Transcription
PaperJ R Coll Physicians Edinb 2013; 43:15–23http://dx.doi.org/10.4997/JRCPE.2013.104 2013 Royal College of Physicians of EdinburghA Dobbin, 2J Dobbin, 3SC Ross, 4C Graham, 5MJ Ford1Honorary Fellow, School of Clinical Sciences and Community Health, University of Edinburgh; 2Research Analyst, Foundation for PositiveMental Health, Edinburgh; 3Clinical Hypnotherapist, Foundation for Positive Mental Health, Edinburgh; 4Statistician, Wellcome TrustClinical Research Facility, Western General Hospital, Edinburgh; 5Retired Consultant Gastroenterologist, University of Edinburgh1ABSTRACT Irritable bowel syndrome (IBS) is a common disorder associated withprofoundly impaired quality of life and emotional distress. The management ofrefractory IBS symptoms remains challenging and non-pharmacological therapeuticapproaches have been shown to be effective. We compared brief interventions withbiofeedback and hypnotherapy in women referred by their GP with refractory IBSsymptoms. Patients were randomised to one of two treatment groups, biofeedbackor hypnotherapy, delivered as three one-hour sessions over 12 weeks. Symptomassessments were undertaken using validated, self-administered questionnaires. Twoof the 128 consecutive IBS patients suitable for the study declined to consider nonpharmacological therapy and 29 patients did not attend beyond the first session. Ofthe 97 patients randomised into the study, 21 failed to attend the therapy session; 15of 76 patients who attended for therapy dropped out before week 12 post-therapy.The mean (SD) change in IBS symptom severity score 12 weeks post-treatment inthe biofeedback group was –116.8 (99.3) and in the hypnotherapy group –58.0(101.1), a statistically significant difference between groups (difference –58.8, 95%confidence interval [CI] for difference [–111.6, –6.1], p 0.029 ). In 61 patients withrefractory IBS, biofeedback and hypnotherapy were equally effective at improving IBSsymptom severity scores, total non-gastrointestinal symptom scores and anxiety anddepression ratings during 24 weeks follow-up. Biofeedback may prove to be themore cost-effective option as it requires less expertise.Correspondence to A Dobbin,Foundation for Positive MentalHealth, 24 Boswall Road,Edinburgh EH5 3RN, UKtel. 44 (0)131 552 th.comKeywords Irritable bowel syndrome, hypnotherapy, biofeedbackDeclaration of Interests Professor Ford’s institution received an endowmentfor this study. S Ross received a pay grant from the Leith Hospital DevelopmentFund for work undertaken during the preparation of this study.IntroductionIrritable bowel syndrome (IBS) is defined as abdominalpain or discomfort on three or more days each monthfor 12 weeks, relieved by defaecation and associatedwith a change in stool frequency or stool appearance.1,2Other symptoms include urgency of defaecation, a senseof incomplete rectal evacuation and abdominal bloating.It is common, affecting 17% of the population, andaccounts for 50% of gastroenterology outpatientworkload and 5% of GP consultations.3,4 It is associatedwith profoundly impaired quality of life and emotionaldistress.5,6 Non-gastrointestinal (GI) symptoms arecommon and fatigue and anxiety reduce quality of lifemore than GI symptoms.6–8 Patients with IBS cost twiceas much in healthcare expenditure than average and havethree times the likelihood of absenteeism from work.9,10Reducing symptom severity in IBS outpatients significantlyreduces healthcare costs.10The cause of IBS is unknown, although alteredgastrointestinal motility, visceral hypersensitivity, hypervigilance and psychological factors have all beenimplicated.2 There is strong evidence of links betweenIBS, emotional distress and adverse life events anddifficulties;11–13 62% of IBS patients have an anxietydisorder14 and anxiety and IBS have many biologicalfactors in common.14 Studies of pain sensitivity toballoon inflation in the colon and oesophagus haveshown increased pain perception with stress induction,suggesting central modulation of visceral pain perceptionspecifically by the effect of threat.15,16 During rectosigmoidal balloon distension, functional magneticresonance imaging (MRI) studies show greater activationof areas regulating affective and sensory processes,including the amygdala, insula, cingulate, and prefrontalcortex in IBS patients than in control subjects.17,18Hyperactivity of the amygdala may play an importantrole in the altered central processing of visceral15clinicalRandomised controlled trial of brief interventionwith biofeedback and hypnotherapy in patientswith refractory irritable bowel syndrome
