Printer Friendly

The value of the chest x-ray in making the diagnosis of bronchial asthma.

INTRODUCTION

Although bronchial asthma is one of the most common chronic illnesses of children and young adults (Kaliner, Eggleston, & Mathews, 1987; Proceedings of the Asthma Mortality Task Force, 1987), with a prevalence of 3-5% (Proceedings of the Asthma Mortality Task Force, 1987), it remains underdiagnosed (Anderson, Bailey, Cooper, et al., 1981; Spaight, Lee, & Hey, 1983). Underdiagnosis of asthma can lead to excess morbidity as well as missed days of school or work. Any test that would enhance diagnosis clearly would be of value.

Currently, spirometry is the standard diagnostic test for bronchial asthma at the primary care level. However, between 1952 and 1974, the chest X-ray also was considered important in making the diagnosis of asthma (Royle, 1952; Simon, 1970; Rebuck, 1970; Simon, Connolly, Littlejohns, et al., 1973; Hodson, Simon, & Batten, 1974; Eggleston, Ward, Pierson, et al., 1974); but since 1976, the chest X-ray as a useful diagnostic test has fallen into disfavor (Snider, 1976; Janower, 1976; Findley & Sahn, 1981; Zieverink, Harper, Holden, et al., 1982; Gershel, Goldman, Stein, et al., 1983; Kaliner, Eggleston, & Mathews, 1987). For example, in 1987, the Primer on Allergic and Immunologic Diseases asserted that "chest roentgenograms are not particularly valuable . . . |and~ are indicated only when complications are suspected" (Kaliner, Eggleston, & Mathews, 1987). In 1991, the U.S. Department of Health and Human Services issued a special expert panel report (Guidelines for the Diagnosis and Management of Asthma, 1991); its judgment on the chest X-ray was no less favorable: "|The chest X-ray is useful only to~ rule out other causes of airway obstruction . . . |which~ is especially important for children." It is curious that prior to 1976, the chest X-ray was considered valuable in making the diagnosis of asthma, but since 1976 it has not been considered valuable even though no evidence was produced to evoke this radical change in view. Clearly, the value of the chest X-ray for diagnosing asthma needs to be resolved.

The purpose of this paper is to assess the value of the chest X-ray. This was done by comparing chest X-ray and spirometric findings among asthmatic, bronchitic, and normal patients with the goal of improving the sensitivity of diagnostic criteria in making the diagnosis of bronchial asthma at the primary care level.

METHOD

We studied a nonselected sample of the medical records of three groups of patients, aged 18-40, seen at the Harvard University Health Services (UHS) from 1982 to 1987, in which there were reports in the records of chest X-rays taken at UHS during the same period and in which one of the following three diagnoses was noted: (1) bronchial asthma, (2) bronchitis, or (3) normal or healthy. The group of patients with bronchial asthma (also known as "asthma," "bronchospasm," or "reactive airway syndrome") consisted of 70 cases. Curiously, the diagnosis was not made on the basis of spirometric findings or on chest X-ray findings; in fact, in not a single case did a clinical note even make reference to these. Rather it seemed to have been made on the basis of symptomatology and/or clinical response to bronchodilators or anti-inflammatory agents. We made no attempt to include or exclude patients on the basis of any set of diagnostic criteria; we simply accepted the diagnoses of the UHS clinicians who had taken the histories and examined the patients. We did exclude patients with the following diagnoses: specifically induced asthmatic syndromes (such as exercise-induced, seasonally associated, or cat-induced asthma), cardiac asthma, asthma with associated pectus excavatum, "chest pain syndrome," "chest wall syndrome," pneumonia, pneumothorax, and pneumomediastinum.

Our group of asthmatic patients may best be described as having mild to moderate asthma because only 18 of 70 (26%) had a history of ever coming to an emergency room for asthma; only 7 of these 18, or 7 of 70 (10%), had a history of ever having been hospitalized for asthma; only 10 of 70 (14%) had ever used oral steroids; and none had been referred to a pulmonologist or an asthma treatment center. This group is to be clearly differentiated from the group of severe asthmatic patients of whom most, if not all, have come on at least one occasion to emergency rooms and/or have been admitted to hospitals, and most of whom use steroids as part of their routine treatment.

The group with bronchitis (acute or recurrent) consisted of 40 patients. (Patients with "asthmatic bronchitis" and "chronic bronchitis" were excluded.) The group of normal or healthy patients consisted of 41 cases without respiratory symptoms. These patients came to UHS for a routine physical examination or an examination required for joining the Peace Corps, the Harvard University Police Department, or for immigration or emigration. Occasionally, concurrent nonrespiratory diagnoses were also present in all three groups, such as hemangioma, lipoma, anxiety, concussion, paroxysmal atrial tachycardia, obesity, dysmenorrhea, gastroenteritis, or cystitis. None of these conditions was used as grounds for exclusion.

We looked for "abnormal" chest X-ray findings in all groups. An abnormal chest X-ray was defined as one having (1) "increased markings" and/or (2) "low diaphragm." Some radiologists use the terms "peribronchial" or "peribronchiolar cuffing" instead of increased markings, and "hyperinflation" instead of low diaphragm. The chief of radiology at UHS, a coauthor of this paper, has chosen not to use the latter terms on grounds that they are based upon alternative assumptions, which, although possibly correct, have not been confirmed by scientific studies. The same radiologist, who read most of the X-rays in the study, was unaware that a hypothesis was being tested or that an experiment was in progress at the time of the study. (When all the data were collected and analyzed and reported to her, she was just as astonished by the results as were the other authors.) Finally, mild orthopedic findings on the chest X-ray, such as mild scoliosis, kyphosis, or lordosis, not uncommon findings, were for the purposes of this study classified as normal, and were not excluded.

Comparisons were made on X-ray findings between (1) the bronchial asthma and normal groups, (2) the bronchitis and normal groups, and (3) the bronchial asthma and bronchitis groups. The three comparisons were made by means of the Mantel-Haenszel test (Rosner, 1990) after stratifying the data by age and sex thus: 18- to 29-year old males, 18- to 29-year-old females, 30- to 40-year-old males, and 30- to 40-year old females. Comparisons were also made in the asthmatic group between positive and negative spirometric and positive and negative X-ray findings by means of McNemar's test for correlated proportions (Rosner, 1990). Abnormal spirometry was defined in terms of abnormalities in the forced one-second expiratory volume (|FEV.sub.1~) and/or the peak expiratory flow rate (PEFR). A positive (abnormal) |FEV.sub.1~ was defined as an |FEV.sub.1~ that was less than 80% of that predicted from standard normal tables and/or an |FEV.sub.1~ that increased more than 20% after administration of a bronchodilator. A positive (abnormal) PEFR was defined as one that was less than 85% of that predicted from standard normal tables. Spirometry test results were considered abnormal if either the |FEV.sub.1~ or the PEFR was abnormal; in other words, in those cases where both an |FEV.sub.1~ and a PEFR test result were recorded and the |FEV.sub.1~ was abnormal but the PEFR was normal (or vice versa), spirometry was considered positive (abnormal). The study was therefore biased in favor of positive rather than negative spirometry.

The study was approved by the administration of the UHS and the chairman of the UHS Research Committee, who agreed that it did not need special approval of either the UHS Research Committee or the Harvard University Committee on the Use of Human Subjects in Research, since human rights--specifically, informed consent, confidentiality, and patient risk--were not at issue.

RESULTS

Table 1 shows the overall chest X-ray findings stratified by age and sex. In the group with bronchial asthma (BA), 52 out of 70 patients (74%) were males; in the group with acute bronchitis (B), 16 out of 40 patients (40%) were males; and in the normal group (N), males (51%) and females (49%) were in about equal numbers. Among 18- to 29-year-old males, abnormal findings were present in 63% of the BA group, 55% of the B group, and 0% of the N group. Among 18- to 29-year-old females, abnormal findings were present in 33% of the BA group, 39% of the B group, and 0% of the N group. Similar differences were apparent among the three diagnostic groups for 30- to 40-year-old patients. The Mantel-Haenszel test revealed that both the BA (p |is less than~ .001) and the B (p |is less than~ .001) groups had significantly more abnormal findings than the N group, but the BA group did not differ significantly from the B group.
TABLE 1 Overall Chest X-Ray Findings Stratified by Age and Sex

