|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the UMD-New Jersey Medical School, Division of Pulmonary Medicine (Drs. Lavietes, Matta, and Cherniack), Newark, NJ; Asthma and Allergy Research Center (Dr. Bielory), UMD-New Jersey Medical School, Newark, NJ; Chronic Fatigue Syndrome Research Center (Dr. Natelson), Veterans Administration Medical Center, Newark, NJ; and School of Psychology (Dr. Tiersky), Fairleigh Dickinson University, Hackensack, NJ.
Correspondence to: Marc H. Lavietes, MD, FCCP, UMD-New Jersey Medical School, University Hospital, I-354, 100 Bergen St, Newark, NJ 07103-2406
| Abstract |
|---|
|
|
|---|
Objective: This study tests the following two hypotheses: (1) that measures of the tendency of a patient to somatize will reduce the variation in the report of dyspnea not explained by airway function; and (2) that plethysmography is a better tool with which to estimate the degree of dyspnea associated with asthma.
Design: A prospective laboratory study carried out over one study session.
Participants: Forty asthmatic subjects who had withheld bronchodilator (BD) therapy overnight.
Interventions: We performed spirometry, plethysmography, and an assessment of dyspnea (ie, modified Borg scale) on all subjects before and after they received BD therapy. Standard questionnaires pertaining to psychological state and trait were administered as well.
Results: The change in specific airway conductance with BD therapy correlated with a decline in the Borg score (r = 0.47; p = 0.007). By contrast, neither spirographic measures nor measures of static lung volumes correlated. Correlation with the Borg scale score was not improved by adding indexes of either somatization or psychological state or trait.
Conclusion: The relief of dyspnea reported by patients with mild asthma after BD therapy is related to dilatation of the central airways.
Key Words: asthma dyspnea psychophysiology somatoform disorders
| Introduction |
|---|
|
|
|---|
Both pathophysiologic and psychological factors have been proposed to explain this variability in the perception of the magnitude of dyspnea among asthmatic subjects. Asthmatic subjects whose airway pathology features inflammation and eosinophilia more than bronchospasm, for example, report less dyspnea with a given fall in FEV1 than do other asthmatic subjects whose obstruction is transient and readily reversible with bronchodilator (BD) treatment and whose sputum eosinophilia is low.5 6 This may be attributed to the prolonged period of adaptation associated with chronic progressive airway inflammation.
Two lines of observation suggest that psychological factors may influence symptom perception as well. First, it is known that asthmatic patients scoring high on a scale measuring defensiveness are less accurate than others in their psychophysical magnitude estimates of external, inspiratory resistive loads. These subjects also perform poorly when asked to judge the quality of respiratory sensations associated with loading. These data indicate that the link between load recognition and the report of dyspnea is weak, at least in subjects with certain psychological traits.7 A study of asthmatic children showed no correlation between the precision with which the child identified an external load and his or her sense of shortness of breath.8 A second line of observation holds that suggestion may play a role in the provocation of bronchospasm.9 10 For example, some subjects who are responsive to a given asthma trigger (eg, ragweed) will wheeze when presented with a photograph of a ragweed plant.
Tools are available to assess the tendency of an individual to "somaticize," that is, to exaggerate responses to benign impulses associated with normal physiologic function and, thus, to misconstrue these responses as symptoms of organic disease.11 A persons tendency to somaticize is not related to any psychological state, trait, or psychiatric diagnosis.
