How effective are antibiotics for the treatment of irritable bowel syndrome?

KEYWORDS:

• Antibiotics
• fecal microbiome
• irritable bowel syndrome
• small bowel bacterial overgrowth

1. Introduction

Irritable bowel syndrome (IBS) is a chronic functional bowel disorder affecting 4% to 10% of people worldwide [1], with a substantial impact on the quality of life. The diagnosis is made in patients meeting symptom-based diagnostic criteria, called the Rome criteria, which define IBS according to presence of abdominal pain associated with a change in bowel habit [2]. Patients with IBS may experience predominant diarrhea (IBS-D), constipation (IBS-C), or a mixture of the two (IBS-M). Treatment aims to improve the patient’s most troublesome symptom, and a range of options are available, such as dietary modification, psychological therapies, or drug treatments, including antibiotics.

The rationale for using antibiotics to treat IBS is based on two factors. Firstly, the observation that a proportion of patients with IBS have a positive hydrogen breath test (HBT), suggestive of small intestinal bacterial overgrowth (SIBO), and secondly the hypothesis that the gut microbiome, particularly bacteria, plays a role in the pathophysiology of IBS [3]. It is important, however, to recognize some key uncertainties. The prevalence of SIBO may be no higher among those with IBS compared with healthy controls [4], and a positive HBT result may reflect rapid small intestinal transit in some people, the rise in breath hydrogen corresponding to degradation of the ingested carbohydrate by colonic bacteria, rather than the presence of SIBO [5]. Furthermore, whether there is a specific microbial signature in IBS is uncertain [6], and the relevance of the gut microbiome in any individual patient remains ambiguous. Enteric infection, whether bacterial, viral, or parasitic, is a recognized risk factor for developing IBS, which, in this context, is referred to as post-infection IBS [7]; however, there is no evidence that antibiotics are more effective for treating this patient group. It should also be noted that the use of antibiotics in general terms has been reported to be associated with the development of IBS [8].

One of the earliest randomized controlled trials (RCT) of antibiotics in IBS used a 10-day course of neomycin 500 mg twice-daily compared with placebo in 111 patients with either IBS-D or IBS-C [9]. Neomycin resulted in a significantly greater improvement in overall symptoms (35.0% vs. 11.4%, p< 0.05) 7 days post-treatment. Important limitations include the small sample size and the short duration of follow-up. However, the most widely studied antibiotic in IBS is rifaximin. The first published RCT compared a 10-day course of rifaximin 400 mg three times daily with placebo in 87 patients, demonstrating a significant improvement in IBS symptoms with rifaximin after 10 weeks of follow-up (36.4% vs. 21.0%, p= 0.020) [10]. A second RCT recruited 124 patients with abdominal bloating, of whom 70 met criteria for IBS, and compared 10 days of rifaximin 400 mg twice-daily with placebo [11]. Among those with IBS, there was a benefit of rifaximin over placebo after 10 days of follow-up (27.0% vs. 9.1%, p = 0.05). Each of these trials also included patients with either IBS-D or IBS-C.

Following the results of these small-scale studies, two large phase-three RCTs of identical design, TARGET 1 and 2, were conducted. These compared a 14-day course of rifaximin 550 mg three-times daily with placebo, and followed patients up for a total of 12 weeks [12]. Patients with IBS-C were excluded. Despite the 12-week follow-up, the primary endpoint was adequate relief of IBS symptoms for at least 2 of the first 4 weeks after treatment. In total, 1260 patients were recruited and, in both studies, significantly more patients treated with rifaximin achieved the primary endpoint (40.8% vs. 31.2%, p = 0.01; and 40.6% vs. 32.2%, p = 0.03), although the absolute differences were modest. A greater proportion of patients in the rifaximin group than the placebo group continued to report adequate symptom relief throughout the 12-week study period. Rifaximin was also significantly more effective than placebo for bloating in both these trials. The safety profile appeared good.

A systematic review and meta-analysis pooled the results from five trials of rifaximin in those naïve to treatment, incorporating a total of 1805 patients [3]. Overall, this confirmed the modest efficacy of rifaximin, with a relative risk of IBS symptoms persisting of 0.84 (95% confidence interval 0.79 to 0.90), and a number needed to treat over placebo of 9. Another RCT, TARGET 3, has investigated the merits of repeated treatment with rifaximin in patients who initially responded to therapy, but whose symptoms had relapsed [13]. Among 1074 patients with IBS-D who responded to an initial 14-day course of rifaximin 550 mg three-times daily, 692 relapsed during an 18-week observation period. Of these, 636 patients were randomized to repeat treatment with two further 14-day courses of rifaximin or placebo. The proportion of responders, defined as a composite decrease in abdominal pain and loose stools from baseline, was significantly greater with rifaximin for both repeat courses, but the absolute difference was again small (38.1% vs. 31.5%,p= 0.03). Rifaximin was well tolerated, with low rates of adverse events in both groups.

Based on the results of these more recent RCTs, rifaximin has been licensed as a treatment option for IBS-D in some countries. Nevertheless, although the drug appears efficacious when compared with placebo, it is also important to understand how it performs relative to other available drugs for treating IBS-D, to assist patients and clinicians in selecting the most appropriate treatment. No head-to-head trials have been conducted, but network meta-analysis can offer some insights, facilitating comparison of rifaximin with other drugs used in IBS-D, including alosetron, ramosetron, and eluxadoline [14]. Overall, alosetron ranked first for efficacy, based on the Food and Drug Administration (FDA)-recommended composite endpoint for trials in IBS of improvement in abdominal pain and stool consistency. It also ranked first for effect on global IBS symptoms and effect on stool consistency. Rifaximin was the drug least likely to lead to adverse events, including constipation, but it ranked last for efficacy according to the FDA-recommended endpoint, and showed limited efficacy overall.

Current evidence supports the use of rifaximin in the treatment of IBS, with RCTs demonstrating a consistent, but modest, benefit. Importantly, however, the mechanism by which rifaximin, and antibiotics more generally, might work in IBS remains unclear. One study demonstrated an acceleration of ascending colon transit in patients taking rifaximin, a finding at odds with its ability to improve diarrhea [15]. Antibiotics may also influence central neurotransmission in the brain, alter brain physiology, and induce behavioral changes [16], which could be relevant to their beneficial effects in some patients with IBS, although more research is required to explore these theories further. The larger trials of antibiotics in IBS have only evaluated the results of HBTs in a subset of patients, making it difficult to gauge the relevance of the SIBO hypothesis. Moreover, breath testing studies have also identified that excess methane production is associated with IBS-C [17]. There is limited evidence that antibiotics might be effective in this patient group [18], although methane production appears to be driven by archaea [17], rather than bacteria, and all recent trials of antibiotics in IBS have focussed on IBS-D or IBS-M. Finally, changes in the microbiome of patients taking rifaximin appear limited and are not sustained [19]. It is also unclear whether these changes are responsible for any therapeutic effects of the drug. Negative effects on the microbiome, such as bacterial resistance or increased Clostridium difficile infection rates, do not seem to be a major concern, but require further assessment in IBS patients taking multiple courses of antibiotics longer-term [3]. In essence, future research needs to focus on uncovering why antibiotics appear effective for treating IBS, and which patients are most likely to benefit.

2. Expert opinion

Without doubt, antibiotics are one of the single most important medical discoveries of the last century and have transformed clinical practice. Alarmingly, however, we now face a rise in multidrug-resistant organisms due, in large measure, to imprudent use of antibiotics where they may be of little or no benefit. Accordingly, the need to be judicious in their prescription has never been clearer, and they should be deployed only where there is good evidence of an infective pathology, ideally guided and rationalized using microbiological testing.

Viewed from this perspective, the use of antibiotics to treat IBS is somewhat anomalous, given that they are being administered empirically to treat symptoms, often without evidence of microbial dysbiosis. Even if an HBT for SIBO is performed, whether positive results are sufficient grounds to use rifaximin remains uncertain, particularly given that patients with IBS can respond to treatment despite a negative result. Recent years have seen a huge surge of interest in the gut microbiome, coupled with growing recognition of the pivotal role that it might play in many aspects of human health and disease. Nevertheless, our understanding of this field is in its infancy and examining the fecal microbiome remains a tool for researchers, not clinicians. We are certainly not in a position where we can reliably interpret an individual’s fecal microbiome, understand how this might relate to the pathophysiology of IBS, and target antibiotic therapy appropriately. Similarly, we cannot be sure, despite the seemingly low risk of microbial resistance with antibiotics like rifaximin, that we are not storing up future problems, particularly if patients are treated multiple times. It is also worth considering that manipulating the gut microbiome using fecal microbial transplantation might prove to be preferable to using antibiotics [20].

The largest trials of antibiotics focus on symptom response to evaluate efficacy but fail to assess underlying mechanisms of action. IBS is a symptom-based diagnosis, but the disorder has a complex and incompletely understood pathophysiology, so it is likely that the underlying cause of symptoms varies between patients. Currently, therefore, antibiotic use in IBS is a blunt instrument. It is possible that, if these drugs could be targeted at patients most likely to benefit, therapeutic gain over placebo might increase. Future studies of antibiotics for treating IBS need to explore this question in more detail to ensure we tailor treatment to the individual patient, thereby prescribing antibiotics as responsibly as possible, and with a clear rationale.

Declaration of interest

The authors have no 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. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has not been funded.