British Society of Gastroenterology guidelines for the ...
Division ofGastroenterology, British Society of Gastroenterology guidelines forUniversity Hospital,Nottingham, UK the management of the irritable bowel syndromeJ JonesLiterature review anddocument preparation for the J Jones, J Boorman, P Cann, A Forbes, J Gomborone, K Heaton, P Hungin, D Kumar,working party was done byJ Jones. Members of the G Libby, R Spiller, N Read, D Silk, P Whorwellworking party are given belowand detailed in the appendix.Letchworth, HertsSG6 1DG, UKJ Boorman 1.0 Preface comprehensive literature search by Dr J Jones, 1.1 PURPOSE OF GUIDELINES specialist registrar in the Department ofCleveland GeneralHospital, These guidelines were compiled by a multidis- Gastroenterology, University HospitalMiddlesborough, ciplinary group at the request of the chairman Nottingham. This involved a review of personalCleveland, UK of the British Society of Gastroenterology’s and electronic databases including Medline,P Cann Clinical Services Committee. The prime tar- PubMed and Ovid using keywords such asSt Mark’s Hospital, gets for these guidelines are consultant gastro- “functional disease”, “dyspepsia”, “irritableNorthwick Park, UK enterologists, specialist registrars in training, bowel syndrome”, “spastic colon”, and “irrita-A Forbes and general practitioners. The purpose is to ble colon”. Further information was obtainedDigestive Diseases identify and inform the key decisions to be from references in quoted papers and byResearch Centre, St made in the management of patients thought to contacting relevant pharmaceutical companies,Bartholomew’s have functional diseases of the gut. As these and a total of 2521 relevant papers were iden-Hospital, London, UKJ Gomborone comprise the commonest conditions seen by tiﬁed. This preliminary document was modi- gastroenterologists, the working party repre- ﬁed after several reviews by members of theDepartment of sented a wide spectrum of practitioners in gas-Medicine, University of working party and submitted to the ClinicalBristol, Bristol, UK troenterology, including gastroenterologists Services Committee for independent review.K Heaton from both district general hospitals and tertiary Members of the British Society of Gastroenter- referral centres, as well as primary care practi- ology Council then further reviewed the docu-Centre for HealthStudies, University of tioners, psychiatrists, psychologists, and diet- ment. Comments from these reviewers andDurham, Durham, UK itians. representatives of the IBS patient group, theP Hungin IBS network, have been incorporated into theSt George’s Hospital, 1.2 SPECIFIC DIFFICULTIES ﬁnal versionTooting, London, UK Compared with producing guidelines for theD Kumar management of well deﬁned diseases such as 1.4 CATEGORIES OF EVIDENCEDepartment of peptic ulcer where there is a clear disease The strength of evidence used in the formula-Gastroenterology and entity, an obvious end point, and highly tion of these guidelines was graded accordingPsychological eVective treatments, drawing up guidelines for to the following system, which has been used inMedicine, St functional gastroenterological disorders hasBartholomew’s previous British Society of GastroenterologyHospital, London, UK had many diYculties. Clinical trials have been (BSG) guidelines. However, in the context ofG Libby diYcult to design as the conditions being functional diseases it should be recognised that treated are highly variable with many possible this tends to over value the contribution of ran-Division ofGastroenterology, end points, and most therapies only marginally domised, double blind, placebo controlledUniversity Hospital, more eVective than placebo. Early trials were trials at the expense of studies of psychologicalQueen’s Medical diYcult to evaluate because of inadequate treatments, which are diYcult or impossible toCentre, Nottingham, patient deﬁnition so that many questions have double blind.UK yet to be addressed with good quality ran-R Spiller Grade Ia: evidence obtained from meta- domised controlled clinical trials. Most of our analysis of randomised, double blind,Department of Human recommendations are therefore supported by placebo controlled trials.Physiology and clinical experience rather than randomisedNutrition, University of Grade Ib: evidence obtained from at least one controlled clinical trials. Finally, because func- randomised, double blind, placeboSheYeld, NorthernGeneral Hospital, tional diseases, although potentially debilitat- controlled trial.SheYeld, UK ing, are non-fatal there are few uniformly avail- Grade IIa: evidence obtained from at least oneN Read able audit measures such as mortality or well designed placebo controlled studyDepartment of survival times by which to judge or compare without randomisation.Gastroenterology and diVerent treatment regimens in diVerent areas Grade IIb: evidence obtained from at least oneNutrition, Central of clinical practice. other type of well designed quasi-Middlesex Hospital, experimental study.London, UK 1.3 PROCESS OF GUIDELINE CREATIOND Silk Grade III: evidence obtained from well de- The co-chairmen were approached by the signed non-experimental descriptiveDepartment of chairman of the British Society of Gastroenter-Medicine, Withington studies such as comparative studies,Hospital, University ology’s Clinical Services Committee and in- correlation studies, and case studies.Hospital of South vited to form a working party. Members wereManchester, UK chosen to be broadly representative of clini-P Whorwell cians and academics with a long term interest Abbreviations used in this paper: BSG, BritishCorrespondence to: and publication record in the ﬁeld of functional Society of Gastroenterology; IBS, irritable bowelDr R Spiller, Division of syndrome; GI, gastrointestinal; FGD, functionalGastroenterology, C Floor, bowel disease. A preliminary document was gastrointestinal disorders; FD, functional dyspepsia;South Block, University produced and subsequently modiﬁed during FBC, full blood count; ESR, erythrocyteHospital, Nottingham several meetings of the working party. TheNG7 2UH, UK. Email: sedimentation rate; CBT, cognitive behaviouralRobin.Spiller@nottingham.ac.uk initial document was further developed after a therapy; ACC, anterior cingulate cortex.
Grade IV: evidence obtained from expert 2.0 Summary committee reports or opinions, clinical 2.1 OVERVIEW experiences, or respected authorities. Functional gastrointestinal disorders (FGD) are the result of disordered GI function in the absence of known pathology of structure. FGD1.5 GRADING OF RECOMMENDATIONS are among the commonest medical conditions;The strength of each recommendation depends functional dyspepsia (FD) and irritable bowelon the category of the evidence supporting it, syndrome (IBS) account for 40–60% ofand is graded according to the following referrals to gastroenterology outpatient clinics.system:Grade A: requires at least one randomised 2.2 FOCUS OF THIS REPORT controlled trial as part of the body of The speciﬁc recommendations which follow literature of overall good quality and refer to IBS but because of extensive overlap, consistency (evidence categories Ia, much of the general recommendations also Ib). apply to other functional disorders includingGrade B: requires the availability of clinical non-ulcer dyspepsia and non-cardiac chest studies without randomisation (evi- pain. However, to avoid lack of focus in this dence categories IIa, IIb, III). report, speciﬁc recommendations for these lat-Grade C: requires evidence from expert com- ter conditions will be addressed in separate mittee reports or opinions, or clinical guidelines. experience of respected authorities, in the absence of directly applicable clini- cal studies of good quality (evidence 2.3 DIFFERENCES BETWEEN PRIMARY AND SECONDARY CARE category IV). Most published studies of IBS are from academic units describing referred patients1.6 SCHEDULED REVIEW OF GUIDELINES who diVer signiﬁcantly from those seen in gen-It is proposed that these guidelines be pre- eral practice, being less likely to accept asented on the BSG world wide web page and be psychological explanation of their symptomsavailable for comment. They should be re- and more convinced they have organic disease.viewed at 2–3 year intervals taking into accountfeedback from both public and profession, as 2.4 EPIDEMIOLOGYwell as new scientiﬁc evidence. Comments on IBS is common, aVecting 9–12% of the popu-these guidelines should be sent to Dr R C lation with a female/male ratio ranging fromSpiller or Dr A Forbes. 1.1 to 2.6 depending on the weight given to individual symptoms. Age and race have no1.7 SUMMARY OF AUDIT GOALS consistent eVect on incidence of symptoms.Audit of the management of non-fatal condi-tions requires assessment of somewhat subjec- 2.5 AETIOLOGYtive quality of life parameters rather than the Psychological morbidity. Most cases seen in gen-familiar morbidity and mortality statistics. Ide- eral practice do not have major psychologicalally patients with functional gastrointestinal morbidity. However, those who progress todisease would have a diagnosis established with outpatients have a higher incidence of psycho-the minimum of investigations without missing logical symptoms and psychiatric disease.signiﬁcant alternative diagnoses. They would Role of stress. Studies of hospital outpatientsthen enter a treatment programme with high suggest that approximately 50% attribute theeYcacy which reduced the need for further onset of their symptoms to a stressful event,consultations and procedures. Speciﬁc audit and one third report sexual and/or physicalgoals might include: abuse both in childhood and subsequent adultx Achieving an acceptably low proportion of life. missed non-functional diagnoses during one Consulting behaviour. Approximately half those year of follow up, which would be lower for suVering from symptoms consult a doctor. Those who do consult report more severe serious diagnoses such as cancer (<1%) symptoms and an increased level of psychologi- while for less serious diagnoses such as cal disturbance (anxiety, depression as well lactose intolerance <10% might be accept- sleep disturbance) compared with those who able. do not.x Ensuring a minimum number of patients Abnormal illness behaviour. Patients with IBS aged <45 years undergo negative barium have an increased incidence of multiple so- studies during diagnostic work up. matic complaints and frequent consultationsx Reducing the number of work days missed for minor illnesses. Patients with IBS are over through ill health after functional diagnosis represented in gynaecology and surgical out- made compared with before diagnosis. patients and are more likely to undergox Reducing the frequency of physician visits inappropriate surgery. for both gastrointestinal (GI) and non-GI Gut motility. There is no consistent evidence of related complaints after diagnosis and treat- abnormal motility. ment. Visceral hypersensitivity. Patients with FGDx Improving quality of life after consultation, exhibit evidence of altered CNS processing of investigation, and management. visceral pain.
