| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Center for Psychobiological and Psychosomatic Research, University of Trier, Trier, Germany (A.H.B., S.F., D.H.H.); and Institute of Psychology, Clinical Psychology and Psychotherapy, University of Zürich, Zürich, Switzerland (U.M.N., U.E.).
Address correspondence and reprint requests to Ulrike Ehlert, PhD, Institute of Psychology, Department of Clinical Psychology and Psychotherapy, University of Zürich, Zürichbergstr. 43, CH-8044 Zürich, Switzerland. E-mail: u.ehlert{at}psychologie.unizh.ch
| ABSTRACT |
|---|
|
|
|---|
Methods: Thirty subjects who fulfilled the Rome Diagnostic Criteria for either irritable bowel syndrome or nonulcer dyspepsia and 24 healthy controls took part in the study. Free salivary morning cortisol and diurnal cortisol profiles were obtained for all subjects. On a second day, a low-dose dexamethasone suppression test was applied. Additionally, in all subjects a corticotropin-releasing hormone (CRH) challenge test was performed.
Results: The results show attenuated unstimulated cortisol levels in patients compared with controls. After CRH challenge, blunted adrenocorticotropic hormone and cortisol responses were observed. These findings suggest lower pituitary and adrenocortical activity in patients with functional gastrointestinal disorders.
Conclusion: The observed pituitary-adrenal reactivity in these patients is discussed as a possible consequence of lower adrenocortical activity, possibly resulting in a disinhibition of CRH in the brain.
Key Words: irritable bowel syndrome nonulcer dyspepsia HPA axis corticotropin-releasing hormone cortisol dexamethasone suppression test
Abbreviations: CRH = corticotropin-releasing hormone; NUD = nonulcer dyspepsia; IBS = irritable bowel syndrome; ACTH = adrenocorticotropic hormone; HPA = hypothalamic-pituitary-adrenal; ANOVA = analysis of variance; AUCtot = area under the curve total.
| INTRODUCTION |
|---|
|
|
|---|
It has been well established that chronic stress, anxiety, and negative affectivity are related to the two most prevalent functional gastrointestinal disorders, nonulcer dyspepsia (NUD) (2022) and irritable bowel syndrome (IBS) (2326). This is especially true in the context of critical life events and psychological trauma such as sexual, emotional, or physical abuse (2730).
Because functional gastrointestinal disorders have been related to both CRH in the brain and stress, we were interested in exploring pituitary adrenal activity under both stimulated and unstimulated conditions.
| METHODS |
|---|
|
|
|---|
Out of the total of 45 persons interviewed, 30 met our criteria. Ten patients fulfilled all criteria for IBS, 5 for NUD, and 15 for both conditions (IBS + NUD). Sociodemographic variables and the distribution of diagnoses are depicted in Table 1. No differences between the three patient groups were observed. This finding justified the combination of the three groups into one patient group.
|
A total of 24 control group participants took part in the study. All control subjects were recruited by advertisements in the local newspaper, at the University of Trier, and at the local blood donation center of the Red Cross. All were generally free of gastrointestinal symptoms. They were matched with respect to age, gender distribution of the sample, number of household members, children, level of education, smoking, and oral contraceptives (Table 1).
Patients with functional gastrointestinal disorders (18 women and 12 men) did not differ from control subjects (15 women and 9 men) with respect to the aforementioned characteristics.
Methods and Biochemical Analyses
Psychometric Testing
For the assessment of trait anxiety, the State-Trait Anxiety Inventory was used (34). For depressed mood, the general depression scale was applied (Allgemeine Depressionsskala) (35). In addition, a clinical psychological DSM-IV (36) based interview (37) was performed.
Biochemical Methods
The CRH challenge test is indicative of dynamic alterations of pituitary-adrenal reactivity (38). It involves the intravenous injection of a certain dose of CRH (here, 1 µg/kg body weight human CRH; Ferring, Kiel, Germany) that leads to an increase in the level of adrenocorticotropic hormone (ACTH) followed by an increase of cortisol, which in turn exerts a negative feedback on the ACTH release at the level of the pituitary. In addition to salivary cortisol measurements, blood was sampled during the CRH challenge test in order to obtain the plasma levels of ACTH and cortisol.
To test the feedback sensitivity of the hypothalamic-pituitary-adrenal (HPA) axis, a low-dose (0.5-mg) dexamethasone suppression test (39) was performed on a different day.
For the analysis of cortisol, saliva was collected by salivettes (Sarstedt, Rommelsdorf, Germany). Intra-assay and interassay variabilities were below 5% and 10%, respectively. Cortisol concentration in saliva was determined using a time-resolved immunoassay with fluorometric detection as described in detail elsewhere (40). Plasma ACTH and cortisol from the blood samples collected during the CRH challenge test was analyzed by a two-site commercial chemiluminescence assay (Nichols, Bad Nauheim, Germany).
Procedure
For all tests, written informed consent was obtained, and participants were fully informed about the risks involved in the pharmacological tests.
According to a routine protocol from this laboratory (41), basal free salivary morning cortisol levels were measured immediately after awakening (t1) and after 30 (t2), 45 (t3), and 60 (t4) minutes. The diurnal profile measures were obtained at 08:00 (t1), 11:00 (t2), 15:00 (t3), and 20:00 (t4) hours of the day. On this first day, participants were instructed to take 0.5 mg dexamethasone (Jena-Pharm, Jena, Germany) at 23:00 hours in the evening. All salivary cortisol measures were repeated the next day to estimate the effects of dexamethasone on HPA activity. To enable participants to obtain these measures in their natural setting at home or at work, they were briefed in detail verbally and in writing on how to handle the salivettes for the collection of saliva and the intake of dexamethasone. Two sets of labeled salivettes were handed out to the participants. They were instructed to chew the salivettes for one minute for each collection. For the morning cortisol saliva collection, they were instructed to stay in bed for at least 15 minutes and to refrain from having breakfast within the first half hour after awakening. Furthermore, they were asked to avoid juices and fruits containing high levels of acid and not to collect saliva immediately after brushing their teeth. Finally, participants were asked to send the saliva samples to our institute by regular mail the day after sampling. The samples were stored at 20°C until the data from all participants were complete.
The CRH challenge test was performed at our laboratory between 15:00 and 18:00 hours in the afternoon of a day within 2 weeks after basal and dexamethasone suppression measurements, with a minimum of 5 days between dexamethasone suppression and the challenge test. Afternoon is the time recommended for this test because the endogenous secretions of ACTH and cortisol are low at this time of day (38,39). The physician in charge performed a medical investigation to assess suitability for the experimental protocols. Participants were then weighed in order to determine the dose of hCRH to be applied (1 µg/kg) (see 42). The indwelling intravenous catheter for blood sampling was then placed in the participants forearm vein. This was followed by a 30-minute rest period to allow the participant to adapt to the indwelling catheter. After this, the first blood (5 ml) and saliva (by salivette) samples were taken immediately before CRH injection (t0). Postinjection samples of blood and saliva were then collected at +20, +30, +45, +60, +90, and +120 minutes. Participants were allowed to read magazines during the time of testing and were constantly monitored. Blood samples were cooled in an icebox and centrifuged within 30 minutes (at 3000 rpm for 10 minutes at 4°C). Plasma samples and the salivettes were then stored at 20°C until biochemical analyses were performed.
The protocol followed the criteria of the declaration of Helsinki and was approved by the Ethical Committees of the University of Trier and the Chamber of Physicians of the State of Rhineland-Palatinate, Germany.
Statistical Methods
For ACTH and cortisol, area under the total response curve, expressed as area under all samples, was calculated using the trapezoid formula (43). Data were tested for normal distribution and homogeneity of variance using a Kolmogorov-Smirnov and Levenes test before statistical procedures were applied. Analysis of variance (ANOVA) was computed to reveal possible variance. To test for significance of possible differences of hormone concentrations and profiles between the two groups, a two-factorial multivariate analysis of variance with repeated measures design was used. Covariates were age and gender (multivariate analysis of covariance). Students t-tests were computed for comparison of the groups. For all analyses, the significance level was
= 5%. p-Values are two-tailed. Unless indicated, all results shown are means ± SEMs.
| RESULTS |
|---|
|
|
|---|
|
Dexamethasone Suppression
Figure 1 shows salivary cortisol profiles after awakening (basal and after dexamethasone-induced suppression) for patients and controls.
|
At the time of awakening, patients had slightly lower cortisol levels than the control group, although this difference did not reach significance (t48 = 1.6; p = .12). The difference increased during the first hour after awakening, because the patients showed a blunted increase of cortisol compared with the steep increase observed in healthy controls. Sixty minutes after awakening, the difference in cortisol levels was in the range of 10 nmol/l (t48 = 2.76; p < .01). ANOVA revealed significant differences for the area under the curve total (AUCtot; F[1,49] = 6.0; p < .05). There were no group differences for cortisol after awakening in the dexamethasone suppression measures (F[1,47] = 0.001; p = .97).
Figure 2 depicts salivary diurnal cortisol profiles (basal and after dexamethasone-induced suppression) for patients and control groups. As was to be expected from our morning cortisol measures, patients started off with significantly lower cortisol levels at 08:00 hours (t48 = 2.48; p < .05). At later time points, the curves then became parallel to each other, and finally converged in the evening. Both groups displayed a regular cortisol decrease over the course of the day, with the lowest measures at 20:00 hours in the evening, when the two groups reached similar levels. The difference between the two groups is significant in the ANOVA of the AUCtot values (F[1,49] = 26.0; p < .001), with the patient group showing lower cortisol levels throughout the day. Again, there was no significant group difference in the dexamethasone-induced cortisol suppression (F[1,49] = 3.09; p = .09).
|
Furthermore, we compared the two groups with regard to conditions (basal, dexamethasone suppression) and time of day (morning, diurnal cortisol concentrations) with a multivariate analysis of covariance, with age and gender as covariates. A significant group effect was found (F[4,37] = 4.89; p < .01). The test of between-subjects effects indicated that groups differed with regard to basal cortisol concentrations (morning cortisol: F = 5.6; p < .05; diurnal cortisol: F = 15.26; p < .001), but not with regard to cortisol concentrations after dexamethasone suppression (morning cortisol: F = 0.93; p = .34; diurnal cortisol: F = 0.98; p = .33).
Within-group comparisons of patients comorbidity with regard to endocrine concentrations revealed that comorbid psychiatric diagnoses did not influence cortisol concentrations (all t-tests were nonsignificant, data not shown).
Corticotropin-Releasing Hormone Challenge Test
The plasma levels of ACTH increased significantly in both groups 20 minutes after CRH injection (Figure 3). Patients started off at a slightly lower ACTH level, although the difference was not significant (t44 = 1.88; p = .066), and the controls showed a steeper ACTH increase.
|
Whereas patients already showed declining ACTH levels between the 20-minute and 30-minute measurements, those of the controls remained at a high level for a further 10 minutes. Both groups then showed declining ACTH levels, with the difference between the groups remaining constant at all times of measurement. The difference between the groups reached significance in the ANOVA of AUCtot values for plasma ACTH levels (F[1,44] = 9.25; p < .01). An additional multivariate ANOVA with repeated measures and age and gender as covariates indicated a significant main effect of time (F[6,37] = 2.75; p < .05) and significant interaction effects of time x gender (F[6,37] = 6.8; p < .05) and time x group (F[6,37] = 2.57; p < .05) and a nonsignificant interaction effect of time x age (F[6,37] = 8.7; p = .51).
Figure 4 depicts saliva cortisol levels during CRH challenge. The increase in free salivary cortisol levels in controls was much more pronounced than in patients at all times of measurement after CRH injection. The maximum increase was observed in both groups after 45 minutes with a delay of approximately 15 to 25 minutes to the time of maximal ACTH secretion. Patients and controls differed significantly in the ANOVA of AUCtot salivary cortisol levels (F[1,42] = 15.18; p < .001). The cortisol in plasma, which was also obtained, showed the same results as the salivary measures. The ANOVA of the AUCtot of plasma cortisol was also significant (F[1,45] = 6.69; p < .05). An additional multivariate ANOVA with repeated measures and age and gender as covariates indicated a significant main effect of time (F[6,40] = 6.72; p < .001) and a nearly significant interaction effect of time x group (F[6,40] = 2.26; p = .057) and nonsignificant interaction effects of time x gender (F[6,40] = 1.77; p = .13) and time x age (F[6,40] = 0.97; p = .46) for plasma cortisol. For salivary cortisol, a significant interaction effect of time x group (F[6,42] = 3.15; p < .05) and a nonsignificant main effect of time (F[6,42] = 1.7; p = .15) and nonsignificant interaction effects of time x gender (F[6,42] = 1.64; p = .16) and time x age (F[6,42] = 1.1; p = .38) were found (Figure 5).
|
|
Within-group comparisons of patients comorbidity with regard to endocrine concentrations revealed that comorbid psychiatric diagnoses did not influence endocrine concentrations (all t-tests were nonsignificant; data not shown).
We analyzed all biochemical data with respect to the three diagnosis groups (IBS, NUD, IBS+NUD). The groups did not differ in any of the biological variables, except for ACTH (morning cortisol: F[2,27] = 2.44; p = .11; diurnal cortisol: F[2,27] = 0.39; p = .68; morning cortisol after dexamethasone: F[2,27] = 0.30; p = .75; diurnal cortisol after dexamethasone: F[2,27] = 0.56; p = .58; ACTH after CRH challenge: F[2,24] = 5.1; p = .02; plasma cortisol after CRH challenge: F[2,24] = 0.57; p = .58; saliva cortisol after CRH challenge: F[2,21] = 1.26; p = .31).
| DISCUSSION |
|---|
|
|
|---|
Comparing levels of plasma ACTH, and plasma and saliva cortisol after CRH challenge with the basal morning cortisol values, similar results can be observed, with patients showing a less pronounced increase. This suggests a downregulated sensitivity or a change in numbers of CRH receptors at the level of the pituitary in patients. The timing of the peaks of ACTH and cortisol and the onset of negative feedback of cortisol on ACTH secretion were seemingly intact in patients. Thus, the results of this study suggest that HPA feedback mechanisms are intact in patients with functional gastrointestinal disorders. HPA activity, however, was found to be abnormally low, perhaps as a result of a downregulation of CRH receptors or receptor sensitivity.
A blunted ACTH response to CRH has been observed in major depression (4446), anxiety disorders (47,48), and eating disorders (49,50), whereas cortisol responses were normal in all of these studies. In these articles, the blunted ACTH response to CRH was interpreted as a consequence of enhanced endogenous CRH release, whereas the adrenals seem to counteract this effect. In the present study, however, both a blunted ACTH and cortisol response were observed. If the adrenals are unable to compensate, this may possibly suggest an adrenal insufficiency in these patients. Such a hypothesis would be supported by the low cortisol levels after awakening. As previously reported, the cortisol increase after awakening correlates well with the cortisol response to ACTH stimulation, suggesting an association between the awakening response and adrenal capacity (51).
As previously reported, we observed a high correlation between the cortisol response to awakening and the cortisol response to ACTH. Thus, the cortisol increase after awakening seems to reflect adrenal capacity. Given the fact that adrenal capacity is low, one may possibly expect a disinhibition of CRH in the paraventricular nucleus of the brain, which, in consequence, may promote a similar feature to that observed in anxiety disorders and depression. The blunted ACTH response to awakening suggests a comparable mechanism. If this is the case, it is not unlikely that CRH would also activate autonomic functions, eventually promoting functional gut disturbances as observed in laboratory animals. However, because of a lack of empirical evidence, the possibility of CRH promoting gastrointestinal disturbances is still a hypothesis that remains to be tested. If the hypothesis is correct, the functional gastrointestinal and psychological disorders observed in these patients could be considered to be a secondary effect of adrenal insufficiency (52).
Our study has several shortcomings to overcome. We used a single within-subject period for test and comparison of endocrine responses between groups (eg, CRH test). However, there is evidence that the stress responsiveness and negative feedback regulation of the HPA axis vary throughout the day (53). Thus, an alternative explanation for the differences in hormone responses and their levels may be a differing change in reactivity and receptor sensitivity between patients and controls during the diurnal cycle. Future studies should incorporate several evaluations of endocrine responses during different times of day to test this hypothesis. Furthermore, the recruitment of our patients was diverse. We cannot exclude that self-selection by the patients might undermine the generalizability of our results. In our study, we examined IBS and NUD patients together. However, because the numbers of the respective groups were not high, no statements can be made as to whether there are similarities or differences between the IBS patients alone, the NUD alone, and the combined groups.
Very few studies have investigated HPA parameters in functional gastrointestinal patients, and the small number of studies that have been performed have yielded inconclusive results. Heitkemper et al. (54) found no significant differences in urinary cortisol between IBS patients and non-IBS patients and healthy controls with the exception of afternoon samples, when IBS patients showed significantly higher levels. Mishra and Pandey (55) found no significant differences between functional gastrointestinal patients and controls in plasma levels of cortisol in samples drawn at 09:00 hours in the morning. Alfven et al. (56), on the other hand, found decreased morning plasma cortisol levels in children suffering from recurring abdominal pain. In a more recent study, Patacchioli et al. (57) found higher salivary cortisol levels in IBS patients in the morning and lower levels in the evening compared with controls. Fukudo and coworkers (57a) found an exaggerated response in ACTH after intravenous administration of CRH in IBS patients in comparison with healthy control subjects. However, the authors found no differences in the response of serum cortisol after CRH challenge between the two groups. To our knowledge, the present study was the first to investigate HPA function systematically in functional gastrointestinal patients by combining basal and stimulated measures. The differences in the outcomes of the present study compared with some of the previous studies might be related to differences in methodology. The present study differed from previous studies in that it assessed cortisol after awakening in several samples. Interestingly enough, low levels of cortisol have also been observed in patients with other somatoform disorders such as chronic fatigue syndrome (58,59), chronic postoperative back pain (60), fibromyalgia (61), and chronic pelvic pain in women (6264). Stratakis and Chrousos (2) and, more recently, Heim et al. (65), and Raison and Miller (66) have described psychobiological states, conditions, and disorders that are related to hypocortisolemia. As outlined in the introduction, functional gastrointestinal disorders are associated with psychosocial factors (2030). In our study, the patients showed high trait anxiety and depression scores. Furthermore, comorbidity was high, especially with other somatoform disorders. Data from the present study provide further evidence for the existence of hypocortisolemic disorders and offer the possibility that functional gastrointestinal complaints may occur as a consequence of HPA axis dysfunction.
| NOTES |
|---|
|
|
|---|
Received for publication February 8, 2004; revision received July 20, 2004.
DOI:10.1097/01.psy.0000157064.72831.ba
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Z. FitzGerald, P. Kehoe, and K. Sinha Hypothalamic--Pituitary-- Adrenal Axis Dysregulation in Women With Irritable Bowel Syndrome in Response to Acute Physical Stress West J Nurs Res, November 1, 2009; 31(7): 818 - 836. [Abstract] [PDF] |
||||
![]() |
D. Chatzitheodorou, S. Mavromoustakos, and S. Milioti The effect of exercise on adrenocortical responsiveness of patients with chronic low back pain, controlled for psychological strain Clinical Rehabilitation, April 1, 2008; 22(4): 319 - 328. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |