Modulation of the gut microbiota: a focus on treatments for irritable bowel syndrome

ABSTRACT

Irritable bowel syndrome (IBS), which is characterized by recurrent abdominal pain and disordered bowel habits, is one of the most common functional bowel disorders. IBS is a substantial burden on both patient health-related quality of life and healthcare costs. Several pathophysiologic mechanisms have been postulated for the occurrence of IBS, including altered gastrointestinal motility, visceral hypersensitivity, changes in gut permeability, immune activation, gut-brain dysregulation, central nervous system dysfunction, and changes in the gut microbiota. Of note, both qualitative and quantitative differences have been observed in the gut microbiota of a population with IBS versus a healthy population. Because of the substantial interest in the gut microbiota and its role as a therapeutic target in IBS, this article provides an overview of specific interventions with the potential to modulate the gut microbiota in IBS, including elimination diets, prebiotics, probiotics, synbiotics, and nonsystemic antibiotics. Although probiotics and synbiotics are generally well tolerated, differences in the composition and concentration of different bacterial species and inclusion or exclusion of prebiotic components varies widely across studies and has prevented strong recommendations on their use in IBS. For nonsystemic antibiotics, rifaximin is indicated in the United States for the treatment of IBS with diarrhea in adults and has been shown to be efficacious and well tolerated in well-designed clinical trials. Overall, more consistent evidence is needed regarding the efficacy and safety of elimination diets, prebiotics, probiotics, and synbiotics for the treatment of patients with IBS. Furthermore, additional well-designed studies are needed that examine alterations in the gut microbiota that occur with these interventions and their potential associations with clinical symptoms of IBS.

KEYWORDS:

• Antibiotic
• constipation
• diarrhea
• diet
• irritable bowel syndrome
• microbiota
• prebiotic
• probiotic
• rifaximin

1. Introduction

Irritable bowel syndrome (IBS) is a common functional bowel disorder characterized by recurrent abdominal pain, bloating, and changes in stool form or frequency without the presence of structural, biochemical, or major inflammatory changes [1]. Overall, the reported global pooled prevalence of IBS is 11.2% [2]; however, regional and cultural variations exist [2,3]. A 2017 meta-analysis performed by the Rome Foundation confirms significant heterogeneity and methodological variance discouraging the use of pooled data when describing prevalence rates [3]. Updated Rome diagnostic criteria in 2016 (Table 1) emphasized recurrent abdominal pain for the definition of IBS and set new symptom frequency criteria: on average, at least 1 day per week of abdominal pain during the last 3 months, either related to defecation or associated with a change in stool form or frequency [4]. Furthermore, these 3-month criteria for IBS should be fulfilled with symptoms starting at least 6 months before the diagnosis [4]. In addition, IBS is further characterized by predominant bowel type: diarrhea (IBS-D), constipation (IBS-C), or mixed (IBS-M); these different IBS types have reported rates of 26, 28, and 44%, respectively, in the United States [4,5]. Other symptoms such as bloating and abdominal distention may occur, and the predominant bowel symptom subtype may change (e.g. IBS-D to IBS-M) over time [6].

Table 1. Rome IV criteria for irritable bowel syndrome.

Recurrent abdominal pain, on average, ≥1 d/wk in the last 3 mo, associated with ≥2 of these 3 criteria:	Defecation related
	Change in stool frequency
	Change in stool form (appearance)
Symptoms present for the last 3 mo with onset ≥6 mo before diagnosis

Adapted and used with permission from Gastroenterology, vol 150, Lacy BE, et al. ‘Bowel disorders,’ pp 1393–1407, Copyright Elsevier (2016) [4].

IBS negatively impacts health-related quality of life (QOL) in affected individuals and is associated with decreased work productivity and increased healthcare utilization and healthcare costs [7,8]. For patients with IBS-C, total direct annual costs were estimated at more than $11,000 (2010 US dollars), with 53.7% of costs resulting from outpatient care (e.g. office visits, diagnostic tests, outpatient procedures); hospitalizations, medications, and emergency department visits accounted for 21.8, 19.1, and 5.4%, respectively, of total costs [9]. Although primarily managed in an outpatient setting, IBS accounted for approximately 12,000 US hospitalizations annually (1997–2010), with a mean duration of 3.7 days [10].

IBS can be classified as mild, moderate, or severe depending on various measures that include symptoms (e.g. abdominal pain), the presence of comorbid health conditions (e.g. fibromyalgia, anxiety/depression), health-related QOL, and healthcare utilization [11–13]. Although an older study estimated that 5% of patients had severe IBS [14], findings from subsequent studies have suggested that severe disease may affect 15–50.4% of patients with IBS [15,16].

The patient population seen at various healthcare settings tends to differ by IBS severity, with patients in primary care typically presenting with milder disease [17]. To help healthcare providers determine IBS severity, the Rome Foundation has developed a useful tool called the multidimensional clinical profile (MDCP). This tool can be used to assist healthcare providers in assessing IBS severity and can also be helpful for establishing a treatment approach. The MDCP defines a number of disease features, including determining whether the patient meets Rome criteria for a diagnosis of IBS, IBS subtype, psychosocial factors, effects of IBS on daily activities, and the presence of biomarkers or known physiologic modifiers associated with IBS (e.g. positive breath test results).

IBS is a diverse disorder, and many factors have been implicated in its pathogenesis (Table 2) [18]. Not all factors may be present in one individual, but genetic, central (e.g. gut-brain dysregulation), hormonal, psychological, diet, and gastrointestinal (GI; e.g. altered motility) factors have all been proposed to be potential candidates [18–21]. Further, it is likely that IBS is a result of various interactions between genetic and environmental factors [22]. Besides the impact of diet, research has noted the role of immune-mediated factors, especially the effect of an infection on the gut microbiota (e.g. post-infectious IBS) as well as intestinal dysbiosis (e.g. small intestinal bacterial overgrowth [SIBO]) [21,23]. The gut microbiota appears to play an important role in IBS (Figure 1), as quantitative and qualitative differences have been observed in patients with IBS compared with healthy individuals [24–29]. For instance, a greater magnitude of gut microbiota instability has been associated with greater intensity of symptoms in some patients with IBS [24]. Also, some clinical characteristics of IBS (e.g. bloating, satiety, altered intestinal transit time) have been associated with specific gut microbiota profiles in patients with IBS [30,31] and increases in some proinflammatory cytokines. Because of the substantial interest in the gut microbiota and its role in disease, the purpose of this review is to provide an overview of specific interventions with the potential to modulate the gut microbiota in IBS, including elimination diets, prebiotics, probiotics, synbiotics, and antibiotics.

Table 2. Proposed pathophysiologic factors in irritable bowel syndrome.

Proposed site of defect	Proposed defect or problem
Genetic^a	Abnormality in serotonin receptors Abnormality in sodium ion channels Altered bile acid metabolism Defect in protein involved in immune response
Central nervous system	Central nervous system processing of afferent gut signals Hormonal influences (e.g. estrogen, progesterone) Visceral hypersensitivity
Gastrointestinal	Altered motility Immune-mediated changes secondary to infection Increased levels of mast cells Increased intestinal permeability Diet (e.g. food sensitivity or intolerance) Changes in gut microbiota/microbiome (e.g. dysbiosis, antibiotic administration) Hormonal influences (e.g. estrogen, progesterone)
Psychological	Anxiety Depression Stress (e.g. coping) Somatization Posttraumatic stress disorder Abuse (e.g. early adverse life events, physical, verbal, or sexual)

^aMay be limited to select individuals.

Adapted and used with permission from Gastroenterol Clin N Am, vol 45, Harris LA, et al., Irritable bowel syndrome and female patients, pp 179–204, Copyright Elsevier (2016) [18].

Figure 1. Impact of gut microbiota in IBS. The potential role of gut microbiota in IBS symptoms and putative modulators of gut microbiota (e.g. diet, probiotics, antibiotics).

FODMAP: fermentable oligo-, di-, and monosaccharides and polyols; IBS: irritable bowel syndrome.

2. Methods

A PubMed search of English-language articles available through 24 August 2017 was conducted using the following key words to identify studies performed in adult humans: ‘irritable bowel syndrome AND diet,’ ‘irritable bowel syndrome AND FODMAP,’ ‘irritable bowel syndrome AND prebiotic,’ ‘irritable bowel syndrome AND probiotic,’ ‘irritable bowel syndrome AND Bifidobacterium,’ ‘irritable bowel syndrome AND Lactobacillus,’ ‘irritable bowel syndrome AND VSL 3,’ ‘irritable bowel syndrome AND antibiotic,’ and ‘irritable bowel syndrome AND rifaximin.’ Reference lists from review articles were used to identify additional studies for inclusion.

3. Role of diet in IBS

Exacerbation of symptoms among patients with IBS has been described to occur commonly in relation to food ingestion [32–34]. While true food allergies are considered uncommon in IBS [1], food intolerances or sensitivities are frequently reported [35]. In a general population study, only 19.4% (95% confidence interval [CI], 11.4–27.4%) of patients who believed they had a food intolerance had a positive reaction during a blinded food challenge [36]. Over the last few years, food-related symptoms have received increased attention, and emerging evidence supports the evaluation of diet modification for patients with IBS [37]. In fact, approximately 90% of patients with IBS restrict their diet to prevent or improve GI symptoms [35].

One potential mechanism for the important role diet may play in IBS involves endocrine cells localized to the GI lumen. GI endocrine cells respond to stimuli from both dietary components (e.g. carbohydrates, proteins) and bacterial products by releasing hormones and influencing cell density [38]. Diet has also been shown to modulate the composition and function of the gut microbiota, which, in turn, may be associated with symptoms of IBS (e.g. bloating, altered intestinal transit time) [39].

3.1. Elimination diets

3.1.1. Gluten-free diets

Nonceliac wheat sensitivity is characterized by symptoms similar to those of IBS, including abdominal pain, diarrhea, constipation, nausea, vomiting, headache, and depression, with symptoms typically improving when wheat-containing products are removed from the diet [40,41]. However, before entertaining a diagnosis of nonceliac wheat sensitivity, patients should be properly evaluated with celiac serological screening and an upper endoscopy with biopsies to rule out celiac sprue. Because studies have shown that a gluten-free diet can improve IBS symptoms, at least in some patients, nonceliac wheat sensitivity may comprise a subset of patients with IBS [40,42]. A gluten-free diet was also shown to improve IBS-related QOL from baseline in some patients with IBS [43]. This may partly be related to GI mucosal changes observed in patients with IBS with a suspected food intolerance; these changes include epithelial barrier disruptions, fluid secretion into the GI lumen, and an influx of intraepithelial lymphocytes [44].

The component(s) of wheat responsible for this effect need to be elucidated, although proteins (e.g. gluten) and carbohydrates (e.g. fermentable oligo-, di-, and monosaccharides and polyols [FODMAPs]) are thought to play a role [41]. Results of controlled studies in patients with IBS receiving gluten-free diets have been mixed [42,45], with some suggesting that gluten contributes little to IBS symptomatology and that fructans contained in wheat are actually the culprits [45]. This was concluded in a double-blind, crossover trial of 37 patients with IBS with self-reported nonceliac wheat sensitivity who received a reduced FODMAP diet during a 2-week run-in phase before randomization to a diet high in gluten (16 g/d), low in gluten (2 g/d plus 14 g whey protein/d), or placebo for 7 days [45]. Global and specific symptoms (i.e. abdominal pain, bloating, satisfaction with stool consistency, flatulence, and fatigue) improved significantly from baseline to the second week of the run-in phase (P < 0.0001, for all comparisons). However, global symptoms and abdominal pain worsened from the end of the run-in phase to day 7 of treatment [45]. Conversely, a single-center, controlled study of 45 patients with IBS-D showed that patients randomly assigned to a gluten-free diet had a significant decrease in daily stool frequency from baseline compared with a diet containing gluten after 4 weeks (P = 0.04) [42]. These results raise the possibility that the positive effect of a gluten-free diet in patients with IBS may be a nonspecific consequence of reducing the intake of FODMAPs, given that wheat is one of the possible sources of FODMAPs. Another possibility is the potential of the gluten-free diet to alter the composition of the gut microbiota, which has been demonstrated in healthy individuals [46]. However, data are currently lacking regarding the effects of a gluten-free diet on gut microbiota of patients with IBS.

3.1.2. Low FODMAP diet

FODMAPs contain carbohydrates that are poorly absorbed in the small intestine and can pass unabsorbed into the colon where they increase luminal water levels through osmotic activity; further, these unabsorbed carbohydrates may undergo fermentation by colonic bacteria, which induces gas production [47,48]. Gas production, in turn, can lead to luminal distension, which induces GI symptoms in susceptible individuals [47,49]. Fecal samples obtained from patients with IBS who consumed a 3-week low FODMAP diet had a significantly lower absolute abundance of bacteria compared with those who consumed a typical diet (P < 0.001), including the butyrate-producing bacteria Clostridium coccoides (P < 0.001) and mucus-degrading bacteria Akkermansia muciniphila (P < 0.001); the abundance of mucus-degrading bacteria Ruminococcus torques was significantly greater with a low FODMAP diet versus a typical diet (P = 0.001) [50]. Patients with IBS following a low FODMAP diet had significant decreases from baseline in the quantity of fecal bacteria (i.e. Mycoplasma hominis, Bifidobacterium, Actinobacteria) after 4 weeks [51]. However, the implications of a long-term low FODMAP diet on the gut microbiota in IBS are currently unknown [50].

A randomized, single-blinded, controlled study (n = 75) published in 2015 compared a low FODMAP diet with a traditional IBS controlled diet (e.g. ingestion of three meals and three snacks daily; decreased consumption of fatty or spicy foods, coffee, alcohol, cabbage, onions, and beans; avoidance of carbonated beverages) in patients with IBS. The study reported that the intensity of IBS symptoms was significantly decreased from baseline with both interventions after 29 days (P < 0.001) [52]. A meta-analysis of 10 prospective clinical studies of the low FODMAP diet in patients with IBS showed that a standard IBS diet (high FODMAP) and low FODMAP diet were both efficacious for improving IBS symptoms [53]. Differences between these two groups did not achieve statistical significance, possibly because of study heterogeneity. However, a crossover clinical study of 30 patients with IBS noted that a low FODMAP diet significantly decreased overall IBS-related symptoms in as early as 7 days, with effects maintained for the duration of the intervention (up to 3 weeks), compared with a typical Australian diet (P < 0.001) [54]. The low FODMAP diet also significantly improved bloating and abdominal pain and decreased dissatisfaction with stool consistency in patients with IBS compared with the typical diet [54]. Furthermore, a systematic review of dietary fiber supplementation (3 studies) or a low FODMAP diet (5 studies) found these dietary interventions to be effective management options for patients with IBS-C, although the authors concluded that not all patients would benefit from these interventions because of the wide variation in both the pathophysiology of the condition and the symptoms experienced by patients [55].