Postinfective bowel dysfunction. A total of 10– 2.8 WHEN TO REFER 20% of patients relate onset of symptoms to an If symptoms are atypical, the history short, or acute gastrointestinal illness. the patient over 45, it is usually appropriate to Diet. True food allergy is rare but many perform further investigations, often via hospi- patients believe that food intolerances cause tal referral. symptoms. These beliefs may have either a rational or an emotional basis. The commonest 2.9 INVESTIGATION AFTER REFERRAL intolerances reported in the UK are wheat, fol- Sigmoidoscopy. Those referred to hospital will lowed by dairy products, coVee, potatos, corn, usually require a sigmoidoscopy if there are and onions. Lactose intolerance is found in colonic symptoms. Any abnormality noted 10% of IBS patients but lactose exclusion should be biopsied, as should all patients with rarely cures IBS. diarrhoea to detect unsuspected microscopic colitis. Speciﬁc further investigation. Thyroid function, 2.6 CLINICAL FEATURES antiendomysial antibodies, stool microscopy, Gastroenterological. These include recurrent and a urinary screen for laxatives will reveal a abdominal pain associated with disturbed limited number of abnormalities (1–2% for bowel habit. Various symptomatic criteria have each test). If such tests are done, they are best been deﬁned (see table 1) for clinical trial pur- performed on the ﬁrst visit, avoiding repetitive, poses but do not match the symptoms of all anxiety provoking serial testing. patients. Other criteria such as disturbed Lactose tolerance testing. This reveals lactose defecation are supportive but not essential. malabsorption in 8–25% of cases depending on Non-gastroenterological. Lethargy, poor sleep, the racial composition of the population but is ﬁbromyalgia, backache, urinary frequency, and only indicated if the patient consumes substan- dyspareunia are more frequent in IBS and sup- tial amounts (>0.5 pint/280 ml) of milk per portive of the diagnosis. Anxiety, depression, day. and somatisation are frequent but do not Colonic imaging. Patients with a family history reliably discriminate between IBS and other GI of colon cancer or who are older than 45 years diseases. at symptom onset should be considered for either a barium enema or colonoscopy if they have colonic symptoms. 2.7 DIAGNOSIS Working diagnosis. This can usually be safely 2.10 PROGNOSIS made in general practice on the basis of typical Once a functional diagnosis is established the symptoms, a normal physical examination, and incidence of new non-functional diagnoses is absence of sinister features (weight loss, rectal extremely low. bleeding, nocturnal symptoms, or anaemia). This diagnosis should be conﬁrmed in general 2.11 MANAGEMENT AND TREATMENT practice by observation over time. 2.11.1 Explanation Supportive features. The diagnosis is more likely Positive diagnosis and reassurance. Most patients if the patient is female, aged <45 with a history will be managed in general practice. The main- >2 years, and has attended frequently in the stay of management should be a positive diag- past with non-gastrointestinal symptoms. nosis with an explanation of symptoms and their possible causes, in language the patientTable 1 Criteria for diagnosing irritable bowel syndrome can understand with reassurance of a benignManning criteria7 prognosis. (1) Abdominal pain relieved by defecation Listening to the patient’s concerns. It is important (2) Looser stools with onset of pain to ask the patient what their fears and beliefs (3) More frequent stools with onset of pain (4) Abdominal distension are, simply listening may help reduce anxiety. (5) Passage of mucus in stools Lifestyle advice. Identifying food fads or deﬁ- (6) Sensation of incomplete evacuation ciencies, including excess or lack of dietary Factor analysis shows the ﬁrst three symptoms correlate well but are not related to (4), (5) and (6).8 ﬁbre, lack of exercise, and not allowingRome I criteria9 adequate and suitable time for regular defeca- At least three months of recurrent symptoms of: tion is particularly important at the ﬁrst (1) Abdominal pain or discomfort relieved with defecation, or associated with a change in stool frequency, or associated with a change in stool consistency and consultation in primary care. Most patients (2) Two or more of the following on at least 25% of occasions or days: referred to hospital will have already tried and Altered stool frequency failed with such measures. Altered stool form Altered stool passage Placebo response. This is usually substantial Passage of mucus (50%) and gives a false impression of the Bloating or distension It should be recognised that these criteria were drawn up with the support of the eYcacy of any treatment initially, although thispharmaceutical industry to allow greater comparability between studies of drug eVects. They are wears oV in the following months.a consensus and should not become a straitjacket to prevent scientiﬁc enquiry. Many patientswith abdominal pain and disturbed bowel habit do not exactly ﬁt these criteria, yet their clinical 2.11.2 Dietary manipulationcourse is similar. The Rome criteria have recently been revised as follows.Rome II criteria10 Diet advice. Self-imposed dietary restrictions to 12 weeks or more in the last 12 months of abdominal discomfort or pain that has two of the avoid pain or diarrhoea are common but mayfollowing three features: be inappropriate. True allergy is rare but intol- (1) Relieved by defecation (2) Associated with a change in frequency of stool erance of poorly absorbed carbohydrate, espe- (3) Associated with a change in consistency of stool cially lactose and fructose, is well recognised. The second group of criteria included in Rome I are now considered supportive rather thanmandatory in the diagnosis. Excessive caVeine containing beverages may be responsible for some symptoms.
Exclusion diets. Performed under supervision of discomfort and, particularly in hospital pa-an enthusiastic dietitian these may be helpful to tients, a range of non-speciﬁc symptoms sucha limited number of patients. However, not all as lethargy, anxiety, disturbed sexual function,oVending food items so identiﬁed prove to and disordered sleep. The majority of patientscause symptoms under double blind testing. have some features of psychological morbidity,This suggests that some of the beneﬁt lies in the particularly mood disorder.reassurance and sense of control such regimens As there are no speciﬁc disease markers,provide. FGD have been categorised according to the likely site of the principal disorder. This ranges2.11.3 Psychological therapies from the oesophagus in functional dysphagia,Identify psychological disorders. This involves a the upper gastrointestinal tract in FD, thecareful history of psychological features, in- colon in IBS, to the anorectum in proctalgiacluding disorders of mood and sleep and any fugax and obstructed defecation. Attemptsassociation of thoughts or feelings with symp- have been made to further subdivide these dis-toms. orders into ulcer-like, dysmotility-like, orRelaxation therapy. This may help those whose reﬂux-like in functional dyspepsia (FD) andsymptoms appear to be “stress related”. into diarrhoea predominant or constipationBiofeedback, hypnotherapy, cognitive behavioural predominant in the case of IBS. These distinc-therapy, and psychotherapy. These may all be tions reﬂect the poor understanding of func-used depending on the main features. Those tional disorders rather than evidence of diVer-without marked psychiatric abnormalities do ent pathological processes. In reality there isbest. frequent symptom overlap and poor sitePsychiatric referral. If a careful history reveals speciﬁc correlation with functional investiga-signiﬁcant psychiatric disease this should be tions, such as manometry, balloon distensiontreated on its own merit. Bowel symptoms may threshold,1 2 intestinal transit, and gastric emp-well remit with successful therapy. tying studies. The same patients may report symptoms typical of both IBS and the various2.11.4 Pharmacological treatments types of FD with variable prominence of theDrug treatments. These have a substantial short diVerent symptoms over time.3 It followsterm response rate, most of which is due to a therefore that the overall approach to thesenon-speciﬁc placebo component. Speciﬁc ben- conditions should be similar.eﬁt is seen in only a small proportion ofpatients. 3.1 SOCIAL IMPACTAbdominal pain. Antispasmodics may help, Despite the benign nature of these disorders,with those with an anticholinergic eVect many functional symptoms such as vomiting,appearing to be most eVective. Alternatively, choking, bloating, faecal urgency, inconti-antidepressant therapy can be given, the nence, diarrhoea, ﬂatulence, and borborygmieYcacy of tricyclics being supported by large can restrict social activities and substantiallyclinical trials reduce quality of life. Chronic food related painUrgency and diarrhoea. This responds well to may lead to refusal of social invitations, whileloperamide or codeine. fears about the need for frequent defecationConstipation. Usually responds to an increase in may substantially restrict travel and work. Overdietary ﬁbre. Some patients appear to be 40% of patients report avoidance of somespeciﬁcally intolerant of wheat bran but activities including work, travelling, socialising,ispaghula is often better tolerated. sexual intercourse, domestic and leisure pur-Other drugs. Although commonly used, most suits, and eating certain foods as a consequencehave not been shown to have a greater eVect of their symptoms.4 Average work days lost inthan placebo. the USA per year by patients with FGDs were 14.8 compared with 8.7 in the asymptomatic2.12 FUTURE RESEARCH NEEDS population.5 It is this reduction in their qualityCompared with other ﬁelds, the evidence base of life, rather than individual symptoms, whichis weak and much of the evidence quoted here most determines how patients rate the severityis at the level of clinical consensus only. Much of their functional bowel diseases.6more research is needed into these common There is often a complex relationshipconditions before we can give conﬁdent between symptoms (table 1) and restrictedanswers to many important clinical questions. social activities, with some patients in eVect hiding (usually subconsciously) behind these3.0 Spectrum of functional symptoms to avoid situations they ﬁnd diY-gastroenterological disorders cult. Patients may experience anxiety andFunctional gastroenterological disorders disturbed sleep, with associated lethargy and an(FGD) are deﬁned by symptoms in the absence “inability to get on with their lives”, such thatof known structural pathology. They have no in the worse cases the condition comes tospeciﬁc disease marker and their symptoms dominate their existence. DiYculty in conﬁrm-overlap with those of other diseases. Experi- ing the diagnosis may lead to further worry andenced clinicians often diagnose these disorders doubt, with numerous visits to doctors andon symptoms alone but as functional disorders repeated unpleasant tests. A further burden,are so much more common than organic especially in women, is the risk of unnecessarydiseases, any diagnostic strategy is likely to have surgery such as cholecystectomy or hysterec-a deceptively high positive predictive value. tomy, which may aggravate the existing disor-Typical symptoms include abdominal pain or der, as well as adding their own speciﬁc
Table 2 Prevalence (%) of irritable bowel syndrome by Most studies have used either the Rome Isex and number of Manning criteria15 criteria or three Manning criteria (which No of symptoms produce closely comparable diagnostic rates). This results in a prevalence of 1.5–12.1% for 1 2 3 4 5 6 men and 5.2–19.1% for women (table 4).Male 27 10.7 5.0 2.3 1.3 0.9Female 46.8 24.0 13.1 6.0 2.9 1.4 4.3 ETHNIC DIFFERENCES Ethnic diVerences have been found in a fewTable 3 Prevalence (%) of irritable bowel syndrome by studies that have made direct comparisons. IBSsymptom frequency and type of pain14 appears to be more common in Japan than Holland (25% v 9%)16 and in Whites compared Symptoms Symptoms >once a <once a with Hispanics in the USA (21.8% v 16.9%)17 month month but similar in US Whites and Blacks.8 One study of students in Nigeria showed a particu-Male 11 54Female 20 66 lar high prevalence of symptoms (48% inMale+pain relieved by defecation 5 46 women and 24% in men using two ManningFemale+pain relieved by defecation 7 55 criteria) but this may have been due to the high incidence of gastrointestinal infections in thispostoperative complications such as scar pain, population.18 By contrast, subjects from ruraladhesions, and surgery related changes in Thailand appear to have a much lower risk.19bowel habit. Cultural factors including diet and socioeco- nomic status are important; thus in the3.2 FOCUS ON THE IBS AS A MODEL FOR FGD bicultural city El Paso on the US/Mexican bor-These guidelines concentrate on IBS as this der, US Whites are more likely to report symp-symptom complex is the commonest and best toms than Hispanics, but after controlling forstudied of the FGD. However, the principles of socioeconomic and dietary diVerences thisinvestigation and management, particularly of ethnic diVerence was no longer signiﬁcant.20its psychological features, are applicable to allfunctional disorders. Epidemiology and possi-ble aetiology are discussed in some detail as the 5.0 Aetiology 5.1 OVERVIEWmost important part of management is expla-nation, reassurance, and dealing with the asso- It is highly likely that within the group ofciated psychological problems. patients with functional bowel disease there are as yet unrecognised infectious and other4.0 Epidemiology organic causes of bowel disturbance. We4.1 SEX AND AGE should not therefore expect all patients to showIBS symptoms are about twice as common in similar features or predisposing factors. How-women as men (tables 2–4). The variability in ever, for many patients the two most consist-the sex ratio (1.1–2.6) may depend on the ent, and probably interrelated, characteristicsweight given to various symptoms as all studies are psychological morbidity and visceral hyper-agree that straining and passage of hard stools sensitivity. A substantial minority may relateare commoner in women while frequent and the onset of symptoms to an acute gastro-loose stools are commoner in men.11 Although intestinal illness, while a further minoritythe frequency of those reporting abdominal report that speciﬁc dietary components pre-pain together with two or more Manning crite- cipitate symptoms.ria declined with age over 45 in moststudies,5 12–14 the inﬂuence of age appears small 5.2 PSYCHOLOGICAL MORBIDITYand was not seen in a recent large UK study Most cases seen in general practice do not havewhich included over 1800 subjects.15 Hence major psychological morbidity. Those whoadvancing age should certainly be no bar to the progress to outpatients have a higher incidencediagnosis of IBS although the increasing of psychological symptoms and psychiatric dis-incidence of other diseases with similar symp- ease, the most ﬂoridly abnormal being found intoms argues for greater caution in making the long term follow up in those attendingdiagnosis in the elderly. academic departments.24 Compared with healthy controls, these IBS patients have higher4.2 SEVERITY scores for anxiety, hostile feelings, sadness,As the number or frequency of symptoms depression, interpersonal sensitivity as well asrequired for making the diagnosis increases, more sleep disturbance.25–28 However, part ofthe calculated prevalence falls (table 2, 3). this is due to the fact that more anxiousTable 4 Prevalence (%) of irritable bowel syndrome in the USA using three Manning or the Rome criteria Total (%) Men (%) Women (%)Reference No Group characteristics n Diagnostic criteria IBS IBS IBS21 US White 835 3 Manning 12.8 12.1 13.615 UK White urban 1896 3 IBS symptoms 9.5 5.0 13.08 US students 1344 3 Manning 15.5 9.6 18.312 US elderly 328 3 Manning 10.9 NA NA22 US health examinees 1264 3 Manning 3.6 1.5 5.223 US students 789 3 Manning 11.7 NA NA5 US 95% White 5430 Rome 9.4 7.7 14.5
patients are more likely to seek a second opin- or increased rectal sensitivity relate so poorly toion as IBS suVerers who do not consult any symptoms but relate better to psychologicaldoctor are not psychologically diVerent from features.45controls.29 Compared with outpatients with Patients’ complaints of ill health may reﬂectorganic gastroenterological disease, there are their experience of others with similar symp-no consistent diVerences and psychiatric fea- toms or previous experience of the secondarytures cannot be used reliably to distinguish gain associated with being ill themselves. Stud-functional from organic disease.30 31 ies of children with chronic abdominal pain Several studies have investigated the preva- have found an association with poor health andlence of adverse life events and in particular emotional disorders in their parents.46 47 Peoplesexual abuse in these patients. More than 50% who recalled being given gifts or special foodslinked the onset of their symptoms to a stress- when they were unwell as a child were moreful event such as employment diYculties, fam- likely to exhibit chronic illness behaviour andily death, a surgical procedure, or marital more likely to have IBS as adults.43stress, and a similar proportion reportedconcurrent social problems relating to work,ﬁnances, housing, and personal relationships.4 5.4 EFFECT OF MOOD ON GI FUNCTIONA history of sexual abuse, often combined with Most people have, at some time or other, exper-physical abuse, both in childhood and subse- ienced the eVect of anxiety on gut function,quent adult life has been reported in 20–30% including cramps and diarrhoea. Animal stud-of patients with IBS, signiﬁcantly more com- ies have shown that stress inhibits small bowelmonly than in the general population (<10%) transit while accelerating colonic transit andor in patients with organic disease (14%).32 33 causing increased stool frequency.48 DepressedAdverse life events prior to the onset of IBS are patients have delayed small bowel and wholeas common as in patients with deliberate gut transit, with a correlation between transitself-poisoning and signiﬁcantly more common time and severity of depression, while anxiety isthan in patients with organic gastrointestinal associated with accelerated small bowel tran-disease.34 Signiﬁcantly, the psychiatric illness or sit.49 Acute stress is diYcult to model in anadverse life event preceded the onset of the ethical experiment but in healthy volunteersbowel disorder in two thirds of patients.28 acute stress disrupts normal fasting motor pat- terns50 and accelerates small bowel transit.51 It5.3 ABNORMAL ILLNESS BEHAVIOUR also stimulates the colon of both normalNot all patients with symptoms consistent with subjects and IBS patients, although untilIBS consult a doctor; consultation rates vary recently it has been diYcult to demonstrate anyfrom 10% to 50% depending on age and consistent diVerence between the twosex.13 15 35 As expected, patients with more groups.52–54 Over the past decade evidence hassymptoms and more severe pain were morelikely to consult,15 as were those with more psy- accumulated showing that the cathartic eVectchological symptoms.36 37 This relationship of severe stress in rats is mediated largelybetween psychosocial disorders and frequent through release of corticotrophin releasingattendance at outpatient clinics is true for factor.55–57 This has made it possible to mimicmany disorders and is not unique to IBS.30 As the eVect of severe stress on the human colonexpected, those who reach outpatients are by using an infusion of corticotrophin releasingmore likely to believe that their illness is not factor which increases descending colon motil-stress related and are therefore more fearful of ity indices and induces abdominal pain. Whenorganic disease.38 this was done, the IBS patients’ colonic Patients with FGD also consult their doctors responses were greater and they experiencedmore often for non-GI complaints than pa- more pain than normal subjects,58 an interest-tients without FGD.5 Non-gastroenterological ing ﬁnding which needs conﬁrming.features such as lethargy, poor sleep, ﬁbromyal- Stress has not been convincingly shown togia, backache, urinary frequency, and dys- alter perceptual thresholds to balloon disten-pareunia are more frequent in IBS and sion59 but relaxation and hypnosis can raise thesupportive of the diagnosis. Patients with IBS threshold for discomfort, while hyperventila-and lower abdominal pain are over represented tion has been shown to lower discomfortin gynaecology and surgical outpatients39 but thresholds.60are less likely to have recognisable pathology40and more likely to undergo surgery.41 42 Studies 5.5 ABNORMAL AUTONOMIC REACTIVITY IN IBSof this abnormal illness behaviour have found arecord of multiple somatic complaints and Altered autonomic reactivity has been noted ininappropriate consultations for minor ill- IBS, with decreased vagal tone associated withnesses.43 There were signiﬁcantly higher abnor- constipation61 62 and increased sympatheticmal illness behaviour scores in IBS patients activity associated with diarrhoea.63 Thesecompared with those with organic disease or observations provide a mechanism wherebypatients consulting speciﬁcally for depression.27 psychological abnormalities could be trans-This negative interpretation of innocent sensa- lated into diVerences in transit. The potentialtions is consistent with the observation that role of autonomic dysfunction in IBS is madepatients with IBS were biased towards remem- more plausible by the report from the Mayobering terms with negative connotations.44 This Clinic of eight patients with acute autonomicmay well explain why objective measures of neuropathies who presented with apparentlydisturbed function such as rapid colonic transit typical IBS symptoms.64
5.6 EVIDENCE OF ABNORMAL GUT MOTILITY IN IBS ity.86 These studies suggest that IBS patientsEarly studies suggested abnormal electrical describe gut stimuli as unpleasant or painful atcontrol activity in the colon65 but this was not lower intensity levels compared with normals, asubstantiated by later workers.66 67 As with phenomenon which is likely to originatenon-IBS constipated patients, constipation centrally rather than peripherally.predominant IBS patients have been reported The central processing of visceral aVerentsto have decreased high amplitude propagated has been assessed using positron emissioncolonic contractions.68 However, sigmoid con- tomography (PET scanning) and more re-tractility is increased in some patients69 and this cently functional magnetic resonance imagingmay cause increased resistance to caudal ﬂow. to measure the resulting regional cerebralExaggerated response to emotion has also been blood ﬂow. Most of the relevant studies have asreported70 but this diVerence from controls was yet only been presented in abstract form. Thenot consistent,71 perhaps due in part to the dif- one published study compared the eVects ofﬁculty in inducing strong emotions reliably actual and sham distension of the rectum inwhile remaining within boundaries set by ethi- healthy volunteers and IBS patients on cerebralcal constraints (see preceding section). Inad- blood ﬂow. Perception of pain during bothequate means of scoring and assessing colonic actual and simulated delivery of painful stimulipressure proﬁles may also contribute to the in healthy subjects was associated with activa-diYculty in showing consistent diVerences as tion of the anterior cingulate cortex (ACC)transit studies have generally shown fast and whereas no ACC response was seen with non-slow transit in diarrhoea and constipation pre- painful stimuli. IBS patients in this studydominant IBS, respectively.72 However, it showed no consistent activation during eithershould be noted that in spite of fast transit, painful or non-painful distension but demon-most stool weights in IBS patients lie within the strated signiﬁcant activation of a diVerentnormal range73 74 even in those with diarrhoea region, the left prefrontal cortex, when antici-as their main complaint.75 76 pating painful stimulation due to sham disten- Interest in possible small bowel abnormali- sion.87 This and other evidence suggests thatties were stimulated by initial reports that normally the ACC has an important role indiscrete clustered contractions were commoner mediating the aVective qualities of visceralin IBS and associated with symptoms.77 78 pain, both intestinal and cardiac, and that thisHowever, these have not been conﬁrmed by response is abnormal in IBS and other painfulothers79 although the later study examined only functional disorders such as ﬁbromyalgia. Sub-fasting activity when symptoms are less fre- sequent studies have produced conﬂicting dataquent. so plainly this area is still evolving. Abnormal Recently there has been an increasing central processing may provide a mechanismemphasis on altered sensation as the basis for which could explain the association betweensymptoms as it has become clearer that symp- IBS and mood, psychological stressors as welltoms correlate poorly with objective measures as disease beliefs and expectations.such as stool weight and transit.5.7 VISCERAL HYPERSENSITIVITY 5.8 POSTINFECTIVE BOWEL DYSFUNCTIONPatients with FGD exhibit decreased pain A subgroup of IBS patients report that theirthresholds to balloon distension of the gut. symptoms began after an acute gastrointestinalThis was ﬁrst described in the rectum of illness, a group which appeared to have apatients with IBS 25 years ago80 and subse- slightly better prognosis in two retrospectivequently conﬁrmed by others81 and is often analyses.88 89 Persistent bowel dysfunction wasnoted with air insuZation during colonos- noted in 25% of patients following docu-copy.82 Similarly, patients with FD have a lower mented Campylobacter, Shigella, and Salmo-threshold to balloon distension of the stom- nella90 91 gastroenteritis. Two separate studiesach.83 This visceral hypersensitivity is not site reported that 38% and 29% of patients withspeciﬁc and has been demonstrated in the enteritis developed IBS.45 91 Factors predispos-oesophagus of patients with IBS as well as in ing to persisting symptoms included a physi-the rectum of patients with FD.84 These cally more severe acute illness90 as well aschanges are speciﬁc to gut stimulation as greater anxiety and adverse life event scores insomatic pain thresholds to extreme cold or the six months leading up to the acute illness.45transcutaneous electrical stimulation are either Increased sensitivity to rectal distension wasnormal or even increased in some studies.81 84 85 also reported after the infectious illness.45This was thought to indicate an abnormality of While macroscopically normal, microscopicmucosal sensitivity in the gut but as studies abnormalities are detectable in rectal biopsieshave become more sophisticated to try to using special stains, the signiﬁcance of which iseliminate external inﬂuences on patient per- under investigation. This phenomenon is notception, this opinion has changed. IBS patients unique to IBS, cystitis being another exampleare much more likely to show an increased of a disease in which inﬂammation appears tosensitivity when the rectum is distended in a increase visceral sensitivity.92 The conclusionspredictable sequence of increasing volumes of these studies have recently been supportedthan when it is distended with volumes chosen by a prospective study of over 584 000 patientsin a random iterative method. This indicates a in whom it was shown that when a range ofresponse bias, which may be related to a demographic details were examined, a bout ofpatient’s apprehension of pain rather than any culture positive bacterial gastroenteritisperipheral and objective increase in sensitiv- emerged as the strongest predictor of new
onset IBS, with a relative risk of 11.9 (95% CI The validity of these studies of food intoler-6.7–21).93 ance is hard to evaluate as a placebo response cannot be excluded unless a double blind food challenge is performed. Such studies, which5.9 DIET involve blind challenge with blended foodsPatients often relate their functional symptoms passed down nasogastric tubes, bypass theto certain foods and some have considerably important social, psychological, and physicalrestricted their diet by the time they consult. aspects of eating, which are likely to be at leastThe patient’s beliefs may have either a rational as important as the direct eVects of individualor emotional basis. The evidence that the gut is food constituents on the gut. They are reallyin some way sensitive to particular foods is lim- only valid in identifying food allergy as theited. Food is chemically highly complex and relatively small amounts of material instilledresponse to food exclusion is poorly reproduc- are not enough to elicit symptoms in cases ofible, leading desperate patients to more and food intolerance. An early study using nasogas-more restricted and illogical diets. Such tric delivery of suspected food reported six ofcircumstances make patients easy prey to 25 consecutive IBS patients who correctlyunscrupulous practitioners and there are many identiﬁed food triggers and showed an increase“fringe” practitioners beneﬁting from the con- in prostaglandins in rectal dialysate.95 However,fusion. a subsequent study of 13 patients who had Studies that have used dietary restriction identiﬁed a food intolerance by means of anfollowed by sequential introduction of single exclusion diet found a high placebo responsefoods have reported speciﬁc food intolerance in with only three patients showing a signiﬁcant33–66% of IBS patients.94 95 The commonest ability to identify food triggers when adminis-intolerance reported in the UK is to wheat, fol- tered double blind.99lowed by dairy products, especially cheese, True food allergy is much less common andyoghurt and milk, coVee, potatos, corn, onions, usually not diYcult to recognise if foodbeef, oats, and white wine.94 Ingestion of ingestion is associated with urticaria, asthma,osmotically active, poorly absorbed fermenta- eczema, angioedema, and rhinorrhoea with able carbohydrates such as lactulose is known to high incidence (70%) of positive skin prick orcause typical IBS symptoms such as bloating, high RAST scores.100 Such patients usually seecramps, and diarrhoea. A recent study showed an immunologist rather than a gastroenterolo-increased colonic hydrogen production in IBS. gist and are not usually thought to have IBS.An exclusion diet was reported to reduce both When symptoms were purely gastroenterologi-symptoms and gas production in response to a cal, only 15 of 88 who believed they were aller-standard dose of lactulose implying that the gic actually had their perceptions conﬁrmed bydiet modiﬁed the fermentation capabilities of double blind trial.101 It is worth noting thatcolonic bacteria.96 The indirect nature of this those who respond immediately to food inges-response to diet may explain why the clinical tion are more likely to have positive skin testsbeneﬁt varies as the bacterial ﬂora is itself so than those who report symptoms which comevariable. on some hours after food ingestion.102 Positive Adult acquired hypolactasia is common in skin prick testing for common food antigensthe UK, with an incidence of 10% in those of has been reported in up to a third of patientsNorthern European descent, rising to 60% in with IBS and these patients have been reportedAsians, and 90% in Chinese patients. Regional to respond better to elimination diet and type IdiVerences in dairy intake may account for the hypersensitivity inhibitors such as sodium cro-variable beneﬁt reported with lactose free diets moglycate.103 These results need conﬁrmingin IBS. Thus in Denmark, with a traditionally before deﬁnitive conclusions can be drawn.104 Ithigh intake of dairy products, a low lactose diet should be recognised that in only a minority ofhas been reported to produce improvement in cases is the patient’s beliefs conﬁrmed objec-13 of 20 Danish adults with symptoms of IBS tively, so some of their response must beand objective evidence of lactose malabsorp- psychologically determined. Recent studiestion.97 However, only subjects ingesting a sub- showing that mast cell degranulation can bestantial amount of lactose (equivalent to more psychologically triggered105 together with evi-than 0.5 pint of milk per day) can expect to dence that food allergy patients degranulatebeneﬁt from lactose restriction as lower jejunal mast cells in response to cold stress106amounts do not cause symptoms, even in open the way for a possible explanationlactose malabsorbers.98 whereby stress or patient’s beliefs about food An initial study using elimination diets (that might trigger a gastrointestinal response.is, diets that eliminate all but a single fruit,meat, vegetable, etc) improved symptoms in67% of those who completed the study. Morepracticable exclusion diets, which make less 6.0 Diagnosisdemands on the patient, have been developed Functional gastroenterological disorders arewhich only exclude foods that had commonly common but only about half actually consultbeen implicated in food intolerance from the their general practitioner107 and of these onlyearlier studies (for example, wheat, milk, about one in ﬁve are referred to a hospital con-coVee, potatos, corn, onion, beef, oats, cheese, sultant in any given year. Most assessment andand white wine). Such studies had a lower suc- management is therefore carried out in generalcess rate (48.2–50%) with similar compliance practice. Unfortunately, most published evi-rates. dence relates to patients referred to hospital.
6.1 DIAGNOSIS AND MANAGEMENT IN GENERAL 6.2 WHEN TO REFER PRACTICE Patients presenting for the ﬁrst time in later life A careful and detailed history, often accrued and those with atypical symptoms normally over several short interviews and sometimes warrant hospital referral. However, some pa- over many months or even years, is required. tients in whom the general practitioner has This will take account of psychological factors, made a conﬁdent diagnosis of FGD develop past family and personal history, as well as the further symptoms or worsening anxiety, often social circumstances of the patient, which the related to adverse life events such as bereave- ment or separation.28 These patients also general practitioner is uniquely placed to warrant referral to help exclude an alternative assess. The patient aged less than 45 years who diagnosis and to provide more deﬁnitive describes typical symptoms (ﬁg 1) without sin- reassurance than the primary care physician ister features, such as weight loss, rectal bleed- can oVer. The general practitioner is well ing or symptoms responsible for night time placed to understand the illness in the wider waking, probably has FGD. This should be context of the patient’s life, and it is important supported by a normal physical examination that this information is transmitted frankly and including, where relevant, rectal examination comprehensively to specialists when referral is and no evidence of anaemia. The diagnosis is needed. The skilled general practitioner will more likely if the patient is female, has a history recognise that these patients commonly have of greater than two years, and has attended fre- complaints relating to several systems, and will quently in the past with non-gastrointestinal avoid the fragmentation of care that so easily symptoms such as malaise and backache. If occurs if every new complaint results in referral symptoms are typical then no further investiga- to a diVerent specialist. tions are necessary to establish a working diag- nosis. However, if there are atypical features or 6.3 DIAGNOSIS IN HOSPITAL SETTING the history is short, it may be appropriate to Although functional disorders account for perform some of the screening tests referred to 36–50% of all outpatient consultations, the ﬁl- below. tering process means that the incidence of other diseases is higher than in general practice A typical history with or without negative and therefore further investigations are often test results should lead to a ﬁrm diagnosis with indicated. detailed explanation and reassurance, prefer- Symptom criteria such as those devised by ably without new medication, followed by fur- Manning diVerentiate IBS reasonably well ther review if symptoms continue. Simple from normal subjects or patients with peptic pharmacological or dietary interventions may ulcer or reﬂux7 31 but do not reliably distinguish be appropriate for some patients at that time. IBS from inﬂammatory bowel disease.108–110 Most symptoms will resolve, or remain un- Thus symptoms alone cannot be relied upon changed but acceptable, and need no further but must be augmented by physical examina- attention. tion, demographic data such as age and sex, together with the progression of symptoms Symptoms over time, all of which strongly inﬂuence the a priori probability of the diVerent diseases from < 45 pain ± altered > 45 pain ± altered which IBS must be distinguished. The most Age bowel habit bowel habit important diagnosis not to be missed is cancer Longstanding Onset Recent onset and as age and family history are the main risk Fluctuating Natural history Progressive factors, these will have a strong inﬂuence in Not present Sinister features Present deciding who to investigate. Present Other functional symptoms Absent A careful dietary and drug history is vital to identify unusual dietary habits or new medica- tions whose use may have preceded the High Increased development of symptoms. Particular attention probability risk of organic should be given to low/excess intake of dietary of IBS disease ﬁbre, or excess of poorly absorbed sugars such as fructose or sorbitol or stimulants such as coVee or tea. Similarly, a wide range of drugs Investigation can cause bowel disturbance with diarrhoea General for organic and/or abdominal discomfort, such as angio- practice Hospital management disease based tensin inhibitors, blockers, antibiotics, management chemotherapeutic agents, proton pump inhibi- Progression/ FBC, ESR, TFT tors, or NSAIDs, while constipation may be endomysial Ab change related to opiate analgesics, calcium channel in symptoms Management Sigmoidoscopy/colonoscopy blockers, or antidepressants with anticholiner- of presumed ± biopsy ± barium enema gic eVects, to mention just a few. Although it is IBS common to ﬁnd a positive family history of IBS Negative Simple diarrhoea/malabsorption this is not unexpected in so common a investigations screen condition and no study has shown this to be of Other imaging studies any diagnostic help. Lactose challenge/ tolerance test Patients with conventional IBS symptoms Laxative screen such as those described by Manning, or thoseFigure 1 Stages in the evaluation of the irritable bowel syndrome (IBS). that fulﬁl the Rome I criteria who have no
alarm symptoms and no abnormal ﬁndings on sial antibodies, but there is no publishedphysical examination in the hospital setting evidence as to their yield, which is likely to behave a 52–74% chance of having IBS.31 111 low (1–2%). However, it should be borne inNon-gastroenterological features such as leth- mind that cheap tests with a low yield may yetargy, poor sleep, ﬁbromyalgia, backache, fre- be cost eVective.quency and urgency of micturition, nocturia, Patients with high stool weight (>200 gincomplete bladder emptying, an unpleasant daily) should have a laxative screen, which intaste in the mouth, early satiety, and dyspareu- some series is positive in about 15–26% of suchnia are all commoner in IBS than controls and cases.121 122supportive of the diagnosis.112–114 Many authorshave drawn attention to the striking disparity 6.5 IMAGINGbetween the proclaimed severity of symptoms Colonic cancer is not reliably excluded110 byand patient’s desperation with their otherwise history, and patients with a positive family his-healthy appearance. Phrases like “ symptoms tory or who are older than 45 years at symptomruling my life”, “ desperation”, “you must do onset (when the incidence of sporadic colonsomething” will strike a cord with many expe- cancer begins to rise steeply) should be consid-rienced practitioners. ered for either a barium enema or colonoscopy. Although as already indicated, abnormal Ultrasound rarely detects a relevant alternativelevels of anxiety, depression, and somatisation diagnosis in patients with suspected IBS and isare features of many patients who are referred not recommended as it uncovers coincidentalto hospital, these features do not discriminate asymptomatic abnormalities such as gall stonesbetween IBS and other GI diseases.30 Inquiring and ﬁbroids in 8%.123 This may easily lead toabout these emotionally disturbing features is inappropriate surgery with no beneﬁt to symp-usually inappropriate at the ﬁrst visit but may toms. Small bowel Crohn’s disease in its earlybe worth exploring when initial tests are unre- stages is easily confused with IBS and bariumwarding. follow through should be considered for Although speciﬁcity of diagnosis after a patients with worsening symptoms or suspicionhistory and physical examination has been of an abdominal mass, particularly if there isreported to be improved to over 95% by using anaemia or elevation of ESR or C reactive pro-a scoring system that includes full blood count tein. However, it should be remembered that(FBC) and erythrocyte sedimentation rate this examination exposes the ovaries to appre-(ESR),115 others have not found such good dis- ciable radiation and it should be used sparinglycrimination.116 Sigmoidoscopy, which can be in young females.done at the ﬁrst clinic visit, should excludethose with ulcerative colitis or rectal cancer. If 6.6 FUNCTIONAL TESTSthe rectum appears macroscopically normal, Various measures of gut function, includingroutine rectal biopsy does not usually add any- bile acid absorption (SeHCAT seven daything.117 However, in those with diarrhoea as a retention) and gut transit have been shown tomajor complaint, it should be performed as it be abnormal in functional diarrhoea but aremay provide evidence of microscopic colitis not widely used in typical IBS. One studywhich may alter management signiﬁcantly.118 reported that ﬁve of 42 patients with functional diarrhoea retained <8% SeHCAT and re-6.4 FURTHER INVESTIGATIONS sponded to cholestyramine.124 More recently,How many further investigations beyond a patients with unexplained diarrhoea and stoolsimple blood count are performed depends on weights >200 g were shown to have reducedwhat is considered to be an acceptable level of bile acid retention.125 Ileal and colonic biopsiesmissed diagnoses. A “screen” including thyroid have yielded inconsistent results and at presentfunction, stool microscopy for ova, cysts, para- it seems likely that low retention in most casessites and fat globules, and ﬂexible sigmoidos- is non-speciﬁcally related to fast small bowelcopy with colonic biopsy, together with lactose transit.126 Rare isolated defects in bile salttolerance testing in a large (1452 patients) absorption have been described127 but areAmerican study of patients ﬁtting IBS criteria unlikely to be responsible for more than a verygave a yield of 6% thyroid abnormalities (3% few cases of IBS.hyperthyroid, 3% hypothyroid), occult inﬂam-matory bowel disease in 1%, and evidence of 6.7 PROGNOSISlactose malabsorption in 21–25%.119 Patient Once the diagnosis is established the incidencereports of lactose intolerance relate poorly to of new signiﬁcant diagnoses is extremely low.objective evidence of lactose malabsorption Harvey et al found no signiﬁcant new diagnosesand cannot be relied upon.120 It would therefore in 104 patients followed for ﬁve years, the diag-be logical to perform a breath hydrogen test for nosis being largely based on symptoms, as onlylactose malabsorption on IBS patients who are 12% of these had radiological studies.89 An-regular consumers of more than 0.5 pint (280 other study of 112 patients in which the major-ml) of milk or equivalent dairy products, espe- ity had extensive radiological studies reportedcially if they come from a racial group with a only two initial misdiagnoses of IBS (onehigh incidence of lactose malabsorption. Alter- chronic pancreatitis and one carcinoma of thenatively, the response to lactose exclusion may pancreas). Five years later one case of thyro-be helpful although the result is usually less toxicosis and one of gall stones had becomeclear cut than the breath hydrogen test. Other apparent, values probably no diVerent from thesimple screening tests which are logical include expected incidence of disease in initiallyESR, calcium and albumin, and antiendomy- healthy controls over a ﬁve year period. Thus
7.2 LISTENING TO THE PATIENT Recommendations Simply listening to the patient and accepting x Young patients (<45 years) with typical that their symptoms are real and valid may functional symptoms, no alarm symp- help, especially if previous consultations have toms or family history of colonic cancer, been unsatisfactory. It is important to ask the and a normal examination can be safely patient about their fears and beliefs. A high given a working diagnosis of IBS without proportion believe there is some serious further tests and their response to reas- disease, in particular cancer. The condition surance and lifestyle advice observed. needs to be explained simply using analogies (Recommendation grade B.) with which a layman can relate. Cramps and x Those referred to hospital with more spasms are easily accepted as causes of pain. severe symptoms usually require further Most can understand how anxiety, such as investigation including at least sigmoidos- before a test or examination, can cause copy, FBC, and ESR. (Recommendation diarrhoea. This can be used to introduce the grade B.) idea of brain-gut interactions. Explanation of x Patients with diarrhoea should be fully possible mechanisms such as “sensitive gut” or evaluated with non-invasive investigations reaction to infection, if put in simple terms, such as serum B12, red cell folate, reduces anxiety caused by unexplained symp- ferritin, thyroid function, antiendomysial toms and is usually highly valued by the antibodies, calcium, albumin, and micro- patient. At a minimum, this prevents further scopy of the stool together with a rectal unnecessary referrals and possibly hazardous biopsy and where appropriate barium fol- treatments, such as hysterectomy or cholecys- low through. Severe diarrhoea warrants a tectomy. Although accepted by many clini- full colonoscopy to exclude microscopic cians, these concepts have not been subjected colitis. (Recommendation grade C.) to proper randomised controlled trials. x Older patients with recent onset of symp- toms or younger subjects with a family 7.3 LIFESTYLE ADVICE history of colon cancer usually justify This will be much more important at ﬁrst pres- imaging of their colon. Progressive symp- entation in primary care than in hospital prac- toms in any age group should prompt tice, when most will already have tried and re-evaluation of the need for further failed such measures. This will include a care- imaging. (Recommendation grade C.) ful dietary and lifestyle history, identifying food fads or deﬁciencies, including excess or lack ofthe chance of remaining free of serious disease dietary ﬁbre. Lack of exercise and not allowingin IBS is excellent. adequate and suitable time for regular defeca- The prognosis for continuing abdominal tion are common problems which are espe-symptoms is however less good and depends on cially relevant to constipated IBS suVerers.the criteria used, with about 30% still sympto- Keeping a two week diary of symptoms,matic at ﬁve years in Harvey’s study89 but only stresses, and dietary intake may identify aggra-5% of patients completely symptom free in a vating factors and will be helpful in discussingDanish ﬁve year follow up study.14 Symptoms management. Those with constipation/vary both in severity and quality with time. diarrhoea need advice about intake of “ﬁbre”Thus a substantial proportion of individuals or poorly absorbed non-starch polysaccha-with IBS symptoms in the community experi- rides, fructose, sorbitol or lactose, which mayence loss of IBS symptoms over 12 months but be either increased or decreased with beneﬁt.may develop other functional symptoms such Intake of drugs and herbal medicines, whichas FD.3 Factors that have been shown to may aVect the bowels, should also be noted.worsen prognosis include more prominentpsychological symptoms88 and a longer history 7.4 PLACEBO RESPONSEof illness24 as well as previous abdominal Deﬁning the best treatment in IBS has beensurgery.128 diYcult, at least in part because the placebo response is so marked, averaging 47% in a recent survey of 25 randomised controlled7.0 Management and treatment drug trials. This eVect was approximately threeMost of this will be carried out in general prac- times larger than the additional drug eVect,tice (ﬁg 1). The mainstay is explanation and which was 16%.129–132 However, the longer thereassurance in terms that the patient can follow up the smaller the placebo eVectunderstand, together with sensible lifestyle becomes and as yet long term beneﬁt has onlyadjustments relating to diet, medications, and been shown for psychological and dietarystressors, which appear to precipitate symp- treatments. The high placebo response duringtoms. clinical trials may reﬂect the eVect of the greater contact between the patient and health7.1 POSITIVE DIAGNOSIS AND EXPLANATION care professionals. Compared with routineMaking a deﬁnite diagnosis helps both doctor outpatient clinics, much more time is availableand patient by reassuring them that it is for explanation, reassurance, and general dis-unlikely that another alternative diagnosis will cussion. The value of reassurance in IBS hasemerge over the ensuing years.89 However, this not been studied systematically but in FD thenot does make the symptoms disappear and the patients’ responses to reassurance that they dopatient may continue to need a supportive not have serious disease depends on psycho-understanding relationship with a physician. logical factors. Thus while patients with low or
moderate anxiety do well with the reassuranceprovided by negative endoscopy, the beneﬁt to Recommendationsthose with marked anxiety is short lived.133 x Simple dietary advice will beneﬁt someThese patients, whose quality of life remains patients with diarrhoea who have exces-poor despite reassurance and explanation and sively large intakes of indigestible carbo-in whom psychological features appear promi- hydrate, fruits, or caVeine. (Recommen-nent, may respond to more formal psychologi- dation grade C.)cal treatments. Several forms of therapy have x Constipated patients with low ﬁbre intakebeen studied in IBS but studies which have not should be given a trial of a high ﬁbre diet.used a suitable placebo are diYcult to interpret (Recommendation grade C.)as in the short term at least, any form of x Those with diarrhoea, whose intake ofincreased patient contact has a non-speciﬁc lactose is substantial (>0.5 pint (280 ml)beneﬁcial impact.134 milk/day) may beneﬁt from a trial of lac- tose exclusion and/or a lactose tolerance7.5 DIETARY FACTORS test. (Recommendation grade B.)Many patients believe that some dietary item is x Formal exclusion diets may be useful inresponsible for symptoms and some have controlling symptoms in some patients.adopted inappropriately restrictive diets. (Recommendation grade B.)Equally, some patients have excessively largeintakes of indigestible carbohydrate, fruits or bowel disorders, reducing autonomic arousal bycaVeine, and these patients may beneﬁt from relaxation reduces symptoms and induces asimple dietary advice. Others, particularly sense of well being, allowing the patient to feelpatients of non-European descent, may have more conﬁdent and in control. Patients arehypolactasia. Those with a substantial intake of taught to exclude sources of tension and relax.lactose (>0.5 pint (280 ml) milk/day) may Unfortunately, there are few controlled studiesbeneﬁt from a low lactose diet. Rarely, and most include very small numbers of patientsexcessive intake of fructose may cause symp- but Blanchard et al showed that 10 sessions oftoms due to slow or incomplete absorption progressive muscle relaxation over eight weekswhich could cause gut distension to which IBS reduced symptoms compared with a placebopatients appear especially sensitive.135 Bloating control of symptom monitoring visits alone.137 Ais an extremely common symptom in the further study showed that relaxation trainingnormal population, being reported as frequent reduced symptoms and, most importantly, theby 10–20%, with an excess in women.12 14 It number of medical consultations during a 40responds poorly to drugs but may respond to month follow up.138the dietary measures outlined above. Exclusion diets may beneﬁt some patients 7.8 BIOFEEDBACKbut are arduous and must be supervised by an This, in theory at least, depends on feedback ofenthusiastic dietitian. Treatment begins with a a measure of visceral function to show thedetailed diet history, followed by a strict exclu- patient a disturbance in physiology so that theysion diet supported by a food and symptom can learn to correct it. It has been used mostdiary, and telephone contact with the dietitian. commonly in the treatment of incontinenceThe exclusion diet omits a range of foods136 (for and constipation. Therapy aims to make theexample, dairy, citrus, and grains), including patient more sensitive to rectal sensation andany the patient believes to provoke symptoms avoid inappropriate straining. It also provides afor a period of two weeks. This is followed by detailed explanation of normal physiology andreintroduction of single foods to identify a re-education concerning the optimum def-which, if any, precipitate their symptoms. An ecatory patterns.139 The supportive relationshipindividualised diet can be produced for each with the therapist undoubtedly contributes topatient, avoiding foods to which they are intol- the overall positive eVect. Interestingly, thiserant. Two large studies using this approach report found no relation between the observedhave found long term remission in approxi- beneﬁt and any demonstrable physiologicalmately 50% of patients. There was however no defect. There are few randomised studies, butsymptom monitoring or attention placebo con- using a mixture of eight weeks of progressivetrols so the results may have been non- relaxation therapy, biofeedback, and copingspeciﬁc.94 136 strategies, two small (n=19 and n=21) trials showed a global improvement in symptoms7.6 PSYCHOLOGICAL THERAPIES compared with symptom monitoring con-These range in their depth and ambition from trols.140 141 The improvement in symptoms overa limited attempt to control symptoms with pretreatment levels was maintained at twosimple behaviour therapy, through hypnosis, to years but no placebo group was available forinsight oriented psychotherapy. It should be comparison at this time.142 Importantly, a sub-recognised that the availability of the more time sequent placebo controlled study of 60 patientsconsuming techniques is limited in the NHS showed no speciﬁc beneﬁt of relaxation,and their use should therefore be restricted to thermal biofeedback, and cognitive therapyonly the most diYcult cases. compared with an attention-placebo control (pseudo-meditation and EEG alpha suppres-7.7 RELAXATION THERAPY sion biofeedback). This strongly suggests thatThis is the simplest form of “psychotherapy” the beneﬁt was non-speciﬁc and due towhich can easily be taught to patients by audio- attention,134 a conclusion supported in a recenttapes. The logic is that if stress causes functional review of the literature.143
7.9 HYPNOTHERAPYHypnosis is used to induce a state of relaxation Recommendationsand to alter the underlying abnormalities of gut x Positive diagnosis, explanation of symp-motility and/or sensation in the presence of the toms, their cause, and their relationship totherapist, with the ultimate aim of enabling physical, dietary, or psychological factorspatients to control symptoms on their own. together with a supportive understandingSuccess depends very much on the enthusiasm relationship should be the mainstay ofof the therapist. An early controlled trial of management. (Recommendation gradehypnotherapy involving 30 patients with refrac- C.)tory IBS treated over a three month period x Those with anxiety but without psychiat-showed that patients treated with hypno- ric disease who do not respond satis-therapy improved signiﬁcantly compared with factorily to the above may beneﬁt froma group of patients receiving a similar contact relaxation therapy. (Recommendationtime spent discussing emotional problems and grade B.)stress.144 A follow up study conﬁrmed long x Those patients with prominent psychiat-term eYcacy and suggested that treatment was ric morbidity may respond to psycho-more likely to be successful with younger therapy or cognitive behavioural therapypatients and those without serious psychopa- or require conventional psychiatric treat-thology.145 A further study from another centre ment, while those with less psychopathol-had a lower improvement rate (61%) but found ogy may respond well to hypnotherapy.group therapy equally as eVective as individual (Recommendation grade B.)therapy.146 Unfortunately, hypnosis is time con-suming and expensive to provide but is costeVective in severe refractory cases. ties through the safe and contained relationship with the therapist. There is undoubtedly an7.10 COGNITIVE BEHAVIOURAL THERAPY important placebo (non-speciﬁc) eVect, whichCognitive behavioural therapy (CBT) is based the therapist seeks to exploit.on the assumption that IBS in some patients is There are only three reported trials ofa behavioural disease generated by responses to dynamic psychotherapy and relaxation in addi-life events. CBT involves helping the patient to tion to standard medical therapy. Theserecognise maladaptive patterns of thinking and showed long term improvement in about twobehaviour. It encourages them to change how thirds of patients, particularly in scores ofthey interpret bodily sensations and changes in abdominal pain and altered bowel habitvisceral function by seeing them, not so much compared with medical treatment alone.149–151as symptoms of disease which need to be The patients had either been symptomatic for atreated, but more as expressions of anxiety that year151 or suVered from chronic symptomsare associated with particular life events. Treat- unresponsive to conventional treatment.149 150ment is essentially an exercise in identiﬁcation Two publications probably referred to the sameand solving of problems which facilitates a patients.149 150 The treatment protocols in-greater sense of control and autonomy in the cluded long and frequent appointments (45patient. CBT insists that the patient takes some minutes to four hours) and there was noresponsibility for the illness and helps him/her attempt to control for this. On subgroup analy-ﬁnd a more healthy way of dealing with the sis, females and patients with overt psychiatricunderlying problem. While widely used in symptoms or pain precipitated by stress didother ﬁelds, there have been few controlled best.studies of eYcacy in IBS. Two small studies(n=20 and n=34) from the same group 7.12 PSYCHIATRYcompared cognitive therapy with untreated Not infrequently a careful history revealscontrols. Both studies used eight weeks of important psychiatric illnesses, particularlytherapy with symptom monitoring visits as pla- depression and anxiety, which may with timecebo controls. Both found a signiﬁcant reduc- come to overshadow the gut symptoms. Suchtion in abdominal pain, diarrhoea, constipa- illnesses need to be treated on their own meritstion, belching, and nausea for up to three and psychiatric referral is then appropriate.months. Sixty six per cent had a generalisedanxiety disorder and there was little response toplacebo.147 148 8.0 Pharmacological treatments7.11 DYNAMIC PSYCHOTHERAPY Current pharmacological treatments have lim-Analytical psychotherapy attempts to provide ited value. Speciﬁc beneﬁt is only seen in a lim-the patient with an insight into why particular ited proportion, and although the immediatesymptoms have developed and what they might placebo response is high, this wears oV withmean/represent in the light of changes in key time, causing repeated consultations. Drugsrelationships. There is an implicit assumption may be counterproductive in patients withthat this insight will cause long lasting changes major psychological problems as their prescrip-in attitude and behaviour. Symptoms often tion may reinforce abnormal illness behaviourseem to stem from signiﬁcant life changes and prevent patients dealing eVectively with(often a loss of a relationship) set against the underlying psychological problems. However,background of a fragile personality who has for those patients who require therapy for spe-diYculty coping with separation. The patient ciﬁc symptoms, the following treatments haveworks through his or her interpersonal diYcul- evidence to support them.