This uncommissioned review article was subject to full peer-review.
Irritable bowel syndrome (IBS) is a prevalent gastrointestinal disease with a substantial social and economic burden. Treatment options remain limited and research on the aetiology and pathophysiology of this multifactorial disease is ongoing.
To discuss the potential role of gut microbiota in the pathophysiology of IBS and to identify possible interactions with pathophysiologic targets in IBS.
Articles were identified via a PubMed database search [‘irritable bowel syndrome’ AND (anti-bacterial OR antibiotic OR flora OR microbiota OR microflora OR probiotic)]. English-language articles were screened for relevance. Full review of publications for the relevant studies was conducted, including additional publications that were identified from individual article reference lists.
The role of gut microbiota in IBS is supported by varying lines of evidence from animal and human studies. For example, post-infectious IBS in humans is well documented. In addition, certain probiotics and nonsystemic antibiotics appear to be efficacious in the treatment of IBS. Mechanisms involved in improving IBS symptoms likely go beyond mere changes in the composition of the gut microbiota, and accumulating animal data support the interplay of microbiota with other IBS targets, such as the gut–brain axis, visceral hypersensitivity, mucosal inflammation and motility.
The role of the gut microbiota is still being elucidated; however, it appears to be one of several important factors that contributes to the aetiology and pathophysiology of the irritable bowel syndrome.
Irritable bowel syndrome (IBS) is a chronic disease that may affect up to 20% of the population and has a negative social and economic impact on patients.[1-5] Globally, the pooled prevalence of IBS is approximately 11% but varies considerably, both by geographical location and by diagnostic criteria applied. Compared with individuals without IBS, patients with IBS symptoms report significantly reduced health-related quality of life and reduced work productivity.[1-3] Despite the substantial burden of IBS, treatment options remain limited and research on the aetiology and pathophysiology of this multifactorial disease is ongoing.[5, 7-9]
The gut microbiota has emerged as an important factor that may contribute to the pathophysiology of IBS. Proposed beneficial effects of the gut microbiota include maintenance of intestinal homoeostasis, maintenance of peristalsis, intestinal mucosal integrity, and protection against pathogens through bacterial antagonism and priming of host immune responses.[11, 12] The putative role of gut microbiota in IBS is supported by several lines of evidence with varying degrees of literature support. These include the differences between microbiota in IBS and non-IBS populations, the development of IBS after intestinal infection (i.e. post-infectious IBS), the interesting preliminary data on colonic faecal microbiota transplantation (FMT) and the efficacy of certain probiotics, prebiotics, synbiotics and nonsystemic antibiotics in the treatment of IBS.[13-23]
This narrative review will discuss the role of the gut microbiota in IBS, including possible interactions with other pathophysiologic targets in IBS. Articles included in this review were identified via the PubMed database through December 2013 using the following search string: ‘irritable bowel syndrome’ AND (anti-bacterial OR antibiotic OR flora OR microbiota OR microflora OR probiotic). The search was limited to English-language articles with the screening of >800 citation titles and abstracts to determine potential relevance. Review of more than 200 publications was conducted and included additional publications that were identified from citations within individual articles.
Microbiota as a target in IBS
Microbiota in IBS vs. the healthy human gut
Numerous studies have reported differences in the mucosal and/or faecal microbiota of patients with IBS compared with healthy controls.[13-16, 24-31] For example, Kassinen et al. extracted bacterial genomic DNA from faecal samples of patients with IBS (n =24) and controls (n =23); not only was the microbiota significantly altered in patients with IBS, but the composition varied depending on the predominant form of IBS. These findings were extended by a study examining the composition of faecal microbiota in patients with IBS (n =62) compared with healthy controls (n =46), in which researchers reported significant differences based on global and deep molecular analysis with duplicate microarray and quantitative polymerase chain reaction. In another investigation, bacterial DNA from faecal samples of patients with diarrhoea-predominant IBS (IBS-D, n =23) were compared with healthy controls (n =23), and taxonomic, structural and diversity differences in gut microbiota were reported. In particular, IBS-D was associated with significantly higher levels of Enterobacteriaceae and significantly lower levels of Faecalibacterium prausnitzii vs. healthy controls, which suggests an imbalance of advantageous and potentially harmful intestinal bacteria.