 Age

 18-29 30-40

 Ab(**) Total (%) Ab(**) Total (%)

 Males

BA(*) 24 38 (63) 11 14 (79)
B(*) 6 11 (55) 3 5 (60)
N(*) 0 16 (0) 0 5 (0)

 Females

BA 3 9 (33) 3 9 (33)
B 7 18 (39) 1 6 (17)
N 0 17 (0) 0 3 (0)

Comparison of Specific Groups(***)

BA vs N Chi-square = 30.64 p |is less than~ .001
B vs N Chi-square = 20.12 p |is less than~ .001
BA vs B Chi-square = 0.28 NS

* BA = bronchial asthma, B = acute bronchitis, N = normal.

** Ab = abnormal. An abnormal chest X-ray is defined as one
having increased markings and/or low diaphragm.

*** Comparison of specific groups, by Mantel-Haenszel test,
after controlling for age-sex.


Findings for "increased markings" are shown in Table 2. Among 18- to 29-year-old males, 45% of the BA group, 45% of the B group, and 0% of the N group had increased markings. Among 18- to 29-year-old females, 22% of the BA group, 39% of the B group, and 0% of the N group had increased markings. Similar results were found for the 30- to 40-year-old age group. The Mantel-Haenszel test showed that both the BA (p |is less than~ .001) and the B (p |is less than~ .001) groups had a significantly greater prevalence of increased markings when compared with the N group, but no significant differences were apparent between the BA and B groups.

Table 3 shows the findings for "low diaphragm." Low diaphragm was not as common a finding as increased markings. Among 18- to 29-year-old males, 24% of the BA group, 18% of the B group, and 0% of the N group had low diaphragm. Among 18- to 29-year-old females, 22% of the BA group, 0% of the B group, and 0% of the N group had low diaphragm. The Mantel-Haenszel test revealed significant differences between the BA and N groups (p = .005), but no significant differences between the B and N groups (p = .12) or between the BA and B groups.
TABLE 2 Chest X-Ray Finding of Increased Markings Stratified by
Age and Sex

 Age

 18-29 30-40

 Ab(**) Total (%) Ab(**) Total (%)

 Males

BA(*) 17 38 (45) 9 14 (64)
B(*) 5 11 (45) 2 5 (40)
N(*) 0 16 (0) 0 5 (0)

 Females

BA 2 9 (22) 1 9 (11)
B 7 18 (39) 1 6 (17)
N 0 17 (0) 0 3 (0)

Comparison of Specific Groups(***)

BA vs N Chi-square = 17.60 p |is less than~ .001
B vs N Chi-square = 16.56 p |is less than~ .001
BA vs B Chi-square = 0.004 NS

* BA = bronchial asthma, B = acute bronchitis, N = normal.

** Ab = abnormal. An abnormal chest X-ray is defined as one
having increased markings and/or low diaphragm.

*** Comparison of specific groups, by Mantel-Haenszel test,
after controlling for age-sex.
TABLE 3 Chest X-Ray Finding of Low Diaphragm Stratified by Age
and Sex

 Age

 18-29 30-40

 Ab(**) Total (%) Ab(**) Total (%)

 Males

BA(*) 9 38 (24) 3 14 (21)
B(*) 2 11 (18) 1 5 (20)
N(*) 0 16 (0) 0 5 (0)

 Females

BA 2 9 (22) 2 9 (22)
B 0 18 (0) 1 6 (17)
N 0 17 (0) 0 3 (0)

Comparison of Specific Groups(***)

BA vs N Chi-square = 7.86 p = .005
B vs N Chi-square = 2.46 p = .12
BA vs B Chi-square = 0.75 NS

* BA = bronchial asthma, B = acute bronchitis, N = normal.

