| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLE |
From the Department of Psychosomatic and Behavioral Medicine (S.B., U.F.M.) and Department of Medicine (S.B.), National Hospital, Oslo, and the Department of Medicine (S.S., M.B.J.), Oestfold County Hospital, Fredrikstad, Norway.
Address reprint requests to: Svein Blomhoff, MD, Department of Psychosomatic and Behavioral Medicine, National Hospital, N-0027 Oslo, Norway. Email: svein.blomhoff{at}rikshospitalet.no
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Eleven female patients with IBS with comorbid phobic anxiety disorder were compared with 22 age-matched female IBS patients without such comorbidity. The groups were compared with respect to event-related potentials (ERP), auditory-presented words with emotional contents, barostat-assessed visceral sensitivity thresholds, and symptom levels the last week before assessment.
RESULTS: The comorbid group had a significantly enhanced first negative ERP wave (N1) to all stimuli, indicating increased use of brain attentional resources. It also had increased visceral threshold for the sensation of gas, and reduced gas-stool and gas-discomfort tolerances compared with the noncomorbid group. Enhanced N1 amplitude at the frontal electrode and reduced gas-stools tolerance significantly predicted subjective gas complaints, explaining 47% of the symptom variation.
CONCLUSIONS: The study suggests an association between information processing in the frontal brain and visceral sensitivity characteristics in IBS patients, and indicates that subjective disease-related symptomatology is predicted by brain perceptual characteristics. The findings indicate that an interaction between IBS-related and anxiety-related hyperreactivity in the frontal brain may constitute a psychophysiological mechanism for the contribution of psychiatric comorbidity to severity and duration of the irritable bowel syndrome.
Key Words: Irritable bowel syndrome, • psychiatric comorbidity, • event-related potentials, • visceral sensitivity, • brain-gut.
Abbreviations: IBS = irritable bowel syndrome;; ERP = event-related potentials;; VAS = visual analogue scale;; GI = gastrointestinal;; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, fourth edition;; IBS-C = IBS patients without comorbid psychiatric disorder;; IBS-PA = IBS patients with phobic anxiety disorder;; HAD = Hospital Anxiety and Depression scale;; EPQ = Eysenck Personality Questionnaire;; N1 = first negative wave of ERP complex;; P3 = third positive wave of ERP complex;; Fz = frontal midline electrode;; Cz = central midline electrode;; Pz = parietal midline electrode;; CNS = central nervous system;; GLM = general linear model.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Even if previous research has failed to demonstrate a clear-cut intestinal dysfunction in IBS, most researchers agree that the brain-gut axis in some way is involved (3, 4). The altered perception of visceral stimuli observed in IBS subjects has been linked to a theory of hyperexcitability of neurons in the dorsal horn (5, 6). Additionally, some studies indicate that patients with IBS also have aberrant cerebral function (7–9).
Cerebral evoked potentials are evoked by all stimuli processed by the brain. In IBS, these potentials have previously been used to study cerebral perception of rectal pain stimuli (10). Cerebral evoked potentials may also be used to study information processing of sensory stimuli, and then named event-related potentials (ERP). One of the advantages of ERPs compared with brain imaging techniques like positron emission tomography and functional magnetic resonance imaging is a better temporal resolution, allowing the assessment of psychophysiological functions during milliseconds.
In a previous ERP study, we found that IBS patients without comorbid psychiatric disorders were characterized by cerebral hyperreactivity to auditory stimuli in the frontal brain region compared with healthy controls (11). This finding may suggest that the hypersensitivity to rectal stimuli observed in IBS patients at least partly may be explained by this hyperreactivity (5).
Previous studies on the physiology of IBS have not addressed the possible contaminating role of concurrent psychiatric disorders for their findings. Because psychiatric disorders are known to influence both the motility of the gastrointestinal tract and neurophysiological processing of sensory stimuli (6, 12, 13), this may be considered a major weakness in the present knowledge.
