| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Psychiatry, University of Ottawa, and the University of Ottawa Institute of Mental Health Research, Royal Ottawa Hospital, Ottawa, Ontario, Canada.
Address correspondence and reprint requests to Dr. Diana Koszycki, Stress and Anxiety Clinical Research Unit, Royal Ottawa Hospital, 1145 Carling Avenue, Ottawa, Ontario K1Z 7K4. E-mail: dkoszyck{at}rohcg.on.ca
| ABSTRACT |
|---|
|
|
|---|
Methods: Eight psychiatrically healthy IBS patients, 8 PD patients with no history of IBS, and 12 normal controls received a bolus injection of CCK-4 and placebo on two separate days in a double-blind, randomized fashion.
Results: Consistent with previous findings, panicogenic sensitivity to CCK-4 was enhanced in PD patients relative to controls. In contrast, IBS patients exhibited a response that was comparable to controls. Interestingly, CCK-4-induced nausea and abdominal distress were decreased in IBS patients relative to the other groups. No diagnostic difference was noted for cardiovascular response to CCK-4.
Conclusion: These data indicate that IBS patients with no lifetime psychiatric history do not share the CCK-2 receptor dysfunction implicated in the pathophysiology of PD and that this dysfunction may not be a common mechanism for both CNS and enteric nervous system disorders. Nevertheless, the results suggest that a dysfunction of the CCK system may be involved in the pathophysiology of some enteric symptoms associated with IBS.
Key Words: irritable bowel syndrome panic disorder panic attacks cholecystokinin-tetrapeptide gutbrain interaction
Abbreviations: IBS = irritable bowel syndrome; PD = panic disorder; CCK = cholecystokinin; NTS = nucleus tractus solitarius; CBF = cerebral blood flow; NC = normal controls; PSS = Panic Symptom Scale; CCK-4 = cholecystokinin-tetrapeptide; VAS = visual analog scale.
| INTRODUCTION |
|---|
|
|
|---|
Stimulation of central CCK receptors with the selective CCK-2 receptor agonist, CCK-tetrapeptide (CCK-4), reliably and dose-dependently provokes panic attacks in humans, with PD patients being exquisitely sensitive to CCK-4s panicogenicity (1416). The CCK-4 induced panic attacks are accompanied by gastrointestinal symptoms that characterize DSM-IV panic attacks, including nausea and abdominal distress. Furthermore, CCK-4-induced anxiety is accompanied by marked biological alterations, including increases in heart rate, blood pressure, and minute respiration as well as activation of the hypothalamicpituitaryadrenal axis (16,2123).
Electrophysiological monitoring with topographic spectral electroencephalography and brainstem evoked potential recordings indicate that CCK-4 penetrates the human brain and has robust effects on the brainstem (2426), which is a proposed primary site of action of CCK-4 (27). Neuroimaging studies have also shown that CCK-4-induced panic anxiety is associated with cerebral blood flow (CBF) changes. Benkelfat et al. (28) reported that CCK-4-induced panic in healthy volunteers was associated with rCBF increases in anterior cingulate cortex, cerebellum, and bilateral regions including insula, claustrum, and amygdala. Javanmard et al. (29) extended this work with a protocol of sampling the induced panic episodes at early and late time points (1 or 2 minutes after CCK-4 bolus injection). They found that the early time was associated with rCBF increases in the hypothalamus, the late time point was associated with rCBF increases in the claustruminsular region, and both phases were associated with rCBF decreases in medial frontal cortex. These imaging data indicate that the panicogen CCK-4 is associated with activation of brain regions of particular relevance in the control of gut function and perhaps the pathophysiology of IBS (30), including amygdala, hypothalamus, and prefrontal cortex.
We have proposed that IBS and PD may share a common dysfunction of the CCK system (31,32). To investigate this hypothesis, we compared behavioral and cardiovascular responses to a submaximal dose of systemic CCK-4 in IBS patients with no lifetime psychiatric history, PD patients with no history of IBS, and normal controls (NC).
| METHODS |
|---|
|
|
|---|
Materials
Cholecystokinin-tetrapeptide (CCK-4), consisting of an amino acid chain of TRP-MET-ASP-PHE-NH2, was synthesized by Peninsula (Belmont, CA) and prepared by GIS Medicament (Nantes, France) as intravenous solutions according to previous protocols (15). A 20-µg dose of CCK-4 was used in this study. The rationale for dose of the CCK-4 was based on data from doseresponse studies, which showed that a dose of 20 µg CCK-4 produces a high panic rate in PD patients and readily discriminates between PD and NC subjects (15,16). Placebo consisted of an equal volume of saline. CCK-4 and placebo were administered intravenously in a bolus injection (i.e., less than 5 seconds).
Assessment
Behavioral Evaluation
A DSM-IV-derived Panic Symptom Scale (PSS) was used to characterize behavioral response to the challenge (15). Subjects were directed to rate the severity of 18 panic symptoms as absent (0), mild (1), moderate (2), severe (3), or extremely severe (4). Two separate scores were obtained from this scale: (a) a score reflecting the total number of symptoms reported (i.e., number of symptoms with scores
1) and (b) a sum intensity score (i.e., the sum of the intensity ratings). Five physical sensations rarely associated with experimentally induced panic were included in the PSS to evaluate response bias toward item overendorsement. These items were earache, itchy nose, stuffy nose, low back pain, and itchy feet. The response bias items were tallied and analyzed separately. To meet criteria for a panic attack, subjects had to achieve a score of two or higher on the anxiety/fear/apprehension item of the PSS and a score of one or higher on each of at least four other items. In addition, PD patients had to report that the CCK-4-induced panic attack was similar to their naturally occurring panic.
Cardiovascular Evaluation
Blood pressure and heart rate were recorded using automated equipment (DinamapTM, Critikon, Canada).
Assessment of IBS Symptoms
A Visual Analog Scale was used to measure 11 gastrointestinal symptoms (GI-VAS). Subjects were directed to indicate the intensity of these symptoms by transecting the 100-mm line with a vertical line. The length in millimeters from the extreme 0 to the vertical line made by the subjects was used to calculate GI-VAS scores. The GI-VAS was administered to the IBS group only.
Procedure
Subjects arrived at the research unit at 8:00 AM after an overnight fast. They were not permitted to smoke or use caffeine- or xanthine-containing substances within 48 hours before each day of testing. CCK-4 and placebo were administered in a randomized order on two separate days under double-blind conditions. The subject sat in a reclining chair, a venous catheter was installed into an antecubital vein, and a physiological saline drip was initiated. A blood pressure cuff with a pulse monitor attached to an automatic sphygmomanometer was applied to the other arm. After baseline measures were obtained, subjects received a single intravenous bolus injection of the study solution (CCK-4 or placebo). Subjects were asked to describe any symptoms they experienced after the injection. Vital signs were recorded immediately before (time 0) and every 20 seconds for the first 5 minutes after the injection. Immediately after the symptoms had abated, the symptoms subjects experienced were evaluated with the PSS. VAS ratings of bowel symptoms were obtained from IBS patients at baseline and after the challenge. Subjects rated their bowel symptoms on the VAS to reflect how they felt at the peak effect after each challenge.
Data Analyses
Analysis of variance appropriate for a repeated measures design (ANOVAR) was performed on all baseline and postchallenge continuous data. Challenge (CCK-4 or placebo) was the within-subject variable, and diagnosis was the between-subject variable. For cardiovascular measures, peak change scores (CCK-4/placebo peak score minus time 0) were used in the analyses. The main interest in the ANOVAR was whether sensitivity to CCK-4 varied as a function of diagnosis. This would be reflected in a challenge by diagnosis interaction. If significant interactions were observed, one-way ANOVAs followed by the post hoc Bonferroni test were conducted separately for the placebo and CCK-4 conditions to examine simple main effects of diagnosis. Because few subjects reported symptoms during the placebo challenge, scores for the onset and duration of symptoms were analyzed by ANOVA for the CCK-4 challenge only. Categorical data were analyzed by chi-square analysis. For cells with expected frequencies less than five, Fishers exact test was used to analyze individual 2 x 2 tables. All data are presented as the mean ± SD. Statistical significance was assumed as p < .05 (two-tailed tests). Trends (p < .10) are also reported.
| RESULTS |
|---|
|
|
|---|
Placebo challenge did not provoke a panic attack in any of the subjects. Panic attack frequency after CCK-4 challenge was 75% (6/8) for PD patients, 17% (2/12) for NCs, and 0% (0/8) for IBS patients. The differences in panic frequency after CCK-4 challenge was statistically significant, with the frequency being higher in PD versus IBS and NC subjects (Fishers exact test p = .007 and p = .019, respectively). ANOVAR performed on PSS data revealed significant challenge main effects (F = 96.45, df = 1,25, p < .001 and F = 48.01, df = 1,25, p < .001), diagnosis main effects (F = 5.65, df = 2,25, p < .01 and F = 7.70, df = 2,25, p < .01), and challenge by diagnosis interactions (F = 4.91, df = 2,25, p < .05 and F = 6.60, df = 2,25, p < .01) for the number and sum intensity of panic symptoms, respectively. As shown in Table 1, post hoc Bonferroni tests revealed that PD patients had higher scores on these measures than both IBS and NC subjects after the CCK-4 challenge. The onset and duration of symptoms after CCK-4 challenge did not differentiate the groups. No significant effects were detected for the number of reported response bias items.
|
Mean ± SD intensity rating scores for individual panic symptoms after CCK-4 and placebo challenge are shown in Table 2. Statistically significant challenge main effects were found for all 18 panic symptoms, with scores being higher after the CCK-4 than placebo challenge. Significant challenge x diagnosis interactions were also found for 9 of 18 PSS symptoms, including unsteadiness (p < .05), faintness (p < .01), trembling/shaking (p < .05), nausea (p < .01), paresthesia (p < .001), flushes/chills (p < .01), anxiety/fear/apprehension (p < .01), fear of dying (p < .05), and fear of losing control (p < .01). Post hoc Bonferroni tests indicated that these symptoms were rated as more intense by PD compared with IBS and NC subjects after the CCK-4 challenge.
|
Interestingly, intensity ratings of CCK-4-induced nausea and abdominal distress were lower in IBS patients compared with the other two groups. Within-group analysis using paired t tests failed to detect a placeboCCK-4 difference on these PSS items for IBS patients. In contrast, significant placeboCCK-4 differences were detected for PD and NC subjects: nausea (t = 4.23, df = 7, p < .01, and t = 2.86, df = 11, p < .05, respectively) and abdominal distress (t = 2.39, df = 7, p < .05 and t = 2.80, df = 11, p < .05, respectively). The proportion of IBS patients who reported having symptoms of nausea and abdominal distress was also lower compared with the other two groups; 87% (7/8) and 62% (5/8), respectively, for the PD group, 58% (7/12) and 58% (7/12), respectively, for the NC group, and 25% (2/8) and 37% (3/8), respectively, for the IBS group. Differences between IBS and PD patients on the frequency of reported nausea was statistically significant (Fishers exact test, p = .041).
VAS ratings of gastrointestinal symptoms obtained at baseline and after the placebo and CCK-4 challenge for the IBS subjects are shown in Table 3. Ratings after both the placebo and CCK-4 injection were low. Paired t tests revealed no significant difference between placebo and CCK-4 for any of the IBS symptoms.
|
No significant main effects or interactions were found for baseline heart rate or blood pressure scores. ANOVARs performed on peak change scores yielded a significant challenge main effect for heart rate (F = 60.22, df = 1,25, p < .001), systolic blood pressure (F = 8.94, df = 1,25, p < .01), and diastolic blood pressure (F = 7.24, df = 1,25, p < .05), but no significant diagnosis main effects or interactions. As anticipated, the increase in heart rate and blood pressure was greater after the CCK-4 than placebo challenge. Mean ± SD maximum change scores for cardiovascular measures are shown in Table 1.
| DISCUSSION |
|---|
|
|
|---|
An interesting result of this study was the decreased gastrointestinal response to CCK-4 in the IBS group relative to PD and normal control subjects. In humans, systemic CCK-4 has been consistently reported to induce nausea, cramping, and abdominal distress (1416,22), and acute pretreatment with the selective CCK-2 receptors antagonist L-365260 attenuated CCK-4-induced abdominal distress in PD patients (38). These data suggest that CCK-2 receptors play a role in eliciting gastrointestinal symptoms associated with panic anxiety. It is not clear at present why CCK-4 had no demonstrable effect on gastrointestinal symptoms in our IBS patients. One explanation is that CCK-2 receptors are down-regulated in these patients, rendering them less sensitive to CCK-4-induced anxiety and gastrointestinal symptoms. Support for this view comes from a study demonstrating that antagonism of CCK-2 receptors after chronic treatment with CI-988 worsens gastrointestinal symptoms in anxious patients with comorbid IBS versus anxious patients with no IBS comorbidity (39). It may be that chronic administration of CI-988 unmasked a hypoactive CCK-2 receptor system.
It is also conceivable that decreased gastrointestinal sensitivity to CCK-4 in our IBS patients is attributable to an imbalance between CCK-1 and CCK-2 receptors. IBS has been associated with hyperactivity of CCK-1 receptors, and CCK-1 receptor antagonists are currently under clinical development as therapeutic agents for this syndrome (40). Animal studies have shown that there is a functional relationship between central and peripheral CCK receptors. In rats, agonism of CCK-1 and CCK-2 receptors results in opposite firing activity in NTS neurons (41) and opposite behavioral effects (42). These findings have led to the hypothesis that there may be a mutually inhibitory relationship between the CCK-1 and CCK-2 receptor systems (42). Thus, one might speculate that the decreased gastrointestinal response to systemic CCK-4 in our IBS patients resulted from an inhibitory action of the CCK-1 system on the CCK-2 system. A fruitful line of research will be to investigate whether CCK-1 and CCK-2 receptor antagonists differentially affect behavioral and gastrointestinal sensitivity to CCK-4 in IBS patients, and to combine such behavioral work with neuroimaging (43).
Methodological limitations of this study require mention. First, interpretation of the results are limited by the small sample size and the unequal gender distribution across groups, particularly the greater number of females in the IBS group and greater number of males in the PD group. In addition, our IBS patients were recruited from local advertisement and may not be representative of the type of patients seen in specialized gastrointestinal clinic. It is possible that had we recruited psychiatrically healthy IBS patients from a specialized gastrointestinal clinic, we might have obtained different findings. Also, we failed to measure severity of IBS symptoms. Some of our patients may have had a mild form of the illness, and this could have affected their response to stimulation of CCK-2 receptors with CCK-4. Indeed, it is conceivable that IBS pathophysiology is distinct in patients with mild versus severe IBS, with milder form of the syndrome being associated with greater peripheral versus central contribution to symptomatic process.
In summary, these preliminary data indicate that IBS patients with no lifetime history of PD do not share the CCK-2 receptor dysfunction implicated in the pathophysiology of PD and that this dysfunction may not be a common mechanism for the two disorders. Further research is needed to replicate these findings. It would also be of considerable interest to evaluate whether CCK-4 differentially affects IBS patients with and without PD comorbidity and PD patients with and without prominent GI symptoms. Such research would further elucidate the neurobiological link between IBS and PD and between the brain and the gut, and clarify whether PD associated with IBS patients is the same as PD without IBS identified in psychiatric samples.
We are grateful to Maria Pizzi, BSc, RN, Alyson Robins, BSc, RN, and Graham Machacek, BA for research assistance. We also thank Pauline Bazinet, MD for her assistance in the study.
| NOTES |
|---|
|
|
|---|
This study was supported in part by the Medical Research Council of Canada, Group Grant 13306.
DOI:10.1097/01.psy.0000160465.79007.65
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |