdominal pain, but it is not possible to draw definitive conclusions about their efficacy. However, there was significant heterogeneity between studies, and evidence of publication bias or other small study effects, in some analyses. We found evidence to support the use of combinations of probiotics as a group, and for particular combinations, although in small numbers of RCTs. In terms of individual probiotics, Lactobacillus plantarum DSM 9843, E. coli DSM1752 and Streptococcus faecium, also appeared beneficial, although the latter two were only used in one RCT each. There was also a trend towards a beneficial effect of Bifidobacterium, in terms of improvement of global IBS symptoms and pain scores, although which particular strain or species may be of benefit remains unclear. The largest trial was a dose-ranging study of Bifidobacterium infantis 35 624, and demonstrated efficacy, in terms of global symptoms and abdominal pain, at a dose of 1 × 108 CFU.[63] Overall, rifaximin was also superior to placebo for the treatment of nonconstipated IBS, with a NNT of 9. There was only one trial each of norfloxacin and neomycin, making it difficult to draw any firm conclusions regarding their efficacy. The RR of adverse events was not significantly greater with either probiotics or antibiotics. Data for both prebiotics and synbiotics were sparse, with neither appearing to be of particular benefit in IBS, albeit in only five trials in total.

We used rigorous and reproducible methodology when conducting this systematic review and meta-analysis. We reported our search strategy in full, and performed the assessment of eligibility and data extraction independently, and in duplicate. We used an intention-to-treat analysis and pooled data with a random effects model, to minimise the likelihood that treatment effect would be overestimated. We also contacted investigators of potentially eligible studies to either obtain dichotomous data and continuous data. This inclusive approach has provided us with access to data for >5500 IBS patients treated with probiotics. Finally, we performed subgroup analyses in an attempt to assess treatment effect according to combinations of, and individual, probiotics used and we extracted and pooled adverse events data, where reported.

This updated meta-analysis identified a further 18 RCTs of probiotics and three trials of prebiotics since the previous iteration 4 years ago, but it is still not possible to draw clear inferences from the data concerning the efficacy and safety of either prebiotics or synbiotics. For probiotics, it remains unclear whether a particular combination of probiotics, or a specific species or strain, is more likely to be effective, or whether there is a particular IBS subtype that is more likely to benefit. Other limitations of this systematic review and meta-analysis arise from the nature of the studies available for synthesis. The risk of bias of many of the trials we identified was unclear, and there was evidence of heterogeneity between RCTs and publication bias in some of our analyses of probiotics. However, there was no heterogeneity between studies when only the five RCTs of rifaximin of similar design conducted in nonconstipated IBS were included, although the treatment effect in favour of rifaximin in these studies was modest.

The fact that there have been another 18 RCTs of probiotics conducted since the last version of this meta-analysis, only 4 years ago, underlines the continuing interest in the manipulation of the GI microbiome as a potential therapy for IBS. This systematic review provides support for the use of some probiotics to achieve this, but there are still insufficient data to recommend a specific species or strain of organism. In addition, there has been a further trial of rifaximin in IBS conducted in the last 2 years,[91] and the drug is now licensed for the treatment of IBS with diarrhoea in the US. This latter RCT studied the efficacy and safety of a further two 14–day courses of rifaximin in IBS with diarrhoea, following 2 weeks of open-label treatment with the drug, demonstrating that repeat treatment led to a durable and reproducible symptom response, which was superior to placebo in the original trial. However, the efficacy was modest after each course of treatment, and the long-term safety of repeated courses of rifaximin, and how many times to re-treat patients whose symptoms recur remains uncertain.

The rationale for the use of antibiotics in patients with IBS was based on diagnostic confusion between IBS and SIBO, with patients in the initial studies undergoing hydrogen breath testing to confirm the presence of SIBO prior to enrolment.[93,99] However, in the pivotal RCTs of rifaximin breath testing was only undertaken in a subset of individuals, and the results were not reported in full.[91,97] In addition, the mechanism of action of rifaximin in IBS remains unclear. A small mechanistic trial found no difference in terms of the faecal microbiome, intestinal permeability or faecal bile acid levels between individuals with IBS randomised to rifaximin or placebo,[100] but demonstrated an acceleration in ascending colon emptying times among those allocated to rifaximin. Given the drugs beneficial effects in patients with IBS with diarrhoea, this would seem paradoxical. Studies that have evaluated the effect of rifaximin on the microbiome, show that any changes are limited, and are not sustained.[100–102] Although the limited research regarding rates of C. difficile infection and microbial resistance are reassuring, continued monitoring of patients receiving repeated courses of the drug will be required. Additionally, advances in molecular techniques may provide further insight into the faecal microbiome of patients with IBS, which may in turn improve the understanding of the role of antibiotic therapy in the treatment of this complex disorder.

The mechanism of action of individual probiotics in improving symptoms in IBS also remains speculative. There have been previous studies conducted that have suggested that some probiotics, such as Lactobacillus acidophilus NCFM, have the ability to modify the expression of pain receptors in the gut in both mice and humans.[103,104] In addition, in one of the trials we identified, Bifidobacterium infantis 35 624 had the ability to normalise interleukin levels in patients with IBS.[60] More recently, the probiotic Bifidobacterium longum NCC3001 has been demonstrated to have a beneficial effect on depression scores among patients with IBS in a RCT.[47] Brain activation to fearful stimuli, seen on functional magnetic resonance imaging, was also reduced among patients allocated to the probiotic in this study. Interestingly, both this effect and the improvement in depression scores appeared to be most pronounced among those with adequate relief of their IBS symptoms. However, it is unlikely that these are class effects of probiotics, and further research in humans is required to identify species and strains of probiotics that are consistently beneficial, as well as to elucidate how these benefits are achieved.

In summary, this meta-analysis has demonstrated little evidence for the use of prebiotics or synbiotics in IBS. Amongst combination probiotics, LacClean Gold and the seven-strain combination of three Bifidobacterium, three Lactobacillus and one Streptococcus were associated with significant improvements in global symptoms, and there was a trend towards an improvement in global symptom scores or abdominal pain scores with VSL#3. Among individual probiotics, Lactobacillus plantarum DSM 9843, Escherichia coli DSM17252, and Streptococcus faecium also had beneficial effects on global symptoms. We could not show any evidence of benefit for any particular combination, strain or species of probiotics for the other endpoints of interest. Overall, therefore, it remains unclear which combination, species, or strain should be preferred in the individual patient. Five trials of similar design that used rifaximin demonstrated a consistent, although modest, benefit in IBS with a NNT of 9. Both probiotics and antibiotics appeared to be safe in IBS, but the longer terms effects of repeated treatment with the latter on the microbiome, and the safety of this approach, remains unclear.