Biopsy studies have shown reduced mucosal microbiota diversity in patients with IBS compared with controls,[26, 28] and one study showed that the number of mucosal-associated bacteria in patients with IBS negatively correlated with number of stools passed. In a related study, two patients with IBS-D followed up for 6–8 weeks experienced temporal instability in faecal bacteria content that correlated with changes in symptomatology. A comparison of microbial diversity in luminal and mucosal niches of patients with IBS-D (n =16) and healthy controls (n =21) identified significantly lower faecal microbial biodiversity in patients with IBS-D. Compared with mucosal samples, microbial diversity was significantly higher in faecal samples in both patient groups. Interestingly, the difference in biodiversity in faecal vs. mucosal samples was greater in the healthy controls than in patients with IBS-D. Reduced diversity in mucosal-associated bacteria in IBS was also reported in a second study.
In contrast, another investigation reported no significant differences in mucosal and faecal microbial diversity in patients with IBS (n =47). This investigation did, however, support the concept of significantly greater variability in diversity in faecal samples from healthy controls (n =33) compared with those from patients with IBS. Variations in study findings may be related to differences in patient populations and methods of sample collection.
Small intestinal bacterial overgrowth (SIBO) has been associated with some cases of IBS; however, issues such as inconsistency of study findings, trial heterogeneity, methodological problems, lack of validation of small bowel culture techniques, lack of validation of breath tests and concerns regarding bias have led to the clinical relevance of SIBO in IBS being questioned.[18, 19, 33-35] A key issue is the lack of a gold standard for the diagnosis of SIBO; in particular, it has been demonstrated that the commonly used lactulose breath test does not reliably distinguish patients with IBS from healthy controls. Interestingly, scintigraphy studies suggest that breath testing measures variations in small bowel transit time occurring in patients with IBS, rather than the presence of SIBO. Therefore, SIBO may be a comorbid condition in a small subset of patients with IBS instead of having a direct pathophysiologic link to the development of IBS.
One of the most compelling arguments for microbiota involvement in IBS pathophysiology is that in some patients, an acute episode of gastroenteritis precedes the onset of IBS. A meta-analysis of eight studies reported an odds ratio (OR) of 7.3 [95% confidence interval (CI): 4.7–11.1] for developing IBS after a gastrointestinal infection, supporting a link between the two. Additional studies have supported that intestinal infection was strongly associated with a subsequent emergence of IBS symptoms. When data from nine prospective studies were pooled, the OR for developing post-infectious IBS was 5.9 (95% CI: 3.6–9.5), and the risk remained increased for up to 3 years post-infection. In the meta-analysis, factors associated with a greater risk of developing IBS post-infection were younger age; psychological disturbance (i.e. increased anxiety and/or depression as observed in four studies reporting Hospital Anxiety and Depression Scale scores); and the nature of the gastrointestinal illness (i.e. prolonged fever during the acute episode, possibly indicative of more severe illness). In a 2013 study, the faecal microbiota of patients with post-infectious IBS resembled that seen in idiopathic IBS-D.
IBS therapeutic approaches targeting the gut microbiota
Several treatments, including those with effects on the gut microbiota, may be potentially useful in the management of IBS, based on the current knowledge of IBS pathophysiology (Figure 1). Although any therapy that changes intestinal motility may have indirect effects on microbiota, the therapeutic approaches that are known to directly target the microbiota include FMT, probiotics and nonsystemic antibiotics.
Faecal microbiota transplantation
Preliminary data in patients with IBS suggest a favourable response to colonic FMT, which is consistent with the concept that the gut microbiota has a role in the pathogenesis of IBS. In one report, 10 patients (n =5 with IBS) received antibiotics and bowel lavage followed by a colonic infusion of faeces from a healthy donor. Examination of stool indicated that FMT resulted in a novel microbiota mainly composed of the bacterial species from the donor, and the microbiota composition remained generally stable over 24 weeks. Although of interest, the published evidence for FMT in IBS is limited, and larger comparative studies are needed to clarify the potential role for FMT in the management of IBS.
Probiotics, prebiotics and synbiotics in IBS
The efficacy of probiotics in IBS has been reviewed in several publications.[39-43] Overall efficacy has been modest, with variation in strains of probiotics showing potential benefit.[40-42] One area of concern is that many of the marketed probiotics have not been adequately evaluated in well-designed clinical trials. While meta-analyses have been positive, the pooled relative risks may be influenced by the limitations of study methodologies and the possibility of bias.[22, 41] Also, given the heterogeneity of both probiotic composition and patient populations studied, the ability to extrapolate findings is limited.[22, 45] A 2013 meta-analysis of probiotics in IBS attempted to address some of these limitations. The authors identified 10 studies meeting relatively robust criteria for inclusion (e.g. randomised placebo-controlled design, IBS defined by Rome criteria). As expected, efficacy varied by probiotic species as well as by the IBS symptom evaluated. Also, not all probiotic strains significantly improved pain, distention and flatulence, whereas improvements in other symptoms such as stool frequency, urgency and straining were not significant, and effects on quality of life were inconsistent with probiotic therapy.
The mechanism of action of probiotics in IBS may extend beyond the modulation of the microbiota composition.[46-48] For example, 4-week administration of a yogurt containing probiotic, with known benefits in IBS, to patients with IBS (n =19) did not alter the composition of their microbiota. However, the authors suggest that probiotic effects on the functionality of the microbiota may be of importance. In one study, probiotic administration has been reported to normalise cytokine levels [e.g. interleukin-10 (IL-10) and IL-12] in patients with IBS. A pilot neuroimaging study reported that healthy females (n =12) who ingested a fermented milk product with probiotic for 4 weeks had significant reductions in the activity of brain regions related to a sensory brain network compared with females who ingested a nonfermented milk control (n =11) or had no intervention (n =13).
While probiotics have been more widely studied, the potential benefits of prebiotics and synbiotics (formulations combining probiotics with prebiotics) have also been evaluated in the treatment of IBS.[51, 52] For example, the first investigation of a prebiotic (trans-galactooligosaccharide mixture) in patients with IBS was an 18-week study in 44 patients randomly assigned to one of three treatment groups: group 1 (n =16), which received placebo for 6 weeks followed by 3.5 g prebiotic for 12 weeks; group 2 (n =14), which received placebo for 6 weeks followed by 7.0 g prebiotic for 12 weeks; and group 3 (n =14), which received placebo for 18 weeks. The prebiotic treatment groups had qualitative changes in faecal flora compared with the placebo group (i.e. relative proportions of Bifidobacterium spp.) and significant improvements compared with the placebo group in terms of stool consistency, flatulence composite scores and patient subjective global assessments.
A synbiotic preparation (n =132) containing a probiotic [viable lyophilised Lactobacillus paracasei B21060 (5 × 109 colony-forming units)] and prebiotic [xylooligosaccharides (700 mg) and glutamine (500 mg)] or a control formulation (n =135) was administered for 12 weeks in patients with IBS based on Rome II criteria. The synbiotic preparation alleviated IBS symptoms but failed to demonstrate a significant benefit vs. the administration of a control formulation, except in a subgroup of patients with diarrhoea predominance (n =47; 27 received synbiotic, 20 received control). These results contrast with those of a previous uncontrolled open-label study of Bifidobacterium longum W11 and an oligosaccharide-based synbiotic preparation, which improved intestinal function in patients with constipation-variant IBS based on Rome II criteria.
Nonsystemic antibiotics in IBS
Most randomised, placebo-controlled studies have evaluated nonsystemic agents (e.g. neomycin, rifaximin) for the treatment of IBS (Table 1).[54-58] A randomised, double-blind study in patients with IBS meeting Rome I criteria compared neomycin [n =55; 25 patients with IBS-D, 18 with constipation-predominant IBS (IBS-C), 10 with other/unknown forms of IBS] with placebo (n =56; 21 patients with IBS-D, 20 with IBS-C and 15 with other/undetermined forms of IBS). A significantly greater reduction from baseline in IBS symptoms (primary endpoint: composite score based on abdominal pain, diarrhoea, constipation) was observed with neomycin compared with placebo (35.0 ± 5.0% vs. 11.4 ± 9.3%; P <0.05). A subanalysis of patients with IBS-C (neomycin, n =19; placebo, n =20) indicated that neomycin treatment resulted in a significantly greater global percentage improvement in IBS compared with placebo (36.7 ± 7.9% vs. 5.0 ± 3.2%; P <0.001).
Table 1. Published randomised, placebo-controlled, clinical data for antibiotics in the management of IBS
Two randomised, double-blind, placebo-controlled trials (n =1260) evaluated rifaximin 550 mg three times a day in patients with IBS-D meeting Rome II criteria. After 14 days of treatment, adequate relief of global IBS symptoms for ≥2 of the first 4 weeks post-treatment (i.e. primary endpoint) was achieved in significantly more patients in the rifaximin group compared with those in the placebo group in each study and in a pooled analysis (40.7% vs. 31.7%, pooled data; P <0.001), and relief was sustained for at least 10 weeks post-treatment. Improvement post-treatment was consistent with another smaller randomised, double-blind, placebo-controlled study of a lower dose of rifaximin (400 mg three times a day; n =43) vs. placebo (n =44) for 10 days in patients with IBS meeting Rome I criteria. Patients treated with rifaximin experienced a greater mean percentage global improvement in IBS at 10 weeks post-treatment compared with those receiving placebo (36.4 ± 31.5% vs. 21.0 ± 22.1%; P =0.02).
Another randomised, double-blind, placebo-controlled study (n =124 patients, with 20%, 38.3% and 41.7% receiving IBS-D, IBS-C and alternating IBS diagnoses respectively) evaluating rifaximin for chronic bloating and flatulence indicated that, in a subgroup of patients with IBS meeting Rome II criteria (n =70), a significantly larger response was observed with rifaximin 800 mg/day after 10 days compared with placebo (40.5% vs. 18.2% respectively; P =0.04). A meta-analysis of randomised, placebo-controlled studies evaluating rifaximin for IBS reported that global symptoms of IBS improved significantly with rifaximin vs. placebo (OR: 1.57; P <0.001), with a number needed to treat of 10.
In a 2012 observational study, patients with IBS (total n =106; 88 IBS-D, 7 IBS-C, 11 alternating IBS) were treated with rifaximin; patients responded (i.e. experienced reduction in bloating, flatulence, diarrhoea and pain) after 2 weeks, and results were maintained for at least 3 months. In addition, several retrospective chart reviews of rifaximin treatment for IBS suggest a low potential for loss of efficacy with repeated courses of rifaximin. In two retrospective chart reviews (both in patients with nonconstipated forms of IBS), clinical improvement was documented in ~75% of patients without loss of response during retreatments.[61, 62] Similarly, another retrospective chart review of rifaximin-treated patients meeting Rome I criteria for IBS showed that 69% of patients had a clinical response. This was greater than the response reported with other antibiotics (i.e. 38% of 24 patients treated with neomycin experienced a response). Retreatment with rifaximin (n =16) was associated with clinical improvement in all cases, whereas retreatment with other antibiotics (e.g. doxycycline, amoxicillin and clavulanate acid, neomycin) was effective in only two of eight cases (25%).
As suggested with probiotics, the action of nonsystemic antibiotics in IBS may extend beyond direct effects on the composition of the gut microbiota. For example, additional mechanisms of action for rifaximin have been proposed, such as a reduction in bacterial pathogen virulence, the stabilisation of gut mucosa (i.e. protection against bacterial infection), anti-inflammatory properties, and the preservation of normal colonic flora. Other potential pathogenic processes in IBS that may be modulated by rifaximin include visceral hypersensitivity and mucosal-immune reactivity (based on animal model data).