To investigate whether a comorbid psychiatric disorder influenced brain processing of auditory-elicited ERPs in patients with IBS, in this study we compared ERPs in IBS patients without a comorbid psychiatric disorder and IBS patients with phobic anxiety comorbidity. Phobic anxiety patients are characterized by an enhanced attentional awareness of situation-specific threats, psychophysiologically fairly consistent across all types of phobia. We thus hypothesized that patients with both IBS and phobia would have a more pronounced enhanced N1, indicating enhanced selective attention compared with non-comorbid IBS patients. Considering the association between IBS symptomatology and comorbid psychiatric disorders, we also wanted to determine if the two IBS groups differed in thresholds of visceral sensations, measured by a rectal barostat or in subjective symptom levels.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients
Eleven female IBS patients with comorbid phobic anxiety disorder (mean age 34.2 ± 6.7) constituted the experimental group (IBS-PA), and were age-matched in a 1:2 design with 22 female IBS patients without any current psychiatric disorder (mean age 33.8 ± 6.4) in the comparison group (IBS-C). In the phobic-anxiety group, seven subjects had a specific phobia (height, snakes, or spiders), two subjects had agoraphobia without panic, and two subjects had a nongeneralized social phobia. Fourteen subjects had experienced a previous major depressive episode, seven in each group.
Clinical Assessments and Baseline Characteristics
The gastroenterological assessment included clinical examination, rectoscopy, and a double-contrast barium enema or coloscopy unless performed during the 2 previous years.
To reliably assess psychiatric diagnosis according to the DSM-IV system, the psychiatric examination included the MINI International Neuropsychiatric interview (16). All participants filled out Hospital Anxiety and Depression scale (HAD) to assess self-reported levels of anxiety and depression, and the neuroticism subscale of the EPQ, because neuroticism is known to be associated with gastrointestinal psychophysiological reactions (17, 18). They also recorded the daily number of evacuations and rated daily levels of pain and gas discomfort on a visual analogue scale ranging from 0 (no symptoms) to 10 (maximal severity) for 1 week immediately before the investigation.
The mean physician-assessed VAS score in the IBS-PA group was 6.36 (±1.12) and 6.05 (±1.15) (NS) in the IBS-C group. Duration of present episode was 9.2 (±11.2) years in the IBS-PA group, 16.1 (±12.9) years in the IBS-C group (NS).
No differences between the groups were observed concerning the level of pain [IBS-PA: 3.7 (±2.1), IBS-C: 4.1 (±2.2)] or mean daily number of evacuations [IBS-PA: 2.1 (±1.3), IBS-C: 2.3 (±1.8)]. However, the IBS-PA group experienced significantly less gas discomfort than the IBS-C group (3.3 (±2.1) vs. 4.9 (±1.8); t = 2.303, p = .028).
No significant differences were observed with respect to scores on the EPQ neuroticism scale (IBS-PA: 11.7 (±4.8), IBS-C 10.4 (±4.4)), HAD anxiety subscale (IBS-PA 7.2 (±3.5), IBS-C 6.1 (±3.3)) or HAD depression subscale (IBS-PA: 3.5 (±3.6), IBS-C 2.6 (±3.0)).
ERP Assessments
The ERP procedure is described in more detail elsewhere (13). Briefly, the paradigm was an ERP oddball distracter paradigm consisting of three conditions, each containing 200 stimulus presentations. A standard tone (800 Hz, 60% probability), a target tone (1200 Hz, 20% probability), and a distracter (everyday words, 20% probability) were in each condition presented by headphones. The distracters were Norwegian words with emotional content, selected from validated translations of scales for anxiety, depression, and irritability (19–21). The ERPs from all three distracter conditions were pooled together in the analysis. Continuous EEG activity was recorded from the frontal (Fz), central (Cz), and parietal (Pz) midline electrodes (22).