** Ab = abnormal. An abnormal chest X-ray is defined as one
having increased markings and/or low diaphragm.

*** Comparison of specific groups, by Mantel-Haenszel test,
after controlling for age-sex.


Since our data showed that the chest X-ray was useful in discriminating patients with bronchial asthma from normal patients (and also patients with bronchitis from normal patients), we proceeded to investigate the relative sensitivity of the chest X-ray versus spirometry in patients with bronchial asthma. In the total sample of 70 patients with bronchial asthma, 58 had both a chest X-ray and spirometry recorded in the medical records; the remaining 12 had one or the other or neither. A comparison of chest X-ray and spirometric findings is presented in Table 4.

Of these asthmatic patients, 21 of 58 (36%) were positive for spirometry, 34 of 58 (59%) had positive chest x-rays, and 44 of 58 (76%) had either positive spirometry or a positive chest X-ray. Further, 23 cases had positive chest X-rays and negative spirometry, whereas only 10 cases had negative chest X-rays and positive spirometry. McNemar's test showed this to be a significant (p = .03) difference. This indicates that the chest X-ray may be a more sensitive test than spirometry among persons with mild asthma. Unfortunately, but not unexpectedly, spirometric data were not available among patients in the bronchitis or normal groups, thus making it impossible to assess the relative specificity of the two tests.
TABLE 4 Comparison of Spirometric and Chest X-Ray Findings in
Asthmatic Patients

Spirometry Chest X-ray n Chi-square(*) p

Abnormal Abnormal 11

Abnormal Normal 10
 4.73 .03
Normal Abnormal 23

Normal Normal 14

* Determined by McNemar's test of correlated proportions by
comparing the discordant pairs (10 vs 23)


DISCUSSION

Although spirometry is valued highly in making the diagnosis of bronchial asthma at the primary care level, we found that it had surprisingly limited sensitivity in a nonselected population of mild to moderate asthmatic patients. In 1991, a statement to this effect was made by an expert panel from the National Institutes of Health (Guidelines for the Diagnosis and Management of Asthma, 1991), but the statement was not supported by data or references. (It is curious that our paper, which contains data to support such a statement, was rejected in the period 1988-1992 by the Lancet, JAMA, Radiology, and other journals for one of three reasons: (1) spirometry is the sine qua non for making the diagnosis of asthma, (2) a chest X-ray could not possibly have any value in making the diagnosis of a pathophysiological illness such as asthma |asthma was not discovered to be pathoanatomical as well as pathophysiological until 1989 (Barnes, 1989)~, (3) a prospective study is superior to a retrospective one.) Only 21 of 58 (36%) asthmatic patients were positive for spirometry. On the other hand, 34 of 58 (59%) had abnormal chest X-rays. The diagnosis of asthma was made in these cases not on the basis of spirometry or chest X-ray findings but rather on the basis of symptomatology and clinical response to asthma therapy.

The finding of low diaphragm has generally been associated with patients with emphysema, a disease heretofore associated with an older age group than the patients in our study. Our study showed that this finding was neither specific for emphysema nor characteristic only of the elderly. A similar observation was made by Royle (1952) in an uncontrolled study of patients with bronchial asthma. It is unfortunate that a smoking history was not taken in the medical records of our patients, since it would be interesting to learn whether low diaphragm is as characteristic of young smokers as it seems to be of elderly smokers. Or, is smoking irrelevant, and is low diaphragm, contrary to current belief, specific not for emphysema but for asthma? And when we see low diaphragm in a patient with emphysema, should it make us think seriously of concurrent asthma? Whether low diaphragm and increased markings are reversible in young adults with asthma, as has been suggested (Simon, 1970; Rebuck, 1970), is another question that needs to be answered. Finally, does the similarity of chest X-ray findings in bronchitic and asthmatic patients imply that bronchitis is a variant of asthma (or vice versa)?

At the clinical level, even though in our study the chest X-ray did not discriminate between asthma and bronchitis (although there was a suggestion that low diaphragm was more characteristic of asthma than of bronchitis), clinicians should, in most cases, be able to discriminate between the two diseases. Acute bronchitis is usually flagrant, since it has the characteristic symptom of relentless coughing of colored--green, yellow, brown, or grey--sputum, whereas mild asthma is usually subtle, since it has variable symptoms--dry cough, chest tightness, chest ache, or shortness of breath. Further, in mild asthma, the symptoms are not only variable, but also intermittent or recurrent, all of which may frequently account for many capable clinicians' failure to make the diagnosis. We suggest to those clinicians who are not specialists in allergy or pulmonary medicine that when they come across a patient in whom they have made the diagnosis of acute bronchitis but the patient does not respond, or only partially responds, to antibiotic therapy, or in whom the bronchitis is recurrent, they should entertain the diagnosis of mild asthma and give the patient a trial of bronchodilator and/or anti-inflammatory treatment. These physicians will be as gratified as they are astonished to see how many of such patients respond to such treatment. Naturally, prior to the initiation of treatment, the physician should also make a spirometry measurement and take a chest X-ray.

How does one explain that the chest X-ray was considered valuable between 1952 and 1974 (Royle, 1952; Simon, 1970; Rebuck, 1970; Simon, Connolly, Littlejohns, et al., 1973; Hodson, Simon & Batten, 1974; Eggleston, Ward, Pierson et al., 1974), but since 1976 has been considered not useful at all (Snider, 1976; Janower, 1976; Findley & Sahn, 1981; Zieverink, Harper, Holden, et al., 1982; Gershel, Goldman, Stein, et al., 1983; Kaliner, Eggleston, & Mathews, 1987)? We suggest that the fall from favor of the chest X-ray has come about for two separate reasons: first, because of an increasing faith in, and reliance upon, spirometry, and second, because of a failure to discriminate between using the X-ray for making the diagnosis of asthma and using it in acute crisis intervention during an asthmatic attack.

Until recently, crisis intervention was a major focus of investigation. Crisis intervention concerned the treatment of patients with severe asthma who presented to the emergency room (Findley & Sahn, 1981; Zieverink, Harper, Holden, et al., 1982; Gershel, Goldman, Stein, et al., 1983; Kaliner, Eggleston, & Mathews, 1987; Proceedings of the Asthma Mortality Task Force, 1987). By the time the diagnosis of severe asthma is made in the emergency room (which in most cases occurs even as the patient is being admitted), an X-ray becomes superfluous except "when complications |are~ suspected" (Kaliner, Eggleston, & Mathews, 1987). Gershel et al. (1983) did not publish their own chest X-ray data in their acute intervention study on grounds that their X-ray findings were irrelevant, being either "benign" or "negative." Further, they advised that if a chest X-ray did not provide guidance in the treatment of severe asthma, it had no value at all, but merely exposed patients to unnecessary radiation. This point of view had already been expressed by other investigators also studying patients with severe asthma who presented to emergency rooms (Findley & Sahn, 1981; Zieverink, Harper, Holden, et al., 1982), and this view prevailed when a task force convened in 1986 for the purpose of studying asthma mortality (Proceedings of the Asthma Mortality Task Force, 1987). When discussing methods for diagnosing asthma, the task force did not mention the chest X-ray at all.

A study by the British Thoracic Association (BTA; 1982) found that most individuals who died of asthma throughout Great Britain did not appear at emergency rooms during their fatal attack; neither did they all have severe antecedent illness (surprisingly, about 10% had mild asthma), nor was the diagnosis always self-evident. The BTA concluded that failure to diagnose asthma precludes effective treatment. And yet, the BTA, too, did not mention the chest X-ray as a possible tool in diagnosis.

We suggest, on the basis of our study, that the chest X-ray may be useful in identifying patients with mild asthma. A reasonable question is: how can an anatomical test (the chest X-ray) have any value in diagnosing bronchial asthma, a disease that is pathophysiological and as a result should require a physiological test (spirometry) to measure its presumed basic pathophysiological mechanism, bronchospasm? A reasonable explanation is that asthma is at least as pathoanatomical as it is pathophysiological; that is, inflammation may be as important an element in the illness as is bronchospasm (Barnes, 1989). We suggest that the increased markings of the chest X-ray may be a manifestation of this inflammatory response. In any event, further study involving larger populations is needed to reproduce the findings reported here and to assess the relative benefit of the chest X-ray as compared with spirometry in helping to make the diagnosis of mild asthma.