clinicalA Dobbin, J Dobbin, SC Ross, C Graham, MJ Fordinformation common in IBS patients.18–20 Anxiety-likebehaviour and visceral hypersensitivity are associatedwith changes in corticotrophin releasing factor (CRF)messenger RNA (mRNA) expression in the amygdalawhich are inhibited by CRF receptor antagonists.18–20There is also an interaction between oestrogen andamygdala CRF sensitivity which may, in part, explain theincreased preponderance of IBS symptoms in women.21,22These factors suggest a possible mechanism for theco-morbidity of psychiatric disorders and IBS as well asother medically unexplained symptoms.A healthy gastrointestinal tract is regulated bysympathetic and parasympathetic pathways of the centralnervous system (CNS), mediated by their interactionwith the enteric nervous system (ENS). Alteredautonomic function has long been recognised in patientswith IBS and is thought to contribute to the changes invisceral sensitivity and gastrointestinal motility seen inresponse to stress.23–26 Cardio-vagal autonomic tone, asexpressed by heart rate variability (HRV) during deepbreathing, is attenuated in patients with IBS comparedwith matched healthy control subjects, reflecting anincrease in the ratio of parasympathetic to sympathetictone.23 There is also evidence of generalised autonomicdysfunction particularly in women with IBS.24 Whilecardiovascular autonomic tone does not necessarilyreflect changes in the autonomic control of the ENS,there is a close association between cardiac autonomicfunction and gastrointestinal function.25,26Pharmacological approaches to the management of IBShave proven disappointing; meta-analyses and doubleblind controlled trials do not achieve sustained andclinically useful improvements in either symptom controlor health-related quality of life.27 Non-pharmacologicaltherapeutic approaches have therefore been studied as away of improving symptoms.28,29 Cognitive behaviouraltherapy (CBT), hypnotherapy and biofeedback offertherapeutic promise in patients with refractory IBS. Gutfocused hypnotherapy has been shown to be the mosteffective and is now widely used throughout the UK.28,29Hypnotherapy has major physiological effects on theautonomic nervous system within the CNS and theENS.30–35 Hypnosis has been shown to affect HRV andperipheral skin conductance, increasing parasympatheticactivity and reducing sympathetic tone.30–37Biofeedback, CBT and relaxation training have all been welldescribed in the management of IBS and other functionalgastrointestinal disorders.38–47 Heart rate variabilitybiofeedback (breathing retraining while monitoring changesin HRV during deep breathing) has been shown to beeffective in reducing symptomatology and autonomicnervous dysfunction in hyperventilation and otherfunctional disorders.42–47 Given the evidence ofsympathovagal dysfunction in IBS patients,23 vagal tonemight diminish IBS symptomatology.47,48 Though there is no16direct evidence that resting vagal tone mediates changes inIBS symptoms or in work-related stress,48 studies appearto support a possible role for decreased vagal tone in theevolution and maintenance of IBS symptoms.47–49Meta-analyses of IBS treatment studies includinghypnotherapy have identified substantial methodologicalproblems with baseline measurements, inappropriatelyselected control groups and poorly documented orinadequate therapies for control subjects.50–52 Given theevidence of efficacy of hypnotherapy in IBS, we designeda trial to compare brief interventions of biofeedbackwith hypnotherapy. Our hypothesis was that briefintervention using HRV biofeedback would be at least aseffective as gut-directed hypnotherapy in achievingsignificant symptomatic improvement in patients withrefractory IBS.MethodsPopulationWomen aged 18–60 years referred by their generalpractitioner (GP) to outpatient gastroenterology clinicsin the Lothian University Hospital NHS Trust wererecruited. Irritable bowel syndrome was diagnosed by anexperienced consultant gastroenterologist, in accordancewith the Rome 111 criteria1,2 and after appropriateinvestigation. Exclusion criteria comprised a clinicalhistory of cardiovascular, neurological, renal or endocrinedisease, major psychiatric disorder or ingestion ofprescribed medications known to influence cardiacautonomic tone. Patients were asked to attend every twomonths to be seen by one consultant gastroenterologistthroughout the six month period of follow-up.TreatmentsAll patients were treated along conventional lines by oneclinic doctor; treatment included an explanation of thenature of IBS (verbal and written), simple advice andinformal counselling together with drug therapy ifrequired (loperamide and/or low-dose amitriptyline).Patients with refractory symptoms (no symptomaticimprovement eight weeks after the initial clinic visit)were given verbal and written details of the treatmentsunder study, and with the written agreement of patientsand their general medical practitioner, randomised toone of two treatment groups: biofeedback orhypnotherapy. Block randomisation with a block size ofeight was undertaken in order to achieve equal groupsizes. Patients were referred to one study therapist andseen individually for three one-hour therapy sessions inthe clinic during the following 12-week period.Biofeedback groupGiven the normal HRV observed during respiration, it ispossible to manipulate resting cardio-sympathetic andvagal parasympathetic activity by adjusting the breathingJ R Coll Physicians Edinb 2013; 43:15–23 2013 RCPE
RCT of brief intervention with biofeedback and hypnotherapy in patients with refractory IBSSession 1: Patients were taught to breathe using thediaphragm, placing one hand on the stomach and pushingthe hand out during inspiration. Using a transducer linkedto electrocardiogram (ECG) electrodes at the wrists, theECG was relayed to a laptop computer. Fast Fouriertransformation of the cardiac RR-interval data allowedthe HRV to be displayed as three columns, the highfrequency [0.15–0.4Hz], low frequency [0.05–0.15Hz]and very low frequency [0.005–0.05Hz] oscillations ofthe RR-interval reflecting levels of sympathetic andparasympathetic activity during respiration. Patients wereasked to breathe for two minutes at several specific rates,ranging between four and seven breaths per minute usingsoftware to provide both a visual and auditoryrepresentation of breathing. At the end of this firstsession the trace was analysed to establish the breathingrate which produced the maximal respiratory sinusarrhythmia (RSA) i.e. the resonant frequency. Patientswere asked to practice breathing at this rate for tenminutes, twice daily using a training CD-ROM whichhelped retrain their breathing to achieve optimal restingcardiac sympathetic and vagal tone.Session 2: repeated the exercises in session 1 to checkthe breathing rate for maximising RSA; in addition, anexternal stressor (maths test: say the seven times tablebackwards) was used to show the patient the responseto the stressor in real time and the recovery broughtabout by breathing at their resonant frequency. Thepatient was encouraged to focus on breathing at theiroptimal rate and asked to use this slow breathingtechnique in their everyday lives.Session 3: focused on observing how thinking aboutinternal stressors (a negative/worrying thought ormemory) influenced breathing and autonomic functionand how, when the individual breathed slowly and calmly,as they had been training themselves to do, the worryingthought was no longer (or much less) worrying.Hypnotherapy groupOur method derived from standard hypnotic techniques54but also incorporated gut-directed hypnotherapyimagery as described by Whorwell29 (personalcommunication).Session 1: introduced the patient to hypnosis, explainingthe process in order to demystify it. The 25-minutehypnotic session aimed to achieve a state of calmrelaxation of mind and body. Specific techniques includedinduction, deepening, ego-strengthening and visualisationsof a healthy gut. Patients were given a 20-minute audiorecording of a muscle relaxation technique to practicewith once a day.J R Coll Physicians Edinb 2013; 43:15–23 2013 RCPESession 2: consolidated the relaxation of session 1 andinvited individuals to explore a possible source of theirdiscomfort. Self-hypnosis techniques were introducedand patients were given a 20-minute audio recording topractice with at home once a day.Session 3: focused on developing self-managementthrough imagery-based techniques, such as metaphorwork and visualisations, to reduce discomfort and anxiety.Primary outcome measuresSymptom assessments were undertaken at eight weeksbefore the start of the programme (–8), at the start (0), andat 12 and 24 weeks, during clinic visits, using well-validated,self-administered questionnaires: the IBS symptom severityscore (IBS-SSS)55 comprising five visual analogue scores(VAS) of pain, distension, bowel dysfunction, quality of lifeand global well-being, and the Hospital Anxiety andDepression Score (HADS),56 in order to estimate theconfounding influence of psychological state. In addition,patients completed VAS of non-GI symptoms (othersymptoms – OS) common in IBS patients.7,8Statistical AnalysesControl data were obtained from the data recorded eightweeks before the programme and at the point of entryinto the study (weeks –8 and 0). Improvement in the IBSSSS was defined as a difference in the mean score of –30.The original study validating the IBS-SSS confirmed that amean change of 50 could be regarded as ‘reliableimprovement’, while a mean change of 83 was indicativeof clinically significant improvements classed as‘considerably better’.55 Our study was powered on aprimary outcome comparing changes in the IBS-SSS atweek 0 and 12 weeks post-treatment in the twotreatment arms. Data were subsequently comparedwithin and between the groups using two-sample pairedand unpaired Student’s t tests as appropriate, togetherwith standard multivariate tests. Using these data, it wascalculated that given 30 patients per group, differences inmean IBS-SSS of 29.4 (standard deviation [SD] 40) couldbe detected between the groups at the 5% level with 80%power. Further statistical analyses were undertaken usinga mixed model fitted using statistical analysis software(SAS) with IBS-SSS, HADS, HADS anxiety scores (HA),HADS depression scores (HD), and non-alimentary othersymptoms (OS) scores as outcomes (each modelledseparately with treatment and time point as fixed effectsand patients as a random effect).Ethical ApprovalAll patients and their general medical practitionersparticipating in this study were asked for their writtenconsent following receipt of written details of the studyoutline, content and purpose. This study was approved17clinicalrate, to identify the maximal amplitude of respiratorysinus arrhythmia (RSA) known as the resonant frequency,using methods described by Lehrer.49,53
clinicalA Dobbin, J Dobbin, SC Ross, C Graham, MJ Fordby the Lothian NHS Regional Ethics Committee (NREC#06/S1104/26) and the NHS Lothian Research andDevelopment Committee (R&D #2006/W/GI/05) andfunded by a local medical research and developmentNHS endowment fund.128 recruited2 declined at week –8126 entered into trialThis study was powered on a primary outcome ofcomparing the baseline at entry to 12 weeks posttreatment change in IBS-SS in the two treatment arms(Figure 3).The mean (SD) change in IBS-SSS from entry to12 weeks post-treatment in the biofeedback group was–116.8 (99.3) and in the hypnotherapy group –58.0(101.1), a statistically significant difference between groups(difference –58.8, 95% CI for difference [–111.6, –6.1],p 0.029). Statistical analysis of a mixed model fitted usingSAS revealed modest evidence of a difference over time;however when treatment was taken into account, therewas no evidence of a significant interaction between timeand treatment.29 DNA at week 097 randomised21 DNA for therapy76 attended treatment61 completed trialBiofeedbackn 3115 drop outs 7 allocatedbiofeedback 8 allocatedhypnotherapyHypnotherapyn 30DNA Did not attendfigure 1 Recruitment flow chart.ResultsRecruitment: Two of the 128 consecutive IBS patientssuitable for the study declined to consider nonpharmacological therapy and 29 patients did not attend(DNA) for the first clinic review (week 0). Of 97 patientsrandomised into the study, 21 failed to attend thetherapy session; 15 of 76 patients who attended fortherapy dropped out before completing therapy (sevenpatients randomised to receive biofeedback and eightpatients to hypnotherapy) (Figure 1). The mean age ofthe study participants was 40.4 years and did not differsignificantly between the two treatment groups.Data: The following series of plots show how individualpatient scores changed over time (Figures 2,4 and 5)in relation to the IBS-SSS, HADS and total OS. Time 0was recorded at eight weeks prior to entry, time 1 atentry, time 2 at week 12 post-treatment and time 3 atweek 24 post-treatment. In each plot there are two‘panels’: these show the two treatment groupsseparately, biofeedback on the left and hypnotherapyon the right. The different colours and symbolsindicate individual patients (these can be seen at es in IBS-SSS and HADS between week 0 and12 weeks post-treatment are plotted in Figure 6. Inorder to illustrate any effect that anxiety and depressionratings may have had, the groups are represented bydifferent symbols. Though there appears to be arelationship, from the distribution of the coloured dots,the effects are similar for both treatment groups.DiscussionThis study has shown that in 61 patients with refractoryIBS, the effects of hypnotherapy and biofeedback aresimilar and equally effective at improving symptoms. At12 weeks, both treatments significantly improved IBSSSS, total non-GI symptom scores and anxiety anddepression ratings. There was a modest differencebetween the two treatment groups (p 0.029), butrepeated measures analysis revealed a non-significanttrend in favour of biofeedback (p 0.079). In a previousstudy, mean changes in the IBS-SSS of 83 correlated withclinically significant improvements.55 There were smallbut significant changes in the HAD scores (Table 1);changes in psychological parameters have previouslybeen noted as a possible confounding factor but wecould not exclude the possibility that the effects oftreatment could at least in part be explained by thedirect effects of therapy on anxiety and depressionratings. This reduction itself could also help improvesymptomatology, particularly if psychological factors playa role in triggering or exacerbating symptoms.50,57 It canbe argued however, that the apparent beneficial effect ofthe two treatments represents a combination of specific,treatment-related effects and less specific, uncontrolledplacebo effects. Apart from hypnotherapy andbiofeedback, CBT is probably the most commontreatment used in patients with refractory IBS. TheCochrane Collaboration suggested that CBT andinterpersonal psychotherapy may be temporarilyeffective but it is unclear whether the effects aresustained in the longer-term.57J R Coll Physicians Edinb 2013; 43:15–23 2013 RCPE
RCT of brief intervention with biofeedback and hypnotherapy in patients with refractory IBSIBS-SSclinicalTimefigure 2A Irritable bowel syndrome symptom severity scores (IBS-SS).Type 3 tests of fixed effectsNum DF Den DFF valuePr FTreatment1602.760.1020Timepoint316929.73 .0001TreatmentXtimepoint31692.300.0791Num DF Numerator degrees of freedomDen DF Denominator degrees of freedomPr Probabilityfigure 2B Type 3 tests of fixed effects (irritable bowelsyndrome symptom severity scores).Trial design, selection bias and placebo effectsTrials comparing psychological and other therapies inIBS are not usually double-blind and the confoundingissues of patient and therapist expectancy, investigatorbias, doctor-patient relationship and rapport assumemajor importance.50–51, 57–60 A recent study of ‘warm’versus ‘cold’ therapists and dummy acupuncturetreatment in IBS identified large placebo effects.59 Thiseffect in IBS therapies can be large (40–60%).57–60 If thetherapist routinely uses one of the treatments undertest, an unconscious bias is likely to inflate any positiveeffects of that treatment and reduce treatment effects inthe control group. In our study, despite exposure to onlyJ R Coll Physicians Edinb 2013; 43:15–23 2013 RCPE200Twelve weeks post-treatmentbaseline therapyfigure 3 Biofeedback vs hypnotherapy (irritable bowelsyndrome symptom severity scores [IBS-SS]): 12 weeks posttreatment.one experienced gastro-enterologist with an interest inIBS and one highly-trained therapist, the drop-out rate(37%) was significant. Studies comparing treatmentefficacies should adopt an ‘intention to treat’ analysis, asthe exclusion of any patient may significantly overestimatetreatment effects. Given similar drop-out rates in thetwo groups and the absence of any follow-up data ondrop-outs, we accepted that the only data that could beanalysed were for those patients who actually attendedtherapy sessions. Selection bias may also occur when19
Other symptoms total scoresclinicalA Dobbin, J Dobbin, SC Ross, C Graham, MJ FordTimefigure 4A Total non-gastrointestinal symptom (other symptoms) scores.Type 3 tests of fixed effectsEffectNum DF Den DFF valuePr tmentXtimepoint31661.040.3762Num DF Numerator degrees of freedomDen DF Denominator degrees of freedomPr Probabilityfigure 4B Type 3 tests of fixed effects (other symptomscores).patients are recruited by advertising or by direct contactwith patient groups; such patient populations are verydifferent from that obtained by random allocation fromoutpatient gastroenterological clinics. In the formerthere may be a considerable effect of self-selection onexpectancy, increasing placebo effects.59 Expectancy andmotivation are key components of suggestibility;61 higherpatient expectancy in self-selected groups and lowermotivation in hospital GI clinic groups result in significantdifferences in compliance, suggestibility and placeboeffects.62 Meta-analyses of randomised trials usinghypnosis have recommended that future studies shouldemploy credible therapies (i.e. behavioural orpsychological) as control groups, not just waiting lists or20medical treatment.50–53 In our study the use of a therapistexperienced in using both therapies under study for IBSshould minimise this particular problem. Poor studydesign or patient selection, lack of long-term follow-up,inappropriate control groups and inattention to patientand therapist expectancy can produce unrecognisedplacebo effects which can inflate the apparent efficacy ofspecific treatments, apart from long-term follow-up,again our study design should have minised these effects.The effects of interpersonal relationships in modifyingtherapeutic outcomes have been well described.63,64ConclusionBiofeedback and hypnotherapy delivered by briefintervention achieved similar clinically significantimprovements in refractory IBS symptoms whichpersisted with no additional treatment during the period12–24 weeks. Heart rate variability biofeedback requiresless training and expertise than hypnotherapy and offerseconomic advantages over hypnotherapy and CBT.Treatment with HRV biofeedback may therefore be acost-effective option, avoiding patient antagonismtowards psychological therapies and the potential formisunderstanding of the practice of medical hypnosis.J R Coll Physicians Edinb 2013; 43:15–23 2013 RCPE
RCT of brief intervention with biofeedback and hypnotherapy in patients with refractory IBSHADSclinicalTimefigure 5a Total hospital and anxiety depression scores (HADS).Type 3 tests of fixed effectsEffectNum DF Den DFtable 1 Hospital anxiety and depression scoresF valuePr FTreatment1600.190.6640Timepoint31729.95 .0001TreatmentXtimepoint31721.330.2671Num DF Numerator degrees of freedomDen DF Denominator degrees of freedomPr Probabilityfigure 5B Type 3 tests of fixed effects (total hospital andanxiety depression [HADS] scores).REFERENCES12345Longstreth GF, Thompson WG, Chey WD et al. Functional boweldisorders. Gastroenterology 2006; 130:1480–91. http://dx.doi.org/10.1053/j.gastro.2005.11.061Rome 111:The functional gastrointestinal disorders. 3rd ed. Lawrence,Kansas: Allen Press; 2006.Ferguson A, Sircus W, Eastwood MA. Frequency of ‘functional’gastrointestinal disorders. Lancet 1977; 2:613–4. vey RF, Salih SY, Read AE. Organic and functional disorders in2000 gastroenterology outpatients. Lancet 1983; 1: 632–4. der SL, Locke GR, Talley NJ et al. Impact of functionalgastrointestinal disorders on health-related quality of life: apopulation-based case-control study. Aliment Pharmacol Ther 2004;19:233–42. J R Coll Physicians Edinb 2013; 43:15–23 2013 RCPETime periodBiofeedbackMean (SD)n 31HypnotherapyMean (SD)n 30TotalAnxietyDepressionTotalAnxietyDepression17.8 (8.0)15.7 (7.9)11.3 (4.4)10.0 (4.9)6.5 (4.6)5.8 (4.2)19.6 (7.4)17.5 (6.9)12.7 (4.5)11.0 (4.4)6.9 (4.1)6.6 (3.4)12 weeks* posttreatmentT214.7 (6.7)15.1 (8.9)9.8 (4.1)9.1 (5.3)4.9 (3.7)5.9 (4.9)24 weeks posttreatmentT314.7 (7.2)15.7 (9.2)9.6 (4.0)9.2 (4.8)5.1 (4.2)6.5 (5.6)8 weeks pretreatmentTOAt entryT1*t tests: Total hospital and anxiety depression score (HADS)*Biofeedback vs hypnotherapy: T1; t 1.15, p 0.26Biofeedback: T1 vs T2; t 2.73; p 0.008Hypnotherapy: T1 vs T2; t 1.17; p 0.2421
clinicalA Dobbin, J Dobbin, SC Ross, C Graham, MJ ��300–30–200–1010HADSfigure 6 Irritable bowel syndrome symptom severity score (IBS-SS) vs hospital anxiety and depression score (HADS).678910111213141516Spiegel BM, Gralnek IM, Bolus R et al. Clinical determinants ofhealth-related quality of life in patients with irritable bowelsyndrome. Arch Intern Med 2004; 164:1773–80. well PJ, McCallum M, Creed FH et al. Non-colonic featuresof irritable bowel syndrome. Gut 1986; 27:37–40. http://dx.doi.org/10.1136/gut.27.1.37Nyhlin H, Ford MJ, Eastwood J et al. Non-alimentary aspects of theirritable bowel syndrome. J Psychosom Res 1993; 37:155–62. sman D, Li Z, Andruzzi E et al. US householder survey offunctional gastrointestinal disorders. Prevalence, sociodemography,and health impact. Dig Dis Sci 1993; 38:1569–80. http://dx.doi.org/10.1007/BF01303162Talley NJ, Gabriel SE, Harmsen WS et al. Medical costs in communitysubjects with irritable bowel syndrome. Gastroenterology 1995;109:1736–41. sman DA, Creed FH, Olden KW et al. Psychosocial aspectsof the functional gastrointestinal disorders. Gut 1999; .ii25Ford MJ, Miller PM, Eastwood J et al. Life events, psychiatric illnessand the irritable bowel syndrome. Gut 1987; 28:160–5. http://dx.doi.org/10.1136/gut.28.2.160Roelofs K, Spinhoven P. Trauma and medically unexplained symptomstowards an integration of cognitive and neuro-biological accounts. ClinPsychol Rev 2007.07.004Myers B, Greenwood-Van Meerveld B. Role of anxiety in thepathophysiology of irritable bowel syndrome: importance of theamygdala. Front Neurosci 2009; 3:47. http://dx.doi.org/10.3389/neuro.21.002.2009Ford MJ, Camilleri M, Zinsmeister AR et al. Psychosensorymodulation of colonic sensation in the human transverse andsigmoid colon. Gastroenterology 1995; 109:1772–80. lips ML, Gregory LJ, Cullen S et al. The effect of negativeemotional context on neural and behavioural responses tooesophageal stimulation. Brain 2003; 126: 669–84. http://dx.doi.org/10.1093/brain/awg0652217 Wilder-Smith C, Schindler D, Lovblad K et al. Brain functionalmagnetic resonance imaging of rectal pain and activation ofendogenous inhibitory mechanisms in irritable bowel syndromepatient subgroups and healthy controls. Gut 2004; 02851418 Myers B, Greenwood-Van Meerveld B. Corticosteroid receptormediated mechanisms in the amygdala regulate anxiety andcolonic sensitivity. Am J Physiol Gastrointest Liver Physiol 2007;292:G1622–9. http://dx.doi.org/10.1152/ajpgi.00080.200719 Myers DA, Gibson M, Schulkin J et al. Corticosterone implants tothe amygdala and type 1 CRH receptor regulation: effects onbehavior and colonic sensitivity. Behav Brain Res 2005; 3.00120 Shepard JD, Myers DA. Strain differences in anxiety-like behavior:association with corticotropin-releasing factor. Behav Brain Res2008; 186:239–45. http://dx.doi.org/10.1016/j.bbr.2007.08.01321 Naliboff BD, Berman S, Chang L et al. Sex-related differences in IBSpatients: central processing of visceral stimuli. Gastroenterology2003;124:1738–47. http://dx.doi.org/10.1016/S0016-5085(03)00400-122 Weiser MJ, Handa RJ. Estrogen impairs glucocorticoid dependentnegative feedback on the hypothalamic-pituitary-adrenal axis viaestrogen rece
the biofeedback group was –116.8 (99.3) and in the hypnotherapy group –58.0 (101.1), a statistically significant difference between groups (difference –58.8, 95% confidence interval [CI] for difference [–111.6, –6.1], p 0.029 ). In 61 patients with refractory IBS, biofeedback and hypn