In general, exclusion diets (e.g. gluten-free, low FODMAP) appear to be efficacious because of improvements in global and individual IBS symptoms in patients, but more evidence is needed before dietary restriction can be strongly recommended for these patients [56]. Although clinical trial data support that the reduction of FODMAPs is beneficial for patients with IBS [52,54,57], long-term adherence to a strict low FODMAP diet is not recommended because of nutritional concerns and potential effects on the gut microbiota [58]; rather, excluded components of the diet should be slowly reintroduced until satisfactory symptom control is maintained by the individual patient [59]. Interestingly, a study conducted in England reported that primary care physicians (PCPs) generally recommend dietary modification first to patients with IBS [60], but based on the very low quality of evidence, the American College of Gastroenterology (ACG) makes only a weak recommendation for specialized diets in the management of IBS [1].

4. Prebiotics, probiotics, and synbiotics in the treatment of IBS

4.1. Prebiotics

Prebiotics are dietary products not digested in the human GI tract, but selectively fermented by specific genera of resident gut microbiota [61]. Prebiotics promote the growth of host bacteria when ingested, potentially affecting the composition and function of the gut microbiota [62,63]. Bifidobacteria concentrations have been found to be lower in patients with IBS compared with healthy controls [64,65]; therefore, as a therapeutic target, specific prebiotic-stimulated growth of bifidobacteria is a potential treatment option. The two most investigated carbohydrates that fulfill the criteria for prebiotics are inulin-type fructans and the galacto-oligosaccharides (GOS) [61]. Inulin-type fructans are fructose polymers linked by β bonds with a terminal α-linked glucose [61,66]. Longer chains are inulin, and shorter chains are oligofructose/fructo-oligosaccharides (FOS) [61]. Use of a prebiotic mixture comprised of trans-GOS, obtained through enzymatic galactosylation of lactose (Clasado, Milton Keynes, UK), showed beneficial effects in a randomized, single-blinded, placebo-controlled, crossover study of patients with IBS (n = 44 completed study) [67]. Patients with IBS who received the prebiotic mixture 3.5 g/d experienced significant improvements in stool consistency, flatulence, bloating, composite symptom score, and subjective global assessment (SGA) compared with baseline after 4 weeks of treatment (P < 0.05, for all vs. baseline) [67]. The prebiotic mixture 7 g/d significantly improved SGA and anxiety levels from baseline in patients with IBS after 4 weeks of treatment (P < 0.05, for both vs. baseline) [67]. Mild nausea was reported by one patient who received the prebiotic mixture 3.5 g/d [67]. Compared to baseline levels, both doses of the prebiotic mixture significantly increased fecal levels for the beneficial bacteria Bifidobacterium (P < 0.05 for both) [67]. Similarly, fecal Bifidobacterium levels significantly increased with the prebiotic mixture 3.5 g/d and 7 g/d compared with placebo after 4 weeks of treatment (P < 0.005 for both) [67]. FOS obtained by enzymatic synthesis from sucrose are naturally found in onions, bananas, and artichokes [61,68]. A randomized, double-blind study of healthy individuals with mild functional bowel symptoms demonstrated that regular consumption of short-chain FOS 5 g/day reduced the frequency and intensity of digestive symptoms and improved intestinal discomfort and QOL compared with placebo after 6 weeks [68]. Others have proposed that use of prebiotics may lead to short-term, potentially deleterious alterations in the microbiota, as was suggested by an increased number and volume of anal gas evacuations 2 days after initiation of supplementation with GOS by healthy individuals; however, after 3 weeks, the number and volume of evacuations decreased to pretreatment levels, suggesting that gut microbiota adapted to the prebiotics [69]. In this study, bifidobacteria abundance increased significantly from baseline to days 3 and 21 (P = 0.04 and P = 0.03, respectively, vs. baseline) for healthy individuals with bifidobacteria abundance <0.5% of total bacteria at baseline [69]. Using prebiotics to enhance the growth of bifidobacteria seems logical to reduce symptoms of IBS; however, studies are limited by the type and dose of prebiotic used. The ACG has determined that there was insufficient evidence to recommend prebiotic use in patients with IBS [1].

4.2. Probiotics

Probiotics are composed of live or attenuated microorganisms (i.e. bacteria, yeast) that may be beneficial to human health when ingested by, for example, boosting host immunity and inhibiting bacterial growth and viral adhesion [63,70,71]. Probiotics differ from each other in a number of ways, including bacterial composition, number of viable organisms, and activity [70]. Bifidobacterium and Lactobacillus strains are commonly used in probiotic formulations [72], with some commercially available.

A survey-based study of community and academic gastroenterologists and PCPs reported that all survey respondents considered probiotics to be safe; 98% of those surveyed believed that probiotics had a role in treating GI disorders, although some respondents also cited a lack of evidence to support the use of probiotics in patients with GI disorders [72]. Overall, 98% of physicians surveyed indicated using probiotics for patients with IBS, with 91% of survey participants recommending Bifidobacterium infantis 35624 (Align®, Proctor and Gamble, Cincinnati, OH) for patients with IBS [72]. Bifidobacteria have been shown to be inversely correlated with abdominal pain in healthy individuals [73], and patients with IBS have lower concentrations of bifidobacteria compared with healthy individuals [64,65], thus supplementation with bifidobacteria may alleviate symptoms in patients with IBS.

Results of a meta-analysis of 11 randomized, controlled clinical studies reported that short-term (i.e. 10–28 days) probiotic therapy could reduce intestinal transit time, although the magnitude of treatment effect was dependent on the bacterial strain of the probiotic, the presence of constipation, and older age of the patient [74]. Overall, the ACG concluded that probiotics improve symptoms of IBS; however, given the lack of consistent data among individual species, strains, and preparations, the ACG weakly recommended the use of probiotics in patients with IBS [1].

4.2.1. Bifidobacterium infantis 35624

Results from three randomized, controlled studies of B infantis 35624 indicated that some baseline symptoms of IBS improved after a 4- or 8-week course of daily B infantis 35624 therapy [75–77]. In a double-blind, placebo-controlled study, patients with IBS who were randomly assigned to receive B infantis 35624 capsules daily did not experience significant improvements in individual symptoms of IBS (i.e. abdominal pain, bloating, urgency, incomplete evacuation, straining, gas) compared with placebo after 8 weeks of treatment; furthermore, only fecal levels of the C coccoides-Eubacterium rectale group of bacteria, but not other bacteria evaluated, were significantly higher compared with placebo after 8 weeks of treatment (P = 0.02) [75]. In this study, the overall intensity of symptoms improved from baseline in both groups, but differences between groups were not significant [75]. B infantis 35624 was generally well tolerated, with the greatest frequency of adverse events (AEs) reported by patients with IBS receiving placebo (38%) compared with 33 and 32% of patients with IBS or healthy individuals receiving B infantis 35624, respectively [75].

In a randomized, double-blind, placebo-controlled, dose-ranging study of B infantis 35624 capsule administered once daily (range, 1 × 10⁶ to 1 × 10¹⁰ cells/dose) for 4 weeks, females with IBS (n = 362) who received B infantis 1 × 10⁸ cells/dose (n = 90) experienced significant improvement from baseline in abdominal pain/discomfort after 4 weeks of treatment (P = 0.02) [76]. Further, patients who received B infantis 1 × 10⁸ cells/dose experienced significant improvement from baseline in bloating/distension, sense of incomplete evacuation, flatulence, straining, and satisfaction with bowel habit compared with placebo [76]. In a separate study, patients with IBS randomly assigned to receive a once-daily malted milk drink containing B infantis 35624 (1 × 10¹⁰ live cells/dose) experienced lower composite symptom scores for pain/discomfort, bloating/distension, and bowel movement difficulty for most weeks during the 8-week treatment course compared with patients assigned to receive a drink containing either Lactobacillus salivarius UCC4331 or placebo (i.e. malted milk drink alone) [77].

4.2.2. Bifidobacterium lactis DN-173 010

Bifidobacterium lactis DN-173 010 (Activia®, Dannon, White Plains, NY) was shown to be efficacious in a female IBS-C population [78] and a healthy population with digestive symptoms [79,80]. A randomized, double-blind, controlled study examined the efficacy of fermented milk-containing B lactis DN-173 010 administered twice daily for 4 weeks in females with IBS-C (n = 34) who experienced bloating at least twice weekly [78]. Compared with patients who received a control solution, patients who received B lactis had an improvement from baseline in median percentage change of maximal abdominal distension (P = 0.02), decreases in orocecal (P = 0.049) and colonic (P = 0.03) transit times, and reduced intensity of IBS-related abdominal pain/discomfort (P = 0.04) and global IBS symptoms (P = 0.03) after 4 weeks of treatment.

In a randomized, double-blind, single-center, controlled study in generally healthy females with ≥1 minor digestive symptom (abdominal pain/discomfort, bloating, flatulence, rumbling stomach) in the previous month, fermented milk containing B lactis DN-173 010 (n = 100) or a nonfermented dairy product (n = 97) was administered twice daily for 4 weeks [79]. B lactis significantly improved overall GI well-being from baseline in a greater percentage of individuals compared with control during ≥2 of 4 weeks of treatment (52.0 vs. 36.1%, respectively; P = 0.02). Further, a randomized, controlled, open-label study of healthy adults with abdominal discomfort and stool frequency of 3–21 weekly stools who received either one (n = 144) or two (n = 147) 125-g servings of fermented milk containing B lactis DN-173 010 had significant improvement in digestive comfort compared with the control group (n = 69) after 2 weeks of treatment (82.5 and 84.3% vs. 2.9%, respectively) [80]. None of these three studies of B lactis DN-173 010 examined alterations in the gut microbiota.

4.2.3. Bifidobacterium animalis ssp lactis

Published data regarding the efficacy of B animalis ssp lactis in patients with IBS are lacking. In a study of healthy volunteers, individuals with a low frequency of weekly bowel movements (i.e. 2–4 d/wk) and abdominal discomfort were randomly assigned to receive 4 weeks of treatment with a probiotic containing B animalis ssp lactis BB-12 (1 billion colony-forming units [CFUs; n = 343] and 10 billion CFUs [n = 452]) or placebo (n = 453) [81]. Probiotic administration increased the likelihood of improving the frequency of weekly bowel movements during ≥2 of 4 weeks of treatment (odds ratio [OR], 1.3; 95% CI, 1.0–1.8). In addition, when the two probiotic formulations were analyzed together, probiotic administration significantly improved the mean number of bowel movements per week from baseline during the 4-week treatment period (P = 0.007); in addition, significant improvement versus placebo occurred during each of the 4 weeks, but varied by probiotic dose (1 billion CFUs, weeks 2–4 [P < 0.01]; 10 billion CFUs, weeks 1–2 [P < 0.05]). Authors noted no apparent differences in the frequency of AEs between groups. This study also did not evaluate potential alterations in the gut microbiota.

4.2.4. Lactobacillus casei Shirota

Two studies that examined the efficacy of Lactobacillus casei Shirota (Yakult®, Yakult USA Inc., Fountain Valley, CA) in patients with IBS reported mixed results, with patients achieving improvement in some, but not all, GI-related symptoms; these studies did not assess alterations in the gut microbiota [82,83]. A randomized, double-blind, placebo-controlled study of adults with IBS (n = 80) who received a solution containing L casei Shirota administered twice daily for 8 weeks did not achieve a significant improvement from baseline of ≥30% in mean symptom score nor did they achieve significant improvements in individual symptom scores for discomfort, pain, bloating, and flatulence, after 8 weeks of treatment [82]. However, patients who received the probiotic had an improvement from baseline of ≥30% in mean symptom scores for discomfort and flatulence (P ≤ 0.05) after 8 weeks posttreatment (i.e. 16 weeks) [82]. In a separate study, most (64%) patients with IBS with SIBO (n = 18) who received a solution containing L casei Shirota once daily no longer met the criteria for a SIBO diagnosis after 6 weeks. However, although flatulence improved significantly from baseline at 6 weeks of treatment (P = 0.04), no significant change from baseline was reported in overall abdominal symptoms (P = 0.3) [83].

4.2.5. Saccharomyces

Saccharomyces boulardii, used in food processing and for the treatment of IBS, was shown to significantly improve IBS symptom scores from baseline in patients with IBS-D (n = 12) after 30 days of treatment (P = 0.003) [84]. A meta-analysis of individual data from 579 patients with IBS included in two randomized, double-blind, placebo-controlled studies of Saccharomyces cerevisiae CNCM I-3856 showed that abdominal pain/discomfort and bloating were significantly improved with probiotic therapy versus placebo during month 2 of treatment (P = 0.01, for both comparisons) [85]. However, data are currently lacking to demonstrate a direct effect of yeast on the gut microbiota of patients with IBS.

4.2.6. Multispecies probiotics: VSL#3®

The medical food probiotic mixture VSL#3 (4 species of Lactobacillus, 3 species of Bifidobacterium, and Streptococcus thermophilus; VSL Pharmaceuticals, Inc., Gaithersburg, MD) has been shown to alter the bacterial profile of the rectal mucosa (i.e. 15 cm from the anal verge) of patients with IBS, with twice daily treatment for 4 weeks, leading to an increase from baseline in average abundance of Bifidobacterium, Lactobacillus, and Streptococcus and a decrease from baseline in Bacteroides average abundance to levels comparable with that of the mucosa of healthy individuals [86]. However, Michail et al. [87] showed no differences from baseline in the number of bacteria or the diversity of bacterial species after 4- or 8-week treatment with VSL #3. The efficacy of the probiotic VSL #3 has been examined in four clinical studies of patients with IBS [87–90]; overall, the efficacy of VSL#3 was similar to that of placebo, although flatulence improved significantly with VSL#3 versus placebo after 4 and 8 weeks in one study (Table 3) [87–90].

Table 3. Summary of clinical studies of VSL#3 in patients with IBS.

Study	Treatment	Outcomes
Wong et al. [89] R, DB, PBO-C Pts with IBS	VSL#3^a (n = 20) vs. PBO (n = 22) bid for 6 wk	IBS mean symptom scores (baseline vs. 6 wk):
		VSL#3: IBS-SSS: 224.5 vs. 158.0 Abdominal pain intensity: 18.8 vs. 15.0 Abdominal pain duration: 37.5 vs. 19 (P < 0.05) Abdominal distension intensity: 34.5 vs. 20 (P < 0.05) QOL: 63.8 vs. 52.5 BM satisfaction: 70.1 vs. 55.3 Frequency of defecation: 1.9 vs. 1.6 Stool type: 4.2 vs. 3.9	PBO: IBS-SSS: 226.4 vs. 183.5 Abdominal pain intensity: 27.7 vs. 21.8 Abdominal pain duration: 30.9 vs. 23.6 Abdominal distension intensity: 39.6 vs. 27.3 QOL: 59.6 vs. 50.2 BM satisfaction: 68.6 vs. 55.7 Frequency of defecation: 1.9 vs. 1.5 Stool type: 4.2 vs. 3.9
Michail et al. [87] R, DB, PBO-C Pts with IBS-D (Rome III)	VSL#3 (n = 15) vs. PBO (n = 9) for 8 wk	GSRS-IBS mean scores (baseline vs. 8 wk):
		VSL#3^b: Global: 2.8 vs. 1.5 Pain: 4.0 vs. 1.9 Bloating: 2.7 vs. 1.6 Constipation: 1.8 vs. 1.2 Diarrhea: 3.1 vs. 1.6 Satiety: 2.3 vs. 1.0	PBO: Global: 2.4 vs. 1.7 Pain: 3.0 vs. 2.0 Bloating: 1.6 vs. 1.5 Constipation: 2.2 vs. 1.5 Diarrhea: 3.1 vs. 1.8 Satiety: 1.8 vs. 1.5
Kim et al. [88] R, DB, PBO-C Pts with IBS (Rome II) with bloating	VSL#3 (n = 24) vs. PBO (n = 24) in yogurt for 4 or 8 wk (± 1 wk)^c	VAS mean symptom scores after 4 wk (VSL#3 vs. PBO): Bloating: 32.4 vs. 38.1 (P = 0.19) Pain: 23.7 vs. 27.4 (P = 0.22) Urgency: 20.1 vs. 23.0 (P = 0.28) Flatulence: 30.8 vs. 40.1 (P = 0.01) VAS mean symptom scores during entire treatment period (VSL#3 vs. PBO): Bloating: 31.3 vs. 38.5 (P = 0.11) Pain: 23.0 vs. 26.9 (P = 0.22) Urgency: 19.7 vs. 22.7 (P = 0.32) Flatulence: 29.7 vs. 39.5 (P = 0.01)
Kim et al. [90] R, DB, PBO-C Pts with IBS-D	VSL#3 (n = 12) vs. PBO (n = 13) bid for 8 wk	Percentage of responders (i.e. patients with “yes” answer to the weekly satisfactory relief question for ≥4 of 8 weekly assessments) did not differ between groups (33% vs. 38%, P > 0.9) GI transit: no significant differences in gastric emptying after 2 and 4 h (P = 0.41 and P = 0.86, respectively), colonic filling after 6 h (P = 0.62), or colonic transit after 24 and 48 h (P = 0.70 and P = 0.99) between groups

bid: twice daily; BM: bowel movement; DB: double-blind; GI: gastrointestinal; GSRS-IBS: Gastrointestinal Symptom Rating Scale-irritable bowel syndrome; IBS: irritable bowel syndrome; IBS-D: irritable bowel syndrome with diarrhea; IBS-SSS: Irritable Bowel Syndrome-Symptom Severity Scale; PBO: placebo; PBO-C: placebo-controlled; pts: patients; QOL: quality of life; R: randomized; VAS: visual analog scale.

^aVSL Pharmaceuticals, Inc., Gaithersburg, MD, USA.

^bP value for between-group differences for VSL#3 vs. placebo were not significant.

^cDuration of treatment was amended from 8 wk to 4 wk after 17 patients completed 8 wk of treatment.

4.2.7. Other multispecies probiotics

Clinical studies of probiotics containing a mixture of bacterial species in patients with IBS have reported mixed results (Table 4) [91–100]. Overall, multispecies probiotics appear to have a favorable safety profile. Levels of B lactis, Lactobacillus rhamnosus, and Streptococcus thermophilus increased significantly from baseline after 4-week treatment with a multispecies probiotic mixture (P < 0.01, P < 0.01, and P = 0.04, respectively); authors postulated that the multispecies probiotics were associated with a decrease in abdominal pain and bloating. However, further analyses are needed to definitively demonstrate this potential association between alterations in the gut microbiota and improvement in specific IBS symptoms. Kajander et al. [100] reported that treatment with a multispecies probiotic stabilized the gut microbiota compared with placebo; during the second half of the 5-month treatment period, the stability of the gut microbiota increased with probiotic supplementation and decreased with placebo (P = 0.002). However, a limitation of this study was the small number of patients with samples who could be included in the analysis of gut microbiota (n = 20) [100].

Table 4. Summary of clinical studies of multispecies probiotics in IBS.

Study and patients	Probiotic and dosing	Efficacy outcomes	Safety
Mezzasalma et al. [91] R, DB, PBO-C Pts with IBS-C (Rome III)	(1) Lactobacillus acidophilus (5 × 10^9 CFUs), L reuteri (5 × 10^9 CFUs), or (2) L plantarum (5 × 10^9 CFUs), L rhamnosus (5 × 10^9 CFUs), Bifidobacterium animalis ssp lactis (5 × 10^9 CFUs) Capsules administered once daily for 60 d	Probiotic (1) and (2) vs. PBO: Response (decrease from baseline ≥30% in symptoms for ≥50% of time) for 60 d: (1) 66–78% and (2) 78–90% vs. PBO, 6–36% (P < 0.001) Response during 30-d follow period: (1) 56–74% and (2) 76–82% vs. PBO, 10–40% (P < 0.001)	Not examined
Yoon et al. [92] R, DB, PBO-C Pts with IBS (Rome III)	LacClean Gold-S® (Cell Biotech, Co., Ltd., Gimpo, Korea) Bifidobacterium bifidum, B lactis, B longum, Lactobacillus acidophilus, L rhamnosus, and Streptococcus thermophilus 2 500-mg capsules (5 × 10^9 viable cells per capsule) daily for 4 wk	Probiotic vs. PBO: Overall adequate symptom relief: 74.4 vs. 61.9% of patients, respectively; P = 0.2 Mean change from baseline in IBS symptoms (VAS scores) at wk 4: Pain/discomfort: P = 0.4 Bloating/gas: P = 0.5 Constipation: P = 0.9 Diarrhea: P = 0.02	Not examined
Yoon et al. [93] R, DB, PBO-C Pts with IBS (Rome III criteria)	LacClean Gold-S 1 500-mg capsules (5 × 10^9 viable cells) bid for 4 wk	Probiotic vs. PBO: Primary end point Pts with global relief of IBS symptoms^a after 4 wk: 68.0 vs. 37.5%, respectively; P = 0.03 Secondary end points Mean % decrease from baseline in intensity of abdominal pain/discomfort and bloating^b at wk 4: 37.1 vs. 9.2 (P = 0.07), 28.7 vs. 17.8 (P = 0.4), 22.4 vs. 20.3 (P = 0.9), respectively % Change from baseline in mean number of weekly bowel movements at wk 4: –2.4 vs. 32.6 (P = 0.2) % Change in stool consistency (BSS) at wk 4: 0.5 vs. –3.3 (P = 0.7)	No AEs or SAEs in either group
Ki Cha et al. [94] R, DB, PBO-C Pts with IBS-D (Rome III criteria)	Duolac7 (Cell Biotech, Co, Ltd, Seoul, Korea) Lactobacillus acidophilus, L plantarum, L rhamnosus, Bifidobacterium breve, B lactis, B longum, and Streptococcus thermophilus 1 capsule (5 × 10^9 cells) bid for 8 wk	Probiotic vs. PBO: Primary end point Adequate relief of IBS symptoms ≥5 of 10 wk during tx and posttx periods: 48 vs. 12%, respectively; P = 0.01 Secondary end points Improvement from baseline in mean individual IBS symptom scores: 33.8 vs. 33.3% (P < 0.01 for both groups) Stool frequency: comparable between groups, and did not change significantly with tx Mean difference from baseline in stool consistency (BSS) at wk 8: 0.85 vs. 0.15, P = 0.04 Percentage change from baseline in overall IBS-QOL score at wk 8: 21.3% vs. 9.0%, P = 0.07	Probiotic: no AEs; PBO: 2 pts withdrew because of AEs (worsening abdominal pain, diarrhea)
Min et al. [95] R, DB, C Pts with IBS (Rome III criteria)	Streptococcus thermophilus (≥3 × 10^9 CFUs/bottle), Lactobacillus acidophilus (≥1 × 10^9 CFUs/bottle), Bifidobacterium animalis spp lactis Bb-12 (≥1 × 10^11 CFUs/bottle), Bifidobacterium enhancer, and acacia dietary fiber (Namyang Dairy Products Co. Ltd., Seoul, South Korea) 1 150-mL bottle bid for 8 wk	Probiotic-enriched yogurt (n = 58) vs. standard yogurt (n = 59): Satisfaction with bowel habit and overall IBS symptoms improved significantly from baseline with probiotic vs. standard yogurt at wk 8 (P = 0.01 and P < 0.001, respectively)	No significant AEs reported
Sondergaard et al. [96] R, DB, PBO-C Pts with IBS	Cultura (Arla Foods, Denmark) milk fermented with Lactobacillus bulgaricus and Streptococcus thermophilus and containing ≥5 × 10^7 CFUs/mL each of L paracasei ssp paracasei F19, L acidophilus La5, and Bifidobacterium lactis Bb12; 500 mL/d for 8 wk	Probiotic (n = 27) vs. PBO (n = 25): Primary end points Adequate relief of IBS symptoms at each wk of tx phase (wk 1–8), and at the end of the follow-up phase (wk 16): no significant difference between groups Total mean IBS-SSS score during study: no significant difference between groups Secondary end points Individual mean VAS scores of the IBS-SSS (abdominal pain, abdominal distension, overall satisfaction with bowel habits, and impact of IBS on overall QOL): symptoms decreased over time, but no significant difference between groups	No SAEs reported
Ringel-Kulka et al. [97] R, DB, PBO-C Pts with non-constipation IBS, functional diarrhea, or functional bloating (Rome III)	Lactobacillus acidophilus NCFM and Bifidobacterium animalis spp lactis Bi-07 1 pill containing 1 × 10^11 CFUs bid for 8 wk	Probiotic (n = 31) vs. PBO (n = 29) Primary end points Global relief of GI symptoms^c and satisfaction with treatment^d improved from baseline in both groups, but did not differ significantly between groups Secondary end points Abdominal pain improved from baseline to wk 8 (probiotic, 3.3 vs. 2.5, P = 0.03; PBO, 3.7 vs. 2.5, P = 0.01), but did not differ significantly between groups Bloating improved significantly at wk 4 vs. PBO (score at wk 4, 4.1 vs. 6.2; P = 0.009) and numerically at week 8 (score at wk 8, 4.3 vs. 5.8; P = 0.06) IBS-QOL improved from baseline in both groups, but did not differ significantly between groups	No differences in AEs between groups AEs included cold symptoms, fatigue, abdominal pain, and sinus infection
Simrén et al. [98] R, DB, PBO-C Pts with IBS (Rome II criteria)	Cultura (Arla Foods, Denmark; milk fermented with Lactobacillus bulgaricus and Streptococcus thermophilus and containing ≥5 × 10^7 CFUs/mL each of L paracasei ssp paracasei F19, L acidophilus La5, and Bifidobacterium lactis Bb12); 400 mL/d for 8 wk	Probiotic (n = 37) vs. PBO (n = 37) Primary end point Adequate relief of IBS-related symptoms ≥50% of wk of tx phase achieved by 38 and 27% of pts, respectively (P = 0.3) Secondary end points Overall IBS-SSS score decreased from baseline during 8 wk tx phase in both groups (P = 0.2 vs. control) Pain frequency, pain severity, bloating severity, satisfaction with bowel habits, and interference with daily life improved from baseline in both groups, but did not differ significantly between groups HAD scale score did not decrease significantly from baseline with probiotic, but did with PBO (P = 0.001 vs. baseline) at wk 8 IBS-QOL subscale scores for physical and social functioning improved significantly from baseline for probiotic at wk 8, but no significant differences between groups	No AEs reported
Williams et al. [99] R, DB, PBO-C Pts with IBS by self-report (Rome II criteria)	LAB4 multistrain probiotic (Cultech Ltd., Port Talbot, UK) Lactobacillus acidophilus CUL60 and CUL21, Bifidobacterium lactis CUL34, and B bifidum CUL20 1 capsule (2.5 × 10^10 CFUs) daily for 8 wk	Probiotic (n = 28) vs. PBO (n = 28) Primary end point Overall IBS-SSS score improved significantly from baseline to wk 8 and 10 with probiotic (P < 0.0001 for both time points) and PBO (P < 0.0001 and P = 0.0005, respectively) Secondary end points Improvement in IBS-SSS individual scores from baseline to wk 8 and 10: Abdominal distension/bloating: probiotic (P < 0.0001 and P = 0.008, respectively); PBO (P = 0.003 and P = 0.13) Satisfaction with bowel habit: probiotic and PBO (P < 0.0001 for both time points) Number of d with pain: probiotic (P < 0.0001 for both time points); PBO (P = 0.0004 and P = 0.01) QOL: probiotic (P < 0.0001 for both time points); PBO (P < 0.0001 and P = 0.0001) Abdominal pain: probiotic (P < 0.0001 and P = 0.0008); PBO (P = 0.0009 and P = 0.01)	Probiotic: flatulence (3.6%)
Kajander et al. [100] R, DB, PBO-C Pts with IBS (Rome II criteria)	Probiotic milk-based drink (Valio Ltd., Helsinki, Finland) Lactobacillus rhamnosus GG, L rhamnosus Lc705, P freudenreichii ssp shermanii JS, and B animalis ssp lactis Bb12 1 1.2-dL drink daily containing 1 × 10^7 CFUs/mL each probiotic strain for 20 wk	Probiotic (n = 43) vs. PBO (n = 43) Primary end point Weekly composite IBS symptom score (including abdominal pain, distension, flatulence, and rumbling) decreased from baseline to wk 20 with probiotic and PBO (37% vs. 9% reduction, respectively; P = 0.008) Secondary end points Distension improved significantly with probiotic vs. PBO at wk 20 (36% vs. 10% reduction; P = 0.02) Abdominal pain decreased at wk 20 with probiotic; but difference vs. PBO was not significant (38% vs. 0%; P = 0.052) Flatulence and rumbling decreased in both groups at wk 20, but differences between groups were not significant (P = 0.11 and P = 0.09, respectively) Frequency of soft stools (58% vs. 53%), hard stools (16% vs. 15%), and diarrhea (27% vs. 32%) comparable between groups at wk 20	AEs reported by 23.3 and 34.9% of pts receiving probiotic or PBO, respectively

AE: adverse event; bid: twice daily; BSS: Bristol Stool Scale; C: controlled; CFU: colony-forming unit; DB: double-blind; GI: gastrointestinal; HAD: Hospital Anxiety and Depression; IBS: irritable bowel syndrome; IBS-C: irritable bowel syndrome with constipation; IBS-D: irritable bowel syndrome with diarrhea; IBS-QOL: irritable bowel syndrome quality of life; IBS-SSS, Irritable Bowel Syndrome-Symptom Severity Scale; PBO: placebo; PBO-C: placebo-controlled; posttx: posttreatment; pts: patients; QOL, quality of life; R: randomized; SAE serious adverse event; tx: treatment; VAS: visual analog scale.

^aAssessed by ‘yes’ or ‘no’ response to the question: ‘Compared with your health before treatment started, have overall IBS symptoms improved during the past 7 days?’

^bWorst symptoms over previous 24 h on numeric scale ranging from 0 to 10.

^cChange in patient’s bowel symptoms scored on a 7-point scale ranging from ‘substantially worse’ to ‘substantially improved.’

^dEvaluated using satisfaction survey.

4.3. Synbiotics

Synbiotics contain a mixture of prebiotics and probiotics [63]. While synbiotics are safe for consumption, a number of randomized, controlled, and open-label clinical studies of patients with IBS who received synbiotics have shown mixed results regarding efficacy (Table 5) [63,101–103]. Findings appear to depend on the microorganisms examined, the duration of treatment, and the design of the study. Of note, the clinical studies of synbiotics presented in this review did not evaluate the gut microbiota. The ACG concluded that there was insufficient evidence to recommend the use of synbiotics in IBS [1].

Table 5. Summary of clinical studies of synbiotics in IBS.

Study design	Patients	Treatment(s)	Efficacy outcomes	Safety
Andriulli et al. [63] R, DB, C, MC	IBS (Rome II criteria) IBS-C: n = 17 IBS-D: n = 47 IBS-M: n = 203	Synbiotic (Flortec, Bracco Co, Milan, Italy; Lactobacillus paracasei B21060 [5 × 10^9 CFUs]; n = 132) vs. prebiotic (synbiotic without probiotic; n = 135) bid for 12 wk 2-wk run-in phase, followed by 12-wk tx phase	Synbiotic vs. prebiotic: Percentage of pts with no or mild pain: wk 4, 54% vs. 57.4%, respectively; wk 8, 60.9% vs. 53.9%; wk 12, 53.9% vs. 53.4% (P = NS for all comparisons) Percentage of pts with self-reported improvement in well-being: wk 4, 60.7% vs. 61.7%, respectively; wk 8, 69.2% vs. 63.5%; wk 12, 63.3% vs. 60.9% (P = NS for all comparisons) Pts with IBS-D: decrease in mean daily bowel movements from baseline to wk 12: synbiotic, 3.59 vs. 2.41; prebiotic, 3.4 vs. 2.95 (between-group difference at wk 12, P = 0.05)	Frequency of AEs did not differ between groups AEs led to study discontinuation in 3.8 and 3.0% of pts receiving synbiotic (i.e. diarrhea, abdominal discomfort, vomiting) vs. prebiotic (i.e. abdominal pain, skin rash), respectively
Bittner et al. [101] OL	IBS (Rome II criteria; n = 22 completed)	Synbiotic (Prescript-Assist™, Safer Medical Inc., Fort Benton, MT; 29 soil-based microorganisms [probiotic] and leonardite [prebiotic]) 550 mg bid for 2 wk 1-y extension period consisted of 3 groups: (1) unsupported remission: no further tx for ≥52 wk (n = 6); (2) maintenance remission: synbiotic 550 mg bid for 12 to 20 wk, then decreased synbiotic use (e.g. synbiotic 550 mg every 2–3 d; n = 6); (3) episodic tx: synbiotic use when mild IBS symptoms perceived to occur (n = 10) Follow-up phase up to ~60 wk	Remission of IBS at 52 wk follow up: 81.5–100% (dependent on whether pts lost to follow up included in analysis)	No AEs reported up to ~60 wk follow up
Bittner et al. [102] R, DB, PBO-C	IBS (Rome II criteria; n = 25 completed)	Synbiotic (Prescript-Assist) 500 mg bid (n = 12) vs. PBO (n = 13) for 2 wk	3 IBS subsyndromic factors (i.e. general ill feelings/nausea, indigestion/flatulence, and colitis) significantly decreased from baseline with synbiotic vs. PBO (P < 0.05)	Not evaluated
Colecchia et al. [103] OL, MC	IBS-C (Rome II criteria; n = 636 completed)	Zir Fos® (Alfa Wasserman, Alanno Scalo, Pescara, Italy; Bifidobacterium longum W11 [probiotic], Fos Actilight [prebiotic]) 3 g daily for ≥36 d	For bloating or abdominal pain, compared with baseline, the frequency of moderate to severe symptoms was significantly decreased at 1 mo after tx (bloating, 62.9% vs. 9.6%, respectively, P < 0.0001; abdominal pain, 38.8% vs. 4.1%, P < 0.0001) Stool frequency significantly increased from baseline to 1 mo after tx (2.9 vs. 4.1 times/wk, respectively; P < 0.001)	Only AE reported was mild diarrhea, experienced by 6 pts (0.9%)

AE: adverse event; bid: twice daily; C: controlled; CFU: colony-forming unit; DB: double-blind; IBS: irritable bowel syndrome; IBS-C: constipation-predominant irritable bowel syndrome; IBS-D: diarrhea-predominant irritable bowel syndrome; IBS-M: mixed/alternating irritable bowel syndrome; MC: multicenter; NS: not significant; OL: open-label; PBO: placebo; PBO-C: placebo-controlled; pts: patients; R: randomized; tx: treatment.

5. Antibiotics

Off-label administration of some antibiotics (e.g. neomycin) has been shown to improve symptoms of IBS; however, adverse effects and the development of antibiotic resistance or Clostridium difficile infection have been associated with systemically absorbed agents [1,6,104–107]. The nonsystemic antibiotic rifaximin (Xifaxan®, Salix Pharmaceuticals, Bridgewater, NJ) was approved in 2015 by the US Food and Drug Administration for the treatment of IBS-D in adults [108]. Rifaximin 550 mg three times daily for 2 weeks has been evaluated in two phase III, identically designed, randomized, placebo-controlled IBS-D trials [109]. A significantly greater percentage of patients who received rifaximin either achieved adequate relief of global IBS symptoms for ≥2 of the first 4 weeks posttreatment or achieved adequate relief of bloating for ≥2 of the first 4 weeks posttreatment versus placebo (Figure 2; combined population) [109]. Rifaximin and placebo had comparable safety profiles through 10 weeks posttreatment, with serious AEs reported in 1.6% of patients in the rifaximin group and 2.4% of patients in the placebo group; no patients developed C. difficile infection [109]. A pooled analysis of safety data from these two phase III studies and a phase IIb study reaffirmed the favorable safety profile of rifaximin compared with placebo; no patients in these 3 studies developed C. difficile infection while receiving treatment [110]. In the pooled analysis, the most common AEs with rifaximin 550 mg (n = 1008) vs. placebo (n = 829) were headache (5.5 vs. 6.2%), upper respiratory tract infection (4.5 vs. 5.7%), nausea (4.1 vs. 3.7%), and abdominal pain (4.0 vs. 4.7%) [110]. A meta-analysis determined that one individual with IBS-D would have an AE with rifaximin for every 846 individuals who would benefit (number needed to harm = 8,971; number needed to treat = 10.6) [111].

Figure 2. Patients achieving adequate relief of global irritable bowel syndrome (IBS) symptoms^a and adequate relief of bloating^b during ≥2 of the first 4 weeks posttreatment (combined population from two phase 3 trials).

^aYes response to the following question, which was asked weekly: ‘In regard to all your symptoms of IBS, as compared with the way you felt before you started the study medication, have you, in the past 7 days, had adequate relief of your IBS symptoms?’ ^bYes response to the following question, which was asked weekly: ‘In regard to your symptoms of bloating, as compared with the way you felt before you started study medication, have you, in the past 7 days, had adequate relief of your IBS symptom of bloating?’ Data from Pimentel M et al. N Engl J Med 2011;364:22–32 [109].

Because rifaximin is administered as short-course therapy (i.e. 2 weeks) for IBS-D, a phase III retreatment study was conducted to evaluate if patients with IBS-D who initially responded to rifaximin and then experienced symptom recurrence would respond to 2-week repeat treatment [112]. Patients with IBS-D who responded to an open-label, 2-week course of rifaximin and then relapsed during follow up (up to 18 weeks) were randomly assigned to double-blind treatment with rifaximin (n = 328) or placebo (n = 308). A significantly greater percentage of patients in the rifaximin group (38.1%) responded (≥30% decrease from baseline in mean weekly pain score for abdominal pain and ≥50% decrease from baseline in the number of days with mushy/watery stools [Bristol Stool Scale type 6 or 7]) for ≥2 of 4 weeks posttreatment compared with placebo (31.5%; P = 0.03) [112]. Repeat treatment with rifaximin was generally well tolerated. The most commonly reported AEs during a 22-week period (which included two 2-week treatment courses) occurring more often with rifaximin versus placebo were nausea (3.7 vs. 2.3%), upper respiratory tract infection (3.7 vs. 2.6%), and nasopharyngitis (3.0 vs. 2.9%). C. difficile infection occurred in one patient receiving rifaximin repeat treatment after 37 days; however, this patient had previous infection with C. difficile and had received 10-day treatment with cefdinir for a urinary tract infection immediately before developing this C. difficile infection. The ACG stated that rifaximin is effective at reducing total IBS symptoms and bloating in IBS-D (weak recommendation based on moderate quality of evidence) [1].

In a case report, a patient with IBS-C with constipation and excess methane production was treated with rifaximin and had improved stool frequency and form [113]. Further, a retrospective study of patients with IBS with a positive breath test for methane showed that the combination of rifaximin and neomycin resulted in greater clinical response (85%) compared with neomycin alone (63%; P = 0.2) or rifaximin alone (56%; P = 0.01) [114]. It is possible that antibiotic therapy reduces the effects of methanogenic bacteria, given that patients with IBS, particularly IBS-C, were shown to have increased fecal concentrations of the methanogen Methanobrevibacter smithii compared with healthy individuals; further, methane producers by breath testing had greater concentrations of M. smithii [115]. However, a retrospective study of patients with IBS receiving antibiotic treatment reported that the best therapeutic response occurred in patients who had no increase in measured hydrogen and methane gases in a lactulose breath test [116]. Thus, while some evidence suggests that methanogenic bacteria may play a role in response to antibiotics, further study is needed.

The effect of a nonsystemic antibiotic on the gut microbiota is viewed as a safety concern of rifaximin. The mechanism of action of rifaximin in IBS is unclear, but studies in patients with non–IBS-C receiving a 2-week course of treatment with rifaximin experienced only modest, although detectable, changes in fecal microbial profiles [117,118]. Additional preclinical [119] studies and human IBS fecal sample analysis [118,120,121] studies of rifaximin have supported that any changes in the fecal microbial profile are modest and not sustained. Overall, data supported short courses of therapy (2 weeks per course) with rifaximin in the treatment of patients with IBS-D.

Although these studies currently raise no red flags, future research should continue to monitor bacterial antibiotic resistance and the incidence of C. difficile infection, especially in patients who require repeat courses of rifaximin and potentially other antibiotics. Advances in molecular techniques, such as RNA sequencing, may offer further information regarding the safety of antibiotic therapy in the treatment of IBS.

6. Fecal microbiota transplantation

Fecal microbiota transplantation (FMT) provides the most powerful means of modifying the microbiota from a presumed healthy donor. FMT received FDA approval for the treatment of recurrent C. difficile infection, with a meta-analysis of 611 patients showing pooled efficacy (i.e. cure) to be 91.2% [122]. There is a growing interest in using FMT for a number of GI diseases that are linked to a disturbed gut microbiota composition, including IBS. Reestablishing a healthy intestinal microbiota could be a promising novel treatment option for IBS; indeed, a small, open-label study of FMT in patients with IBS (n = 10) demonstrated significant improvement from baseline in the reestablishment of fecal microbiota diversity 4 weeks after treatment (P = 0.03) and efficacy of FMT was dependent on increased concentrations of bifidobacteria in the donor stool [123]. A small number of clinical trials are currently registered under ClinicalTrials.gov, and to date, there are no published randomized clinical trials investigating the impact of FMT on IBS. A 2017 review compiled the limited number of studies, the majority of which were conference abstracts that included a small number of patients (range, n = 9 to n = 13) [124]. Because of differences in methods of administration, inclusion criteria, assessment of symptoms, and control groups, conclusions must be made cautiously. The authors concluded that there is an ‘at least temporary improvement’ in a large percentage (58%) of FMT-treated patients with IBS without any significant complications. However, there is concern that more than desired microbes are being transmitted during FMT and that donors need to be carefully screened [125,126]. In one study, germ-free mice receiving fecal microbiota samples from patients with IBS-D exhibited increased GI transit, activation of innate immunity, and altered intestinal barrier function after 3 weeks compared with mice receiving fecal microbiota from healthy individuals [127]. Anxiety-like behavior was evident in mice receiving FMT from patients with IBS-D with anxiety, but not in mice receiving FMT from patients with IBS-D without anxiety or from healthy individuals. FMT appears promising; however, further research is needed before FMT can be recommended as a therapeutic modality for patients with IBS.

7. Conclusions

IBS is a common chronic condition characterized by alterations in the gut microbiota. Treatment of patients with IBS may include a number of agents with the potential to modulate the gut microbiota (i.e. dietary modifications, prebiotics, probiotics, synbiotics, and nonsystemic antibiotics). However, the number and diversity of different bacterial species for probiotics or synbiotics, used either alone or in combination, are inconsistent among studies, as is the duration of treatment. Thus, it is not entirely surprising that efficacy findings across studies vary widely. Further research should place an emphasis on identifying species- and strain-specific effects, rather than drawing generalized conclusions from studies with limited comparability.

Overall, in studies that evaluated safety, dietary interventions, prebiotics, probiotics, and synbiotics were associated with a favorable safety profile. For nonsystemic antibiotics, rifaximin has been shown to be efficacious and well tolerated in well-designed clinical trials. However, it is unclear how often a 2-week course of rifaximin repeat treatment might be needed to manage symptoms of IBS, and thus repeat treatment courses should be individualized based on patient symptoms. It also bears mentioning that, for other nonsystemic antibiotics, there are an inadequate number of IBS-D studies in the literature. Furthermore, more consistent evidence is needed regarding the efficacy and safety of elimination diets, prebiotics, probiotics, and synbiotics for the treatment of IBS. Additional well-designed studies that examine alterations in the gut microbiota and potential associations with the clinical symptoms of IBS are also warranted.

Declaration of Interests

LA Harris reports serving as a consultant for Salix Pharmaceuticals, Ironwood Pharmaceuticals, Inc., Allergan plc, Synergy Pharmaceuticals Inc., Synthetic Biologics, Inc., IM HealthScience, QOL Pharmaceuticals, and Napo Pharmaceuticals; as well as performing research with Alvine Pharmaceuticals and Rhythm Pharmaceuticals, Inc. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Acknowledgments

Technical editorial assistance was provided, under the direction of the authors, by Sophie Bolick, PhD, Synchrony Medical Communications, LLC, West Chester, Pennsylvania, USA. Funding for this support was provided by Salix Pharmaceuticals, Bridgewater, NJ, USA.

Additional information

Funding

The authors did not receive any compensation for development of this manuscript. Salix Pharmaceuticals provided funding for editorial support. Salix, the study sponsor, did not actively contribute to the content or have a role in the decision to submit, but reviewed the article for scientific accuracy.