Article Text

Download PDFPDF

Dietary and pharmacological treatment of abdominal pain in IBS
  1. Michael Camilleri1,
  2. Guy Boeckxstaens2
  1. 1Clinical Enteric Neuroscience Translational and Epidemiological Research (CENTER), Mayo Clinic, Rochester, Minnesota, USA
  2. 2Department of Clinical and Experimental Medicine, Translational Research Center for Gastrointestinal Disorders (TARGID), KU Leuven, Leuven, Belgium
  1. Correspondence to Professor Michael Camilleri, Clinical Enteric Neuroscience Translational and Epidemiological Research (CENTER), Mayo Clinic, Charlton Bldg., Rm. 8-110, 200 First St. S.W., Rochester, MN 55905, USA; camilleri.michael{at}


This review introduces the principles of visceral sensation and appraises the current approaches to management of visceral pain in functional GI diseases, principally IBS. These approaches include dietary measures including fibre supplementation, low fermentable oligosaccharides, disaccharides, monosaccharides and polyols diet, and pharmacological approaches such as antispasmodics, peppermint oil, antidepressants (tricyclic agents, selective serotonin reuptake inhibitors), 5-HT3 receptor antagonists (alosetron, ondansetron, ramosetron), non-absorbed antibiotic (rifaximin), secretagogues (lubiprostone, linaclotide), μ-opioid receptor (OR) and κ-OR agonist, δ-OR antagonist (eluxadoline), histamine H1 receptor antagonist (ebastine), neurokinin-2 receptor antagonist (ibodutant) and GABAergic agents (gabapentin and pregabalin). Efficacy and safety are discussed based on pivotal trials or published systematic reviews and meta-analysis, expressing ORs or relative risks and their 95% CIs. Potential new approaches may be based on recent insights on mucosal expression of genes, and microRNA and epigenetic markers in human biopsies and in animal models of visceral hypersensitivity.

The objectives of this review are to appraise the physiology and anatomy of gut sensation and the efficacy in the relief of visceral pain (typically in IBS) of several classes of therapies. These include fermentable oligosaccharides, disaccharides, monosaccharides and polyols (FODMAPs) and different classes of medications (box 1).

Box 1

Classes of pharmacological agents for visceral pain

  • Antidepressants (tricyclic agents, selective serotonin reuptake inhibitors)

  • Peppermint oil

  • 5-HT3 receptor antagonists (alosetron, ondansetron, ramosetron)

  • Non-absorbed antibiotic (rifaximin)

  • Secretagogues (lubiprostone, linaclotide)

  • μ-Opioid receptor (OR) and κ-OR agonist and δ-OR antagonist (eluxadoline)

  • Histamine H1 receptor antagonist (ebastine)

  • Neurokinin-2 receptor antagonist (ibodutant)

  • GABAergic agents (gabapentin and pregabalin)

View Full Text

Statistics from

Introduction to the physiology of pain signalling in the gut

Sensory neurons are organised into three pathways to the central nervous system: vagal, thoracolumbar and lumbosacral. Parasympathetic afferents, comprising the majority of nerve fibres in the vagus and pelvic nerves, convey non-conscious sensory information, to the solitary tract nucleus in the brainstem. Visceral afferents that course along sympathetic nerves, also referred to as spinal afferents, convey painful stimuli to the spinal cord via the dorsal roots. They are equipped with a variety of pronociceptive and antinociceptive ion channels and receptors, of which the balance between pain sensing and suppressing signals finally determines the activation status of the nerve ending.1

There are five major structural types of endings of the extrinsic primary afferents signalling in the gut, and they appear to show distinctive combinations of physiological responses. These are: ‘intraganglionic laminar’ endings in myenteric ganglia; ‘mucosal’ endings located in the subepithelial layer; ‘muscular-mucosal’ afferents with mechanosensitive endings close to the muscularis mucosae; ‘intramuscular’ endings within the smooth muscle layers and ‘vascular’ afferents with sensitive endings primarily on blood vessels. Physiological properties of the gut's extrinsic afferent nerves are characterised by variability and plasticity, which can make it difficult to reliably distinguish the classes of sensory neurons that underlie gut sensation. With reference to IBS pain, it is worth noting that current evidence suggests that pain from the rectum is mediated substantially through pelvic pathways, whereas pain from more proximal gut regions is primarily mediated via thoracolumbar spinal afferents, most of which correspond to the vascular afferents. Low-threshold, wide-dynamic range lumbosacral mechanoreceptors from the colorectum are likely to be responsible for activation of pain pathways. The mechanoreceptors include rIGLEs and rectal muscular-mucosal endings, suggesting that both endings in the wall of the bowel as well as serosal or vascular afferents may mediate pain sensation. In the human rectum, sensation to distension was reduced by luminal application of the local anaesthetic, lidocaine, suggesting a possible role for muscular-mucosal afferents with endings close to the mucosal surface (reviewed in ref. 2).

As with somatic sensation, a three-neuron chain3 conveys afferent signals from the first order neuron arising in the viscus with cell bodies in the dorsal root ganglia (DRG), from the second order neurons that ascend through spinothalamic tracts, as well as special columns in the spinal cord such as the dorsal column pain pathway, and from the third order neuron projecting from the thalamus or brainstem to neurons in the cerebral cortex (see figure 1), including specific brain networks (discussed briefly below). The supraspinal projections lead to perception, autonomic and cardiovascular responses to pain.

Figure 1

Physiology of gut sensation (adapted from Delgado-Aros and Camilleri3 and Mayer et al4).

Brain networks are important in the experience of IBS symptoms such as abdominal pain and altered bowel function, as well as the experience of anxiety and poor coping. In fact, there are task-related brain networks that are associated with structural and functional alterations in IBS. The networks involved are the central executive, salience, sensorimotor, emotional arousal and central autonomic. The outputs from these networks that are related to IBS pathophysiology occur in the form of descending pain modulation (eg, originating in the periacqueductal grey) and autonomic nervous system activity.4

Increased pain perception (visceral hypersensitivity) in IBS

Pain is an essential feature of IBS.5 Brain dysfunction, abnormal interaction of the peripheral and central nervous system and aberrant afferent nerve function are potential mechanisms causing hypersensitivity in IBS.6 Visceral hypersensitivity and hyperalgesia are well documented in IBS7–12 and comprise increased perception to distension, as assessed by barostat or chemical components such as acid or lipids. However, hypersensitivity13 ,14 and central nervous system dysfunction15 are not ubiquitous in patients with IBS, and hypersensitivity is associated with greater likelihood to report pain.16 Given the importance of abdominal pain in IBS, it is relevant to consider approaches that address the cause and relieve this symptom in patients with IBS.

Current approaches to manage visceral pain in patients with IBS

In table 1, we have summarised efficacy of interventions on the relief of symptoms in IBS, expressed as relative risk (RR) or OR and CI, based on systematic reviews and meta-analyses from the literature.

Table 1

Efficacy of interventions on the relief of symptoms in IBS: relative risk (RR) or OR and CI based on systematic reviews and meta-analyses

Dietary interventions

Food is often a trigger of abdominal symptoms in patients with IBS,17 and the mechanisms evoked by food ingestion have been reviewed extensively elsewhere.18 In relation to colonic function, these triggers include stimulation of colonic motility through a vagally mediated reflex, inhibition of colonic water absorption and stimulation of colonic transit and high amplitude propagated contractions by carbohydrates that reach the colon and their metabolic products. On the other hand, deficiency of fibre in the diet is often considered as a factor predisposing to constipation,19 which may then cause abdominal pain.

Fibre supplementation

Psyllium, methylcelluose, calcium polycarbophil and wheat dextrin are forms of fibre that are recommended for the treatment of constipation, although the quality of evidence is low.20

An extensive systematic review showed that bran, ispaghula and unspecified fibre treatments were no more efficacious than control treatments for the relief of abdominal pain (RR 0.87 with 95% CI 0.76 to 1.00).21 This contrasts with the general impression that soluble fibre is efficacious in the treatment of constipation.22


There is evidence that the intake of FODMAPs is associated with development of symptoms of IBS, including pain. The proposed mechanisms include increasing water retention in the small intestine through the osmotic effects of the molecular entities and rapid fermentation by intestinal bacteria, leading to production of gas and short chain fatty acids with luminal distension resulting in sensations of pain and bloating and stimulation of abnormal motility.23

A meta-analysis supports the efficacy of a low FODMAP diet in the treatment of functional GI symptoms; however, further research should ensure that studies include dietary adherence, greater numbers of patients and long-term adherence.24 A systematic review and meta-analysis has documented the greater benefit of low FODMAP diet over control treatment for overall symptom severity and severity of abdominal pain and bloating in six trials of 3–6 weeks duration; the total numbers of patients in these studies were 182 on low FODMAP and 172 in the control arms.24 For severity of abdominal pain, the OR based on four trials was 1.81, with 95% CI of 1.13 to 2.88. The mechanism of benefit is assumed to relate to reduced colonic fermentation25 or greater microbial diversity and reduced total bacterial abundance.26 More specifically, a recent study of 20 patients with diarrhoea-predominant IBS (IBS-D) or IBS-M reported that low FODMAP diet decreased serum levels of proinflammatory interleukin (IL)-6 and IL-8, as well as levels of faecal bacteria (Actinobacteria, Bifidobacterium and Faecalibacterium prausnitzii), as well as faecal total SCFAs and n-butyric acid compared with baseline.27 Lower levels of Bifidobacteria were confirmed with low FODMAP diet in another study.28 Conversely, increased Actinobacteria richness and diversity were demonstrated in another study, which also showed minor decrease in H2 production, decreased urinary histamine and reduced IBS symptom severity score.29

However, a more recent randomised, controlled trial has shown that a diet low in FODMAPs reduces symptoms of IBS as well as traditional dietary advice,30 and there was no significant difference in the primary end point of the proportion of patients reporting adequate relief of IBS-D symptoms by ≥50% during intervention weeks 3 and 4 of the 4-week trial.31 In the latter study, a secondary end point was a 30% reduction in mean daily abdominal pain score during 2 of the 4 weeks in the study, and this pain end point was significantly reduced by low FODMAP diet compared with control diet. Overall, the IBS dietary algorithm has been simplified to first-line (healthy eating, provided by any healthcare professional) and second-line (low FODMAP, provided by dietitian) dietary advice.32 An alternative to a low FODMAP diet that has shown equal efficacy is gut-directed hypnotherapy.33 In general, the low FODMAP diet remains controversial, and it still presents short-term and long-term limitations, including a high level of restriction that may be required in individual patients, the need for monitoring by an expert dietitian, potential for developing nutritional deficiencies, potential for changes in gut microbiota, lack of predictors of response as well as relative efficacy compared with other dietary, psychological or pharmacological interventions for IBS.34


Scholarly reviews have appraised the potential effects of the intestinal microbiota on intestinal motility and sensation, autonomic nervous system, hypothalamic-pituitary-adrenal axis, enteric nervous system, mucosal barrier and neuroimmune signalling.35 The interactions between the gut microbiota, stress and the central nervous system have emerged suggesting that visceral pain-related disorders may be candidates for symptom relief through therapeutic alterations of the microbiome.36 These observations suggest that therapeutic interventions that alter the microbiome may have beneficial effects in patients with IBS.

Earlier studies had shown that probiotics affect bowel function and bloating in patients with IBS-D treated with individual probiotics such as Bifidobacterium infantis37 or combination probiotics.38 ,39 Ford et al40 reported benefit of probiotics over placebo (standard mean difference for continuous scale measurement: −0.25 (−0.36 to −0.14) of persisting symptoms) for global symptom or abdominal pain scores. However, the data do not provide sufficient information to specifically assess effect on abdominal pain. A 2015 systematic review and meta-analysis of probiotics in IBS41 included 15 trials, of which only 2 had sufficient data to assess effects on abdominal pain; these studies used combination Escherichia coli (DSM 17252) and Enterococcus faecalis (DSM 16440) or E. coli (DSM 17252) alone as the probiotics compared with placebo.42 ,43 The RR of responders to probiotics based on abdominal pain score in patients with IBS compared with placebo was 1.96 (95% CI 1.14 to 3.36; p=0.01) and borderline heterogeneity was identified (p=0.11).

However, recent randomised controlled trials demonstrated no significant benefit of probiotics preparations over placebo in the treatment of pain in adults with IBS44 in contrast to the benefit observed in some (but not all) trials in children with regard to frequency and intensity of abdominal pain, for example, with a combination of three Bifidobacterial species45 or a single bacterial species.46

In summary, effects of probiotics in the treatment of abdominal pain appear to be strain dependent, and may be more significant in children. Further studies are required to address which bacterial strains and which patients are most likely to respond.

Non-absorbed antibiotic, rifaximin

Rifaximin, administered for 2 weeks, was efficacious in providing adequate relief of global IBS symptoms over a subsequent 10 weeks in two studies.47 A meta-analysis of five trials showed that the number needed to treat (NNT) was 10.2 for global improvement of IBS (OR 1.57, 95% CI 1.22 to 2.01) and 10.1 for relief of bloating (OR 1.55, 95% CI 1.23 to 1.96).48 Durability of benefit in patients with IBS-D responding to a 2-week course of rifaximin was 50% at 10 weeks and 10% at 20 weeks.49 Rifaximin produced significant improvements in core symptoms of IBS-D in patients treated with up to three 2-week courses of therapy. With second repeat treatment, the most significant benefit was the relief of urgency and bloating, with borderline benefit on abdominal pain (p=0.055) and stool consistency (p=0.08).50 ,51 The mechanism of benefit of rifaximin is still unclear; in the past, it was assumed that the benefit reflected beneficial effects on small bowel bacterial overgrowth or direct anti-inflammatory actions that countered effects of bacterial products.52 However, there is evidence that breath hydrogen measurements after lactulose or glucose load may reflect caecal arrival.53 ,54 A recent study appraised several potential mechanisms of action of rifaximin in the treatment of non-constipated patients with IBS, including permeability, expression of barrier proteins and faecal microbiome. There were no significant effects relative to placebo.55 In summary, the efficacy and the mechanism leading to relief with this non-absorbable antibiotic are unclear.


In addition to visceral hypersensitivity, abnormal GI motility is recognised as an important pathophysiological mechanism. Increased or decreased GI transit is indeed reported in IBS-D and IBS-constipation (IBS-C), respectively.56 Moreover, patients with IBS develop increased small bowel motility in response to meal ingestion and the stress hormone, corticotropin-releasing factor (CRF); in some cases, this was associated with episodes of abdominal pain.57 ,58 As antispasmodics reduce GI contractility, these compounds may be beneficial in IBS58 and are indeed widely used in Europe.

Overall assessment of the antispasmodics

A European systematic review59 identified nine placebo-controlled studies of antispasmodics in IBS, but many did not use standardised diagnostic criteria, and all were of low to intermediate quality since they were performed before the development of the Rome criteria for study design. Abdominal pain was improved in seven of the studies, bowel symptoms improved significantly versus placebo in two studies and four of the studies reported global symptom severity improvement. Reviewers concluded that there is level II evidence suggesting that antispasmodics may improve abdominal pain, but that there is lack of evidence to support global symptom improvement.

A Cochrane review concluded that there was weak evidence for the benefit of some antispasmodics for abdominal pain and global symptom relief, although it was unclear which individual classes were effective.60 A more recent meta-analysis of 22 randomised controlled trials of antispasmodics in IBS21 showed them to be superior to placebo in relieving IBS symptoms (61% vs 44%, p<0.001; NNT=5), but with significant heterogeneity between studies. Of the 12 agents studied, the strongest data were for otilonium bromide (OB).

Otilonium bromide

This compound targets L-type and T-type calcium channels, and muscarinic type 2 and tachykinin neurokinin (NK)-2 receptors, possibly contributing to its increased efficacy. The efficacy of OB in IBS has been confirmed in four studies,21 including significant improvement of abdominal pain and bloating severity with OB versus placebo,61 or reduction in the number of pain episodes and severity of abdominal distension, improved well-being and global assessment, but not in bowel symptoms.62 A post hoc analysis of symptom ratings found higher response rates with OB for a wide range of symptoms.63

In a recent international, placebo-controlled trial of 356 patients randomised to OB (40 mg three times a day) or placebo for 15 weeks,64 OB was superior to placebo in reducing the frequency of episodes of abdominal pain, abdominal bloating and in global efficacy, with higher probability of remaining relapse free during 10 weeks' follow-up.

Antispasmodics are generally well tolerated, apart from anticholinergics which can cause atropine-like side effects, including constipation.21

Peppermint oil

Peppermint oil and its active ingredient, l-menthol, are smooth muscle calcium channel antagonists that may cause muscle relaxation65 and, therefore, serve as an antispasmodic. Peppermint oil and menthol also have κ-opioid agonistic properties that may alter gut sensitivity,66 have been reported to possess anti-inflammatory effects67 and have serotonergic (5-HT3) antagonistic properties.68 Menthol, the active component in peppermint oil, is also widely used in medicinal preparations for the relief of acute and inflammatory somatic painful conditions. Recent evidence has indicated that this menthol-induced analgesia is mediated by activation of the temperature sensing ion channel, transient receptor potential ion channel melastatin subtype 8 (TRPM8).69 This same receptor is expressed by nociceptive visceral afferents, where TRPM8 has antinociceptive properties. One can thus anticipate that peppermint oil, if delivered efficiently to these afferent nerve endings, may contribute to a better pain relief compared with standard antispasmodics.

In a systematic review and meta-analysis of five randomised controlled trials of an older formulation of peppermint oil that included 197 patients on the active treatment arm and 195 on placebo, the analysis favoured peppermint oil (RR 2.23 (95% CI 1.78 to 2.81)) over placebo.70 Peppermint oil was significantly superior to placebo for global improvement of IBS symptoms (five studies) and improvement in abdominal pain (five studies).70 However, most of the clinical trials performed were small in size and, therefore, lacked sufficient statistical power to draw definite conclusions. Patients treated with peppermint oil were more likely to experience adverse events like heartburn, dry mouth, belching, peppermint taste and peppermint smell. Such events were, however, mild and transient in nature. A new formulation of peppermint oil with sustained release into the small bowel was tested in a small multicentre trial and showed no superiority over placebo,71 although pain, bloating and urgency were significantly reduced.



The efficacy of antidepressants is best supported by reviewing meta-analyses. One such analysis in 2009 evaluated 13 studies and compared antidepressants with placebo for IBS, with a total cohort of 789 patients, 432 on active therapy and 357 on placebo.72 In the analysis, there was a lower RR of IBS symptoms persisting or remaining unimproved after treatment with antidepressants, compared with placebo. The funnel plot was asymmetrical (p=0.02), suggesting publication bias, but the estimated NNT for antidepressant therapy to prevent IBS symptoms persisting was 4 (95% CI 3 to 6). This experience was updated73 ,74 with the addition of four more papers, for a total of 1100 patients (592 on antidepressants and 508 on placebo); the NNT of 4 was confirmed. Effect on abdominal pain was reported by 7 of the 17 randomised controlled trials that included 182 patients on antidepressants and 169 patients on placebo. The overall effect of antidepressants was significant; however, there was considerable heterogeneity between studies (I2=72.4%, p=0.001).

Nevertheless, these analyses have a number of flaws or inconsistencies in the design, analyses of efficacy, NNT and safety, as documented elsewhere,73–,75 leading to significant questions regarding the generalisability and true NNT for this class of pharmacological agents.


The Ford meta-analysis73 documented that 31.3% of patients taking antidepressants presented with adverse effects compared with 16.5% of those given placebo (RR=1.63, 95% CI 1.18 to 2.25). The number needed to harm was 9 (95% CI 5 to 111).

The recent literature is questioning whether antidepressants are safe drugs when used over the long term for non-psychiatric indications, but several confounding factors may influence this link (eg, severity of depression and other psychiatric comorbidities) and causality has not been proven. For example, selective serotonin reuptake inhibitors (SSRIs) affect bone health,76 and there is epidemiological evidence of dementia (table 2) with long-term antidepressant treatment, based on a population-based, retrospective, case-control analysis using the Taiwan National Health Insurance Research Database of patients enrolled from 2005 to 2011.77 Thus, the adjusted OR for dementia was reduced in patients using tricyclic agents, but use of other classes such as SSRIs and other antidepressants was associated with an increased risk of dementia. A separate prospective, nationwide, population-based cohort study in France78 suggests exposure to psychotropics was higher in patients with dementia, but neither of these two population-based studies actually prove causality.

Table 2

Risk of adverse event of dementia: relative risk (RR) or OR and CI based on systematic reviews and meta-analyses

5-HT3 antagonists

The role of 5-HT in GI motility and pain perception is rather complex,79 ,80 and intervention in the 5-HT signalling pathway may, therefore, impact several mechanisms involved in IBS symptoms, including pain. Several studies have documented relief of abdominal pain and discomfort in addition to beneficial effects of this class of medications in the relief of diarrhoea and urgency. For example, meta-analyses that appraised homogeneous patient cohorts with standard study design, duration, end points and dosages have demonstrated alosetron's efficacy, based on eight trials with 3214 patients treated with alosetron and 1773 with placebo, with an average of 623 patients per trial, which is 10 times higher than the average size of antidepressant trials.81 The overall pooled estimated RR for the relief of abdominal pain and discomfort was 1.30 (1.22 to 1.39) in favour of 5-HT3 antagonist treatment. The calculated NNT was 7.7, and the overall risk difference was 0.13 (0.1 to 0.16). The results were consistent across studies (I2 22%).

These results were confirmed in a subsequent meta-analysis of 10 trials of alosetron (6232 patients), which showed RR of response of 1.23 (1.15 to 1.32) for relief of abdominal pain and discomfort, and 1.55 (1.40 to 1.72) for the four trials (1732 patients) that assessed the global improvement of IBS symptoms.82 Alosetron has also been tested for clinical effectiveness and impact on quality of life. It is only approved for severe IBS in women, after reports of ischaemic colitis (∼1 in 800) and complications from constipation. The latter is easily managed in clinical practice by individualisation of the dose used (MC, personal observation). An analysis of the Federal Adverse Event Reporting System documented more recent data on the reports of ischaemic colitis in patients treated with alosetron, although this medication was responsible for <1% of the cases in the database.83 Overall, this risk has to be viewed in relation to the epidemiological evidence as well as medical claims data for an association between IBS and ischaemic colitis independent of therapy with serotoninergic agents.84 ,85 One randomised, crossover, placebo-controlled trial of ondansetron included the option of dose titration in order to reflect common clinical practice of modifying the drug dose to minimise adverse effects while benefitting from the drug.86 In this trial, pain scores did not change significantly with ondansetron; however, IBS symptom severity score fell more with ondansetron than placebo, and the RR of adequate relief response was higher with ondansetron than with placebo (RR 4.7, 95% CI 2.6 to 8.5, p<0.001).

Another 5-HT3 antagonist, ramosetron, which is efficacious in both males and females, has been tested predominantly in Japan where it is approved and marketed.87 ,88 Among male patients,87 there was no significant group difference between ramosetron and placebo in severity of pain and discomfort over the entire 12-week trial, although such a difference was recorded at week 5 of treatment. In the trial conducted in female patients with IBS-D, the monthly responder rate (each month over the 3-month trial and aggregated over the 3 months) for abdominal pain/discomfort was significantly higher in the ramosetron compared with the placebo group, except for month 2.88 Beneficial effects on global assessment of symptoms, adequate relief and quality of life related to IBS were recorded in both genders.


Effects of the chloride channel activator, lubiprostone, on bowel function and pain scores in IBS-C have generally shown consistent efficacy for spontaneous bowel movement rate over 3 months and somewhat reduced efficacy in relief of reduction in IBS pain and discomfort in the third month of a 3-month clinical trial.89 These results were generally replicated in a systematic review and meta-analysis,90 which showed the 3 months combined standardised difference in mean change of 0.181 (95% CI 0.212 to 0.573; p=0.37).

Linaclotide, a guanylate cyclase C receptor agonist, has beneficial effects on bowel function in IBS-C and, in addition, it appears to increase the proportion of adequate relief and global relief responders, as well as improving weekly frequency of bowel movements and reducing pain. It also relieved pain in patients with severe symptoms (eg mean daily score of >3 in worse pain, based on 11 point scale), as well as discomfort and bloating ratings during 12 weeks of treatment.91 ,92 The effect on abdominal pain has been shown experimentally to result from linaclotide's inhibition of colonic nociceptors and relief of abdominal pain via guanylate cyclase-C and extracellular cyclic guanosine 3′,5′-monophosphate, which reaches the visceral afferents in the lamina propria or submucosa.93 Based on the US Food and Drug Administration's (FDA) recommended end point for IBS-C of an increase from baseline of one or more complete spontaneous bowel movements/week and a 30% or more reduction from baseline in the weekly average of daily worst abdominal pain scores for 50% of the treatment weeks, the RR for response to treatment with 290 μg linaclotide compared with placebo was 1.95 (95% CI 1.3 to 2.9).94 In the same study, the RR for ≥30% decrease in worst abdominal pain for ≥75% weeks was 1.58 (95% CI 1.02 to 2.46); however, there were only two trials at the time of that report and the I2 heterogeneity was ≥80% for both end points.

µ-Opioid agonist

There is evidence that µ-opioid agonists are used during acute exacerbations of pain in patients with IBS95; a combined European and the US study documented use of opioids in 35% of attacks either alone or in combination with other drugs, and that patients with IBS-D were more likely to use opioids during pain attacks (32% of attacks) than were patients with IBS-C (20% of attacks) or IBS-M (19% of attacks). There are no randomised controlled trials of the use of µ-opioid agonists in the treatment of chronic pain in patients with IBS. Given that ∼4% of adults in the USA are on opioids for at least 3 months for chronic non-cancer indications, there are several public health initiatives96 to try to end the epidemic of opioid use for different pain indications. It is important to reiterate that there is no evidence for use of µ-opioid agonists for the pain of IBS; however, recent data show that among the patients with no cancer attending the Mayo Clinic Emergency Department with acute abdominal pain, ∼19% (442/2354) were on µ-opioid agonists for chronic pain. The indication for the use of opioids was abdominal pain in 21% (93/442) of these patients, suggesting that despite the lack of evidence of efficacy or safety, µ-opioid agonists are being prescribed for patients with non-cancer-related abdominal pain, which likely includes IBS.97

The κ-opioid agonists, fedotozine and alvimopan, are no longer in development for IBS indication.

A mixed opioid agent, eluxadoline

Eluxadoline is a µ-OR and κ-OR agonist and a δ-OR antagonist with minimal oral bioavailability. Eluxadoline, at 100 and 200 mg, resulted in greater improvements in bowel movement frequency and urgency, global symptoms, IBS symptom severity score, IBS quality of life and adequate relief.98 Results on the primary efficacy end point (combined bowel function and pain) were generally confirmed in pivotal trials with an NNT of ∼8,99 although the abdominal pain scores were not significant for either of the two trials at either the 75 or 100 mg dosages. The adverse events of pancreatitis and sphincter of Oddi spasm (SOS), each in 0.3% of patients, led to specific exclusions of patients with a history of bile duct obstruction, pancreatitis, severe liver impairment or severe constipation, and in patients who drink more than three alcoholic beverages per day. An updated analysis of safety of eluxadoline in the phase II and III trials shows that clinically apparent SOS events were observed in eluxadoline-treated patients without a gallbladder, and the majority were observed with the higher (100 mg) dose of eluxadoline.100 Nevertheless, the FDA Adverse Event Reporting System received information on 99 cases of pancreatitis, 39 cases of SOS and 220 cases of abdominal pain within 10 months of the availability of the medication to patients with IBS-D.101

Histamine H1 receptor antagonist, ebastine

Mast cells and their mediators, in particular histamine, serotonin and proteases, are increasingly recognised as contributing to the pathogenesis of IBS (for review, see ref. 102). Of interest, histamine is released by IBS colonic biopsies and induces visceral hypersensitivity to colorectal distension in murine models. Recently, evidence was reported that histamine sensitises TRPV1 on neurons from murine DRG and on human submucosal neurons in rectal biopsies via activation of H1 receptors (HRH1).103 Moreover, supernatant of IBS biopsies sensitised murine DRG neurons, an effect also mediated via HRH1. Based on these findings, 51 patients with IBS were treated with ebastine, a non-sedating antagonist of HRH1. Ebastine reduced visceral hypersensitivity in those with visceral hypersensitivity prior to treatment and reduced overall IBS symptoms and abdominal pain in patients with IBS.103

NK2 receptor antagonist, ibodutant

NKs (eg, substance P) and NK2 receptors are abundantly expressed in the GI tract and mediate robust and long-lasting contractions of smooth muscle in the gut. Moreover, NK2 receptor activation is involved in stimulation of sensory nerves and activation of visceral reflexes. Based on these characteristics, NK2 receptor antagonists have been evaluated as treatments for IBS. In a phase II, dose-finding study, ibodutant, a highly selective NK2 antagonist with high oral bioavailability, revealed improvement in patients with IBS-D and a baseline pain severity score >1.104 A more recent multinational, double-blind, placebo-controlled study in 559 patients showed dose-dependent improvement of overall symptoms, abdominal pain and stool pattern in IBS-D in females, but not in males, in a phase II, randomised controlled trial; the best efficacy was observed with a 10 mg dose.105 The tolerability of the compound was reported to be excellent.

GABAergic agents

GABAergic agents are α2δ ligands that generally bind potently to an auxiliary protein associated with voltage-gated calcium channels, reducing depolarisation-induced calcium influx at nerve terminals which reduces the release of several excitatory neurotransmitters including glutamate, noradrenaline, substance P and calcitonin gene-related peptide, which are involved in pain mechanisms.

Forty patients with IBS-D were randomised for 5-day treatment with gabapentin, 300 and 600 mg/day; rectal sensory thresholds were increased through attenuating rectal sensitivity to distension and enhancing rectal compliance.106 Pregabalin has been tested in pharmacodynamics studies in healthy controls107 and in patients with IBS, with inconsistent results on effects on colonic compliance and sensation thresholds or ratings;108 ,109 however, no clinical trials are reported to date with either of these agents.

Visceral analgesics: a look to the future

Several chemical and molecular factors in the intestine have a potentially significant role in IBS, particularly in IBS-D, including mast cell products such as histamine, proteases and tryptase, and mucosal messenger RNAs (mRNA), proteins and microRNAs (miRNA); these are reviewed elsewhere.110 The mast cell stabiliser, ketotifen, was tested in a prior single-centre trial,111 and it increased the threshold for discomfort on rectal balloon distension in patients with IBS with visceral hypersensitivity, reduced global IBS symptoms and severe abdominal pain, and improved health-related quality of life. However, the number of CD117-positive and tryptase-positive mast cells, spontaneous release of histamine and tryptase and the threshold of discomfort at baseline did not predict the response to treatment with ketotifen. The potential unblinding resulting from sedating effects of ketotifen raised questions about its therapeutic potential. As ketotifen also possesses HRH1 antagonistic properties and HRH1 has been implicated in visceral hypersensitivity, a follow-up study evaluating the HRH1 antagonist, ebastine, indeed showed a significant improvement in global relief and abdominal pain.103 Of interest, a recent study detected histamine levels in urine samples of patients with IBS that were modulated by a low FODMAP diet.29 Detection of histamine in urine or other samples may represent an interesting approach to select patients responding to HRH1 antagonists.

Future medications may target the genetic variations that are associated with increased visceral pain or sensitivity, such as the ion channel Nav 1.5, serotonin-reuptake (transporter) protein, cannabinoid 1 receptor or the cannabinoid metabolism (such as fatty acid amide hydrolase modulation).

Another potential target in visceral pain modulation is TRPM8, which is activated by cold temperatures and cooling agents such as menthol and which may conceivably be a mechanism of the benefit of peppermint oil in IBS. TRPM8 has been identified on colonic primary afferent neurons, TRPM8 mRNA in colonic DRG neurons and TRPM8 protein on nerve fibres throughout the wall of the colon.112 Given the mucosal expression of P2RY4 in human colonic mucosa as well as diverse miRNA (eg, −24, −199) and epigenomic changes (methylation, acetylation) in animal models of visceral hypersensitivity and their interactions with 5-HT, brain-derived neurotrophic factor, CBR1, TRPV1 mechanisms (reviewed elsewhere, ref. 110), their potential application augurs well for new approaches to visceral analgesic development.

Conclusions and a look to the future

There is still considerable unmet need for effective and safe visceral analgesics. The most efficacious current therapies for visceral pain are directed (primarily) at bowel dysfunction (eg, 5-HT3 antagonists). Antidepressant efficacy for visceral pain is questionable, based on flawed meta-analyses. Advances will only materialise with identification of biomarkers to better select the subpopulations that will respond to a certain compound directed at the visceral hypersensitivity, and with collaborative application of basic neurobiology and pharmacology to develop peripheral visceral analgesics without central adverse events. The recent discoveries of mucosal expression of genes, and miRNA and epigenetic markers in human biopsies and in animal models of visceral hypersensitivity suggest that future therapies may be developed to target new mechanisms involved in abdominal pain.


View Abstract


  • Contributors Both authors contributed to the writing and revisions of the manuscript.

  • Funding MC is supported by National Institutes of Health grant R01-DK92179. GB received a research grant from Takeda and is supported by a KU Leuven University Grant (Global Opportunities for Associations GOA 14.011).

  • Competing interests GB receives research funding from Takeda Pharmaceuticals for IBS pharmacology research unrelated to drugs discussed in this manuscript. He has participated as an invited speaker at a Menarini symposium. On the subject of IBS, MC received research grants from EnteraHealth, Novartis, NGM Pharmaceuticals and NPS Pharma. Also on the subject of IBS, MC performed consulting for Theravance, Takeda, Elobix AB, GlaxoSmithKline, Allergan, Rhythm, Novartis and EA Pharma and participated as an invited speaker at a Menarini symposium with all fees going to his employer, Mayo Clinic.

  • Provenance and peer review Commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.