Brief Communications
_
Chest Radiography and High Resolution Computed Tomography of the Lungs in Asthma1,2
F. PAGANIN, V. TRUSSARD, E. SENETERRE, P. CHANEZ, J. GIRON, P. GODARD, J. P. SENAC, F. B. MICHEL, and J. BOUSQUET
In 1962, the American Thoracic Society defined asthma as a reversible airway disease with increased airway responsiveness, and it noted that some asthmatics, particularly those who have had asthma for many years, may have persistent airflow obstruction (1). The pathophysiologic mechanisms of asthma include bronchospasm, hypersecretion, and chronic airway inflammation, but, more recently, attention has been focused on bronchial fibrosis and structural defects of the bronchi, implying a third component of asthma, i.e., "bronchial destruction." This component, already proposed in the ATSdefinition in 1962, has been clinically suggested by the "accelerated decline" of the pulmonary function of asthmatic patients as well as by the poor reversibility of airway obstruction in patients with a long history of the disease. However, many patients with uncomplicated bronchial asthma have a normal chest roentgenogram and do not show structural lesions. Computerized tomography (CT scan) provides a high degree of anatomic detail, and it has been used in the diagnosis of airway diseases such as emphysema (2) and bronchiectasis (3). However, CT scans in nonABPA (allergic bronchopulmonary aspergillosis) asthmatic patients with chronic obstructive pulmonary disease have been examined in only one study, where the focus was on emphysema (4). In an attempt to better characterize the possible bronchial abnormalities of asthma and to discover whether irreversible lesions are associated with this disease, high resolution CT scans with millimetric slices were examined in 57 asthmatic patients with variable disease severity, and in 10 sex- and age-matched normal subjects. None of these participants smoked. Ten subjects had a CT scan during asthma exacerbation and again after appropriate treatment to define labile abnormalities. We studied 61 asthmatic patients IS to 67 yr of age (mean ± SO, 37.6 ± 14.8 yr). Patients were randomly chosen for the study, and there was no particular selection criteria. Asthma was defined according to the criteria of the American Thoracic Society (l), and all patients had reversible airway obstruction characterized by an increase of 15070 of FEY 1 after inhalation of 200 ug of salbutamol or a positive inhalation challenge with carbachol. None of the subjects had a history of allergic bronchopulmonary aspergillosis as defined by the crite-
1084
SUMMARY CTscans have been studied only In asthmatics who were smokers, and no such study has been performed in patients with chronic uncomplicated asthma where a permanent bronchial destruction is likely to occur after a long course of the disease. The object of the study was to characterize CT-scan abnormalities and determine whether bronchial destructive lesions may be observed. Fifty-seven adults with chronic asthma of variable severity and etiology and 10 normal subjects were studied. None of the subjects smoked. Chest radiographs and HR·CT scans were performed in all patients. Todiscriminate between reversible and Irreversible CT-scan abnormalities, two examinations were made In 10 patients with acute asthma both before and 2 wk after parenteral high dose corticosteroid treatment. The chest radiographs showed the expected abnormalities of asthma in 37.8% of the asthmatics. CT scans were abnormal In 71.9% of the asthmatics. Reversible abnormalities Included mucoid impactions, acinar pattern, and lobar collapse. Irreversible abnormalities included bronchiectasis, bronchial wail-thickening, sequellar line shadows, and emphysema. Most of these abnormalities are likely to be related to bronchial destruction. AM REV RESPIR DIS 1992; 146:1064-1087
TABLE 1 ROENTGENOGRAPHIC ABNORMALITIES IN ASTHMA" Aas Score
Patients tested, n Normal chest radiograph
2
3
15 10 (66%)
20 13 (65%)
0 0 0 0
0 1 (6%) 0 0
0 0 0 2 (10%)
0 0 0 3 (23%)
2 (22%)
4 (26%)
7 (35%)
6 (46%)
0 0
0 0
0 0
0 2 (15%)
9 7 (78%)
4-5 13 6
(46%)
Mucoid impaction Pneumomediastinum Pneumothorax Consolidation andlor infiltrates (23%)
Bronchial wall-thickening (46%)
Bronchiectasis Emphysema (15%)
• Results are expressed in number and in percentage of patients (shown in parentheses).
ria of Rosenberg and coworkers (5). In particular, they had no serum Aspergillus precipitins or IgE (Phadebas RAST~, Pharmacia Diagnostics AB, Uppsala, Sweden), and skin test results with a nonstandardized Aspergillus allergen extract (Stallergenes Laboratories, Fresnes, France) were negative. The clinical severity of chronic asthma was quoted according to the score of Aas, which is used to grade chronic asthma from very mild forms (score of I) to incapacitating disease requiring medication (score of 5) (6). Ten normal healthy volunteers 22 to 60 yr of age (mean ± SO, 40.1 ± 14.1 yr) were used as a control group. Their pulmonary function was within normal range. They were nonallergic and had never had asthma.
None of the subjects or patients tested were current or previous smokers. The study was carried out after informed consent of the patients and approval by the Ethics Committee of the Hospital were obtained.
(Received in original/arm August 13, 1991 and in revised form April 27, 1992) 1 From the Clinique des Maladies Respiratoires and the Departernent de Radiologie, MontpellierCedex, France. 2 Correspondence and requests for reprints should be addressed to Dr. J. Bousquet, Clinique des Maladies Respiratoires, 34059 MontpellierCedex, France.
BRIEF COMMUNICATION
1085 TABLE 2 COMPUTED TOMOGRAPHIC ABNORMALITIES IN ASTHMA' Aas Score 2
Patients, n Normal CT scan Mucoid impaction Acinar pattern Collapse Bronchiectasis Cylindrical Varicose Cystic Bronchial wall-thickening Emphysema Paraseptal Centrolobular Bullous Pneumomediastinum Pneumothorax
9 5 (55.5%) 1 (11%) 0 0
4 (44%)
15 6 2 6 2
(40%) (15%) (40%) (15%)
3 20 3 6 8 4
4-5
(15%) (30%) (40%) (20%)
13 4 3 5 1
(31%) (23%) (38%) (8%)
0 0 0
6 (40%) 0 0 2 (15%)
16 (800/0) 3 (15%) 0 4 (20%)
6 1 1 3
(46%) (8%) (8%) (23%)
0 0 0 0 0
0 0 0 1 (6.6%) 0
1 (5%) 1 (5%) 0 1 (5%) 0
4 3 1 1 0
(31%) (23%) (8%) (7.6%)
, Results are expressed as the nu mber of patients, with percentages of patients shown in parentheses.
Posteroanterior and lateral chest radiographs were made for all patients, and they were examined blind by three independent readers (FP, VT, and ES). Pulmonary hyperinflation was not considered to be an abnormality because of its common occurrence in asthmatics and its lack of specificity. CT scans of the chest were performed using a Somaton DR H (Siemens, Munich, Germany) in high resolution technique according to Mayo and coworkers (7). The matrix size was 512 x 512 at a pixel size of 0.5 mm. Gantry (20 degrees) was used to improve the ability to visualize the subsegmental bronchi, and scanning time ranged from 2.5 to 3 s. The patients wereexamined in the supine
position during mild inspiration, with their arms over their heads. Images were recorded at a window width of 1,600HU and a window levelof - 600 HU and intravenous contrast media were never administered. Section cut 1mm thick at I5-mm increments were obtained through the lungs. The scans wereinterpreted independently by three doctors who had no knowledge of the disease or of the chest radiographs (FP, VT, and ES). agreement among the three reviewers was considered to be satisfactory. The first assessment was a subjective examination of the quality of the test followed by an evaluation of each lobe of the pulmonary parenchyma. The images were interpreted as follows. The diagnosis of bronchiectasis and mucoid impactions was
based on the criteria of Naidich and coworkers (8) and Grenier and colleagues (3), with high resolution CT scans. Emphysema was characterized according to Foster and coworkers (9), whereas atelectasis, acinar pattern (10), and proximal wall-thickening were defined according to Naidich and colleagues (Il). Bronchial wall-thickening and/or peribronchial fibrosis (11, 12) were characterized by CT scans according to Naidich and colleagues (11) and Zerhouni (l2). Other abnormalities assessed were pneumothorax and pneumomediastinum. All subjects underwent the same radiographic and CT-scan examinations. Moreover, since some lesions are likely to be reversible, two CT scan examinations were made with a 2-wk interval between them in 10 patients with acute asthma who had undergone systemic corticosteroid treatment (daily dose of methylprednisolone, 2 mg/kg of body weight) associated with intravenously administered bronchodilators (salbutamol, 0.05 mg/kg of body weight). CT scan lesions were classified as either permanent or reversible and correlated (1) using Kendall's 't test, with the severity of asthma defined by the clinical score of Aas, and (2) by Spearman's rank test, with the duration of asthma or FEV,.
Among the 61 asthmatic patients included in the study, only 57 were studied because disagreements were noticed among reviewersin four patients. Among the patients studied, nine had Aas scores of 1, 15 had scores of 2, 20 had scores of 3, and 13 had scores of 4 or 5. FEV, ranged from 31 to 106070 of predicted values (mean ± SD, 70.12 ± 22.9%). The etiologic investigations showed that 63.1% of patients had perennial allergy, 5.2% had premenstrual syndrome, 21% had chronic sinusitis, 5.2% had aspirin intolerance, and 5.2% had gastroesophageal reflux; 14% wereconsidered to have intrinsic asthma. Chest radiographs werecompletely normal in all the control subjects and in 63.1% of the asthmatics (table 1). CT scans were completely normal in all the control subjects and in 28.1% of the asthmatics (table 2). When CT scans were compared with chest radiographs there was a significantly (p < O.ot,chi-square test) increased number of asthmatics who had abnormal imaging patterns. All CT scan patterns, except pneumothorax, were observed in the asthmatics (figures 1 to 3). In the 10 asthmatics who had undergone
TABLE 3 EVOLUTION OF COMPUTED TOMOGRAPHIC ABNORMALITIES IN 10 PATIENTS AFTER CORTICOTHERAPY' Before Mucoid impaction Acinar pattern Lobar collapse Bronchiectasis Cylindrical Varicose Cystic Bronchial wall-thickening Emphysema Fig. 1. HR-CT scan of asthmatic patient (Aas score, 3) showing an acinar pattern of left lower lobe and filled cylindric bronchiectasis of the right lower lobe.
3 (2) 2 (2) 5 (3) 25 11 0 16 12
(7) (3) (4) (4)
After 0 0 0 25 11 0 16 12
(7) (3) (4) (4)
, Results are expressed as cumulative numbers of lobes involved, with the numbers of patients shown in parentheses.
1086
BRIEF COMMUNICATION
two CT scans within a 1- or 2-wk interval, it was observed that mucoid impaction, acinar patterns, and lobar collapses were reversible abnormalities, as these disappeared in all subjects (table 3). However, there was no change in bronchiectasis (figures 1 and 2), bronchial wall-thickening, or emphysema. There was no significant correlation between the severity of asthma and CT scan abnormalities, but patients with the most severe asthma tended to display more irreversible abnormalities.
* * *
Fig. 2. HR-CT scan of the same patient as in figure 1 after 2 wk of 120 mg per day of methylprednisolone showing cylindric bronchiectasis but no acinar pattern.
Fig. 3. HR-GT scan of patient with severe asthma (Aas score, 4) showing cylindric and varicose bronchiectasis in both lower lobes and com plete destruction of the middle lobe.
This study showed the expected chest radiographic abnormalities in asthmatics, whereas CT scans revealed that a large proportion of the asthmatics had lesions. These lesions included such reversible abnormalities as mucoid impaction, acinar pattern, and lobar collapse, and irreversible abnormalities such as bronchiectasis, emphysema, and bronchial thickening. The use of high resolution CT scan in the diagnosis of lung diseases is now well established, and in this study none of the CT scans analyzed displayed any of the artifacts described by Traver and coworkers (13). Bronchiectasis has been studied by many investigators who compared CT scans and bronchography. Their findings showed that CT scans are specific in greater than 90010 and sensitive in greater than 95% of cases when the high resolution CT-scan technique was used (3). The diagnosis of emphysema assessed by pathologic examination was correlated with high resolution CT scans (2). In asthma, as suspected,abnormalities were more frequently observed on CT scans than on chest radiographs. CT scans often revealed reversible abnormalities that were not correlated with the severity of the disease. Mucoid impaction, acinar patterns, and lobar collapse disappeared after anti-inflammatory treatment. Radiographic studiesof the asthmatics have disclosed bronchial wall-thickening, and this has been considered as a possibly reversible abnormality (14). In the present study, CT scans revealed such lesions, but they there was no reversibility, suggesting that radiographs and CT scan images of bronchial thickening may represent different pathologic abnormalities.The lesions 0 bservedby radiography may correspond to bronchi in which the bronchial lumen is filled in by secretion. On the other hand, using C'I-scananalysis, bronchial wallthickening indicates the abnormal visualization of bronchi in the periphery of the lung (11, 12). The image of bronchi seen in this area may reflect either bronchial inflammation increasing the thickness of the bronchial wall or peribronchial fibrosis (11, 12). These identified CT-scan lesions were not reversible in the 10 patients treated by corticosteroids. Peribronchial fibrosis wasobserved by Soboya (15) in pathologic specimens of asthmatic lungs, suggesting that the second hypothesis is more relevant in this study. Other permanent abnormalities in asthma
1087
BRIEF COMMUNICATION
revealed by CT scans include bronchiectasis and, more rarely, emphysema. These lesions were observed in nonsmoking patients who had never suffered from the pulmonary diseases that might account for their presence (4, 10). In the present study, emphysema revealed by CT scan may not be due to a parenchymal destructive disease per se, but rather to an extensive peribronchial fibrosis that might relate to cicatricial emphysema since in the pathologic examination of asthmatic lungs, destructive emphysema has been observed in some studies (16). Bronchiectasis has been observed in the pathologic examination of asthmatic lungs (16) as wellas in livingasthmatics by bronchography, and, more recently, in a small series of patients by CT scan (17). However, in most of these studies it was not possible to determine the incidence of bronchiectasis in asthma. In the present study, we confirmed the presence of this abnormality in greater than 50070 of the patients. Cylindric bronchiectasis occurred frequently, and it was found in patients with variable severity. Varicose and cystic bronchiectasis were rare, and they were observed only in the patients with the most severe asthma. These may be sequelae of mucoid impactions and bronchial hypersecretion. This study has shown that destructive lesions are observed in asthma. Most of these lesions are related to bronchial abnormalities.
Acknowledgment The writers thank Dr. 1. P. Daures and 1. Y.Lacoste and all the staff of the Departement de Radiologie for their kind help, and Dr. A. M. Campbell for reviewing the manuscript.
References 1. Meneely GR, Rensetti AD 1r, Steele 10, Wyatt IP, Harris HW. American Thoracic Society:chronic bronchitis, asthma, and pulmonary emphysema. Am Rev Respir Dis 1962; 85:762-8. 2. Hruban RH, Meziane MA, Zehrouni EA, et al. High resolution computed tomography of inflation-fixed lungs. Pathologic-radiologic correlation ofcentrilobular emphysema. Am Rev Respir Dis 1987; 136:935-40. 3. Grenier P, Maurice F, Musset 0, Menu Y, Nahum H. Bronchiectasis: assessment by thin-section CT. Radiology 1986; 161:95-9. 4. Kinsella M, Muller NL. Hyperinflation in asthma and emphysema. Assessmentby pulmonary function testing and computed tomography. Chest 1988; 94:286-9. 5. Rosenberg M, Patterson R, Mintzer R, Cooper B, Roberts M, Harris K. Clinical and immunologic criteria for the diagnosis of allergic bronchopulmonary aspergillosis. Ann Intern Med 1977; 86:405-14. 6. Bousquet 1, Chanez P, Lacoste lY, et al. Eosinophilic inflammation in asthma. N Engl 1 Med 1990; 323:1033-40. 7. Mayo JR, Webb WR, Gould R, et at. High resolution CT of the lungs: an optimal approach. Radi-
ology 1987; 163:507-10. 8. Naidich DP, McCauley 01, Khouri NF, Stitik FP, Siegelman SS. Computed tomography of'bronchiectasis. J Comput Assist Tomogr 1982; 6:437-44. 9. Foster WL, Pratt P, RoggliV,Godwin 10, Halvorsen RA, Putman CEo Centrilobular emphysema/CT-pathologic correlation. Radiology 1986; 159:27-32. 10. Fraser RG, Pare JAP, Genereux GP. Roentgenologic signs in the diagnosis of chest disease. Diagnosis of the diseases of the chest. Philadelphia: WB Saunders, 3rd ed. 1989; 458-693. 11. Naidich DP, Zerhouni EA, Siegelman SS. Computed tomography of the thorax. New York: Raven Press, 1984. 12. Zerhouni EA. Semiologie tomodensitometrique pulmonaire. In: GrenierP, ed. Imageriethoraciquede l'aduite. Paris: Flammarion, 1988; 197-204. 13. Traver R, Conces 01, Godwin D. Motion artifacts on CT simulate bronchiectasis. AJR 1988; 151:1117-9. 14. lanover ML. Radiographic findings in asthma. In: Weiss EB, Segal MS, Stein M, eds. Bronchial asthma, mechanisms and therapeutics. 2nd ed. Boston: Little, Brown, 1985; 562-75. 15. Soboya RE. Quantitative structural alterations in long-standing allergic asthma. Am Rev Respir Dis 1984; 130:289-92. 16. Dunhill MS. Asthma. In: Dunhill MS, ed. Pulmonary pathology. Edinburgh: Churchill Livingstone, 1987; 61-80. 17. Neeld DA, Goodman LR, Gurney lW, Greenberger PA, Fink JA. Computerized tomography in the evaluation of allergicbronchopulmonary aspergillosis. Am Rev Respir Dis 1990; 142:1200-5.
_
Chest Radiography and High Resolution Computed Tomography of the Lungs in Asthma1,2
F. PAGANIN, V. TRUSSARD, E. SENETERRE, P. CHANEZ, J. GIRON, P. GODARD, J. P. SENAC, F. B. MICHEL, and J. BOUSQUET
In 1962, the American Thoracic Society defined asthma as a reversible airway disease with increased airway responsiveness, and it noted that some asthmatics, particularly those who have had asthma for many years, may have persistent airflow obstruction (1). The pathophysiologic mechanisms of asthma include bronchospasm, hypersecretion, and chronic airway inflammation, but, more recently, attention has been focused on bronchial fibrosis and structural defects of the bronchi, implying a third component of asthma, i.e., "bronchial destruction." This component, already proposed in the ATSdefinition in 1962, has been clinically suggested by the "accelerated decline" of the pulmonary function of asthmatic patients as well as by the poor reversibility of airway obstruction in patients with a long history of the disease. However, many patients with uncomplicated bronchial asthma have a normal chest roentgenogram and do not show structural lesions. Computerized tomography (CT scan) provides a high degree of anatomic detail, and it has been used in the diagnosis of airway diseases such as emphysema (2) and bronchiectasis (3). However, CT scans in nonABPA (allergic bronchopulmonary aspergillosis) asthmatic patients with chronic obstructive pulmonary disease have been examined in only one study, where the focus was on emphysema (4). In an attempt to better characterize the possible bronchial abnormalities of asthma and to discover whether irreversible lesions are associated with this disease, high resolution CT scans with millimetric slices were examined in 57 asthmatic patients with variable disease severity, and in 10 sex- and age-matched normal subjects. None of these participants smoked. Ten subjects had a CT scan during asthma exacerbation and again after appropriate treatment to define labile abnormalities. We studied 61 asthmatic patients IS to 67 yr of age (mean ± SO, 37.6 ± 14.8 yr). Patients were randomly chosen for the study, and there was no particular selection criteria. Asthma was defined according to the criteria of the American Thoracic Society (l), and all patients had reversible airway obstruction characterized by an increase of 15070 of FEY 1 after inhalation of 200 ug of salbutamol or a positive inhalation challenge with carbachol. None of the subjects had a history of allergic bronchopulmonary aspergillosis as defined by the crite-
1084
SUMMARY CTscans have been studied only In asthmatics who were smokers, and no such study has been performed in patients with chronic uncomplicated asthma where a permanent bronchial destruction is likely to occur after a long course of the disease. The object of the study was to characterize CT-scan abnormalities and determine whether bronchial destructive lesions may be observed. Fifty-seven adults with chronic asthma of variable severity and etiology and 10 normal subjects were studied. None of the subjects smoked. Chest radiographs and HR·CT scans were performed in all patients. Todiscriminate between reversible and Irreversible CT-scan abnormalities, two examinations were made In 10 patients with acute asthma both before and 2 wk after parenteral high dose corticosteroid treatment. The chest radiographs showed the expected abnormalities of asthma in 37.8% of the asthmatics. CT scans were abnormal In 71.9% of the asthmatics. Reversible abnormalities Included mucoid impactions, acinar pattern, and lobar collapse. Irreversible abnormalities included bronchiectasis, bronchial wail-thickening, sequellar line shadows, and emphysema. Most of these abnormalities are likely to be related to bronchial destruction. AM REV RESPIR DIS 1992; 146:1064-1087
TABLE 1 ROENTGENOGRAPHIC ABNORMALITIES IN ASTHMA" Aas Score
Patients tested, n Normal chest radiograph
2
3
15 10 (66%)
20 13 (65%)
0 0 0 0
0 1 (6%) 0 0
0 0 0 2 (10%)
0 0 0 3 (23%)
2 (22%)
4 (26%)
7 (35%)
6 (46%)
0 0
0 0
0 0
0 2 (15%)
9 7 (78%)
4-5 13 6
(46%)
Mucoid impaction Pneumomediastinum Pneumothorax Consolidation andlor infiltrates (23%)
Bronchial wall-thickening (46%)
Bronchiectasis Emphysema (15%)
• Results are expressed in number and in percentage of patients (shown in parentheses).
ria of Rosenberg and coworkers (5). In particular, they had no serum Aspergillus precipitins or IgE (Phadebas RAST~, Pharmacia Diagnostics AB, Uppsala, Sweden), and skin test results with a nonstandardized Aspergillus allergen extract (Stallergenes Laboratories, Fresnes, France) were negative. The clinical severity of chronic asthma was quoted according to the score of Aas, which is used to grade chronic asthma from very mild forms (score of I) to incapacitating disease requiring medication (score of 5) (6). Ten normal healthy volunteers 22 to 60 yr of age (mean ± SO, 40.1 ± 14.1 yr) were used as a control group. Their pulmonary function was within normal range. They were nonallergic and had never had asthma.
None of the subjects or patients tested were current or previous smokers. The study was carried out after informed consent of the patients and approval by the Ethics Committee of the Hospital were obtained.
(Received in original/arm August 13, 1991 and in revised form April 27, 1992) 1 From the Clinique des Maladies Respiratoires and the Departernent de Radiologie, MontpellierCedex, France. 2 Correspondence and requests for reprints should be addressed to Dr. J. Bousquet, Clinique des Maladies Respiratoires, 34059 MontpellierCedex, France.
BRIEF COMMUNICATION
1085 TABLE 2 COMPUTED TOMOGRAPHIC ABNORMALITIES IN ASTHMA' Aas Score 2
Patients, n Normal CT scan Mucoid impaction Acinar pattern Collapse Bronchiectasis Cylindrical Varicose Cystic Bronchial wall-thickening Emphysema Paraseptal Centrolobular Bullous Pneumomediastinum Pneumothorax
9 5 (55.5%) 1 (11%) 0 0
4 (44%)
15 6 2 6 2
(40%) (15%) (40%) (15%)
3 20 3 6 8 4
4-5
(15%) (30%) (40%) (20%)
13 4 3 5 1
(31%) (23%) (38%) (8%)
0 0 0
6 (40%) 0 0 2 (15%)
16 (800/0) 3 (15%) 0 4 (20%)
6 1 1 3
(46%) (8%) (8%) (23%)
0 0 0 0 0
0 0 0 1 (6.6%) 0
1 (5%) 1 (5%) 0 1 (5%) 0
4 3 1 1 0
(31%) (23%) (8%) (7.6%)
, Results are expressed as the nu mber of patients, with percentages of patients shown in parentheses.
Posteroanterior and lateral chest radiographs were made for all patients, and they were examined blind by three independent readers (FP, VT, and ES). Pulmonary hyperinflation was not considered to be an abnormality because of its common occurrence in asthmatics and its lack of specificity. CT scans of the chest were performed using a Somaton DR H (Siemens, Munich, Germany) in high resolution technique according to Mayo and coworkers (7). The matrix size was 512 x 512 at a pixel size of 0.5 mm. Gantry (20 degrees) was used to improve the ability to visualize the subsegmental bronchi, and scanning time ranged from 2.5 to 3 s. The patients wereexamined in the supine
position during mild inspiration, with their arms over their heads. Images were recorded at a window width of 1,600HU and a window levelof - 600 HU and intravenous contrast media were never administered. Section cut 1mm thick at I5-mm increments were obtained through the lungs. The scans wereinterpreted independently by three doctors who had no knowledge of the disease or of the chest radiographs (FP, VT, and ES). agreement among the three reviewers was considered to be satisfactory. The first assessment was a subjective examination of the quality of the test followed by an evaluation of each lobe of the pulmonary parenchyma. The images were interpreted as follows. The diagnosis of bronchiectasis and mucoid impactions was
based on the criteria of Naidich and coworkers (8) and Grenier and colleagues (3), with high resolution CT scans. Emphysema was characterized according to Foster and coworkers (9), whereas atelectasis, acinar pattern (10), and proximal wall-thickening were defined according to Naidich and colleagues (Il). Bronchial wall-thickening and/or peribronchial fibrosis (11, 12) were characterized by CT scans according to Naidich and colleagues (11) and Zerhouni (l2). Other abnormalities assessed were pneumothorax and pneumomediastinum. All subjects underwent the same radiographic and CT-scan examinations. Moreover, since some lesions are likely to be reversible, two CT scan examinations were made with a 2-wk interval between them in 10 patients with acute asthma who had undergone systemic corticosteroid treatment (daily dose of methylprednisolone, 2 mg/kg of body weight) associated with intravenously administered bronchodilators (salbutamol, 0.05 mg/kg of body weight). CT scan lesions were classified as either permanent or reversible and correlated (1) using Kendall's 't test, with the severity of asthma defined by the clinical score of Aas, and (2) by Spearman's rank test, with the duration of asthma or FEV,.
Among the 61 asthmatic patients included in the study, only 57 were studied because disagreements were noticed among reviewersin four patients. Among the patients studied, nine had Aas scores of 1, 15 had scores of 2, 20 had scores of 3, and 13 had scores of 4 or 5. FEV, ranged from 31 to 106070 of predicted values (mean ± SD, 70.12 ± 22.9%). The etiologic investigations showed that 63.1% of patients had perennial allergy, 5.2% had premenstrual syndrome, 21% had chronic sinusitis, 5.2% had aspirin intolerance, and 5.2% had gastroesophageal reflux; 14% wereconsidered to have intrinsic asthma. Chest radiographs werecompletely normal in all the control subjects and in 63.1% of the asthmatics (table 1). CT scans were completely normal in all the control subjects and in 28.1% of the asthmatics (table 2). When CT scans were compared with chest radiographs there was a significantly (p < O.ot,chi-square test) increased number of asthmatics who had abnormal imaging patterns. All CT scan patterns, except pneumothorax, were observed in the asthmatics (figures 1 to 3). In the 10 asthmatics who had undergone
TABLE 3 EVOLUTION OF COMPUTED TOMOGRAPHIC ABNORMALITIES IN 10 PATIENTS AFTER CORTICOTHERAPY' Before Mucoid impaction Acinar pattern Lobar collapse Bronchiectasis Cylindrical Varicose Cystic Bronchial wall-thickening Emphysema Fig. 1. HR-CT scan of asthmatic patient (Aas score, 3) showing an acinar pattern of left lower lobe and filled cylindric bronchiectasis of the right lower lobe.
3 (2) 2 (2) 5 (3) 25 11 0 16 12
(7) (3) (4) (4)
After 0 0 0 25 11 0 16 12
(7) (3) (4) (4)
, Results are expressed as cumulative numbers of lobes involved, with the numbers of patients shown in parentheses.
1086
BRIEF COMMUNICATION
two CT scans within a 1- or 2-wk interval, it was observed that mucoid impaction, acinar patterns, and lobar collapses were reversible abnormalities, as these disappeared in all subjects (table 3). However, there was no change in bronchiectasis (figures 1 and 2), bronchial wall-thickening, or emphysema. There was no significant correlation between the severity of asthma and CT scan abnormalities, but patients with the most severe asthma tended to display more irreversible abnormalities.
* * *
Fig. 2. HR-CT scan of the same patient as in figure 1 after 2 wk of 120 mg per day of methylprednisolone showing cylindric bronchiectasis but no acinar pattern.
Fig. 3. HR-GT scan of patient with severe asthma (Aas score, 4) showing cylindric and varicose bronchiectasis in both lower lobes and com plete destruction of the middle lobe.
This study showed the expected chest radiographic abnormalities in asthmatics, whereas CT scans revealed that a large proportion of the asthmatics had lesions. These lesions included such reversible abnormalities as mucoid impaction, acinar pattern, and lobar collapse, and irreversible abnormalities such as bronchiectasis, emphysema, and bronchial thickening. The use of high resolution CT scan in the diagnosis of lung diseases is now well established, and in this study none of the CT scans analyzed displayed any of the artifacts described by Traver and coworkers (13). Bronchiectasis has been studied by many investigators who compared CT scans and bronchography. Their findings showed that CT scans are specific in greater than 90010 and sensitive in greater than 95% of cases when the high resolution CT-scan technique was used (3). The diagnosis of emphysema assessed by pathologic examination was correlated with high resolution CT scans (2). In asthma, as suspected,abnormalities were more frequently observed on CT scans than on chest radiographs. CT scans often revealed reversible abnormalities that were not correlated with the severity of the disease. Mucoid impaction, acinar patterns, and lobar collapse disappeared after anti-inflammatory treatment. Radiographic studiesof the asthmatics have disclosed bronchial wall-thickening, and this has been considered as a possibly reversible abnormality (14). In the present study, CT scans revealed such lesions, but they there was no reversibility, suggesting that radiographs and CT scan images of bronchial thickening may represent different pathologic abnormalities.The lesions 0 bservedby radiography may correspond to bronchi in which the bronchial lumen is filled in by secretion. On the other hand, using C'I-scananalysis, bronchial wallthickening indicates the abnormal visualization of bronchi in the periphery of the lung (11, 12). The image of bronchi seen in this area may reflect either bronchial inflammation increasing the thickness of the bronchial wall or peribronchial fibrosis (11, 12). These identified CT-scan lesions were not reversible in the 10 patients treated by corticosteroids. Peribronchial fibrosis wasobserved by Soboya (15) in pathologic specimens of asthmatic lungs, suggesting that the second hypothesis is more relevant in this study. Other permanent abnormalities in asthma
1087
BRIEF COMMUNICATION
revealed by CT scans include bronchiectasis and, more rarely, emphysema. These lesions were observed in nonsmoking patients who had never suffered from the pulmonary diseases that might account for their presence (4, 10). In the present study, emphysema revealed by CT scan may not be due to a parenchymal destructive disease per se, but rather to an extensive peribronchial fibrosis that might relate to cicatricial emphysema since in the pathologic examination of asthmatic lungs, destructive emphysema has been observed in some studies (16). Bronchiectasis has been observed in the pathologic examination of asthmatic lungs (16) as wellas in livingasthmatics by bronchography, and, more recently, in a small series of patients by CT scan (17). However, in most of these studies it was not possible to determine the incidence of bronchiectasis in asthma. In the present study, we confirmed the presence of this abnormality in greater than 50070 of the patients. Cylindric bronchiectasis occurred frequently, and it was found in patients with variable severity. Varicose and cystic bronchiectasis were rare, and they were observed only in the patients with the most severe asthma. These may be sequelae of mucoid impactions and bronchial hypersecretion. This study has shown that destructive lesions are observed in asthma. Most of these lesions are related to bronchial abnormalities.
Acknowledgment The writers thank Dr. 1. P. Daures and 1. Y.Lacoste and all the staff of the Departement de Radiologie for their kind help, and Dr. A. M. Campbell for reviewing the manuscript.
References 1. Meneely GR, Rensetti AD 1r, Steele 10, Wyatt IP, Harris HW. American Thoracic Society:chronic bronchitis, asthma, and pulmonary emphysema. Am Rev Respir Dis 1962; 85:762-8. 2. Hruban RH, Meziane MA, Zehrouni EA, et al. High resolution computed tomography of inflation-fixed lungs. Pathologic-radiologic correlation ofcentrilobular emphysema. Am Rev Respir Dis 1987; 136:935-40. 3. Grenier P, Maurice F, Musset 0, Menu Y, Nahum H. Bronchiectasis: assessment by thin-section CT. Radiology 1986; 161:95-9. 4. Kinsella M, Muller NL. Hyperinflation in asthma and emphysema. Assessmentby pulmonary function testing and computed tomography. Chest 1988; 94:286-9. 5. Rosenberg M, Patterson R, Mintzer R, Cooper B, Roberts M, Harris K. Clinical and immunologic criteria for the diagnosis of allergic bronchopulmonary aspergillosis. Ann Intern Med 1977; 86:405-14. 6. Bousquet 1, Chanez P, Lacoste lY, et al. Eosinophilic inflammation in asthma. N Engl 1 Med 1990; 323:1033-40. 7. Mayo JR, Webb WR, Gould R, et at. High resolution CT of the lungs: an optimal approach. Radi-
ology 1987; 163:507-10. 8. Naidich DP, McCauley 01, Khouri NF, Stitik FP, Siegelman SS. Computed tomography of'bronchiectasis. J Comput Assist Tomogr 1982; 6:437-44. 9. Foster WL, Pratt P, RoggliV,Godwin 10, Halvorsen RA, Putman CEo Centrilobular emphysema/CT-pathologic correlation. Radiology 1986; 159:27-32. 10. Fraser RG, Pare JAP, Genereux GP. Roentgenologic signs in the diagnosis of chest disease. Diagnosis of the diseases of the chest. Philadelphia: WB Saunders, 3rd ed. 1989; 458-693. 11. Naidich DP, Zerhouni EA, Siegelman SS. Computed tomography of the thorax. New York: Raven Press, 1984. 12. Zerhouni EA. Semiologie tomodensitometrique pulmonaire. In: GrenierP, ed. Imageriethoraciquede l'aduite. Paris: Flammarion, 1988; 197-204. 13. Traver R, Conces 01, Godwin D. Motion artifacts on CT simulate bronchiectasis. AJR 1988; 151:1117-9. 14. lanover ML. Radiographic findings in asthma. In: Weiss EB, Segal MS, Stein M, eds. Bronchial asthma, mechanisms and therapeutics. 2nd ed. Boston: Little, Brown, 1985; 562-75. 15. Soboya RE. Quantitative structural alterations in long-standing allergic asthma. Am Rev Respir Dis 1984; 130:289-92. 16. Dunhill MS. Asthma. In: Dunhill MS, ed. Pulmonary pathology. Edinburgh: Churchill Livingstone, 1987; 61-80. 17. Neeld DA, Goodman LR, Gurney lW, Greenberger PA, Fink JA. Computerized tomography in the evaluation of allergicbronchopulmonary aspergillosis. Am Rev Respir Dis 1990; 142:1200-5.