Interplay of microbiota and other treatment targets in ibs
Gut microbiota is one of many factors contributing to the pathophysiology of IBS, and each represents a potential target for therapy (Figure 1). Given the complexity of the interactions between these factors/targets, it is possible that therapies directly targeting the microbiota may also impact other pathways either directly or indirectly through their effects on the gut microbiota. The majority of the evidence supporting the potential interplay of the gut microbiota with other targets is derived from animal studies (where there is no suitable IBS model), meaning that in the absence of human studies, clinical relevance cannot be established. Nonetheless, these pre-clinical studies have led to the development of hypotheses on gut microbiota interaction and are laying the groundwork for future investigation.
The important role of the brain–gut/gut–brain axis in IBS and in other functional gastrointestinal diseases has been previously reported on,[68-72] while the concept of a microbiota–gut–brain axis in health and disease is emerging.[73-75] Of related interest, in patients with cirrhosis (n =25), differences in gut microbiota were observed vs. healthy controls (n =10), and there was a direct correlation between several bacterial taxa and cognitive function, particularly in 17 patients with hepatic encephalopathy. As noted above, animal data interpretation has limitations, especially given the lack of suitable IBS animal models; however, pre-clinical data can provide insight into the potential relationship between the microbiota and the gut–brain axis. For example, mice administered a mixture of nonsystemic antibiotics had an altered microbiota, increased exploratory behaviour and less apprehensive behaviour compared with controls. After a 2-week antibiotic washout period, these behavioural changes were reversed as the microbiota normalised. The authors noted that, consistent with the behavioural changes, mice receiving the antibiotic mixture also had higher levels of central brain-derived neurotropic factor in the hippocampus and lower levels in the amygdala compared with controls. Collins et al. has provided a comprehensive review of data on the bidirectional interactions between the gut microbiota and the brain (animal-based studies), noting that while animal data support the influence of gut microbiota on brain development and function, information elucidating the possible underlying mechanisms is lacking. Current knowledge on how intestinal microbiota dysbiosis, the gut–brain axis and pathophysiologic changes help to explain the disease pathophysiology in IBS is summarised in Figure 1.
Inflammation and visceral hypersensitivity
The gut microbiota may contribute to the low-grade inflammation and intestinal immune activation described in IBS through effects on cytokine levels and toll-like receptor activity.[24, 78, 79] In a study of patients with IBS (n =77) undergoing colonoscopy to rule out inflammatory bowel disease, patients were categorised based on biopsy as non-inflamed IBS, nonspecific microscopic colitis or lymphocytic colitis. Increases in lymphocytic populations were observed in all patient subgroups, even those with no overt signs of inflammation, suggesting a pathophysiologic role of immune activation in IBS. The authors speculated that bacterial antigens could be one of the factors triggering immune activation.
Visceral hypersensitivity has also been implicated in IBS, and the influence of gut microbiota on both visceral hypersensitivity and inflammation has been suggested in a variety of animal models, several of which are summarised in this section. In one study, the transfer of faecal microbiota of patients with IBS to germ-free rats was accompanied by a transfer of visceral hypersensitivity (assessed by colorectal distention) when these IBS human microbiota-associated (HMA) rats were compared with healthy HMA rats. An investigation of whether changes in gut flora and gut inflammatory cell activity impacted visceral hypersensitivity in mice demonstrated that, in the absence of sterile precautions (i.e. allowing for the fluctuation of gut bacterial content), the mice had a substantial increase in visceral sensitivity over time that was associated with a slightly increased activity of inflammatory cells. When an anti-inflammatory agent (i.e. dexamethasone) was administered, both inflammatory activity and visceral hypersensitivity were reduced, lending further support to the interplay between inflammation and visceral sensitivity. Overall, these results support the hypothesis that perturbations of the gut microbiota are associated with small changes in inflammatory activity in the gut that can change visceral perception; this is a possible rationale for the administration of certain probiotics for IBS. To assess for a protective effect from antibiotic-induced increases in visceral sensitivity, a probiotic (L. paracasei) was coadministered with an antibiotic in mice. Visceral hypersensitivity and histology improved; however, total Lactobacilli populations were undetectable, suggesting that the beneficial effects observed may not simply be related to recolonisation of the microbiota.
In rats with induced post-inflammatory chronic hypersensitivity to colorectal distention, the administration of a probiotic resulted in the normalisation of visceral sensitivity.
The protective effect of three different probiotic strains was assessed in rat models of psychological stress (i.e. chronic maternal deprivation-induced visceral hypersensitivity associated with altered colonic paracellular permeability or acute partial restraint-induced visceral hypersensitivity). Only one of the three probiotic strains, L. paracasei NCC2461 (Lpa), demonstrated reversal of visceral hypersensitivity to colorectal distention and complete restoration of gut paracellular permeability. Thus, effects of probiotics on visceral hypersensitivity appear to be strain specific. In the partial restraint-induced animal model, anti-nociceptive effects on visceral hypersensitivity were observed only when Lpa and the spent culture medium were both present. The authors suggested a potential synergistic interplay between the probiotic and intraluminal bacterial products.
The effect of oral rifaximin on gut microbiota, intestinal inflammation and visceral hyperalgesia was evaluated in a model of visceral hyperalgesia (chronic water avoidance and restraint stress in adult Wistar rats). Rifaximin administration significantly reduced the ileal bacterial load and altered the bacterial composition in the distal ileum (i.e. dominance of Lactobacilli post-treatment). In addition, rifaximin inhibited mucosal inflammation, barrier impairment and visceral hyperalgesia. Although these animal data are interesting, the associations between gut microbiota and inflammation and visceral hypersensitivity remain speculative and require support from clinical investigations in patients with IBS.
In addition, bacterial factors in the gut may cause changes in motor function that may underlie IBS symptoms. Colonisation with different bacterial species in rats showed that changes in the composition of gut microbiota were associated with altered (either impeded or increased) gastrointestinal motility. It has also been reported that bacterial components (e.g., lipopolysaccharides) may interact with specific toll-like receptors and directly affect the contractile function of human colonic smooth muscle cells. Anti-motility agents are often used in IBS-D, and their effect may be related in part to changes in gut flora.
Animal and human data support an interplay between the gut microbiota and pathophysiologic factors targeted by agents evaluated for the management of IBS. Further research on bacterial populations identified in stool and associated with intestinal mucosa of healthy individuals may provide new insight and opportunities to normalise gut homoeostasis in patients with IBS. Currently, several probiotics and nonsystemic antibiotics are efficacious for the treatment of patients with IBS. However, given the complexity of the IBS disease process, it is likely that the mechanisms of these agents are not limited to their direct effects on microbiota composition.[9, 48, 89-92] Gut microbiota appears to be one of several important factors that may contribute to the aetiology and pathophysiology of IBS. Future research on metabolic and/or immunological pathways and microbial genes in the gastrointestinal tract (metagenomics) and their effects on protein expression will help elucidate the role of gut microbiota in this burdensome condition.[19, 75]
While much is known about the pathophysiology of IBS and the role of the microbiota, a great deal more is unknown. Studies are needed to further refine the role of the intestinal microbiota in the gut–brain axis and in the chronic inflammation associated with the disorder, and how these factors influence the important intestinal events in the syndrome. As we define these factors, IBS will move from a functional syndrome to a disease with a defined biochemical and immunological origin.
Guarantor of the article: H. L. DuPont.
Author contributions: H. L. DuPont provided concept and scientific content for the development of the manuscript, critically reviewed and edited all drafts of the manuscript, and approved the final version of the manuscript.
Declaration of personal interests: In 2011, Dr DuPont served as a consultant to Salix Pharmaceuticals, Inc. on the topic of antibiotic resistance with chronic use of rifaximin and has received research grants from Santaurus, Inc., a company that has been acquired by Salix Pharmaceuticals, Inc.
Declaration of funding interests: Writing and editorial support were provided under the direction of the author by Mary Beth Moncrief, PhD, and Kulvinder Singh, PharmD, Synchrony Medical Communications, LLC, West Chester, PA, and funded by Salix Pharmaceuticals, Inc.