In an auditory paradigm, the first negative wave (N1) in the ERP is a correlate of sensory processing and attention (23). Later occurring components, in particular the third positive wave (P3), are involved in stimulus evaluation and interpretation. N1 latency and amplitude to standard tones, target tones, and distracter words were analyzed at all electrodes. To capture the continuous nature of auditory language processing, the mean amplitudes to target and distracter stimuli in consecutive 50-msec time bins from 150 to 600 msec were additionally analyzed (24).
Visceral Sensitivity Assessments
A computer-driven barostat (Synectics Visceral Stimulator; Synectics, Stockholm, Sweden) was used in a double-random staircase procedure to record rectal perception thresholds (25). In this procedure, the sequence of distentions randomly alternates between two distention sequences. In each sequence, the distentions always increase after each trial on which the subject does not report the requested sensation, and they always decrease after each trial on which the subject does report sensation.
An 8-cm long cylindrical plastic bag, noncompliant if the intrabag volume was smaller than 500 ml, was tightly fixed at both ends to a catheter and placed with the distal end of the bag 5 cm from the anal verge. The barostat bag was inflated before use and tested in water to control for leakage. The tube was secured in its proper position with tape. The position was continuously controlled by a three-lumen water-perfused manometry system monitoring the pressure in the anal high-pressure zone. The experiment was performed after a fast period of at least 6-hour duration followed by the application of a 120-ml Klyx enema. The subjects were placed in a left lateral position and were requested not to make unnecessary movements. The distension device was connected to a subject-operated marker device that logged the threshold sensation of gas, first sensation of stools, and the sensation of discomfort. The mean value of three measurements was regarded as the individual’s threshold. The differences in pressure between sensations of stools and gas constituted the ability to discriminate between gaseous and fecal rectal content and were taken as the gas-stools tolerance; the differences between gas and discomfort as the gas-discomfort tolerance. The volume/pressure relationship, defined as the rectal wall compliance, was calculated at all thresholds.
Ethics
Written and verbal informed consent was obtained from all subjects. The protocol was approved by the Regional Committee on Ethics in Medicine. The study was performed according to the Declaration of Helsinki.
Statistical Analysis
The Statistical Package for the Social Sciences (SPSS), version 8.0, was used in all analysis. Group differences were first analyzed using Student’s t test, then reanalyzed using the GLM-general factorial model with duration of present episode and self-reported symptoms of anxiety, depression, and neuroticism as covariates. General linear model (GLM)-repeated measurement statistics were used in the time bin analysis. Greenhouse-Geisser corrections were made when appropriate. Finally, all patients were included in a forward linear-regression analysis where ERP data from all electrodes combined with significant visceral sensitivity data that were used as independent variables: number of evacuations, VAS-levels of pain, gas discomfort, and global severity as dependent variables.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Event-Related Potentials
The ERP differences between IBS patients with and without phobic anxiety disorder are illustrated in Figure 1, showing the waveform at the frontal midline electrode evoked by the distracter words. No significant differences in N1 latency were found at any electrode or stimuli condition. The group-related findings with respect to N1 amplitudes are shown in Table 1.
|
|
When we adjusted for duration of IBS episode, and self-reported anxiety and depression levels, no major changes in significances occurred, with the exception that the distracter condition difference at Fz changed from marginally significant (t = 2.14, p = .040) to marginally insignificant (F = 4.04, p = .054).
In the analysis of the ERP wave to the target tones and distracter words in the 150 to 600 msec time bins no significant differences between the groups were found at any electrode (target: Fz: F = 1.94, df = 2.82, p =.133; Cz: F = 2.58, df = 2.58, p = .058; Pz: F = 1.93, df = 2.80, p = 0.136; distracter: Fz: F = 0.84, df = 3.44, p = .489; Cz: F = 0.86, df = 3.42, p = .474; Pz: F = 1.35, df = 3.21, p = .263).
Visceral Sensitivity
When we adjusted for duration of present episode, anxiety, and depression, the threshold for sensation of gas was significantly increased in the IBS-PA group compared with the IBS-C group (Table 2). No differences in thresholds for sensation of stools or discomfort were found. Additionally, the gas-stools and gas-discomfort tolerances were significantly reduced. No significant group differences with respect to rectal wall compliance (ml/mm Hg) at any of the observed thresholds were detected [gas: IBS-PA: 5.22 (±2.93), IBS-C: 6.12 (±5.00); stools: IBS-PA: 5.32 (±2.81), IBS-C: 7.11 (±4.60); discomfort: IBS-PA: 6.45 (±1.97), IBS-C: 7.53 (±3.35)].
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Influence of Phobic Anxiety Comorbidity on Cerebral Stimulus Perception in IBS Patients
The enhanced N1 wave in the IBS-PA group was present at all electrodes and all stimulus conditions, only marginally influenced by duration of episode and emotional factors. In general, differences between groups that can be seen to all stimuli (tones and words) indicate more generalized group-specific traits. Differences that are seen only in response to a specific stimulus, like an emotional word, may indicate that the aberrant feature is represented by a specific condition initiated by the stimulus. The enhanced N1 wave thus most probably is a characteristic of the IBS group studied.
We did not observe any significant differences between the groups in HAD-assessed anxiety scores. Because the HAD-assessed anxiety primarily assess generalized emotion, the anxiety in the phobic group was restricted to the phobic situation. All subjects were recruited through newspaper advertisement. Most of them had consulted a physician for their complaints, but had in general accepted the symptoms as stress-related or something they had to "live with." However, the group included subjects with agoraphobia, simple phobia, or nongeneralized social phobia. In previous research these groups did not differ with respect to psychophysiological responses to emotional stimuli (27). However, some caution in the interpretation may be needed inasmuch as this is the first ERP study in phobic anxiety.
Additionally, 7 of 11 patients in the IBS-PA group and 7 of 22 in the IBS-C group had experienced a previous depressive episode. We did not observe any statistical association between a previous depressive episode and our findings. In previous ERP research, major depression has mainly been associated with decreased P3 amplitude (28). It is thus unlikely that the increased N1 amplitude observed in the IBS-PA group in our study is caused by a history of previous depressive episode.
Because N1 represents an automated shift in attention toward a new stimulus, the enhanced wave in the IBS-PA group probably represents the environmental scanning for the feared object or situation which is a clinical characteristic of phobic anxiety patients (23). In a previous study, we found that the nonpsychiatric IBS patients had a significantly enhanced frontal N1 wave compared with healthy controls (11). If substantiated in further research, IBS patients with a comorbid phobic anxiety disorder consequently have two disorders that both share an enhanced frontal N1 wave as an essential disease characteristic.
Cerebral Perception, Visceral Sensitivity, and Subjective Complaints
When we compared the groups with respect to visceral sensitivity, the main finding was a significantly increased threshold for the sensation of gas in the IBS-PA group. Even if this finding must be interpreted with some caution because of the fairly small number of subjects included, it is supported by other findings indicating that some IBS patients have impaired tolerance of intestinal gas (29, 30). Our study may indicate that the impaired tolerance is caused by increased visceral sensitivity for the sensation of gas, and suggests an association between this finding and processing of stimuli in the frontal brain area. A possible explanation for this finding may be that the increased threshold is caused by a competition between attentions toward external or internal stimuli caused by the two disorders. The reduced gas-stools and gas-discomfort tolerances are probably consequences of the increased gas threshold. The increased threshold thus seems to represent a neurophysiological alteration in sensory processing. The very low gas-stools tolerance in the comorbid group was remarkable, and may indicate that IBS subjects with phobic anxiety have reduced ability to discriminate between the sensations of gas and stools. This may explain disease characteristics like incomplete evacuations and evacuation urgency.
Even if we found a relationship between N1, visceral sensitivity, and gas symptomatology, no group differences emerged with respect to VAS-assessed level of pain. This is in line with previous research indicating that reduced threshold may be found for one stimuli, but not necessarily for another (6). Because perception of pain has a complex cerebral representation involving different brain regions, differences in location of the cerebral processing of different pain stimuli may be an additional explanation (31).
In the regression analysis, we found that frontal brain information processing alone (N1) was a highly significant predictor of VAS-assessed gas complaints, explaining 32% of the symptom variance in the whole group. If visceral sensitivity data additionally were included (gas-stools tolerance), the model explained 47% of the symptom variance. This finding emphasizes the role of frontal brain information processing in the irritable bowel syndrome and may suggest a causal relationship between brain processing, visceral sensitivity, and subjective IBS symptomatology.
If perceptual disturbance is considered the main feature of the irritable bowel syndrome, the additional impact of phobic anxiety on the N1 wave may maintain the IBS-related enhanced N1 and thus also elucidate one mechanism for the impact of comorbidity on treatment resistance (32). Psychiatric disorders differ with respect to ERP characteristics. In controlled studies, enhanced N1 amplitudes have been demonstrated in panic and somatization disorders (33, 34). Other anxiety disorders may enhance the P3 wave, whereas a major depressive disorder seems to attenuate the same wave (28). It is thus possible that other psychiatric disorders may influence the visceral sensitivity differently, and thus explain some of the variability in previous sensitivity studies. To our knowledge, no previous visceral sensitivity study has controlled their data for psychiatric diagnosis.
Methodological Considerations
Because our study represents a new research paradigm in brain-gut research, the findings must so far be interpreted with caution. This is particularly important inasmuch as the number of subjects included is fairly small. It must be noted that our findings only apply to female nonconstipation-prone IBS patients.
In both study groups, we found considerable within-group variance both with respect to ERP and visceral sensitivity measures. In addition to individual variation, the finding most probably represented a methodological problem in ERP research, caused by the difficulty in assessing CNS function with electrodes located outside the skull (23). However, methodological studies have documented that ERPs have reliable test-retest properties with the stability required for group research, but not necessarily for clinical applications (35, 36).
Many visceral sensitivity studies are primarily focused on the discomfort threshold, and the staircase barostat procedure is well validated with respect to this threshold (37). The gas and stool thresholds are less frequently assessed. Even if the double-staircase method should be an adequate method for these assessments, some caution is needed in the interpretation of these findings.
Brain-Gut Axis in Irritable Bowel Syndrome
Aberrant perception of visceral stimuli has been demonstrated in a series of IBS studies (5, 6). It is generally accepted that the majority of IBS patients primarily have inappropriate perception of physiological events, and possibly altered reflex responses in the gut (6). In controlled studies, IBS patients have a significantly increased proportion of rapid eye movement (REM) sleep, aberrant brain activation pattern to visceral stimuli, and greater activation of CNS pain centers, prefrontal cortices included, during painful rectal distension (7–9, 38).
Together with our studies, these findings indicate that IBS could be regarded as reflecting a hyperreactivity in a brain-gut system that includes receptors in the gut, the enteric nerve system, and CNS. It is possible that this hyperreactivity may be triggered by both psychological and biological factors at different levels in the system.
The main finding in this study is that the described neurophysiological mechanism may constitute a biological link between the neurophysiology of a psychiatric disorder and IBS symptomatology. Phobic anxiety comorbidity seems to interfere with the processing of visceral information in the frontal cerebral region and thus affects visceral sensitivity and IBS symptomatology. The described link may also more generally explain the contribution of psychiatric comorbidity to the severity and duration of intestinal motility disorders. However, because other psychiatric disorders may affect different brain processing components, the contribution may vary depending on the specific psychiatric diagnosis. Further studies will be needed to substantiate this hypothesis.
Received for publication January 19, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E Guthrie, F Creed, L Fernandes, J Ratcliffe, J Van der Jagt, J Martin, S Howlett, N Read, J Barlow, D Thompson, et al. Cluster analysis of symptoms and health seeking behaviour differentiates subgroups of patients with severe irritable bowel syndrome Gut, November 1, 2003; 52(11): 1616 - 1622. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