This study tested the hypothesis that, in asthmatics, much of the variation in reports of dyspnea not explained by airway function might be explained by a variation in the tendency of individuals to somaticize. We asked asthmatic subjects to rate their degree of dyspnea and to perform lung function tests before and after bronchodilation. Subjects also responded to questionnaires designed to evaluate their tendency to somatize. Both spirometry and plethysmography were used to assess airway function. This allowed us to formulate and examine a second hypothesis, namely, that a plethysmographic measure of airway resistance rather than spirometry might best predict the magnitude of dyspnea (or the change in dyspnea associated with BD treatment) perceived by individual asthmatic subjects.
| Materials and Methods |
|---|
|
|
|---|
15% increase in
FEV1 in response to an inhaled BD. The average
(± SD) age was 40 ± 9 years. None had any other cardiopulmonary or
psychiatric disorders. Patients receiving either psychotropic
medications or recreational drugs were excluded from the study. The
study was approved by the local institutional review board. All
subjects signed informed consent forms.
Procedures
Lung Function Testing:
Spirometry, lung volume
determinations, and plethysmographic measurements of airways resistance
were performed with standard equipment (DSII plus; WE Collins;
Braintree, MA).
The modified Borg scale was used to obtain quantitative assessments of dyspnea.12 At prescribed intervals, subjects were presented with the scale and were asked to rate the magnitude of their dyspnea in the following way. They were told that 0 represented no breathlessness at all while 10 represents the "strongest feeling of breathlessness you have ever known." Subjects then were asked to select either a number from 0 to 10 on the printed scale or an adjacent verbal descriptor that would represent their degree of breathlessness at that moment.13
Psychometric Evaluation:
Three tests of psychometric
evaluation were used. The Barsky somatosensory amplification scale
(SSAS) was used to assess the tendency to amplify
symptoms.14
This 10-item questionnaire asks the subject to
estimate the degree of discomfort that he or she would associate with
each of 10 common but minor physical sensations, all of which are
unlikely to reflect major abnormalities (eg, feeling
hunger contractions). Second, sections of the revised Hopkins symptom
checklist (SCL-90-R) were used to examine the following four symptom
dimensions: somatization; psychoticism; anxiety; and
depression.15
The SCL-90-R is a standard tool designed to
measure psychopathology. Finally, the neuroticism subset of the
personality inventory of Costa and McCrae16
was
administered. Neuroticism, thought by Costa and McCrae to be the most
pervasive facet of personality, is ranked for each subject on a range
between emotionally stable and unstable or maladjusted.
Protocol:
Subjects refrained from using
ß2-agonists for 12 h prior to the study. On
reporting to the laboratory, they performed spirometry and
plethysmography, and scored their dyspnea on the Borg scale. To avoid
an effect on our data of the transient bronchoconstriction that may
accompany a deep breath in asthma, the spirometry and plethysmography
procedures were separated by 5 min.17
The first 30
subjects were told that they would receive BD therapy
(ie, aerosolized albuterol) after first performing the
tests and did inhale a BD. Over the next 20 min, psychometric
questionnaires were administered. Then lung function measurements were
repeated, and the Borg scale again was administered.
The protocol differed for the final 10 subjects in that they were told that they would receive either a BD or a placebo (namely, saline solution [SAL]), and in fact they received a nebulized SAL.
Data Analysis:
Standard equations from the literature were
used to convert lung volumes given in liters to a percentage of a
predicted value.18
Standard methods of data analysis
included paired comparisons (paired t test) and
regression analysis (ie, Pearson correlation coefficient
and multiple stepwise analysis). The data for two of the subjects were
incomplete, in that they provided acceptable data for lung function
studies but did not complete the psychometric evaluation
satisfactorily.
To evaluate the data from the 10 subjects in the SAL placebo-treated
group, we formed a comparison subgroup from among the original 30
subjects. This comparison group was composed of 14 subjects whose
change in Borg scale score was identical to that of the SAL-treated
group (ie,
0.10 U).
| Results |
|---|
|
|
|---|
|
The changes in sGaw and Borg scale scores were highly correlated
(r = 0.47; p = 0.007). Inspection of Figure 1
, however, shows a wide range of
Borg scale values in subjects
showing little change in sGaw (eg, those whose sGaw
values increased
0.10 U after therapy). By contrast, larger changes
in airway function always were accompanied by a decrease in the Borg
scale score. There were no significant correlations between other
measures of lung function and Borg scale scores (Table 2
).
|
|
Response to Inhalation of Placebo
Lung function test results for the 10 SAL-treated subjects appear
in Table 3
. The results for these subjects were similar to those for the first
group studied. Two subjects had been hospitalized previously (one
subject had been intubated previously), one subject currently smoked,
and all subjects were receiving inhaled adrenergic or steroid
preparations. Prior to SAL inhalation, the mean FEV1
values, as measured both in liters (1.84 L) and as percent predicted
(66%), were no different from those for the first 30 study
subjects. Similarly, the values for airway resistance and
functional residual capacity (FRC) for the SAL-treated subjects were
similar to those of the 30 subjects who had received BD
treatment. No changes were observed in lung function after or
before SAL inhalation in these 10 SAL-treated subjects.
|
|
| Discussion |
|---|
|
|
|---|
Other authors have examined the relationship between airway function and dyspnea during the induction of bronchospasm. These studies have compared indexes of dyspnea rated first when the patient was in stable condition and rated again after the production of a 20% decline of FEV1 . Given these published data,22 it is possible that the change in FEV1 for our study group (10% of a predicted value) was too small to elicit a change in the perception of dyspnea in our subjects.
The decrease of Borg scale score without a concomitant change in FRC deserves comment. The elevation of FRC by 0.25 L in healthy subjects (which is similar to the change in FRC in our subjects) by positive end-expiratory pressure administered via a mouthpiece is associated with an increase of 1 U on the Borg scale.23 When, in another study,22 bronchoconstriction was induced in patients in stable condition with mild asthma, the change in end-expiratory lung volume rather than the change in airways resistance was the strongest predictor of change in the perception of dyspnea. Bronchodilation (ie, the opposite of bronchoconstriction) is accompanied by the withdrawal of the stimulus for dyspnea; that is, a lessening of the displacement of inspiratory muscles from their normal resting position occurs. Our data may differ from those of others cited in that in those studies the change in dyspnea was brought about by an increase, not a decrease, in airways resistance and lung volume. To our knowledge, however, there is no reason that the changes in the magnitude of dyspnea perceived by a subject should depend on the direction (eg, bronchoconstriction or bronchodilation) of the change in airway function.
Given the inability of a change in FEV1 to predict a change in Borg scale score in this study group, the correlation between the change in sGaw and the change in Borg scale score is of interest. Dynamic airway compression, a physiologic event that is reflected by the FVC maneuver, is thought to be important in the genesis of the sensation of dyspnea. It is unlikely, however, that patients whose FEV1 values exceed 60% of predicted would experience dynamic airway compression during quiet breathing.24 An alternate explanation for our findings might be that the magnitude of the sensation of dyspnea is related to the amplitude of the tidal swing of esophageal pressure.22 This measure of inspiratory pressure would clearly be related to airway conductance but not to FEV1 . The sensation of dyspnea during bronchospasm is mediated by afferent sensory airway receptors. These receptors would likely monitor changes in the tidal swing of esophageal pressure during quiet breathing.25 26 Our study supports the notion of Noseda et al27 that the sGaw measure is the most appropriate measure with which to predict the magnitude of the change of dyspnea accompanying a therapeutic intervention in asthmatic patients with mild airway obstruction.