REFERENCES

Anderson, H. R., Bailey, P. A., Cooper, J. S., et al. (1981). Influence of morbidity, illness label, and social, family, and health service factors on drug treatment of childhood asthma. Lancet, 2, 1030-1032.

Barnes, P. J. (1989). A new approach to the treatment of asthma. New England Journal of Medicine, 321, 1517-1527.

British Thoracic Association. (1982). Death from asthma in two regions of England. British Medical Journal, 285, 1251-1255.

Eggleston, P. A., Ward, B. H., Pierson, W. E., et al. (1974). Radiographic abnormalities in acute asthma in children. Pediatrics, 54, 442-449.

Findley, L. J., & Sahn, S. A. (1981). The value of chest roentgenograms in acute asthma in adults. Chest, 80, 535-536.

Gershel, J. C., Goldman, H. S., Stein, R. E. K., et al. (1983). The usefulness of chest radiographs in first asthma attacks. New England Journal of Medicine, 309, 336-339.

Guidelines for the diagnosis and management of asthma (1991). National Asthma Education Program, Expert Panel Report; National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland. Publication No. 91-3042 (1991), 7.

Hodson, M. E., Simon, G., & Batten, J. C. (1974). Radiology of uncomplicated asthma. Thorax, 29, 296-303.

Janower, M. L. (1976). Radiographic findings in asthma. In B. Weisse & M. S. Segal (Eds.), Bronchial asthma. Boston: Little, Brown.

Kaliner, M., Eggleston, P. A., & Mathews, K. P. (1987). Rhinitis and asthma: Primer on allergic and immunologic diseases. Journal of the American Medical Association, 258, 2851-2873.

Proceedings of the Asthma Mortality Task Force. (1987). Journal of Allergy and Clinical Immunology, 80(3, Pt. 2), 361-514.

Rebuck, A. S. (1970). Radiological aspects of severe asthma. Australian Radiology, 14, 264-268.

Rosner, B. (1990). Fundamentals of biostatistics (3rd ed.). Boston: PWS Publishers.

Royle, H. (1952). X-ray appearances in asthma: A study of 200 cases. British Medical Journal, 1, 577-580.

Rubenstein, H. S., Rosner, B. A., LeMay, M., & Neidorf, R. (1992). Value of the chest X-ray in making the diagnosis of asthma. Journal of Allergy and Clinical Immunology, 89(1, Pt. 2), 332.

Simon, G. (1970). Regression of radiological changes in asthma following treatment. Respiration, 27(Suppl.), 147-151.

Simon, G., Connolly, N., Littlejohns, D. W., et al. (1973). Radiological abnormalities in children with asthma and their relation to the clinical findings and some respiratory function tests. Thorax, 28, 115-123.

Snider, G. L. (1976). The interrelationships of asthma, chronic bronchitis, and emphysema. In B. Weisse & M. S. Segal (Eds.), Bronchial asthma. Boston: Little, Brown.

Spaight, A. N. P., Lee, D. A., & Hey, E. N. (1983). Underdiagnosis and undertreatment of asthma in childhood. British Medical Journal, 286, 1253-1256.

Zieverink, S. E., Harper, A. P., Holden, R. W., et al. (1982). Emergency room radiography of asthma: An efficacy study. Radiology, 145, 27-29.

Howard S. Rubenstein, M.D., Physician and Chief of Allergy, Harvard University Health Services, Cambridge, Massachusetts.

Bernard A. Rosner, Ph.D., Associate Professor, Department of Preventive Medicine (Biostatistics) and Medicine, Harvard Medical School, Boston, Massachusetts.

Marjorie LeMay, M.D., Chief of Radiology, Harvard University Health Services, Cambridge, Massachusetts.

Rhoda Neidorf, B.S., Chief of the Department of Medical Records, Harvard University Health Services, Cambridge, Massachusetts.
COPYRIGHT 1993 Libra Publishers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rubenstein, Howard S.; Rosner, Bernard A.; LeMay, Marjorie; Neidorf, Rhoda
Publication:Adolescence
Date:Sep 22, 1993
Words:4065
Previous Article:The Silent Twins.
Next Article:An ecological approach to evaluating a special education program.
Topics:

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters