| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Firestone Institute for Respiratory Health (R.L., F.E.H.) and the Brain Body Institute (G.M., G.T., J.B.), St. Joseph’s Healthcare, McMaster University, Hamilton, Ontario, Canada.
Address reprint requests to: John Bienenstock, McMaster University, HSC Room 3N26H, Department of Medicine, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5. Email: bienens{at}mcmaster.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: To determine whether suggestion differentially influenced airway tone in preselected suggestible compared with suggestion-resistant asthmatics, we performed a prospective, double-blind, crossover pilot study in which subjects were identified as being suggestible or suggestion-resistant using the Creative Imagination Scale. Responses to inhaled saline were assessed using FEV1 and a modified Borg scale after the suggestion that the saline was a bronchoconstrictor and subsequently a bronchodilator.
RESULTS: Five of eight suggestible subjects compared with one of nine suggestion-resistant subjects demonstrated a fall in FEV1 greater than 150 ml in response to inhaled saline and suggestion of bronchoconstriction (p = .027). Fourteen subjects experienced dyspnea in response to sham bronchoconstrictor, but none reported increased dyspnea after sham bronchodilator.
CONCLUSIONS: This is the first time that subjects with asthma have been categorized as suggestible or nonsuggestible with this distinction then used to predict response to an intervention. The results of this pilot study suggest that a subgroup of suggestible asthmatic patients is more likely to respond to a placebo-like sham bronchoconstriction challenge. The data support earlier observations that psychological factors may influence asthma, and provide insights into the placebo response.
Key Words: asthma, • suggestion, • placebo response.
Abbreviations: CIS = Creative Imagination Scale;; FEV1 = forced expiratory volume in 1 second;; PC20 = provocative concentration of methacholine causing a 20% fall in FEV1.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
No study has yet attempted to characterize this group of placebo responders within an asthmatic population. Lasagna et al. (7) have previously described "reactors" and "nonreactors" among postoperative patients who received either placebo or narcotic analgesics in a double-blind manner, suggesting features that may distinguish placebo responders from nonresponders. There is controversy surrounding the nature of the placebo response, recently highlighted by an article noting that natural fluctuations in illness patterns may be interpreted as placebo responses (8). However, the placebo effect has generally been observed in a wide range of conditions (eg, depression and irritable bowel syndrome), in which its magnitude may be so great as to obscure the clinical effect of the active intervention unless large samples are compared.
There is literature pointing to the importance of suggestion in the pathogenesis of asthma (9–17). In 1889, MacKenzie (18) reported on a woman with a rose allergy who responded to the presence of a silk rose (the "rose cold"). Black (19, 20) demonstrated that allergen-induced immediate-type hypersensitivity skin responses were inhibited by direct suggestion while study subjects were under hypnosis (19) and further demonstrated a similar phenomenon for a delayed-type hypersensitivity response (20). More recently, several studies have reported that suggestion may produce both bronchoconstriction and bronchodilation in a proportion of asthmatic subjects (9–12, 21, 22). However, interpretation of the results from these and other studies may be limited by methodological considerations, including small sample size, lack of masking of study subjects, and a lack of control for both the effect of saline-induced bronchoconstriction and the effect of repeated spirometric testing. Furthermore, none of the previous studies have found a clear association between physiological responses and any predictive subject variable.
On the basis of the work highlighting the possible importance of suggestion in asthma and allergy, we reasoned that the level of suggestibility of people may influence the likelihood of responding to a suggestion for bronchoconstriction—essentially a placebo challenge. People vary in the extent to which they are responsive to nonhypnotic suggestion. Scales of suggestibility, such as the Creative Imagination Scale (23), have been constructed and validated such that nonclinical populations are normally distributed on this measure, with subsets of the population scoring within the highly suggestible or suggestion-resistant ranges. Although it is apparent that the mechanisms by which suggestion and other complex psychosocial events interact with biological systems to facilitate physiological airway dysfunction are not yet understood, we hypothesized that we could use this measure to determine whether suggestion differentially influenced airway tone in a preselected group of highly suggestible asthmatics compared with those identified as suggestion-resistant. We reasoned that, through a carefully designed experiment, we might gain better understanding of how suggestion may influence airway function in asthma.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Study Design
This was a prospective, double-blind, single-center, crossover pilot study. Eighty study subjects attended an initial screening visit, during which the Creative Imagination Scale questionnaire was administered. Seventeen subjects were identified as either suggestible (N = 8) or suggestion-resistant (N = 9) and were asked to attend a second visit, at which time baseline spirometry and three bronchial response tests were performed 45 minutes apart. The first bronchial response test was performed using normal saline as a putative bronchoconstrictor with the accompanying suggestion that it was similar in its action to methacholine. The second test was performed using normal saline as a putative bronchodilator with the accompanying suggestion that it was a new bronchodilator drug, similar in its action to nebulized albuterol. The third test was a true methacholine challenge to confirm the severity of airway hyperresponsiveness and was performed without any accompanying suggestion. Changes in FEV1 and the Borg dyspnea scale were monitored during and after each of the three response tests.
Procedures
Creative Imagination Scale
The CIS is a 10-item (test suggestions) permissive scale that has been shown to be reliable and valid (23). Scores usually follow a Gaussian distribution, with the highest percentiles representing suggestible people and the lowest percentiles the nonsuggestible people. One investigator (G.M.) assessed the CIS questionnaire scores to identify subjects in the appropriate tails of the distribution and referred subjects for bronchial response testing in random order. Other investigators and the technologists present during the bronchial response tests were blind to which subjects were suggestible or suggestion-resistant.
Spirometry
FEV1 and slow vital capacity were measured according to the American Thoracic Society specifications (24). The baseline FEV1 was calculated as the best of three reproducible values (within a maximum change of 5%), using the predicted values of Crapo et al. (25).
Bronchial response inhalation tests
Bronchial response inhalation tests (using normal saline as placebo for the first two tests and methacholine for the third test) were conducted as described by Cockcroft et al. (26). Each subject performed three bronchial response tests 45 minutes apart in the same order with the same standardized verbal suggestion from the same technologist in all cases. The saline was always at room temperature. During the first response test, subjects were told that the inhaled solution might make their airways become tighter and cause them to wheeze. At the end of each 2-minute run, subjects were told that they should be feeling increasing chest tightness, possibly with some wheezing. At this point, they rated their symptoms on a modified Borg scale from 0 to 10 (0 = nothing at all, 10 = maximal) (27).
After a 45-minute interval, participants were given the second response test, again using normal saline, but accompanied by the suggestion that this was a newly developed pharmacological bronchodilator being evaluated for the treatment of asthma. Subjects were told that inhaling the drug should cause their airways to become less tight and allow their breathing to feel much easier. At the end of each 2-minute run, subjects were told that they should be experiencing less chest tightness and that their breathing should be becoming easier, and they rated their dyspnea symptoms on a modified 7-point Likert scale (1 = almost the same/hardly better at all, 7 = a great deal better) (28). A methacholine challenge test was performed 45 minutes after the second response test to confirm the presence and severity of airway hyperresponsiveness. As part of the methacholine challenge, subjects inhaled normal saline at room temperature at the start of the test, but without any accompanying suggestion. The methacholine test was stopped at PC20. At the end of the methacholine test, subjects were given 200 µg of inhaled albuterol via a spacer device, and their FEV1 was monitored until it had returned to within 5% of the baseline value. Subjects who attended visit 2 received individual verbal debriefing at the end of the visit.
Outcomes
The primary outcome measurement was change in FEV1 after bronchial response testing. A priori, a change greater than 150 ml from baseline FEV1 in either direction was considered clinically important. This decision was based on the fact that the coefficient of variation for FEV1 measurements is reported to be less than 5% in normal subjects, with the standard deviation of repeated measurements on the same day 102 ml (29). A secondary outcome measure was the reported changes in the Borg dyspnea scale.
Statistical Analysis
Descriptive statistics were used to summarize clinical and demographic characteristics of the study participants, with qualitative variables expressed as percentages and quantitative variables reported as the arithmetic mean and standard deviation. Statistical analysis to compare proportions was performed using 
2 and Fisher exact tests. Within-subject differences were compared using the paired t test. The Pearson correlation coefficient was used to test for a significant relationship between dyspnea scores after administration of the sham bronchoconstrictor and the methacholine challenge. Two-tailed p values <.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The baseline demographic characteristics of the 17 study subjects attending visit 2 are summarized in Table 1. There were no significant differences between the suggestible and nonsuggestible study groups.
|
|
Overall, an increased sensation of dyspnea, as measured by the subjective Borg dyspnea scale, was experienced by 14 subjects (six suggestible and eight suggestion-resistant) in response to the sham bronchoconstrictor, together with a subsequent improvement in dyspnea scores in response to the sham bronchodilator. In contrast, none of the 17 subjects reported any increase in the Borg dyspnea scale in response to the sham bronchodilator. There was no significant difference in the degree of breathlessness experienced after the sham bronchoconstrictor by either suggestible or suggestion-resistant subjects. However, in the suggestible group, there was a significant correlation between increased dyspnea scores reported after sham bronchoconstriction and after methacholine challenge (r = .81, p < .001); no significant correlation was seen in the suggestion-resistant group (r = .13, p > .05).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
The physical changes observed cannot be explained by a nonspecific bronchoconstrictive response to the inhalation of 0.9% saline. Although it may be argued that the initial fall in FEV1 seen in five of eight suggestible subjects was a result of a nonspecific response to inhaled saline, this response was not seen in suggestion-resistant subjects inhaling the same saline concentration. It was also not evident when the same suggestible subjects inhaled 0.9% saline either with the accompanying suggestion that it was a bronchodilator or as part of the methacholine challenge, when there was no suggestion that it was either a bronchoconstrictor or a bronchodilator.
Although it might have been preferable to randomize the order of the sham bronchoconstrictor and sham bronchodilator inhalation challenges (together with their accompanying instructions) to overcome criticism that the fall in FEV1 was caused by a nonspecific saline effect or situational anxiety, we elected not to randomize the order to minimize any confounding. We reasoned that the primary hypothesis was a between-group difference, and that the lack of randomization of the presentation would not be a major study design flaw.
It is perhaps surprising that there was less evidence of a suggested bronchodilation effect than there was of a bronchoconstriction effect in suggestible subjects (Figure 1, A and B). Previous reports (30, 31) suggested that we might have expected a greater bronchodilator effect in response to suggestion than we observed. It is important to note that our study subjects were not experiencing bronchoconstriction at the time that bronchodilation was suggested; had this been the case, perhaps bronchodilation would have been observed. In the absence of bronchoconstriction it is difficult to speculate how much dilation above base-line we should have expected; rather, this condition is more appropriate for demonstrating that bronchoconstriction did not occur with the procedure alone, in the absence of a specific suggestion for bronchoconstriction.
To our knowledge, this is the first instance in which patients with asthma have been categorized as suggestible or nonsuggestible with this distinction then used to predict response to an intervention. Emotion and stress could be important components in the precipitation or exacerbation of asthma, and there is increasing evidence that neuronal pathways influence asthma control and airway tone (16, 32–35). Although previous studies have addressed aspects of the association between emotion/suggestion and airway hyperresponsiveness, the results have been unclear. This may be because these studies were conducted in general groups of patients rather than in groups preselected for emotionality, suggestibility, or stress-responsiveness. Although participants were recruited from a specialist tertiary center and may not be representative of the asthmatic population in the community, most clinical studies are similarly conducted among mild to moderate asthmatics attending tertiary care centers, and our results are likely generalizable to subjects participating in large clinical drug trials.
In summary, within the asthmatic population, there is a specific subgroup of suggestible patients in whom suggestion influences airway tone and the perception of dyspnea. These patients can be identified a priori on the basis of their response to the Creative Imagination Scale. These data support earlier observations that psychological factors may influence asthma, and they provide a possible explanation, through the differential response of suggestible and nonsuggestible subjects, for discrepancies in previous reports. If replicable in other populations, these data may provide an important insight into how the placebo response may be minimized in clinical trials, not only in asthma but also in other conditions, including functional and psychiatric disorders.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication April 17, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. R. Sutherland Sham Procedure versus Usual Care as the Control in Clinical Trials of Devices: Which is Better? Proceedings of the ATS, October 1, 2007; 4(7): 574 - 576. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
