Doctor-patient communication

• Establishing an effective doctor-patient relationship and a shared understanding is key to the management of IBS. Such a relationship can lead to improved quality of life and symptoms, reduce healthcare visits and enhance adherence to treatment (recommendation: strong, quality of evidence: low).

• Patients with IBS would like increased empathy, support and information from clinicians about the nature of the condition, diagnosis and symptom management options (recommendation: strong, quality of evidence: low).

Diagnosis, investigation and education

• The National Institute for Health and Care Excellence guideline definition of IBS (abdominal pain or discomfort, in association with altered bowel habit, for at least 6 months, in the absence of alarm symptoms or signs) is more pragmatic and may be more applicable to patients with IBS in primary care than diagnostic criteria derived from patients in secondary care, such as the Rome IV criteria (recommendation: weak, quality of evidence: low).

• All patients presenting with symptoms of IBS for the first time in primary care should have a full blood count, C reactive protein or erythrocyte sedimentation rate, coeliac serology and, in patients <45 years of age with diarrhoea, a faecal calprotectin to exclude inflammatory bowel disease. Local and national guidelines for colorectal and ovarian cancer screening should be followed, where indicated (recommendation: strong, quality of evidence: moderate).

• Clinicians should make a positive diagnosis of IBS based on symptoms, in the absence of alarm symptoms or signs, and abnormalities on simple blood and stool tests (recommendation: strong, quality of evidence: moderate).

• Referral to gastroenterology in secondary care is warranted where there is diagnostic doubt, in patients with symptoms that are severe, or refractory to first-line treatments, or where the individual patient requests a specialist opinion (recommendation: weak, quality of evidence: low).

• There is no role for colonoscopy in IBS, other than in those with alarm symptoms or signs, or those with symptoms suggestive of IBS with diarrhoea who have atypical features and/or relevant risk factors that increase the likelihood of them having microscopic colitis (female sex, age ≥50 years, coexistent autoimmune disease, nocturnal or severe, watery, diarrhoea, duration of diarrhoea <12 months, weight loss or use of potential precipitating drugs including non-steroidal anti-inflammatory drugs, proton pump inhibitors, etc) (recommendation: strong, quality of evidence: moderate).

• In those with symptoms suggestive of IBS with diarrhoea, but with atypical features such as nocturnal diarrhoea, or a prior cholecystectomy, 23-seleno-25-homotaurocholic acid scanning or serum 7α-hydroxy-4-cholesten-3-one should be considered to exclude bile acid diarrhoea (recommendation: strong, quality of evidence: low).

• In patients with IBS and coexisting symptoms suggestive of a defaecatory disorder or faecal incontinence, anorectal physiology tests can be considered, where available, to select those who might benefit from biofeedback (recommendation: weak, quality of evidence: low).

• There is no role for testing for exocrine pancreatic insufficiency, or for hydrogen breath testing to rule out small intestinal bacterial overgrowth or carbohydrate intolerance, in patients with typical IBS symptoms (recommendation: strong, quality of evidence: weak).

• The diagnosis of IBS, its underlying pathophysiology and the natural history of the condition, including common symptom triggers, should be explained to the patient. This should introduce the concept of IBS as a disorder of gut-brain interaction, together with a simple account of the gut-brain axis and how this is impacted by diet, stress, cognitive, behavioural and emotional responses to symptoms, and postinfective changes (recommendation: strong, quality of evidence: weak).

First-line treatments

• All patients with IBS should be advised to take regular exercise (recommendation: strong, quality of evidence: weak).

• First-line dietary advice should be offered to all patients with IBS (recommendation: strong, quality of evidence: weak).

• Food elimination diets based on IgG antibodies are not recommended in patients with IBS (recommendation: strong, quality of evidence: moderate).

• Soluble fibre, such as ispaghula, is an effective treatment for global symptoms and abdominal pain in IBS, but insoluble fibre (eg, wheat bran) should be avoided as it may exacerbate symptoms. Soluble fibre should be commenced at a low dose (3–4 g/day) and built up gradually to avoid bloating (recommendation: strong; quality of evidence: moderate).

• A diet low in fermentable oligosaccharides, disaccharides and monosaccharides and polyols, as a second-line dietary therapy, is an effective treatment for global symptoms and abdominal pain in IBS, but its implementation should be supervised by a trained dietitian and fermentable oligosaccharides, disaccharides and monosaccharides and polyols should be reintroduced according to tolerance (recommendation: weak, quality of evidence very low).

• A gluten-free diet is not recommended in IBS (recommendation: weak, quality of evidence very low).

• Probiotics, as a group, may be an effective treatment for global symptoms and abdominal pain in IBS, but it is not possible to recommend a specific species or strain. It is reasonable to advise patients wishing to try probiotics to take them for up to 12 weeks, and to discontinue them if there is no improvement in symptoms (recommendation: weak, quality of evidence: very low).

• Loperamide may be an effective treatment for diarrhoea in IBS. However, abdominal pain, bloating, nausea and constipation are common, and may limit tolerability. Titrating the dose carefully may avoid this (recommendation: strong; quality of evidence: very low).

• Certain antispasmodics may be an effective treatment for global symptoms and abdominal pain in IBS. Dry mouth, visual disturbance and dizziness are common side effects (recommendation: weak, quality of evidence: very low).

• Peppermint oil may be an effective treatment for global symptoms and abdominal pain in IBS. Gastro-oesophageal reflux is a common side effect (recommendation: weak, quality of evidence: very low).

• Polyethylene glycol may be an effective treatment for constipation in IBS. Abdominal pain is a common side effect (recommendation: weak; quality of evidence: very low).

Second-line treatments

• Tricyclic antidepressants used as gut-brain neuromodulators are an effective second-line drug for global symptoms and abdominal pain in IBS. They can be initiated in primary or secondary care, but careful explanation as to the rationale for their use is required, and patients should be counselled about their side-effect profile. They should be commenced at a low dose (eg, 10 mg amitriptyline once a day) and titrated slowly to a maximum of 30–50 mg once a day (recommendation: strong, quality of evidence: moderate).

• Selective serotonin reuptake inhibitors used as gut-brain neuromodulators may be an effective second-line drug for global symptoms in IBS. As with tricyclic antidepressants, they can be initiated in primary or secondary care, but careful explanation as to the rationale for their use is required, and patients should be counselled about their side-effect profile (recommendation: weak, quality of evidence: low).

• Eluxadoline, a mixed opioid receptor drug, is an efficacious second-line drug for IBS with diarrhoea in secondary care. It is contraindicated in patients with prior sphincter of Oddi problems or cholecystectomy, alcohol dependence, pancreatitis or severe liver impairment, and lack of availability may limit its use (recommendation: weak, quality of evidence: moderate).

• 5-Hydroxytryptamine 3 receptor antagonists are efficacious second-line drugs for IBS with diarrhoea in secondary care. Alosetron and ramosetron are unavailable in many countries; ondansetron titrated from a dose of 4 mg once a day to a maximum of 8 mg three times a day is a reasonable alternative. Constipation is the most common side effect. This drug class is likely the most efficacious for IBS with diarrhoea (recommendation: weak, quality of evidence: moderate to high).

• The non-absorbable antibiotic rifaximin is an efficacious second-line drug for IBS with diarrhoea in secondary care, although its effect on abdominal pain is limited. The drug is licensed for IBS with diarrhoea in the USA but is not available for this indication in many countries (recommendation: weak, quality of evidence: moderate).

• Linaclotide, a guanylate cyclase-C agonist, is an efficacious second-line drug for IBS with constipation in secondary care. It is likely to be the most efficacious secretagogue available for IBS with constipation, although diarrhoea is a common side effect (recommendation: strong, quality of evidence: high).

• Lubiprostone, a chloride channel activator, is an efficacious second-line drug for IBS with constipation in secondary care. This secretagogue is less likely to cause diarrhoea than others. However, patients should be warned that nausea is a frequent side effect (recommendation: strong, quality of evidence: moderate).

• Plecanatide, another guanylate cyclase-C agonist, is an efficacious second-line drug for IBS with constipation in secondary care. Diarrhoea is a common side effect and is no less likely than with linaclotide or tenapanor. Although the drug is licensed for IBS with constipation in the USA, it is not yet available for this indication in many countries (recommendation: strong, quality of evidence: high).

• Tenapanor, a sodium-hydrogen exchange inhibitor, is an efficacious second-line drug for IBS with constipation in secondary care. Again, diarrhoea is a frequent side effect. Although the drug is licensed for IBS with constipation in the USA, it is not yet available for this indication in many countries (recommendation: strong, quality of evidence: high).

• Tegaserod, a 5-Hydroxytryptamine 4 receptor agonist, is an efficacious second-line drug for IBS with constipation in secondary care but is unavailable outside the USA. Diarrhoea is a common side effect (recommendation: strong, quality of evidence: moderate).

Psychological therapies

• IBS-specific cognitive behavioural therapy may be an efficacious treatment for global symptoms in IBS (recommendation: strong, quality of evidence: low).

• Gut-directed hypnotherapy may be an efficacious treatment for global symptoms in IBS (recommendation: strong, quality of evidence: low).

• Psychological therapies should be considered when symptoms have not improved after 12 months of drug treatment. Referral can be made at an earlier stage, if accessible locally, and based on patient preference (recommendation: strong, quality of evidence: low).

Management of severe or refractory IBS

• Severe or refractory IBS symptoms should prompt a review of the diagnosis, with consideration of further targeted investigation (recommendation: weak, evidence: very low).

• Severe or refractory IBS should be managed with an integrated multi-disciplinary approach (recommendation: weak, evidence: very low).

• Iatrogenic harms due to opioid prescribing, unnecessary surgery and unproven unregulated diagnostic or therapeutic approaches incentivised by financial or reputational gain should be avoided (recommendation: strong, evidence: very low).

• Use of combination gut-brain neuromodulators, termed augmentation, may be considered for more severe symptoms, with vigilance for risks of serotonin syndrome (recommendation: weak, evidence: very low).

Research

• Successful completion of large clinical trials will require pragmatic inclusion criteria, minimisation of the participant trial burden and effective recruitment strategies that reach into community settings. Virtual (remote access) trial approaches will reduce geographical exclusion.

• A priority-setting partnership would best discern valuable research questions.

• Some future research themes include, but are not limited to:

  • Characterisation of the illness to understand predictors (clinical, genetic, psychological and biological) of outcome and treatment response, determinants of refractory illness and burden of illness (particularly with respect to workplace productivity) by conducting large-scale epidemiological studies with extended observation.

  • Trials of novel treatments, including pharmacological, dietary and behavioural therapies, device-based treatments and faecal microbiota transplantation. There is also a need for development of visceral analgesics. Consideration should be given to stratifying randomised controlled trials by IBS severity and subtype, burden of extraintestinal symptoms and psychological comorbidity.

  • A better understanding of treatment combinations to uncover augmentation effects between therapies, and to assess the value of multidisciplinary approaches.

  • Modulation of pain and psychological responses using pharmacological (eg, serotonin norepinephrine reuptake inhibitors) or behavioural approaches (eg, cognitive behavioural therapy used earlier in the disease course or via digital provision), and comparison of cognitive behavioural therapy with gut-directed hypnotherapy.

  • Med-tech approaches (web-based, apps and devices) to behavioural modification.

Patient summary

These guidelines have been produced on behalf of the British Society of Gastroenterology (BSG) by a team of specialists, and with input from patients with irritable bowel syndrome (IBS). The guidelines are aimed at healthcare professionals who look after patients with IBS.

IBS is a common condition, which is caused by problems arising between the gut and the brain. It consists of symptoms like abdominal pain linked to changes in bowel frequency or appearance of stools, and often bloating. Problems in IBS have been found in the nervous system supplying the gut, often making it more sensitive. Psychological factors (including stress), certain foods and the micro-organisms (bugs) living in the gut can all play a role in triggering symptoms. Occasionally, it can start after a gut infection, or antibiotic use, but more often there is no clear origin.

Some patients with IBS learn to manage their symptoms themselves, by changing their lifestyle or diet, or managing stress differently. Others, however, will consult their general practitioner who can usually make the diagnosis based on the typical symptoms. General practitioners will carry out some blood tests, including one to rule out coeliac disease (an immune reaction to gluten) and, if diarrhoea is present, a stool test to rule out inflammation. If there are concerning symptoms, including bleeding from the back passage, substantial weight loss or anaemia, a strong family history of cancer, or the patient is older, then the general practitioner will refer to a hospital specialist for further tests. The specialist may request a camera test of the large bowel, known as a colonoscopy, or do extra tests to look for other causes of diarrhoea or constipation, especially if the patient’s symptoms are less typical of IBS.

Regular exercise, making some simple dietary changes, and adopting healthy eating patterns will help many patients. Some patients find reducing dietary fibre improves symptoms, while others may find that a soluble fibre supplement helps. Referral to a dietitian can be helpful if these first-line approaches to diet do not help. Taking supplements of probiotics (often referred to as ‘friendly bacteria’) may also help, but these can be expensive.

Some patients may require different medications, depending on their main symptom. Some of these can be obtained over the counter, but others need to be prescribed by a doctor. This guideline has reviewed the evidence for which medications work, and the possible harms they may cause. We have only recommended medications with good evidence that they are effective and have recommended against tests or treatments where the evidence is that they do not help, are harmful or where there is not enough evidence. Some medications have most of their effect on the gut itself, others work both at the level of the gut and the brain (called ‘neuromodulators’ as they help to reduce nerve sensitivity). Some drugs that have good evidence are unfortunately not available, or are too expensive, in some countries.

There is good evidence that psychological treatments directed against IBS symptoms, especially cognitive behavioural therapy (CBT), and hypnotherapy, are helpful for many patients’ symptoms, but unfortunately these are not always readily accessible. Work is being done to improve access to these.

Very severe symptoms that do not respond to some of the above treatments are rare. However, patients whose symptoms do not improve may be left feeling desperate, and therefore vulnerable to approaches which are not proven, expensive or high risk. It is recommended that patients in this position are supported by a multidisciplinary specialist team to help reduce harms, such as unwarranted tests or operations, or harmful drugs.

Although much progress has been made in understanding and treating IBS, there are still many things we do not know about the condition, and there are lots of active areas for research, and therapies that need to be explored. We hope this guideline will also help to highlight and prioritise these areas.

Central nervous system and autonomic nervous system modulation

Symptoms are generated by interoceptive signals from the gut, and memories of such signals, and are modulated by emotional (anxiety, depression), cognitive (attention, beliefs and expectation) and motivational factors.33 Altered activation of brain regions responsible for cognitive processing and emotional and autonomic responses to visceral and somatic stimuli are seen in IBS, consistent with visceral hypersensitivity, hypervigilance and symptom-related anxiety.34 Some patients with IBS have compromised central inhibitory regulation of visceral and somatic stimuli.35 Modulating activity of the brain regions responsible for visceral pain using various therapies has been shown to improve IBS symptomatology.36–38

The autonomic nervous system mediates communication between the gut and brain. In IBS, a reduction in parasympathetic activity and an increase in sympathetic nervous system activity is frequently observed.39 Reduced vagal tone may be caused by stress, and impacts on gut motility and sensitivity, and peripheral inflammation and gut permeability.40 Conversely, the vagus nerve may sense the gut microenvironment indirectly and transfer this information to the brain.

Altered visceral perception

Between 20% and 60% of patients with IBS have enhanced visceral perception to various physiological stimuli (eg, mechanical or electrical).41 Hypersensitivity to mechanical distension of the gut is reported by more patients with IBS-D than IBS-C. However, studies do not distinguish between affective, cognitive and true peripheral, versus central, mechanisms of visceral hypersensitivity. A study conducted in separate patient cohorts from Sweden, Belgium and the USA reported that visceral sensitivity correlated positively with symptom severity, even after adjustment for the tendency to report symptoms or psychological comorbidity.42 Approximately 20% of patients with IBS are viscerally hyposensitive or insensitive to mechanical distension, more commonly those with IBS-C than IBS-D, with one study suggesting that the degree of insensitivity correlated with abdominal distension (ie, a true physical increase in abdominal girth).43

Transit and motility

Colonic transit is abnormal in only 10%–20% of patients with IBS-C and IBS-M, and 25%–45% of patients with IBS-D.44 45 Similar observations are seen for oro-caecal transit.46 47 However, patients with normal transit can still have abnormal fasting and postprandial motility.48 Patients with IBS-C display reduced motility, fewer high amplitude propagating contractions of the colon and delayed transit, whereas those with IBS-D have increased motility, more high amplitude propagating contractions and accelerated transit.1 Colonic transit time correlates inversely with stool consistency and, to a lesser extent, with stool frequency.45 However, symptoms of abdominal pain, bloating and flatulence correlate poorly, or not at all, with colonic transit,45 46 whereas abdominal distension (the physical increase in abdominal girth, rather than the sensation of bloating) correlates with oro-caecal and colonic transit times, and inversely with stool consistency.46 High amplitude propagating contractions in patients with IBS-D are associated with abdominal pain.1 Changes in gastrointestinal motility may be influenced by alterations in serotonin (5-Hydroxytryptamine (5-HT)) metabolism,49 with high levels reported in patients with IBS-D, and low levels in IBS-C.50

Immune regulation, inflammation and epithelial permeability

Low-grade mucosal inflammation may arise from a compromised epithelial barrier, dysbiosis or altered stress levels, and impaired epithelial barrier function from an aberrant stress and immune response, and/or dysbiosis. It can be linked to a previous episode of infectious gastroenteritis induced by bacteria, parasites or viruses, referred to as postinfection IBS (PI-IBS).51 Increased numbers of mast cells, particularly in the descending colon and rectosigmoid region,52 53 and in the small intestine,54 is the most consistent histological finding in IBS. Mast cell hyperplasia is more common in IBS-D and PI-IBS.52 55 The severity and frequency of abdominal pain correlates with the presence of activated mast cells in close proximity to nerve endings in the gut mucosa in some studies,56 but in others mast cell proliferation associates with a reduction,57 or no change,58 in visceral sensitivity. Interleukin-10 mRNA expression and protein levels are consistently reduced in the mucosa and/or peripheral circulation, in patients with IBS-D and PI-IBS, and associate with comorbid anxiety or depression.59 There is lack of consensus on whether numbers of T cells and levels of cytokines in the peripheral circulation of patients with IBS are abnormal.52 53

Increased permeability in patients with PI-IBS and IBS-D correlates with visceral sensitivity60 61 and symptom severity.62 63 However, findings in patients with IBS-C are inconsistent.64 65 The expression and levels of the tight junction protein zonula occludens are significantly reduced in IBS-D, and associated with mast cell activation and symptoms.66 Confocal laser endomicroscopy studies suggest exposure to certain food antigens can disrupt the epithelial barrier in approximately 50% of patients with IBS.67 68 Removal of the reacting antigen from the diet improved symptoms significantly.67

The microbiome

Strong evidence supports a role for bacterial, viral or parasitic infections triggering IBS.51 Antibiotic usage may also associate with either the development,69 or improvement,70 of IBS symptoms. Moreover, changes in the gut microbiome can modify gastrointestinal motility, visceral sensation, intestinal permeability, stool consistency and visceral sensitivity.71 72 In a mouse model, bacterial infection led to an increase in intestinal permeability, which appeared to allow previously tolerated food antigens to activate a localised host immune response in the gastrointestinal tract, via IgE, leading to histamine release, altered motility and visceral hypersensitivity.73 There is no conclusive evidence for a specific IBS gut microbiome profile.74 75 However, an integrated longitudinal multi-omics analysis of the gut microbiome, metabolome, host epigenome and transcriptome, in the context of host symptoms and physiology in patients with IBS-D and IBS-C, identified subtype-specific and symptom-related variations in microbial composition and function.76 It remains unclear whether such microbial changes are secondary and relate to other factors including diet, drugs, altered physiology, including gastrointestinal transit or gastrointestinal water content.

Genetics and epigenetics

Familial clustering of IBS may be attributed to both genetic and shared environmental factors.77 78 In addition, numerous genetic single nucleotide polymorphisms have been described in association with symptom phenotypes, regulation of neurotransmission, barrier function, inflammatory mediators, ion channels and bile acid metabolism in IBS.77 78 However, a meta-analysis of genes associated with inflammatory mediators found no significant associations for most genes assessed.79 There is some evidence for epigenetic changes in IBS, including alterations in DNA methylation, and various miRNAs appear to be associated with increased visceral sensitivity and permeability.80

Overview

Patients may have IBS-type symptoms for many years without presenting to medical care, often self-managing their symptoms without medical input, and some may never consult. Nevertheless, lower gastrointestinal symptoms frequently prompt people to present to primary care,81 accounting for approximately 1 in 12 of all consultations.82 Functional gastrointestinal disorders, such as IBS, are by far the most common diagnosis, but symptoms can be difficult to assess and the possibility of colorectal cancer or inflammatory bowel disease (IBD) may create diagnostic uncertainty for clinicians, and anxiety for patients.83

General practitioners are the first point of contact and provide the diagnosis and medical care for most people with IBS. Management guidelines encourage general practitioners to make a positive diagnosis of IBS, based on symptoms, in the absence of alarm symptoms or signs that warrant referral to exclude colorectal cancer (box 1),84 or abnormalities on simple investigations.10 However, persistent abdominal bloating or distension in female patients should prompt consideration of CA-125 and pelvic ultrasound to exclude ovarian cancer.85 General practitioners provide the majority of long-term medical care for people with IBS, referring only a minority of patients to specialist clinics.86 Compared with hospital settings, general practitioners can, and should, aim to build a long-term relationship with patients, harnessing this to develop a shared understanding of their IBS in the context of their other medical conditions, concerns, priorities and impact on their lives.87 88 This can assist in providing appropriate tailored education, advice and reassurance, shared decision-making and management plans, and be facilitated by an ongoing supportive doctor-patient relationship and prioritisation of continuity of care.

Doctor-patient communication and patient perspectives in primary care

Establishing an effective doctor-patient relationship and a shared understanding is key to the successful diagnosis and management of chronic conditions such as IBS.89 90 Reports from patients with IBS confirm that they would like increased empathy, support and information about the nature of the condition from general practitioners, and options for symptom management.87 They often feel their symptoms are dismissed or trivialised, describe the diagnostic process as confusing, or invasive, and the often-lengthy search for efficacious treatments as frustrating.87 89 91–95

Patients often seek information and support from multiple sources, including internet web forums,96 and may receive conflicting, or incorrect, advice. Clinicians should aim to gain a better understanding of patients’ ideas, concerns and expectations of diagnosis and management. Multiple factors influence both the patient’s decision to consult their doctor with IBS, and their ability and willingness to self-manage symptoms and engage with treatment. These factors include the impact on their own and their family’s lives, social and psychological factors, employment, comorbidities and health beliefs. All should be considered and acknowledged for successful diagnosis and management of IBS in primary care, and good doctor-patient communication generally.

Presentation of IBS to primary care

General practitioners must assess and manage undifferentiated disease, multiple comorbidities, health anxieties and hidden agendas in brief consultations. Multiple factors, described above, influence patients’ decisions to consult. Concerns about serious illness, advice or pressure from friends or relatives, life events and underlying health beliefs can all drive healthcare-seeking behaviour. Understanding reasons for presenting at a particular point in time, especially if symptoms have been present for many years, is important in determining the most appropriate management strategy.

General practitioners’ key skills, especially in relation to chronic disorders such as IBS, are to make a positive diagnosis, including providing a simple explanation of the pathophysiology underlying the symptoms, clarifying the patient’s main concerns and managing current symptoms in the wider context of the patient’s life. The doctor’s relationship with the patient, continuity of care, empathy, including acknowledgement of the impact of symptoms on daily life, a shared understanding of IBS and shared decision-making can assist in providing appropriate education, signposting to reputable online information or peer support, reassurance, advice and management options.

Diagnosis of IBS in primary care

The key to diagnosis starts with skilled, targeted history taking and examination, considering the patient’s medical history and life circumstances. The Rome diagnostic criteria are based on specific symptoms of a defined duration and frequency,2 which have been derived predominantly from secondary care patients, and are rarely used in primary care.97 Their applicability to clinical practice has been challenged as unnecessarily restrictive,98 and only a minority of people diagnosed with IBS in primary care fulfil them.99

This restrictive diagnostic approach to IBS may be unhelpful and overly complicated in this setting, where fundamentals of clinical management are common across all these functional gastrointestinal disorders. Applying rigid criteria potentially leaves many patients with troublesome impactful symptoms without a clear diagnosis, increasing uncertainty and leading to issues with providing appropriate advice and management options. The NICE guideline definition of IBS is therefore preferable.10

Investigation in primary care

A positive diagnosis of IBS can be made on the basis of ongoing characteristic symptoms, after assessing for alarm symptoms or signs, and undertaking relevant blood test results, including full blood count (FBC), C reactive protein (CRP) or erythrocyte sedimentation rate (ESR), and serological tests for coeliac disease.10 The chance of identifying organic disease on the basis of checking FBC, CRP and ESR in suspected IBS is low,100 but the prevalence of abnormal serological testing for coeliac disease is almost three times higher in people with suspected IBS than people without symptoms of IBS, irrespective of predominant stool pattern.101 If all these blood tests are normal, other investigations should be minimised. Abdominal and digital rectal examination can help exclude other diagnoses, and may confirm the consistency of stool, including rectal impaction, or identify dyssynergic defaecation (paradoxical contraction on rectal examination during straining) or low rectal masses.102 An abdominal X-ray can be considered to rule out faecal loading if constipation is the predominant symptom.

The non-invasive marker of intestinal inflammation faecal calprotectin has enabled risk stratification to prioritise access to investigations to exclude IBD in patients with chronic diarrhoea, reducing unnecessary investigations and referrals from primary to secondary care.103 However, calprotectin is not specific to IBD and can be elevated in older age groups (age ≥45 years), obesity, infection, malignancy or by medications, such as proton pump inhibitors (PPIs) or non-steroidal anti-inflammatory drugs (NSAIDs). Local laboratory values vary but, generally, a faecal calprotectin of <100 μg/g can be considered normal, 100–249 μg/g is borderline and should be repeated, with subsequent referral if persistently elevated, and ≥250 μg/g requires urgent referral to secondary care to exclude IBD.103 Faecal occult blood or faecal immunochemical testing are not used routinely for assessing patients with possible IBS in primary care, although they are recommended in current guidelines for colorectal cancer screening.104 Local and national guidelines for colorectal and ovarian cancer screening should be followed, where indicated. Once a diagnosis of IBS has been made, the general practitioner should endeavour to follow-up the patient within the next 2 months to ensure symptoms are not getting progressively worse, which may be indicative of a more sinister underlying disease process.

When to refer patients to secondary care

Most patients with IBS are diagnosed and managed by general practitioners in community settings and are never referred to secondary care, even if they have ongoing troublesome symptoms.82 In a primary care study of patients with refractory IBS, only 10% had ever had a secondary care referral.86 Reasons for seeking a secondary care opinion include: uncertainty about the diagnosis or alarm symptoms or signs; ongoing refractory symptoms that have not improved despite lifestyle changes (including diet and trials of medication), necessitating initiation of therapies that are unavailable in primary care or patient request for a specialist opinion.

Recommendations

• Establishing an effective doctor-patient relationship and a shared understanding is key to the management of IBS. Such a relationship can lead to improved quality of life and symptoms, reduce healthcare visits and enhance adherence to treatment (recommendation: strong, quality of evidence: low).

• Patients with IBS would like increased empathy, support and information from clinicians about the nature of the condition, diagnosis and symptom management options (recommendation: strong, quality of evidence: low).

• The National Institute for Health and Care Excellence guideline definition of IBS (abdominal pain or discomfort, in association with altered bowel habit, for at least 6 months, in the absence of alarm symptoms or signs) is more pragmatic and may be more applicable to patients with IBS in primary care than diagnostic criteria derived from patients in secondary care, such as the Rome IV criteria (recommendation: weak, quality of evidence: low).

• All patients presenting with symptoms of IBS for the first time in primary care should have a full blood count, C reactive protein or erythrocyte sedimentation rate, coeliac serology and, in patients <45 years of age with diarrhoea, a faecal calprotectin to exclude inflammatory bowel disease. Local and national guidelines for colorectal and ovarian cancer screening should be followed, where indicated (recommendation: strong, quality of evidence: moderate).

• Clinicians should make a positive diagnosis of IBS based on symptoms, in the absence of alarm symptoms or signs, and abnormalities on simple blood and stool tests (recommendation: strong, quality of evidence: moderate).

• Referral to gastroenterology in secondary care is warranted where there is diagnostic doubt, in patients with symptoms that are severe, or refractory to first-line treatments, or where the individual patient requests a specialist opinion (recommendation: weak, quality of evidence: low).

The first consultation in secondary care

IBS is one of the most common disorders that gastroenterologists deal with, accounting for at least 10% of consultations in the outpatient clinic.105 Patients presenting to secondary care often attend the consultation with the hope that the specialist can diagnose an organic disease that has been missed in primary care. This may be because a diagnosis of IBS is stigmatised,106 or considered, erroneously, as a purely psychological disorder,107 or due to unexpressed concerns of a missed sinister cause of their symptoms. These aspects should all be considered by the secondary care clinician when managing patient expectations and selecting appropriate investigations.107 It is equally important to build rapport and to build trust in the doctor-patient relationship in secondary care by adopting the principles of empathic listening to optimise the interaction.108 Evidence suggests that 2 minutes of active listening at the beginning of a consultation gives the patient the feeling of having being listened to,109 and therefore having confidence in subsequent decisions around their care. An empathic approach can improve quality of life and symptoms,110 reduce healthcare visits and enhance adherence to treatment.108 111 The principles of history taking are similar to those in primary care. It is useful to screen for potential symptom triggers, including previous acute enteric infection, present in approximately 10% of people with IBS,112 antibiotics or psychological stress. This demonstrates to the patient that the clinician is interested in understanding their disorder, and helps the patient’s understanding of the possible underlying aetiology and validates the diagnosis.

It is important to start the consultation by asking when the patient’s symptoms started (figure 1). A detailed history should confirm presence of the cardinal symptoms of IBS. These include abdominal pain and altered bowel habit (abnormal stool frequency and/or consistency) and, in particular, the relationship between the two, remembering that the location of pain can be in the upper or lower abdomen. Importantly, the clinician needs to assess whether the patient recognises that there is a link between the pain and the alteration in bowel habit. Thus, pain can be relieved or exacerbated by defaecation, or associate temporally with changes in bowel habit (eg, the pain is present when the individual is more constipated or has worsening diarrhoea). Predominant stool pattern, on days when the patient’s stools are abnormal, should be assessed using the Bristol stool chart.113 Attention should also be paid to other gastrointestinal symptoms. The presence of bloating is not required to fulfil the Rome IV criteria but, if present, is highly suggestive of IBS, and is often accompanied by visible abdominal distension.114 Although the Rome IV criteria are the gold standard to define IBS for research purposes, they are probably overly restrictive for use, even in secondary care, and a pragmatic definition in line with that used in the NICE guideline,10 and outlined above, should be preferred.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Diagnostic algorithm detailing the approach to the positive diagnosis of IBS. *If the initial faecal calprotectin level is abnormal (eg, ≥250 μg/g), the suspicion for IBD is high, proceed to colonoscopy; if the initial faecal calprotectin level is indeterminate according to local laboratory values (eg, 100–249 μg/g), repeat the test off non-steroidal anti-inflammatory drugs, proton pump inhibitors, etc, and refer for colonoscopy if the repeat test remains indeterminate or is abnormal. †SeHCAT, 23-seleno-25-homotaurocholic acid.

Coexistent early satiety, postprandial fullness, epigastric pain, nausea or heartburn are common, as functional dyspepsia and gastro-oesophageal reflux frequently overlap with IBS.115–117 Extraintestinal symptoms, such as back pain, bladder and gynaecological symptoms, and insomnia are frequent, as is the presence of other functional somatic disorders, such as fibromyalgia, tension headache or chronic fatigue.118 Common mental disorders, and somatoform-type behaviour, often coexist.119 120 A patient with more severe IBS may volunteer a history of abuse, or respond to a cue when told that ‘some people report abuse as a possible cause…’. It is, therefore, important to consider all these factors when assessing a patient with a possible diagnosis of IBS during the initial consultation, as they add diagnostic value, and predict the degree of functional limitation of the condition, reduction in quality of life and healthcare utilisation.117 This may prevent presentation to multiple other specialities, and avoid iatrogenic harm from unnecessary interventions.121 Objective evidence of weight loss is also important to assess and document. Other relevant items in the clinical history include previous surgical interventions, and a family history of gastrointestinal cancer, IBD, coeliac disease or IBS. Finally, attention should be taken to exclude gastrointestinal symptoms related to a change in diet, drugs that can alter gut motility, such as psychotropic agents or opioids or alcohol excess.

Investigations in secondary care

In a patient with normal investigations from primary care on referral, exhibiting typical symptoms, and in the absence of alarm symptoms or signs, or atypical features, the diagnosis of IBS is secure (figure 1). A validation study of the Rome IV criteria in secondary care demonstrated this was particularly the case for IBS-C and IBS-M.21 Patients meeting these criteria were 21 times more likely to have IBS-C than to not have IBS-C, and 11 times more likely to have IBS-M than to not have IBS-M after limited diagnostic workup. The clinician should, therefore, appear confident and, after clinical assessment is complete, communicate a positive diagnosis of IBS based on symptoms. In those with alarm symptoms or signs, urgent referral for colonoscopy or radiological evaluation of the colon is required,10 although the diagnostic performance of alarm symptoms or signs is modest,122 and up to 80% of patients with IBS in primary and secondary care will report at least one alarm symptom.123 In those with atypical features, such as nocturnal diarrhoea or abdominal pain, or features of obstructive defaecation, further limited investigation may be required to exclude important mimics.124 These include microscopic colitis or primary, or idiopathic, bile acid diarrhoea (BAD) in those with suspected IBS-D, and dyssynergic defaecation and other defaecatory disorders in those with suspected IBS-C.

The yield of colonoscopy in patients with IBS is extremely low,125 and there is no evidence of reassurance being derived by patients from a normal examination.126 However, colonoscopy to exclude microscopic colitis should be considered in patients with diarrhoea. Factors that should alert the clinician to the possibility of microscopic colitis include female sex, age ≥50 years, coexistent autoimmune disease, nocturnal or severe, watery, diarrhoea, duration of diarrhoea <12 months, weight loss or use of potential precipitating drugs including NSAIDs, PPIs, selective serotonin reuptake inhibitors (SSRIs) or statins.127–129

In primary BAD, bile acids enter the colon, enhancing mucosal permeability, inducing water and electrolyte secretion, and accelerating colonic transit. The condition is diagnosed via 23-seleno-25-homotaurocholic acid (SeHCAT) scanning, although this may be unavailable in some countries.130 A serum 7α-hydroxy-4-cholesten-3-one is a reasonable alternative. The current BSG guideline for the investigation of chronic diarrhoea does not recommend a therapeutic trial of a bile acid sequestrant as a diagnostic test for BAD,131 because a lack of response does not exclude the condition.132 Symptoms of BAD can mimic IBS-D, with between one-in-three and one-in-four patients with suspected IBS-D having an abnormal SeHCAT retention,133 134 and response rates to a bile acid sequestrant are higher at retentions of <10% or <5%.135 Predictors of primary BAD are lacking, other than higher body mass index,133 but if nocturnal or severe diarrhoea is present the diagnosis should be considered. BAD should also be suspected in patients with symptoms suggestive of IBS-D with prior cholecystectomy.

Symptoms suggestive of a defaecatory disorder include straining at stool, a sensation of incomplete, or blocked, evacuation and use of digital manoeuvres to facilitate defaecation. However, these symptoms are common in patients with IBS-C, as well as in those with functional constipation, and whether they arise due to different pathophysiological mechanisms in the two disorders is unclear.136 137 In the presence of these symptoms, or of faecal incontinence, physiological testing could be considered, where available, to facilitate selection of patients most likely to benefit from targeted pelvic floor biofeedback therapy to improve anorectal function.138 Particular caution should be given to considering surgical correction of anorectal anatomic alterations in patients with typical symptoms of IBS-C, as no prospective studies have demonstrated surgery improves symptoms.139 Likewise, abdominal pain is considered a relative contraindication to surgical correction of refractory slow transit constipation.139

Some investigators have reported a high prevalence of exocrine pancreatic insufficiency (EPI) in patients with suspected IBS,140 although other studies have not confirmed this.141 The current BSG guideline for the investigation of chronic diarrhoea only recommends screening patients for EPI, via faecal elastase, if features consistent with fat malabsorption are present.131 Similarly, testing patients with suspected IBS-D for EPI is not recommended, unless steatorrhoea is reported. Finally, there is no role for hydrogen breath testing to exclude lactose intolerance or small intestinal bacterial overgrowth in patients with IBS, particularly as these tests may be falsely positive in patients with IBS, due to rapid transit.142 Studies using small intestinal aspiration, considered the gold standard for diagnosing small intestinal bacterial overgrowth, have not demonstrated an increased prevalence of the condition in suspected IBS,143 and the lactulose breath test correlates poorly with small intestinal aspiration.144 In addition, despite this being the rationale for use of non-absorbable antibiotics in IBS,70 a positive breath test result does not predict response to treatment.145 Similarly, variants in the sucrase-isomaltase gene have been reported to be associated with an increased risk of IBS,146 and some investigators have reported evidence of sucrase-isomaltase deficiency on small intestinal biopsy in patients with suspected IBS,147 but at present there is insufficient evidence for consideration of routine testing.

Communicating a positive diagnosis and management plan in secondary care

A diagnosis of IBS needs to be communicated clearly to the patient using simple words and explanations. There is evidence that patient education about the condition can lead to an improvement in symptoms.148 It should be underlined that IBS is a chronic disorder, with recurrent fluctuating symptoms triggered by stress, intercurrent illnesses, drugs and often the act of eating. IBS is not associated with an increased risk of cancer or mortality,149 but affects quality of life to the same degree as organic gastrointestinal diseases, such as IBD.13 The main pathophysiological aspect is related to visceral hypersensitivity, which is also the principal target of many current treatments. Therefore, explaining IBS as a disorder of gut-brain interaction, together with a simple account of the gut-brain axis and how this is impacted by diet, stress, cognitive, behavioural and emotional responses to symptoms and postinfective changes is important. Such an approach may improve patient understanding and acceptance of a diagnosis of IBS, and engagement with a shared management plan to include an explanation of the mechanisms of action, potential side effects and rationale for the use of drugs or psychological and dietary therapies within the context of the gut-brain axis.

Recommendations

• There is no role for colonoscopy in IBS, other than in those with alarm symptoms or signs, or those with symptoms suggestive of IBS with diarrhoea who have atypical features and/or relevant risk factors that increase the likelihood of them having microscopic colitis (female sex, age ≥50 years, coexistent autoimmune disease, nocturnal or severe, watery, diarrhoea, duration of diarrhoea <12 months, weight loss or use of potential precipitating drugs including non-steroidal anti-inflammatory drugs, proton pump inhibitors, etc) (recommendation: strong, quality of evidence: moderate).

• In those with symptoms suggestive of IBS with diarrhoea, but with atypical features such as nocturnal diarrhoea, or a prior cholecystectomy, 23-seleno-25-homotaurocholic acid scanning or serum 7α-hydroxy-4-cholesten-3-one should be considered to exclude bile acid diarrhoea (recommendation: strong, quality of evidence: low).

• In patients with IBS and coexisting symptoms suggestive of a defaecatory disorder or faecal incontinence, anorectal physiology tests can be considered, where available, to select those who might benefit from biofeedback (recommendation: weak, quality of evidence: low).

• There is no role for testing for exocrine pancreatic insufficiency, or for hydrogen breath testing to rule out small intestinal bacterial overgrowth or carbohydrate intolerance, in patients with typical IBS symptoms (recommendation: strong, quality of evidence: weak).

• The diagnosis of IBS, its underlying pathophysiology and the natural history of the condition, including common symptom triggers, should be explained to the patient. This should introduce the concept of IBS as a disorder of gut-brain interaction, together with a simple account of the gut-brain axis and how this is impacted by diet, stress, cognitive, behavioural and emotional responses to symptoms, and postinfective changes (recommendation: strong, quality of evidence: weak).

General overview

The treatment of IBS is generally directed towards the predominant symptom, or symptoms, experienced by the patient. All patients should be advised of the potential benefits of regular exercise, as there is some evidence from RCTs that this can be beneficial,150 151 particularly for constipation,150 with beneficial effects still apparent at 5 years in one trial.152 Otherwise, treatment should commence with dietary therapies or first-line drugs, according to patient choice, with second-line drugs reserved for those whose symptoms do not improve with these measures, due to a combination of the potential side effects, as well as the costs, of some of these agents to the health service. Most second-line drugs are only available in secondary care. Ideally, the efficacy of selected treatments should be reviewed at 3 months, and discontinued if no response, with escalation to the next available therapy (figure 2). Currently, psychological therapies are reserved for patients whose symptoms are refractory to drugs, although more research is required to explore the efficacy of earlier use, and it may be worth mentioning them earlier on so that patients have the option to consider them, and so that they are not viewed as a last resort. There should be a realistic discussion concerning the limitations of all available treatments for IBS to manage expectations. It is important to stress that cure is unlikely, but substantial improvement in symptoms, social functioning and quality of life is achievable. The final decision regarding treatment choices should be made by the patient, with advice and support from the clinician.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Treatment algorithm for IBS. *Review efficacy after 3 months of treatment and discontinue if no response. †As per the National Institute for Health and Care Excellence IBS dietary advice sheet, plus consider ispaghula. ‡TCAs should be first choice, starting at a dose of 10 mg at night, and titrating slowly (eg, by 10 mg/week) according to response and tolerability. Continue for at least 6 months if the patient reports symptomatic response. ±Where available locally, and based on patient preference, psychological therapies can be considered at an earlier stage, but are recommended strongly when symptoms are refractory to drug treatment for 12 months. 5-HT, 5-hydroxytryptamine; CBT, cognitive behavioural therapy; FODMAP, fermentable oligosaccharides, disaccharides and monosaccharides and polyols; IBS-C, IBS with constipation; IBS-D, IBS with diarrhoea; IBS-M, IBS with mixed bowel habits; IBS-U, IBS unclassified; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant.

Recommendations

• All patients with IBS should be advised to take regular exercise (recommendation: strong, quality of evidence: weak).

Methodology for systematic reviews of IBS therapy

To inform this guideline, we updated a series of systematic reviews and trial-based or network meta-analyses conducted by some of the authors.153–163 The aim was to assess the efficacy of dietary modifications and therapies, unlicensed, as well as licensed, pharmacological therapies, and psychological therapies in IBS. We considered RCTs comparing pharmacological therapies with placebo, psychological therapies with either no treatment or standard/usual care or dietary therapies with standard dietary advice, habitual diet or a sham dietary therapy. Cross-over trials were eligible for inclusion, provided extractable data were available at the end of the first treatment period, prior to cross-over. Studies recruited adults from primary, secondary or tertiary care with IBS symptoms diagnosed by any criteria (including clinical impression). Trials had to assess the effect of treatment in terms of either improvement of IBS symptoms, or improvement of abdominal pain, as a dichotomous assessment. It is important to point out that most RCTs of first-line treatments, as well as gut-brain neuromodulators and psychological therapies, used less rigorous end points to judge treatment efficacy, such as improvement in, or satisfactory relief of, global symptoms or abdominal pain. Trials of novel second-line drugs, on the other hand, tend to use Food and Drug Administration (FDA)-approved end points to judge efficacy, consisting of a ≥30% improvement in abdominal pain, an increase in the number of complete spontaneous bowel movements (CSBMs) per week in IBS-C, or a reduction in the number of days with stools of loose consistency in IBS-D, and composites thereof.

We considered the following treatments: soluble or insoluble fibre, a diet low in fermentable oligosaccharides, disaccharides and monosaccharides and polyols (FODMAPs), a gluten-free diet, probiotics, antidiarrhoeals, antispasmodic drugs (including peppermint oil), laxatives, gut-brain neuromodulators (tricyclic antidepressants (TCAs) and SSRIs, previously termed antidepressant drugs), eluxadoline, 5-HT₃ receptor antagonists, antibiotics, secretagogues, 5-HT₄ receptor agonists or psychological therapies (including gut-directed hypnotherapy) (online supplemental table 1). As this was an update of prior meta-analyses,153–163 we searched MEDLINE, EMBASE, EMBASE Classic and the Cochrane central register of controlled trials between January 2017 and September 2020. The search strategy is provided in the online supplemental file 1. No restrictions were applied regarding language of publication. We conducted a recursive search of the bibliography of eligible articles. The lead reviewer (ACF) screened titles and trial abstracts that had been identified by the search strategy for articles that could possibly be eligible for the review. The lead reviewer (ACF) then screened the selected trials to confirm eligibility, using predesigned eligibility forms. A second reviewer (CJB), masked to the initial assessment, also evaluated all identified trials for eligibility. We resolved discrepancies by discussion, with a consensus view taken, and used the kappa statistic to measure the degree of agreement for judging study eligibility.

The literature search identified 4111 citations, of which 46 appeared to be relevant, and 17 were eligible and were incorporated into this guideline.164–180 Fourteen of these were used to update meta-analyses.164–177 Agreement between reviewers for study eligibility was excellent (kappa statistic=0.81). Of these 14 studies, 2 compared linaclotide with placebo,164 165 and were used to update a previous network meta-analysis,155 8 compared various probiotics with placebo,166–173 and were used to update an existing trial-based meta-analysis158 and 4 were RCTs of a low FODMAP diet,174–177 and again were used to update a prior trial-based meta-analysis.161 The remaining 3 RCTs were an 8-week trial of bimodal release ondansetron in IBS-D,178 and 2 phase II trials of minesapride179 180 a novel 5-HT₄ receptor agonist. The results of these latter 3 trials are discussed briefly below. Recommendations for all other treatments are, therefore, made based on the results of existing trial-based and network meta-analyses.

All data for newly identified RCTs were extracted independently by two investigators (ACF and CJB) on to a Microsoft Excel spreadsheet (XP professional edition; Microsoft, Redmond, Washington, USA). We resolved disagreements between investigators by discussion. We extracted data as intention-to-treat analyses, with all dropouts assumed to be treatment failures, wherever trial reporting allowed this. We incorporated data from newly identified trials into existing trial-based and network meta-analyses. As we examined binary outcomes (global IBS symptoms or abdominal pain improved or not improved), we expressed the impact of each intervention as a relative risk (RR) of global IBS symptoms or abdominal pain not improving, together with 95% CIs, where if the RR is <1 and the 95% CI does not cross 1, there is a significant benefit of the intervention over the control. This approach is the most stable, compared with RR of improvement, or using the OR, for some meta-analyses.181

We used Review Manager V.5.4.1 (RevMan for Windows 2020, the Nordic Cochrane Centre, Copenhagen, Denmark) for updates to trial-based meta-analyses. We conducted updated network meta-analyses using the frequentist model, with the statistical package ‘netmeta’ (V.0.9–0, https://cran.r-project.org/web/packages/netmeta/index.html) in R (V.4.0.2). Network meta-analysis usually gives a more precise estimate, compared with results from standard, trial-based meta-analysis.182 183 It can also rank treatments to inform clinical decisions,184 according to their P-score, which is a value between 0 and 1, with higher scores indicating a greater probability of a treatment being ranked as best.185 For both trial-based and network meta-analyses, we pooled data using a random effects model, to give a more conservative estimate of the efficacy of individual therapies,186 and assessed heterogeneity using the I² statistic, which ranges from 0% to 100%, with 0% representing no observed heterogeneity, and larger values indicating increasing heterogeneity. A value ≤50% was chosen to represent low levels of heterogeneity.187

Fibre and dietary therapies

Over 80% of individuals with IBS report food-related symptoms, especially to fermentable carbohydrates and fats.188 189 Patients reporting adverse food reactions experience more severe gastrointestinal symptoms, associated subjective health complaints of musculoskeletal pains and chronic fatigue and reduced quality of life, compared with those without food sensitivities.189–191 Hence, most patients with IBS are keen to explore dietary options, with over 60% wanting to know what food(s) they should avoid, and up to 70% having modified their diet.192 There are multiple mechanisms by which food may trigger symptoms in IBS, including primary effects (eg, osmotic, chemical, immunological, mechanical or neuroendocrine) and secondary effects (eg, fermentation by-products, alterations in intraluminal pH or effects on the gut microbiome).193 194

Patients may seek to undertake dietary manipulations based on tests that suggest potential food intolerances. A prior RCT, comprising 150 patients with IBS and positive IgG antibodies to food, found a significant improvement in symptoms in those allocated to a true-exclusion diet, compared with a sham-exclusion diet.195 However, the effect was modest and there are concerns regarding the poor specificity and applicability of IgG antibody testing.196 For example, IgG antibodies to yeast were reported in 87% of patients yet are rarely responsible for symptoms following dietary rechallenges.195 196 Hence, food elimination diets based on IgG antibodies are not recommended. Although some studies have identified potential food intolerances via leucocyte antigen testing of peripheral blood samples or real-time confocal laser endomicroscopy, this requires further corroboration.68 197

In clinical practice, the last decade has seen a growing interest in the use of three diets for IBS, which are traditional dietary advice, a low FODMAP diet or a gluten-free diet. Of these, traditional dietary advice is considered as first-line, and is based on guidance produced by NICE and the British Dietetic Association (BDA).198 199 Its principles, which do not require formal dietetic input, include adopting healthy eating patterns, such as regular meals, maintaining adequate nutrition, limiting alcohol and caffeine intake, adjusting fibre intake, and reducing consumption of fatty and spicy foods. However, the evidence for this is based on a combination of clinical experience and the potential mechanisms by which these foods may induce gastrointestinal symptoms in IBS, rather than evidence from RCTs of this approach versus a control treatment. With regard to fibre, a systematic review and meta-analysis of 15 RCTs, comprising 946 patients, demonstrated its benefit in IBS (RR of symptoms persisting=0.87; 95% CI 0.80 to 0.94) (online supplemental figure 1).158 However, this effect was limited to soluble fibre, such as ispaghula (RR 0.83; 95% CI 0.73 to 0.94), but not insoluble fibre, like wheat bran, which may exacerbate abdominal pain and bloating. It is generally advised to start with low doses of soluble fibre (3–4 g daily) and build up gradually, as tolerated, to a total dosage of 20–30 g/day, as it increases colonic water content and volume,200 which may aggravate abdominal pain and bloating.

A low FODMAP diet is recommended as a second-line diet for IBS.10 198 FODMAPS are short-chain fermentable carbohydrates that are found in a variety of fruits, vegetables, dairy products, artificial sweeteners and wheat. They increase small intestinal water volume and colonic gas production and, in those with visceral hypersensitivity, induce gastrointestinal symptoms.201 They may also trigger bowel symptoms as they produce short chain fatty acids, which lower colonic pH.202 Hence, the benefits of adopting a low FODMAP diet in IBS seem physiologically plausible and several RCTs have evaluated its efficacy in IBS. The update to the prior systematic review and meta-analysis,161 used to inform this guideline, identified 11 trials comparing a low FODMAP diet with various dietary control interventions, including habitual diet, a high FODMAP diet, traditional dietary advice as recommended by NICE and the BDA or a sham diet, in 658 participants. A low FODMAP diet was associated with a reduction in the risk of remaining symptomatic, compared with all control interventions (RR 0.71; 95% CI 0.61 to 0.83) (online supplemental figure 2). However, as reported in the prior meta-analysis,161 the quality of evidence from these trials was very low, due to small sample sizes, difficulties in blinding and heterogeneity between studies. The latter has mainly been attributed to the various control interventions used. Interestingly, those studies that compared the low FODMAP diet with traditional dietary advice from NICE and the BDA had the least heterogeneity, but also the least magnitude of effect (RR 0.82; 95% CI 0.67 to 1.01), suggesting that the 50%–70% symptomatic benefit reported for a low FODMAP diet in some studies may have been overestimated. Moreover, RCTs have focused solely on the initial ‘elimination’ phase of the low FODMAP diet, which lasts between 4 and 6 weeks, not the subsequent reintroduction and long-term ‘personalisation’ phase. The effect of FODMAP reintroduction to tolerance on IBS symptoms is therefore unclear, although there have been open-label studies reporting the long-term efficacy of an adapted low FODMAP diet ranges somewhere between 50% and 60%.177 203

Finally, some patients with IBS report symptomatic benefit from a gluten-free diet despite no objective evidence of coeliac disease.204 The prior systematic review and meta-analysis identified only 2RCTs,161 comprising 111 participants and noted that although a gluten-free diet was associated with a reduction in global symptoms compared with a control diet, this was not statistically significant (RR 0.42; 95% CI 0.11 to 1.55). There is, therefore, insufficient evidence to recommend a gluten-free diet routinely in IBS, although given its widespread popularity further RCTs are needed. It has also been suggested that the clinical benefit reported with a gluten-free diet is, in the main, not due to the removal of gluten per se, but rather a reduction in dietary fructan content, which is a FODMAP, resulting from wheat exclusion.205 206 Future trials comparing a gluten-free diet head-to-head with a low FODMAP diet with regard to efficacy, convenience, cost and acceptability, may better inform patient choice.

Issues with following a low FODMAP diet, as opposed to traditional dietary advice, include the need for a specialist dietitian to implement it, followed by close monitoring to avoid nutritional deficiencies or the development of overly restrictive eating habits.207–209 The latter emphasises the importance of screening out patients at high risk for such behaviour prior to recommending such diets.207–209 This can be achieved using simple eating disorder questionnaires (eg, SCOFF), applying these carefully and with empathy, to identify those with high levels of psychological distress.210 211 Moreover, a low FODMAP diet may induce detrimental changes to the gut microbiota, with reductions in Bifidobacteria and total bacterial count,212 although the long-term consequences of this are unknown. Future studies should aim to identify actionable biomarkers that might predict response to a given dietary intervention. Preliminary data suggest that response to a low FODMAP diet may be predicted from baseline faecal bacterial profile and metabolomic activity, while a gluten-free diet may benefit those with serum antigliadin antibodies.213–215

Recommendations

• First-line dietary advice should be offered to all patients with IBS (recommendation: strong, quality of evidence: weak).

• Food elimination diets based on IgG antibodies are not recommended in patients with IBS (recommendation: strong, quality of evidence: moderate).

• Soluble fibre, such as ispaghula, is an effective treatment for global symptoms and abdominal pain in IBS, but insoluble fibre (eg, wheat bran) should be avoided as it may exacerbate symptoms. Soluble fibre should be commenced at a low dose (3–4 g/day) and built up gradually to avoid bloating (recommendation: strong; quality of evidence: moderate).

• A diet low in fermentable oligosaccharides, disaccharides and monosaccharides and polyols, as a second-line dietary therapy, is an effective treatment for global symptoms and abdominal pain in IBS, but its implementation should be supervised by a trained dietitian and fermentable oligosaccharides, disaccharides and monosaccharides and polyols should be reintroduced according to tolerance (recommendation: weak, quality of evidence very low).

• A gluten-free diet is not recommended in IBS (recommendation: weak, quality of evidence very low).

Probiotics

The faecal microbiome of patients with IBS may differ significantly from that of healthy individuals.216 The theory that this might, in part, be involved in pathophysiology has led to interest in whether probiotics, which are live or attenuated microorganisms that may have beneficial effects in humans, can be used to alter the microbiome, improving symptoms. We updated a prior meta-analysis of 37 RCTs,158 incorporating data from 8 new trials166–173 and randomising 6352 patients. Subgroup analyses according to type of probiotic used (where more than one trial of a particular group of probiotics was conducted) demonstrated significant effects on global symptoms or abdominal pain for combinations of probiotics (RR 0.79; 95% CI 0.70 to 0.89), Lactobacillus (RR 0.75; 95% CI 0.60 to 0.94), Bifidobacterium (RR 0.80; 95% CI 0.70 to 0.91) and Escherichia (RR 0.86; 95% CI 0.79 to 0.93) (online supplemental figure 3). Adverse event rates were similar in the probiotic and placebo arms. Variations in study design, strain and species of probiotic used, and heterogeneity between studies make it difficult to give specific recommendations. However, it is reasonable to advise patients wishing to try probiotics to take them for up to 12 weeks, and to discontinue treatment if there is no improvement in symptoms.

Recommendations

• Probiotics, as a group, may be an effective treatment for global symptoms and abdominal pain in IBS, but it is not possible to recommend a specific species or strain. It is reasonable to advise patients wishing to try probiotics to take them for up to 12 weeks, and to discontinue them if there is no improvement in symptoms (recommendation: weak, quality of evidence: very low).

Drugs used first line for IBS

Loperamide is a synthetic μ-opioid agonist that reduces myenteric plexus activity, thereby increasing intestinal transit time and enhancing water reabsorption. A prior systematic review identified only 2 RCTs of loperamide in IBS-D and IBS-M containing 42 patients.160 Although the drug improved stool frequency and consistency, it had no effect on global symptoms (RR 0.44; 95% CI 0.14 to 1.42). The incidence of adverse events with loperamide was similar to placebo in these trials. However, in clinical practice, abdominal pain, bloating, nausea and constipation are common side effects, and may limit tolerability. Titrating the dose carefully may improve tolerability.

Antispasmodics are among the most frequently used over-the-counter treatments for IBS, and can be divided, broadly, into antimuscarinics and smooth muscle relaxants. Antimuscarinics, including dicycloverine, propantheline, otilonium bromide and hyoscine butylbromide reduce intestinal motility, whereas alverine and mebeverine are direct-acting intestinal smooth muscle relaxants. The proposed mechanism of action of these agents is based on the assumption that some IBS symptoms are a result of gastrointestinal spasm and dysmotility, which antispasmodics ameliorate.217 A prior meta-analysis identified 26 RCTs, containing 2811 patients, which compared 13 different antispasmodics with placebo.160 Despite significant heterogeneity between trials, presumably driven by differences in antispasmodics studied, patient selection and study design, fewer patients treated with antispasmodics had persistent global symptoms or abdominal pain (RR 0.65; 95% CI 0.56 to 0.76) (online supplemental figure 4). However, these findings should be interpreted with caution, given heterogeneity between trial results and the varying end points studied. In addition, most trials recruited unselected patients with IBS, so whether the proposed reduction in gastrointestinal motility with antispasmodics improves diarrhoea is unclear. Access to some of these drugs is limited, although hyoscine butylbromide is available widely; pooled results from 3 RCTs, containing 426 patients, demonstrated efficacy (RR 0.63; 95% CI 0.51 to 0.78) (online supplemental figure 4).160 In contrast, in this meta-analysis, neither alverine nor mebeverine demonstrated benefit over placebo. The overall rates of adverse events were significantly higher with antispasmodics compared with placebo; most notably dry mouth, visual disturbance and dizziness.160

Peppermint oil is another popular over-the-counter remedy for IBS. Although not completely understood, its putative antispasmodic action is via L-menthol’s blockade of calcium channels.218 A meta-analysis of 8 RCTs, which included 823 patients, has evaluated its efficacy.162 For global symptoms or abdominal pain, peppermint oil was more efficacious than placebo (RR 0.58; 95% CI 0.34 to 0.98) (online supplemental figure 5). It should be stressed that these trials involved specific formulations of peppermint oil. Their results, therefore, cannot be extrapolated to other formulations and, in the largest trial to date included in this meta-analysis, which used two formulations of variable release peppermint oil (small bowel vs ileocolonic) there was no benefit over placebo for the primary end point.219 In addition, low study quality, the lack of consistent use of the Rome criteria to define IBS and heterogeneity between RCTs limit confidence in the data. There is also a lack of information as to which IBS subtype would benefit most. Overall, adverse events with peppermint oil were no more common than placebo,160 although patients can report gastro-oesophageal reflux symptoms due to its effects on the lower oesophageal sphincter. Comparison of peppermint oil and antispasmodics with other unlicensed or ‘traditional’ treatments for IBS, including ispaghula and gut-brain neuromodulators, in a network meta-analysis suggested that, for global symptoms, peppermint oil was ranked first and antispasmodics third, with both superior to placebo (online supplemental figure 6).157 In terms of effect on abdominal pain, antispasmodics ranked second, with peppermint oil third (online supplemental figure 7). Again, both were significantly more efficacious than placebo.

Current NICE guidance for the management of IBS suggests that patients with IBS-C can be treated with laxatives, advising dose titration according to symptoms.10 Although both stimulant and osmotic laxatives are efficacious in the treatment of chronic idiopathic constipation,220 only the latter have been evaluated in 2 RCTs of polyethylene glycol, recruiting 181 patients with IBS-C.221 222 In one trial, there was no significant effect on either abdominal pain or number of bowel movements,221 and in the second the number of bowel movements increased significantly, but with no improvement in abdominal pain.222 Polyethylene glycol was generally well-tolerated with abdominal pain the most frequent adverse event. The long-term efficacy of osmotic laxatives in IBS-C is unknown, as both trials were of only 4 weeks duration.

Recommendations

• Loperamide may be an effective treatment for diarrhoea in IBS. However, abdominal pain, bloating, nausea and constipation are common, and may limit tolerability. Titrating the dose carefully may avoid this (recommendation: strong; quality of evidence: very low).

• Certain antispasmodics may be an effective treatment for global symptoms and abdominal pain in IBS. Dry mouth, visual disturbance and dizziness are common side effects (recommendation: weak, quality of evidence: very low).

• Peppermint oil may be an effective treatment for global symptoms and abdominal pain in IBS. Gastro-oesophageal reflux is a common side effect (recommendation: weak, quality of evidence: very low).

• Polyethylene glycol may be an effective treatment for constipation in IBS. Abdominal pain is a common side effect (recommendation: weak; quality of evidence: very low).

Gut-brain neuromodulators

Dysfunction within the bidirectional gut-brain axis is considered to play an important role in the genesis and maintenance of symptoms in IBS. Although IBS is often considered a functional gastrointestinal disorder, these conditions have, therefore, been re-termed as disorders of gut-brain interaction.32 Patients with IBS often have comorbid anxiety and depression,119 and these are also risk factors for the subsequent development of IBS in healthy people.31 This, together with their peripheral effects on gastrointestinal function,223 is part of the rationale for the use of gut-brain neuromodulators, such as TCAs and SSRIs. In a meta-analysis of 12 RCTs of TCAs, recruiting 787 patients, these drugs were superior to placebo for global symptoms or abdominal pain (RR 0.65; 95% CI 0.55 to 0.77) (online supplemental figure 8), and for abdominal pain alone (RR 0.59; 95% CI 0.42 to 0.83) (online supplemental figure 9).159 SSRIs were also more efficacious for global symptoms or abdominal pain (RR 0.68; 95% CI 0.51 to 0.91) (online supplemental figure 8), but not abdominal pain alone (online supplemental figure 9), and there was significant heterogeneity among the 7 trials, which contained only 356 patients.159 Adverse event rates were significantly higher among patients treated with TCAs or SSRIs, with drowsiness and dry mouth the most common.159 The effect of these drugs on stool pattern is less clear, as very few trials restricted their recruitment to a particular subgroup of patients.

Other gut-brain neuromodulators include serotonin norepinephrine reuptake inhibitors (SNRIs), such as duloxetine, or agents acting on the calcium channel α2δ ligand, including pregabalin. There have been no RCTs of SNRIs in IBS, although there is evidence from case series that duloxetine may improve symptoms and quality of life,224 225 and there is good evidence for use of SNRIs in other chronic painful disorders, such as fibromyalgia and low back pain.226 Pregabalin improved visceral hypersensitivity in one small trial,227 and in a recent RCT recruiting 85 patients with IBS, 12 weeks of pregabalin 225 mg twice daily led to significant improvements in global symptoms, abdominal pain, diarrhoea and bloating versus placebo.228 Blurred vision, dizziness and altered sensation were more common with pregabalin. However, more RCTs are needed, and pregabalin is classed as a controlled drug in some countries. In a network meta-analysis evaluating relative efficacy of gut-brain modulators with other unlicensed or ‘traditional’ treatments,157 TCAs were ranked second and first for their effect on global symptoms and abdominal pain respectively, and were more efficacious than placebo (online supplemental figures 6 and 7). In contrast, SSRIs were ranked fifth and fourth for global symptoms and abdominal pain, respectively, and pregabalin was ranked sixth for global symptoms, with no benefit of either over placebo in these trials.

It is reasonable to consider using TCAs second line to treat global symptoms or abdominal pain or SSRIs second line to treat global symptoms, or if there is coexistent anxiety.226 They can be offered by general practitioners, depending on familiarity of use and expertise. The rationale for the use of gut-brain neuromodulators, as well as their side-effect profile, needs to be explained carefully to the patient, within the context of IBS as a disorder of gut-brain interaction.226 It should be reinforced that these drugs are being used at low doses for their pain modulatory properties and peripheral effects on gastrointestinal function, rather than at a dose that is used to treat common mental disorders. TCAs should be taken in the evening, before bedtime, due to their sedating effects, and may also improve sleep patterns. The patient should be counselled that these drugs take some time to have any benefit and that side effects, such as drowsiness, tend to ameliorate after the first 1 or 2 weeks of treatment. They should be commenced at a low dose (eg, 10 mg of amitriptyline once a day) and titrated relatively slowly in 10 mg increments, to a maximum of 30–50 mg once a day, with follow-up to assess efficacy and tolerability. If beneficial, the drugs are likely to be continued for a minimum of 6–12 months and, in some cases, this may be even longer-term.

Recommendations

• Tricyclic antidepressants used as gut-brain neuromodulators are an effective second-line drug for global symptoms and abdominal pain in IBS. They can be initiated in primary or secondary care, but careful explanation as to the rationale for their use is required, and patients should be counselled about their side-effect profile. They should be commenced at a low dose (eg, 10 mg amitriptyline once a day) and titrated slowly to a maximum of 30–50 mg once a day (recommendation: strong, quality of evidence: moderate).

• Selective serotonin reuptake inhibitors used as gut-brain neuromodulators may be an effective second-line drug for global symptoms in IBS. As with tricyclic antidepressants, they can be initiated in primary or secondary care, but careful explanation as to the rationale for their use is required, and patients should be counselled about their side-effect profile (recommendation: weak, quality of evidence: low).

Drugs used second line for the treatment of IBS-D

For patients with IBS-D who do not experience symptom improvement with antidiarrhoeals, several licensed therapies are available in secondary care. Eluxadoline is a μ-opioid and κ-opioid receptor agonist and δ-opioid receptor antagonist licensed for IBS-D. The drug slows intestinal transit and reduces visceral hypersensitivity.229 Data from a meta-analysis (4 RCTs containing 3122 patients) demonstrated that both 75 mg two times per day and 100 mg two times per day were superior to placebo using the FDA-approved composite end point for IBS-D, consisting of improvement in abdominal pain and stool consistency (RR 0.89; 95% CI 0.84 to 0.94 and RR 0.87; 95% CI 0.83 to 0.91, respectively) (online supplemental figure 10), global symptoms and stool consistency (online supplemental figures 11 and 12).153 Eluxadoline 100 mg two times per day was also superior to placebo for abdominal pain (online supplemental figure 13). Adverse events included constipation, nausea and headache, and adverse events leading to drop out were significantly higher with active drug than placebo. Serious adverse events, including pancreatitis and sphincter of Oddi spasm, have been reported, occurring in 0.5% of patients in these trials.230 The drug is contraindicated in patients with prior sphincter of Oddi problems or cholecystectomy, alcohol dependence, pancreatitis or severe liver impairment. Although licensed for IBS-D, the drug is unavailable in many countries.

Drugs acting as antagonists at the 5-HT₃ receptor are also licensed for IBS-D. These include alosetron and ramosetron, which slow gastrointestinal transit, reduce visceral hypersensitivity and alter rectal compliance.231–233 In a previous meta-analysis,153 both alosetron 1 mg two times per day and ramosetron 2.5 μg or 5 μg once a day were superior to placebo across various end points, including the FDA composite end point for IBS-D (3 RCTs of alosetron 1 mg two times per day, 787 patients, RR 0.69; 95% CI 0.60 to 0.80, and 1 RCT of ramosetron 2.5 μg once a day, 348 patients, RR 0.78; 95% CI 0.67 to 0.91) (online supplemental figure 10). Both drugs were also more efficacious than placebo for global symptoms, abdominal pain and stool consistency (online supplemental figures 11-13). Adverse events included constipation, nausea and headache; patients assigned to both drugs were more likely to report adverse events than with placebo. Alosetron was withdrawn from the market in 2001 due to reports of ischaemic colitis.234 However, it was reintroduced in the USA via a risk evaluation and mitigation strategy, at a lower dose of 0.5 mg two times per day, for women with severe IBS-D. Rates of ischaemic colitis observed since reintroduction are similar to the background rate in female patients with IBS.235 Ramosetron is only available in Asia.153 There have been no reports of ischaemic colitis associated with the drug. Due to the limited availability of both alosetron and ramosetron, RCTs of ondansetron, a widely available 5-HT₃ receptor antagonist with a robust safety profile, have been conducted. A small crossover trial of ondansetron titrated from 4 mg once a day to a maximum of 8 mg three times a day demonstrated significantly higher rates of improvement in urgency, bloating and stool consistency, but not abdominal pain.236 A subsequent RCT of 12 mg once a day of bimodal release ondansetron also demonstrated superiority over placebo for improvement in stool consistency, but not abdominal pain.178 Constipation is the most common side effect. Results from a parallel-group RCT are awaited.237

The efficacy of rifaximin, a non-absorbable antibiotic, has also been tested in IBS-D, on the basis that disturbances in the gastrointestinal microbiota may, in part, be responsible for symptoms. In a meta-analysis of 2 RCTs,153 which recruited 1260 patients, rifaximin 550 mg three times a day for 14 days was more efficacious than placebo for the FDA composite end point for IBS-D (RR 0.92; 95% CI 0.86 to 0.98) (online supplemental figure 10) and for stool consistency alone (online supplemental figure 12), but not for global symptoms or abdominal pain (online supplemental figures 11 and 13). Headache was the most common adverse event, but side effects were no more common with rifaximin than with placebo. Due to the modest efficacy, and concerns over the potential for adverse events (including Clostridium difficile infection and bacterial resistance) with repeated courses of rifaximin, FDA approval was not forthcoming. A subsequent ‘retreatment’ trial was therefore conducted. In this RCT, patients received open-label rifaximin and were then randomised to two repeat 14-day courses of rifaximin or placebo if they experienced symptom relapse. Significantly more patients experienced an improvement in global symptoms with rifaximin after each treatment course, and there were no safety concerns.238 The drug is now licensed for IBS-D in the USA but is not available for this indication in many countries.

A network meta-analysis comparing the relative efficacy of all these licensed therapies for IBS-D, across various end points,153 demonstrated that alosetron 1 mg two times per day ranked first for the FDA composite end point for IBS-D and global symptoms (online supplemental figures 10 and 11), with ramosetron 2.5 μg once a day second. For the FDA composite end point, alosetron was superior to all treatments, except ramosetron 2.5 μg once a day For, abdominal pain, ramosetron 2.5μg once a day and ramosetron 5 μg once a day were ranked first and second, respectively (online supplemental figure 13). Finally, for stool consistency alosetron 1 mg two times per day ranked first, with ramosetron 5 μg once a day second (online supplemental figure 12).

Recommendations

• Eluxadoline, a mixed opioid receptor drug, is an efficacious second-line drug for IBS with diarrhoea in secondary care. It is contraindicated in patients with prior sphincter of Oddi problems or cholecystectomy, alcohol dependence, pancreatitis or severe liver impairment, and lack of availability may limit its use (recommendation: weak, quality of evidence: moderate).

• 5-Hydroxytryptamine 3 receptor antagonists are efficacious second-line drugs for IBS with diarrhoea in secondary care. Alosetron and ramosetron are unavailable in many countries; ondansetron titrated from a dose of 4 mg once a day to a maximum of 8 mg three times a day is a reasonable alternative. Constipation is the most common side effect. This drug class is likely the most efficacious for IBS with diarrhoea (recommendation: weak, quality of evidence: moderate to high).

• The non-absorbable antibiotic rifaximin is an efficacious second-line drug for IBS with diarrhoea in secondary care, although its effect on abdominal pain is limited. The drug is licensed for IBS with diarrhoea in the USA but is not available for this indication in many countries (recommendation: weak, quality of evidence: moderate).

Drugs used second line for the treatment of IBS-C

In patients with IBS-C who do not experience symptom improvement with laxatives, escalation to second-line drugs should be considered in secondary care. These fall into two main classes, secretagogues and 5-HT₄ agonists. Secretagogues, including linaclotide, lubiprostone, plecanatide and tenapanor activate ion channels on the intraluminal surface of enterocytes, resulting in an efflux of ions and water into the intestinal lumen, softening stools and accelerating transit.239 240 Linaclotide is a peptide that acts as a guanylate cyclase-C agonist. Lubiprostone is a prostaglandin E₁ derivative, which activate chloride type 2 channels. Plecanatide is another guanylate cyclase-C agonist that binds in a pH-dependent manner, in contrast to linaclotide, such that the majority of its activity is confined to the proximal small bowel.241 Tenapanor is a small molecule inhibitor of the gastrointestinal sodium-hydrogen exchanger-3. 5-HT₄ agonists, such as tegaserod, have prokinetic effects and also accelerate transit.242 Abdominal bloating is a particularly troublesome symptom in patients with IBS-C,46 and the effects of all these drugs on this symptom, other than plecanatide, has been assessed in some RCTs.

In an update of a previous meta-analysis,155 a dose of 290 μg once a day linaclotide was superior to placebo in 5 RCTs, containing 3193 patients, for the FDA composite end point for IBS-C, consisting of improvement in abdominal pain and an increase of ≥1 CSBMs per week from baseline (RR 0.82; 95% CI 0.78 to 0.87) (online supplemental figure 14), abdominal pain alone (online supplemental figure 15) and an increase of ≥1 CSBMs per week from baseline (online supplemental figure 16). The drug was also superior to placebo in terms of an improvement in abdominal bloating in 4 trials containing 3061 patients (online supplemental figure 17). Adverse events were significantly more common with linaclotide 290 μg once a day, with diarrhoea being the most common. Lubiprostone 8 μg two times per day was superior to placebo for both the FDA composite end point for IBS-C (RR 0.87; 95% CI 0.78 to 0.96) (online supplemental figure 14) and abdominal pain alone (online supplemental figure 15) in this meta-analysis, using a post hoc analysis of data from 2 phase III RCTs, containing 452 patients.155 The drug was superior to placebo for abdominal bloating in these 2 RCTs (online supplemental figure 17). Adverse events were no more common with lubiprostone, except for nausea. In the same meta-analysis,155 both plecanatide 3 μg once a day and 6 μg once a day were superior to placebo for the FDA composite end point (RR 0.88; 95% CI 0.82 to 0.94 for 3 μg once a day in 3 RCTs, recruiting 1632 patients, and RR 0.87; 95% CI 0.81 to 0.93 for 6 μg once a day in 2 RCTs, containing 1461 patients) (online supplemental figure 14) and abdominal pain alone (online supplemental figure 15), but not for an increase of ≥1 CSBMs per week from baseline (online supplemental figure 16). Adverse events were significantly more frequent with plecanatide 3 μg once a day, compared with placebo, and diarrhoea was significantly more likely with both doses. Finally, 3 RCTs of tenapanor 50 mg two times per day, recruiting 1428 patients, were included in this meta-analysis.155 The RR for the FDA composite end point, compared with placebo, was 0.85 (95% CI 0.79 to 0.92) (online supplemental figure 14). The drug was also more efficacious than placebo for abdominal pain (online supplemental figure 15) and an increase of ≥1 CSBMs per week from baseline (online supplemental figure 16). Again, the drug was more likely to improve abdominal bloating than placebo in 3 trials containing 1428 patients (online supplemental figure 17). Except for diarrhoea, adverse events were no more likely with the drug than with placebo.

A previous meta-analysis of 11 RCTs demonstrated that tegaserod was superior to placebo for the treatment of IBS-C in 9242 patients (RR 0.85; 95% CI 0.80 to 0.90) (online supplemental figure 18).163 Diarrhoea was the most common adverse event and was significantly more likely than with placebo. Due to a small excess number of cerebrovascular and cardiovascular ischaemic events in patients taking the drug, it was withdrawn in 2007. Tegaserod was reintroduced in the USA in 2018 for female patients <65 years of age with IBS-C without pre-existing cardiovascular disease, based on a post hoc analysis of 3 large trials reporting efficacy according to the FDA composite end point for IBS-C. In a meta-analysis using data from these 3 trials, containing 2472 patients, the drug was superior to placebo (RR 0.85; 95% CI 0.80 to 0.91) (online supplemental figure 14).154 Finally, tegaserod was superior to placebo for abdominal bloating in 4 RCTs, containing 5132 patients (online supplemental figure 17). Although prucalopride, which is a highly selective 5-HT₄ agonist with no known cardiovascular or cerebrovascular safety concerns, is efficacious in the treatment of chronic idiopathic constipation,220 to date there have been no RCTs in IBS-C.

A network meta-analysis examining the relative efficacy of secretagogues and tegaserod across 18 RCTs, in 10 638 patients, demonstrated that all drugs were superior to placebo.154 155 Linaclotide 290 μg once a day ranked first across all end points, including abdominal bloating (online supplemental figure 14-16 and online supplemental figure 19), but on indirect comparison of active treatments there were no significant differences between individual drugs and dosages.

Recommendations

• Linaclotide, a guanylate cyclase-C agonist, is an efficacious second-line drug for IBS with constipation in secondary care. It is likely to be the most efficacious secretagogue available for IBS with constipation, although diarrhoea is a common side effect (recommendation: strong, quality of evidence: high).

• Lubiprostone, a chloride channel activator, is an efficacious second-line drug for IBS with constipation in secondary care. This secretagogue is less likely to cause diarrhoea than others. However, patients should be warned that nausea is a frequent side effect (recommendation: strong, quality of evidence: moderate).

• Plecanatide, another guanylate cyclase-C agonist, is an efficacious second-line drug for IBS with constipation in secondary care. Diarrhoea is a common side effect and is no less likely than with linaclotide or tenapanor. Although the drug is licensed for IBS with constipation in the USA, it is not yet available for this indication in many countries (recommendation: strong, quality of evidence: high).

• Tenapanor, a sodium-hydrogen exchange inhibitor, is an efficacious second-line drug for IBS with constipation in secondary care. Again, diarrhoea is a frequent side effect. Although the drug is licensed for IBS with constipation in the USA, it is not yet available for this indication in many countries (recommendation: strong, quality of evidence: high).

• Tegaserod, a 5-Hydroxytryptamine 4 receptor agonist, is an efficacious second-line drug for IBS with constipation in secondary care but is unavailable outside the USA. Diarrhoea is a common side effect (recommendation: strong, quality of evidence: moderate).

Psychological therapies

A recent network meta-analysis of RCTs of psychological therapies for IBS demonstrated that several psychological therapies were more efficacious than control interventions.156 However, the most compelling evidence, based on the number of trials and long-term outcomes was for CBT and gut-directed hypnotherapy,156 both of which are recommended by the NICE guideline when symptoms have not improved after 12 months of drug treatment.10

The principles of CBT are based on the five systems model, which suggests that cognitions (thoughts), behaviours (actions), emotions and physiology all interact within the context of the broader environment or social system. By altering any of these systems, others can potentially be modified (eg, changing one’s thoughts can alter one’s emotions, as well as one’s physiological responses). Although there is a core set of defined therapeutic techniques employed in all variants of CBT, the underlying formulations of the therapy differ, depending on the primary outcome. For instance, if reducing depression is the primary outcome the therapy focuses on increasing pleasurable activities and challenging alternative negative thoughts about the self. If anxiety is the outcome, the therapy focuses on reducing avoidance of threatening situations and threat-related thought patterns.

The network meta-analysis of psychological interventions for IBS, which included 15 trials of CBT in 1844 patients, concluded that CBT delivered in several formats was more effective than a control, including education and support, treatment as usual, and a waiting list control (online supplemental figure 20).156 Face-face CBT (10 RCTs, 930 patients, RR 0.62; 95% CI 0.48 to 0.80), self-administered or minimal contact CBT (4 trials, 434 patients, RR 0.61; 95% CI 0.45 to 0.83), therapist-delivered CBT over the telephone (1 RCT, 373 patients, RR 0.50; 95% CI 0.29 to 0.84) and group CBT (2 trials, 50 patients, RR 0.41; 95% CI 0.19 to 0.91) were all superior to a waiting list control.156 There was substantial heterogeneity in some of the estimates, which may, in part, be explained by differences in trial design, sample size and whether patients with refractory IBS were included. Analysis of trials that only included patients with refractory symptoms reduced the heterogeneity, and still demonstrated efficacy for CBT in some formats (online supplemental figure 21).

Other sources of heterogeneity may include hours of therapy time. For example, within the face-to-face CBT groups therapist time ranged from 5 to 12 hours. Of the two web-based interventions, one had eight online interactive sessions with 2.5 hours of telephone therapist support, and the other five online sessions with email support. There were also differences in the skill level of therapists across trials. Most were experienced CBT therapists, but some trials used doctoral-level students, and one RCT trained nurses to deliver CBT. The CBT protocols themselves varied. Some made IBS-specific modifications to existing mental health protocols, and others were based on stress management related to IBS.

The 2 most recent, and largest, RCTs used CBT developed specifically for IBS.86 243 In both, this included education concerning the role of stress in IBS, stress management techniques, cognitive techniques to identify and challenge both unhelpful thoughts associated with IBS and core beliefs around perfectionism, and relapse prevention. The ACTIB trial also included a description of the pathophysiology of key symptoms, and how CBT may work through the gut-brain axis.86 Other sections focused on altering IBS-specific safety and avoidance behaviours (eg, not going out until bowels are empty, or a toilet location is known) and managing negative emotions. The IBSOS trial included problem-solving training focused on coping with IBS stressors.243

The rationale for IBS-specific CBT is further supported by a review of the psychological mechanisms of CBT for IBS.30 Key mechanisms related to reduction in IBS symptom severity appear to be changes in IBS-specific cognitions and gastrointestinal -specific anxiety, rather than changes in general anxiety. With respect to this issue, it is worth noting that in the largest trial of CBT for IBS conducted to date at least 50% of patients met cut-offs for probable common mental disorders at baseline, and both therapist-delivered CBT over the telephone and web-based CBT using IBS-specific protocols reduced anxiety and depression scores at all follow-up points.86 These data suggest that treatment with IBS-specific CBT protocols may benefit both mental health and gastrointestinal symptoms.

The network meta-analysis suggested that therapist-delivered CBT over the telephone had a larger effect on IBS symptoms at follow-up than web-based CBT.156 However, in the ACTIB trial, health economic analysis suggested web-based CBT was the more cost-effective option.244 In this trial, the web-based intervention used the same protocol as the therapist-delivered CBT over the telephone,244 but the eight sessions were delivered on an interactive, tailored, website. Patients worked through this on their own at home, with guided telephone support from the therapist. The network meta-analysis also demonstrated that therapist-delivered CBT over the telephone, web-based CBT, face-to-face CBT and self-administered or minimal contact CBT were all superior to treatment as usual after 12 months of follow-up (online supplemental figure 22).156 It is likely that these approaches have similar efficacy, but more work is needed to determine cost-effectiveness of the various modes of delivery. One advantage of web-based therapies is that they are easy to standardise at scale and monitor usage. However, the disadvantage is that they tend to have lower adherence.86 244

Evidence suggests, therefore, that CBT for IBS is effective in both high-intensity (therapist-delivered) and minimal contact (therapist-guided) formats, as well as self-administered with either bibliographic material or web-based. There is also some evidence for group CBT, but more trials are needed to confirm this. As the ACTIB trial suggested a bigger treatment effect with therapist-delivered treatment, but that this was less likely to be cost-effective,244 a stepped care approach may provide greatest benefit, where patients with more complex needs receive face-to-face CBT, and those with milder symptoms are offered web-based or other guided, supported versions.

Despite an evidence base for use,156 many psychological therapies are not widely available, despite being recommended in the NICE guideline for patients with ongoing symptoms after 12 months of drug treatment.10 However, with the success of the ACTIB and IBSOS trials,86 243 244 there have been improved training opportunities for therapists, and therapist manuals made freely available to Improving Access to Psychological Therapy (IAPT) services in the UK, on completion of a specified training programme. Telephone and web-based delivery of CBT also has the potential to further increase access.86 The IAPT service has, therefore, increased its remit to include CBT for IBS, and patients can be referred via general practitioners or can self-refer. More work is needed to enhance and standardise the training programme providers to ensure IAPT expertise in this area. Therapists without specific IBS training tend to default to using mental health treatments, which can disengage patients with IBS. Therefore, referrals should specify that this is for IBS-specific CBT.

Gut-directed hypnotherapy is one of the psychological therapies for IBS with the largest evidence base for both short-term and long-term efficacy in RCTs.156 The aims of this are to induce a deep state of relaxation in order to teach the patient new skills for self-management and control of their gut function.245 The treatment is delivered using IBS-specific protocols,245 246 which incorporate combinations of a variety of techniques including imagery, metaphors, tactile approaches to alleviate pain and diaphragmatic breathing specifically targeting abdominal bloating and distension. One of the strengths of the treatment is that the content can be tailored according to the patient’s symptom profile. Although the exact mechanisms of its effects in IBS remain uncertain, hypnotherapy modulates the gut-brain axis, with several studies demonstrating positive changes in gut-brain function before, and immediately after, hypnotherapy, including modulation of postprandial gastro-colic reflex activity,247 altered colonic motility,248 reduced visceral hypersensitivity249 and normalisation of gut-brain pain processing signals on functional brain imaging.250 251

Traditionally, hypnotherapy for IBS has been delivered via between 6 and 12 face-to-face weekly sessions of individualised treatment with a trained therapist. This approach has been shown to be efficacious; a meta-analysis of 6 RCTs, recruiting 639 patients, reported a RR of remaining symptomatic of 0.73 (95% CI 0.55 to 0.97) compared with education and/or support and 0.67 (95% CI 0.49 to 0.91) compared with a waiting list control (online supplemental figure 20).156 Moreover, in the largest clinical series to date, including 1000 patients, >75% of patients achieved a clinical response to hypnotherapy, defined as a ≥50-point reduction in IBS symptom severity score. There were also significant improvements in extraintestinal symptoms, and anxiety and depression scores.252 Hypnotherapy has previously only been recommended for patients with IBS when symptoms are refractory to conventional treatments.10 Indeed, a meta-analysis of RCTs has shown that gut-directed hypnotherapy is one of the few treatments that performs better than a control for patients with refractory symptoms (online supplemental figure 21).156 However, its clinical efficacy has also been demonstrated in non-refractory populations,156 and clinical outcomes in children and adolescents with IBS suggest that use of gut-directed hypnotherapy at an earlier stage of the condition may be beneficial.253

One of the barriers to wider scale provision of gut-directed hypnotherapy, and its current restriction to refractory cases, may be the cost of its delivery, including time intensity, and the requirement for a trained therapist. However, intervention with gut-focused hypnotherapy has been shown to have wider socioeconomic benefits including improving general well-being, reductions in healthcare utilisation in both primary and secondary care,254–256 reduced presenteeism at work,255 improved quality of life252 257 and long-term beneficial effects on symptoms,256 making it a potentially cost-effective option. Patients with IBS in tertiary care with severe functional limitations may require individualised hypnotherapy, with the content of sessions customised to their symptom profiles. However, patients in primary or secondary care may benefit from accessing a more 'generic' form of group-delivered hypnotherapy. In a large, multicentre, RCT in patients with IBS in primary or secondary care, group hypnotherapy was shown to be non-inferior to individual hypnotherapy.258 Group hypnotherapy may therefore have a role in primary and secondary care settings, with the potential advantage that this approach could reduce delivery costs and improve access. Early reports of clinical outcomes via video-consultation are also promising, with similar response rates achieved, compared with face-to-face treatment.259

The offer of psychological therapies should not be limited to patients with psychological comorbidities. IBS symptoms are inherently distressing so there are often symptoms of anxiety and depression in IBS, but these are not necessarily at case level for a psychiatric diagnosis. Nevertheless, it may be worthwhile screening for evidence of both in the clinic, using a simple questionnaire, such as the General Anxiety Disorder assessment-7 and the Patient Health Questionnaire-9, and discussing referral to a mental health service or psychiatrist if mood is felt to be the key issue. The majority of the trials of psychological therapies conducted to date include a wide range of patients with IBS, many of whom would also not have met the threshold for a psychiatric disorder. In contrast to psychological treatments used for psychiatric disorders, which focus on mood, the IBS-specific therapies discussed here focus on brain-gut symptom-specific treatment mechanisms. The primary aim of treatment is to reduce severity and impact of abdominal pain and to help regulate bowel habit. They should, therefore, be viewed as behavioural methods for managing and treating IBS symptoms, rather than as psychotherapies.

Recommendations

• IBS-specific cognitive behavioural therapy may be an efficacious treatment for global symptoms in IBS (recommendation: strong, quality of evidence: low).

• Gut-directed hypnotherapy may be an efficacious treatment for global symptoms in IBS (recommendation: strong, quality of evidence: low).

• Psychological therapies should be considered when symptoms have not improved after 12 months of drug treatment. Referral can be made at an earlier stage, if accessible locally, and based on patient preference (recommendation: strong, quality of evidence: low).

Approach to the patient with severe or refractory symptoms

Severe IBS lacks a precise consensus definition, but is considered to be a biopsychosocial composite of patient-reported gastrointestinal and extraintestinal symptoms, degree of disability, illness-related perceptions and behaviours,260 insufficient response to conventional treatments261 and high healthcare utilisation.262 Refractory IBS is a related, but distinct term, again with no consensus definition, which is taken to mean patients whose symptoms have not improved with interventions, some of whom may also have severe symptoms. Validated severity scoring systems to assess impact and severity of IBS symptoms include the IBS severity scoring system,263 the gastrointestinal symptom rating scale-IBS264 and the functional bowel disorder severity index.265 In one European study, approximately one in four individuals with IBS were categorised as severe in a general population setting.266

Although the risk of missing, or subsequently developing, an organic disorder in patients diagnosed with IBS is low, this rate may be increased in those with severe symptoms,267 and should prompt a review of the diagnosis, with consideration of further targeted investigation. Nevertheless, it is important to stress that in most patients a diagnosis of IBS is secure,111 and further repeated investigations have a low yield.268 Severe IBS should also be distinguished from other severe functional gastrointestinal disorders that may have overlapping symptom presentations, including the narcotic bowel syndrome, if the patient is taking long-term opioids, centrally mediated abdominal pain syndrome269 and small intestinal dysmotility.270 Referral to a multidisciplinary chronic pain team to aid pain management and help with opioid reduction should be considered if abdominal pain becomes centrally-mediated or if narcotic bowel syndrome develops.

A large and diverse range of dietary, microbial, traumatic, interpersonal, genetic, psychological, physiological, psychiatric and functional comorbidity factors have been described in patients with severe IBS.75 189 271–276 Consequently, it is unlikely that a single targeted intervention will be transformative and a multidimensional, multisystem and integrated multidisciplinary team approach is usually required. Although there is a limited evidence base to guide management of this group of patients, as most RCTs do not differentiate response to treatment according to baseline symptom severity, there is recent trial data reporting superior outcomes with an integrated approach involving gastroenterologists, dietitians and clinical psychologists, rather than a gastroenterologist alone.277 There is a danger that, in this vacuum, ‘alternative’ therapies with the least evidence for their efficacy and safety are recommended by physicians or other practitioners, or are sought out by patients.278

Patients with IBS with severe symptoms are more willing to accept significant medication risks, for example, a mean 1% chance of sudden death in return for a 99% chance of cure of their symptoms with a hypothetical medication.279 Furthermore, patients are at increased risk of iatrogenic harms, through unnecessary surgery, including hysterectomy, appendicectomy and cholecystectomy,121 and inappropriate prescribing of opioids.90 The first principle of care for this vulnerable population of patients should therefore be primum non-nocere—first do no harm. This includes harm from unregulated and unproven approaches, especially if incentivised by financial or reputational gain.

Interventions with reported efficacy for patients with undifferentiated IBS specifically classified as severe or refractory include CBT, integrative group therapy, gut-directed hypnotherapy, gut-brain neuromodulators or psychodynamic interpersonal therapy.86 243 252 280–282 For severe or refractory IBS-C, surgical management, consisting of a potentially reversible temporary loop ileostomy, with a view to colectomy if stoma function results in improved, rather than worsened, quality of life for the appropriately screened patient,139 the ileal bile acid transporter inhibitor elobixibat,283 or linaclotide,284 all have some evidential support. For severe or refractory IBS-D, alosetron has regulatory approval in the USA for women.285 In the UK, a reasonable alternative might be ondansetron although, to date, this has not yet been confirmed to be efficacious in severe or refractory IBS. For severe or refractory abdominal pain one study reported efficacy with intramuscular hyoscine.286

Other IBS guidelines suggest the use of combination gut-brain neuromodulators, termed augmentation, for more severe symptoms.226 Evidence from a large cohort of patients with severe chronic continuous abdominal pain showed that combinations of neuropathic analgesics (eg, duloxetine plus gabapentin) were more efficacious than monotherapy.287 Vigilance for the development of the serotonin syndrome for some combinations, especially those involving both SSRIs and SNRIs, is required. Symptoms include fever, hyperreflexia, tremor, sweating and diarrhoea. For patients with symptoms that are refractory to these pharmacological therapies, and those who have comorbid conditions or psychological symptoms, a combination of a gut-brain neuromodulator and psychological therapy may be more efficacious than monotherapy with either, drawing parallels with evidence from the depression and chronic headache literature.288 289

Recommendations

• Severe or refractory IBS symptoms should prompt a review of the diagnosis, with consideration of further targeted investigation (recommendation: weak, evidence: very low).

• Severe or refractory IBS should be managed with an integrated multidisciplinary approach (recommendation: weak, evidence: very low).

• Iatrogenic harms due to opioid prescribing, unnecessary surgery and unproven unregulated diagnostic or therapeutic approaches incentivised by financial or reputational gain should be avoided (recommendation: strong, evidence: very low).

• Use of combination gut-brain neuromodulators, termed augmentation, may be considered for more severe symptoms, with vigilance for risks of serotonin syndrome (recommendation: weak, evidence: very low).

Drugs in development

The highly selective 5-HT₄ agonist minesapride has been studied in 2 phase II dose-ranging RCTs in patients with IBS-C.179 180 A dose of 40 mg once a day was superior to placebo, in terms of improvements in number of bowel movements per week, abdominal pain and global symptoms. The drug was well-tolerated, with diarrhoea the most common side effect, and there were no cardiovascular adverse events. Histamine has a potential role in mediating visceral hypersensitivity, and in a small RCT in 45 patients the histamine₁ receptor antagonist ebastine led to significant improvements in both abdominal pain and global symptoms.290 A larger trial in 200 patients is ongoing (NCT01908465). Novel drugs that have been tested successfully in chronic idiopathic constipation, including elobixibat and mizagliflozin, a sodium-glucose cotransporter-1 inhibitor, are likely to undergo testing in IBS-C.283 291 Some secretagogues, including linaclotide, stimulate cyclic GMP production, which can attenuate visceral pain.292 An RCT of delayed-release linaclotide, with action confined to the ileo-caecal region of the gastrointestinal tract, demonstrated significant effects on abdominal pain, with lower rates of diarrhoea than convention-release linaclotide.165 Other novel approaches include drugs that act on cannabinoid receptors, which are expressed in the gastrointestinal tract and may also modulate pain expression. The cannabinoid type-2 receptor agonist, olorinab, has been tested in patients with quiescent Crohn’s disease, and led to reductions in abdominal pain and improved bowel movements293; a trial in IBS is underway (NCT04043455).

Other treatments in development

In recent years, there has been considerable interest in the evaluation of faecal microbiota transplantation (FMT) for IBS. Unfortunately, a meta-analysis of 5 RCTs, containing 267 patients, demonstrated no significant benefit of FMT compared with placebo (RR 0.98; 95% CI 0.58 to 1.66), and in 2 pooled trials placebo capsules administered orally were superior to capsules containing donor stool (RR 1.96; 95% CI 1.19 to 3.20).294 Criticisms of the trials, to date, have included small sample sizes, heterogeneity in IBS subtypes recruited, lack of standardisation of donor samples and suboptimal end points used. There is therefore a need for further, large, high-quality trials of FMT for IBS, perhaps targeting subgroups of patients with evidence of dysbiosis, who may be more likely to benefit. At present, therefore, there is insufficient evidence to recommend FMT for IBS outside of a research setting. Enterosgel, an intestinal adsorbent approved for use in IBS-D and available over-the-counter in the UK is currently the subject of a multicentre RCT in IBS-D.295 For IBS-C, there are ongoing trials of an exo-peristalsis device.296 Future research priorities are outlined below.

Recommendations

• Successful completion of large clinical trials will require pragmatic inclusion criteria, minimisation of the participant trial burden and effective recruitment strategies that reach into community settings. Virtual (remote access) trial approaches will reduce geographical exclusion.

• A priority-setting partnership would best discern valuable research questions.

• Some future research themes include, but are not limited to:

  • Characterisation of the illness to understand predictors (clinical, genetic, psychological and biological) of outcome and treatment response, determinants of refractory illness and burden of illness (particularly with respect to workplace productivity) by conducting large-scale epidemiological studies with extended observation.

  • Trials of novel treatments, including pharmacological, dietary and behavioural therapies, device-based treatments and faecal microbiota transplantation. There is also a need for development of visceral analgesics. Consideration should be given to stratifying randomised controlled trials by IBS severity and subtype, burden of extraintestinal symptoms and psychological comorbidity.

  • A better understanding of treatment combinations to uncover augmentation effects between therapies, and to assess the value of multidisciplinary approaches.

  • Modulation of pain and psychological responses using pharmacological (eg, serotonin norepinephrine reuptake inhibitors) or behavioural approaches (eg, cognitive behavioural therapy used earlier in the disease course or via digital provision), and comparison of cognitive behavioural therapy with gut-directed hypnotherapy.

  • Med-tech approaches (web-based, apps and devices) to behavioural modification.