Dyspnea and Indexes of Psychological State or Trait
Our data do not support the hypothesis that a measure of
somatization will reduce the unexplained variability between any
measure of airway function and the degree of dyspnea reported by
asthmatic subjects. We used two somatization scales (the Barsky SSAS
and the SCL-90-R) because the published data show poor correlation
between these scales. Our correlation between the data derived from the
two instruments (r = 0.43) is similar to that previously
observed.28
This discrepancy between the two instruments
is not unexpected. The Barsky SSAS is designed to assess the awareness
of multiple somatic symptoms, while the SCL-90-R assesses the degree of
physical discomfort associated with a symptom. Neither instrument
served as a second independent variable with the change in sGaw to
predict the change in Borg scale score. This observation is compatible
with other data that show that there is poor correlation
(R2 = 0.16) between a measure of
hypochondriasis and disease severity in asthmatic
subjects.4
Our study group was limited to ambulatory asthmatic patients in stable condition. Because our patients were recruited solely from an outpatient setting, patients with severe airway obstruction were excluded. Observations from other studies29 30 suggest that patients presenting with symptoms that are suggestive of any disease but with no objective findings of that disease are likely to score highly on any somatization scale, whereas patients with objective findings show less of a tendency to somaticize. In this regard, our subjects scored 2.9 ± 0.7 on the Barsky SSAS and 1.1 ± 0.7 on the SCL-90-R. This mean value of 2.9 on the Barsky SSAS would be associated with a high degree of somatization, and, by contrast, the mean value of 1.1 for the SCL-90-R would describe a population less likely to somaticize.15 31 These observations taken together raise the possibility that a wider range of Barsky SSAS scores would be observed if the study group were composed of patients with a greater range of airway function.
Placebo Effect of Albuterol
Our observation (Fig 1)
that many subjects who did not demonstrate
a large increase in sGaw with albuterol nevertheless reported a
decrease in their Borg scale score prompted us to recruit 10 additional
subjects to inhale isotonic SAL (placebo). This group of 10 was matched
to 14 subjects from among the original 30 such that the two resultant
groups had similar initial sGaw values and similar changes in sGaw with
treatment. Table 4
shows that those in the original BD
(albuterol)-treated group were more likely to report symptomatic
relief, despite their lack of physiologic response, than those in the
SAL-treated group. This in fact may have represented a placebo
effect, in that subjects in the albuterol-treated group were told that
they would be receiving a BD while subjects in the SAL-treated group
were told that they would receive either BD or a
placebo.32
An alternate possibility is that albuterol has
some dyspnea-sparing effect beyond its effect as a BD. Selective
ß2-agonists increase the concentrations of
glucose, lactate, and free fatty acids in plasma.33
There
is evidence that during submaximal exercise, partial
ß2-agonist activity attenuates dyspnea when it
has been exacerbated by generalized ß-blockade.34
A
proposed mechanism for this is the increased fatty acid availability
associated with ß-agonism and its resultant glycogen-sparing effect.
Whether or not this mechanism would operate in this protocol when the
patients are breathing at rest and the drug has been administered by
aerosol is unknown.
In summary, this study showed that asthmatic patients with mild airway
obstruction are likely to associate the relief of dyspnea with BD
therapy. For those subjects whose sGaw level increases
0.1 U after
bronchodilation, a linear relationship was seen between the magnitude
of changes in conductance and Borg scale scores. By contrast, such a
relationship was not seen between the change in dyspnea rating and the
change in any spirometric measure of airway function. Some subjects
whose sGaw levels increased minimally (
0.1 U) with bronchodilation
nevertheless reported a decline in their dyspnea score; no subjects
receiving placebo reported such a change. Psychophysiologic measures of
somatization did not explain any variability between airway function
and dyspnea rating.
| Acknowledgements |
|---|
| Footnotes |
|---|
Dr. Natelson is supported by National Institutes of Health Center grant U01A132247.
Received for publication April 10, 2000. Accepted for publication March 2, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Han, Y. Zhu, S. Li, J. Zhang, X. Cheng, O. Van den Bergh, and K. P. Van de Woestijne The Language of Medically Unexplained Dyspnea Chest, April 1, 2008; 133(4): 961 - 968. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. von Leupoldt and B. Dahme Cortical Substrates for the Perception of Dyspnea Chest, July 1, 2005; 128(1): 345 - 354. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. H. Lavietes Listening to Our Patients Chest, June 1, 2005; 127(6): 1877 - 1878. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |