| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Service de Pneumologie et d'Allergologie Pédiatriques, Hôpital
Necker Enfants Malades, 149 rue de Sèvres, 75015 Paris, France.
2 Service de Radiologie Pédiatrique, Hôpital Necker Enfants
Malades, 75015 Paris, France.
3 Service de Biostatistiques et Informatique Médicale, Hôpital
Necker Enfants Malades, 75015 Paris, France.
Received February 11, 2002;
accepted after revision April 22, 2002.
Address correspondence to V. Marchac.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Bronchial sections were counted at five levels in both lungs and at three levels in the right lung on high-resolution CT and plotted against lung function. Findings from 27 children with persistent symptoms of asthma (mean age, 11.4 years; SD, ±3.1 years) that were severe (group A, n = 15) or moderate (group B, n = 12) were compared with findings from 21 control subjects (mean age, 10.8 years; SD, ±3.0 years) using the Student's t test, analysis of variance, and Dunn-Bonferroni test.
RESULTS. A bronchial wall thickening score based on the number of visible bronchi at three levels (three-level score) proved to be as valuable as and simpler to obtain than a score based on the number of bronchi at five levels (five-level score). The three-level scores for groups A and B were similar (mean ± SD, 16.8 ± 4.2 vs 18.4 ± 3.4, respectively; p = not significant), but these scores were significantly higher than those for the control subjects (mean ± SD, 8.2 ± 3.4, respectively; p < 0.001). There was no correlation between the three-level score and forced expiratory volume in 1 sec or forced expiratory flow between 25% and 75% of forced vital capacity. In contrast with adults with severe asthma, our pediatric patients with difficult-to-treat asthma did not have CT evidence of mucoid impaction, emphysema, areas of hyperlucency, bronchiectasis, or sequellar line shadows.
CONCLUSION. Bronchial wall thickening visible on high-resolution CT may constitute an additional criterion of asthma severity in children. CT evidence of bronchial wall thickening might help to identify patients with a higher risk of airway remodeling.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Chronic inflammation of the airways is followed by healing that may lead to structural changes known as airway remodeling [2]. The remodeling process involves a thickening of the airway wall by fibrosis and an increase in smooth muscle and mucus gland mass [3]. Structural changes take place in both the large and small airways in fatal cases of asthma, but are mainly confined to the small airways in non-fatal cases [4]. Airway wall thickening has been observed on high-resolution CT (HRCT) scans in adults [5], along with other findings such as emphysema, bronchiectasis, and line shadows corresponding to sequelae.
The aim of this study was to describe HRCT findings and to establish an objective simple criterion for bronchial wall thickening in children with difficult-to-treat asthma. Children with this condition are likely to have bronchial inflammation and therefore to have a greater risk of remodeling. By analogy with findings in adults, the visualization of abnormalities on HRCT may be another objective criterion of bronchial inflammation useful for asthma management. HRCT findings were therefore compared with clinical status and lung function data for children with difficult-to-treat asthma.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
800 mg/day
beclomethasone dipropionate or budesonide). All patients had difficult-to-treat asthma and were not symptom-free despite undergoing treatment with high doses of antiasthma medication. Children were said to have difficult-to-treat asthma [7] if receiving 800 mg or more of beclomethasone dipropionate or budesonide per day or more than 1 mg/kg body weight of prednisolone on alternate days; requiring use of bronchodilators 3 days every week, in addition to any long-acting ß-agonists, for wheezing with related proven poor spirometry on home recording; having less than 80% forced expiratory volume in 1 sec (FEV1); and requiring additional steroids every month to reverse a proven episode or absence from school for more than 5 days per term because of asthma.
Patients were classified into the following two groups according to the severity and frequency of their symptoms. Group A consisted of children with persistent symptoms that were moderate: cough on exertion or infrequent nocturnal cough. Group B was composed of children with persistent symptoms that were severe: symptoms that required additional ß-agonists more than three times a week, at least one asthma attack requiring hospital treatment, frequent nocturnal cough, peak flow variation of more than 20% during the day, or persistent asthma attacks.
Group C was the control group, consisting of pediatric patients (>6 years old) with normal findings on thoracic CT. Exclusion criteria included a history of wheezing, chronic coughing, or recent bronchitis. HRCT was performed to assess mediastinal lymphadenopathy suspected on the basis of positive results for Mantoux's test on chest radiography [8, 9], to investigate pneumothorax, or to look for evidence of interstitial pulmonary disease suspected from chest radiography of patients with a systemic disease without respiratory involvement (e.g., disseminated lupus erythematosus, vasculitis, scleroderma). One patient was examined on HRCT for isolated intrapulmonary calcification.
For all the patients with asthma, HRCT of the thorax was performed to exclude differential diagnoses, essentially viral sequelae (i.e., bronchiolitis obliterans or bronchiectasis), and lung function tests were performed to assess disease severity. Informed consent was given by parents. Lung CT scans were obtained during a stable period (i.e., no exacerbation of asthma symptoms during the month before scanning), and function tests were performed at approximately the same time as CT (±1 week) with no exacerbation between CT and function testing.
Exclusion criteria were a history of chronic lung disease such as bronchopulmonary dysplasia, cystic fibrosis, or other definite systemic disease; the presence of stridor; suspected inhalation of a foreign body; immunoglobulin deficiency; and age of 6 years or less.
The characteristics of the patients that were recorded included age at the time of the study, age at the onset of symptoms, sex, duration of the disease, high level of total IgE, or positive IgE levels specific for at least one respiratory allergen, and current asthma medication. In addition, we recorded personal atopic status for each patient by indicating whether patients had a history of eczema or positive findings on skin prick tests specific for at least one inhaled allergen as previously described [10].
Imaging
All patients underwent HRCT of the chest on a helical scanner (ProSpeed;
General Electric Medical Systems, Milwaukee, WI). HRCT scans were obtained
during full inspiration in all patients and subjects except in those too young
to hold their breath. The following parameters were used: 1-mm sections at
10-mm intervals, bone E3 enhanced algorithm, 512 x 512 matrix, 140 kVp,
and 130 mAs. The scanning time was 1.0 sec, and scanning was performed from
the apices of the lungs to the lung bases. Lung window settings were
photographed at a width of 1600 H and a level of -600 H. CT scans were
interpreted by two pediatric radiologists who were unaware of the clinical
history of the patients and control subjects.
Bronchial wall thickening was assessed as follows. All visible sections of bronchi were counted at five levels in the pulmonary fields (right and left lungs). All circular (complete circles or at least two thirds of a circle) and longitudinal bronchi except hilar bronchi were included. The five levels were 1 cm above the carina, 1 cm below the carina, the right pulmonary vein, 2.5-3 cm below the top of the pulmonary vein, and above the right side of the diaphragm. The number of visible sections of bronchi at these five levels were added to yield a score (five-level score) for bronchial wall thickening.
A simplified method of examining only the right lung at the following three levels was also used: 1 cm below the carina, 2.5-3 cm below the top of the right pulmonary vein, and above the right side of the diaphragm. The right lung was chosen so that the effects of cardiac motions of the left side could be avoided. The number of visible sections of bronchi at these three levels were added to yield a score (three-level score) for bronchial wall thickening.
In 20 patients, four sections were acquired from the base during expiration to investigate areas of air trapping. In addition, we looked for other abnormalities described in adult patients with asthma [11, 12]: mucoid impaction, collapse, emphysema, bronchiectasis, air-space consolidation, and line shadows corresponding to sequelae. We defined these line shadows as linear densities 2-5 cm long extending through the lung and in some cases coming into contact with the pleura [12,13,14].
Pulmonary Function Tests
Spirometric tests were performed in the respiratory and physiology
laboratory as previously described
[15]. We analyzed
FEV1 and the mean forced expiratory flow between 25% and 75% (i.e.,
during the middle half) of forced vital capacity (FEF25-75).
FEV1 is used to evaluate the large airways, whereas
FEF25-75 reflects the small airways.
Statistical Analysis
The agreement between the two observers for the number of visible bronchial
sections was assessed by the intraclass correlation coefficient.
The means for groups A and B were compared with the mean for group C using the Student's t test.
Differences between groups and subgroups were assessed by factorial analysis of variance. The Dunn-Bonferroni test was used to determine correlations among asthma severity, lung function (FEV1 and FEF25-75), and bronchial wall thickening scores on HRCT.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Groups A and B did not differ in mean age at the onset of symptoms, age at
HRCT, duration of the disease, or personal atopic status. Moreover, the groups
did not differ in treatment received, including the level of inhaled steroids,
or in the duration of treatment before HRCT. All patients received long-term
(6 months) inhaled beclomethasone dipropionate, with 24 (88.9%) of the 27
patients receiving 800 mg/day or more. In addition, 19 (70.3%) of the 27
patients also received long-acting ß2-agonists.
Imaging
The bronchial wall thickening scores based on the number of bronchi counted
on HRCT scans are shown in Table
2. In group A, the five-level score was 52.5 ± 11.0, and
the three-level score was 16.8 ± 4.2. In group B, the five-level score
was 58.0 ± 10.3, and the three-level score was 18.4 ± 3.4. The
five-level score (mean ± SD) for bronchial wall thickening in the
control subjects was 29.8 ± 9.2, and the three-level score was 8.2
± 3.4.
|
The difference between patients (groups A and B combined) and the control subjects (group C) was highly significant (p < 0.001). No significant difference was found between groups A and B (Figs. 1 and 2). The three-level scores for bronchial wall thickening proved to be as valuable as the five-level scores.
|
|
We found no correlation between treatment (inhaled steroids 6 months
with or without long-acting ß2-agonists), age at the time of
the scanning, duration of the disease, and scores for bronchial wall
thickening on HRCT.
For each of the five levels studied on HRCT, a statistically significant difference between the number of bronchial sections seen in patients with asthma (groups A and B) and control subjects (group C) (data not shown) was found. This finding is consistent with the fact that asthma is a diffuse disease of the bronchial tree. The three low sections used for the three-level score were chosen to generate a simplified version of the five-level score for bronchial wall thickening. Similar scores were obtained regardless of which sections were used to yield a score for bronchial wall thickening (data not shown).
Agreement between the two observers was good. The interclass correlation coefficient was 0.997 for the five-level score and was 0.995 for the three-level score.
Figures 3A,3B,3C,3D,3E,3F and 4A,4B,4C,4D,4E,4F show the HRCT scans that were obtained at three predetermined levels (1 cm below the carina, 2.5-3 cm below the top of the right pulmonary vein, and above the right diaphragm) to yield a score for bronchial wall thickening. The visible bronchi, all of which are circular (complete circles or at least two thirds of a circle), and longitudinal bronchi with the exception of the hilar bronchi are shown in schematic diagrams to the right of each HRCT scan.
|
|
|
|
|
|
|
|
|
|
|
|
Figure 3A,3B,3C,3D,3E,3F shows the HRCT scans of a 15-year-old female control subject. The three-level score for bronchial wall thickening was calculated by counting all the redrawn bronchi shown on the HRCT scans obtained at the three predefined levels. Therefore, the three-level score for bronchial wall thickening was 8 (1 + 6 + 1). Figure 4A,4B,4C,4D,4E,4F shows the HRCT scans of an 11-year-old boy with severe, persistent symptoms (group B). The three-level score for bronchial wall thickening in this patient was 18 (6 + 8 + 4).
Table 2 provides the FEV1, FEF25-75, and HRCT scores. There was no significant difference in FEV1 between groups A and B. The difference in FEF25-75 between groups A and B (wide SDs) was not significant, but FEF25-75 tended to be lower in group B.
No correlation could be found between HRCT scores (five-level score and three-level score) and results of lung function tests (FEV1 or FEF25-75).
Additional HRCT Findings
We found no mucoid impaction, emphysema, areas of hyperlucency,
bronchiectasis, or line shadows corresponding to sequelae on HRCT scans
obtained at five predefined levels.
In some patients, septal lines were seen on HRCT. This finding was defined as thin (1-2 mm) linear opacities that were less than 1 cm long and not parallel to the inner chest wall. None of the septal lines were attached to the pleura (one septal line, n = 8; two septal lines, n = 3; more than two septal lines, n = 10) or localized mild air-space consolidation (n = 5; always in the right middle lobe). We observed no air trapping on inspiratory sections. In contrast, air trapping was found in 11 of 20 expiratory sections and was diffuse in eight of these 11 sections. Air trapping was not related to the severity of the symptoms (groups A and B) or to lung function (FEV1 and FEF25-75).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We selected pediatric patients with difficult-to-treat asthma according to the definition of McKenzie [7]. The selected patients required high levels of medication and had persistent symptoms, asthma exacerbations, or airflow obstruction despite taking high doses of anti-asthma medication (groups A and B). This population of patients with asthma is the most difficult to treat, as reported at the American Thoracic Society workshop [10], and probably corresponds to the patients at most risk of developing remodeling sequelae. Patients with mild asthma were not included for ethical reasons.
There is currently no reference method defining bronchial wall thickness for use in adults. Bronchial wall thickness is currently diagnosed on the basis of high visibility of the bronchi in the external third of the pulmonary field on HRCT. This method is highly subjective, and its reproducibility has, to our knowledge, not been studied. In studies using this subjective analysis [12, 16, 17], bronchial wall thickening has been reported in adults with mild to severe asthma with a prevalence of from 30% [12] to 92% [16].
Attempts have been made to analyze bronchial wall thickening quantitatively [18,19,20,21]. Boulet et al. [18] measured the thickness of the intermediate bronchus with an electronic caliper and calculated the ratio of wall thickness to outer diameter. Okazawa et al. [21] measured (in excised dog lungs) the airway wall area and airway wall percentage, which was defined as [(airway wall area / total area) x 100]. Awadh et al. [19] and Niimi et al. [20] applied a slightly modified version of this technique to adults with asthma.
These methods have some limitations. Boulet et al. [18] presented results for only a proximal bronchus (the right intermediate bronchus). Because asthma also affects small airways, this technique may underestimate the extent of peripheral bronchial wall thickening. Awadh et al. [19] and Niimi et al. [20] restricted their analyses to airways for which cross sections were obtained on the basis of the apparent roundness of the airway lumen. This method may lead to significant errors because the magnitude depends on the acuteness of airway angles. This method also requires image amplification.
No data are available concerning the normal thickness of bronchi in children, and direct measurements of the bronchial wall are difficult and subject to error. We therefore established a semiquantitative criterion that was based on known physical spatial resolution data. Spatial resolution is a physically known parameter for a given CT scan and reveals the size of the smallest structure that is visible on the CT scan. By definition, any structure smaller than the spatial resolution will not be visible, whereas all structures larger than the spatial resolution will be visible. This definition is well accepted for interstitial lung disease, in which the thickened structures become visible [5].
Because the spatial resolution of the CT scanner remained constant, we assumed that any increase in the thickness of the bronchial wall would be associated with an increase in the number of visible bronchi. We therefore counted the visible bronchi in the pulmonary parenchyma at three predefined levels and at five predefined levels to generate a three-level score and five-level score for bronchial wall thickening, respectively.
This method yielded consistent radiologic findings and is reliable and reproducible. The three-level score for bronchial wall thickening proved to be as valuable as the five-level score for bronchial wall thickening, but the three-level score is simpler to obtain because visible sections of bronchi are counted at only three levels on the right lung as opposed to five levels on both lungs. We therefore recommend that the three-level score be used for bronchial wall thickening. Because each new machine is likely to have higher definition than previous machines, it is important to reestablish normality criteria for each new type of scan.
Our results show that patients with difficult-to-treat asthma who are symptomatic, with severe (group B) or moderate symptoms (group A), had significantly higher bronchial wall thickening scores on HRCT than control subjects (group C). There was no significant difference in the bronchial wall thickening scores on HRCT scans between children with moderate symptoms (group A) and those with more severe symptoms or attacks (group B).
Studies involving quantitative analysis of HRCT scans in adults [17, 19] have shown that all groups of patients with mild to severe asthma, defined according to clinical [19] or functional [17] criteria, have airway walls that are thicker than those of healthy subjects and that subjects with more severe asthma display a greater amount of airway wall thickening than those with milder asthma. Our results are consistent with these findings and show that bronchial wall thickening occurs in children with persistent asthma.
The use of HRCT for the assessment of bronchial wall involvement in pediatric patients with asthma is associated with potential risks. Recently, a group of researchers suggested that the dose of radiation used should be reduced when examining children [22]. We currently use the recommended setting for CT in children (i.e., 80 mA, 140 kVp, 1-sec acquisition). The settings described in this article correspond to the CT protocol proviously used at our institution.
Bronchial wall thickening that is visible on HRCT may result from edema and transient inflammatory infiltration, an increase in the mass of airway smooth muscle, extracellular matrix, and elastic and collagenous tissues. Bronchial wall thickening visible on HRCT may reflect reversible or irreversible inflammation. Our results suggest that inflammation persists in patients with difficult-to-treat asthma who have persistent symptoms, even in patients with moderate symptoms, and that these individuals may therefore have a higher risk of remodeling and sequelae.
One of the fundamental criteria for the noninvasive surveillance and adaptation of asthma treatment is respiratory function. Our results show that this criterion may underestimate bronchial inflammation. No correlation was observed between HRCT scores and FEV1, which reflects proximal obstruction, or between HRCT scores and FEF25-75, which reflects the obstruction of small airways. We conclude from our analysis that respiratory function (volume and flow) may not reflect all pathologic aspects of asthma and may be an insufficient surveillance criterion.
This conclusion is consistent with findings from a study showing that reducing airway hyperresponsiveness by inhaled steroid treatment leads to more effective control of asthma and alleviation of chronic airways inflammation than the usual practice of adjusting treatment according to lung function and symptoms [23].
Studies measuring an objective criterion of bronchial wall thickening and determining measures reflecting proximal obstruction (FEV1) have shown that bronchial wall thickening occurs in patients with asthma who have a normal FEV1 value [18,19,20]. No study has yet, to our knowledge, compared an objective criterion of bronchial wall thickening and measures reflecting distal obstruction (FEF25-75). We found no individual correlation between bronchial wall thickening scores and distal functions (which were generally lower in the patients than the subjects). In patients with asthma, the large airways are almost always involved to some extent, so determining whether there is any narrowing of the small airways on the basis of tests of forced expiration is difficult [24]. Although patients with asthma present with significantly lower FEF25-75 values (reflecting peripheral airway function) than control subjects, this measure is not specific [25]. Lung attenuation curve measures have been shown to be more sensitive than traditional spirometry for the detection of peripheral airway dysfunction [26], but this technique requires serial HRCT before and after metacholine administration, which would not be ethical in pediatric patients. This constraint shows the value of searching for other objective criteria by which to assess the severity of asthma.
The use of an HRCT score for bronchial wall thickening should be evaluated as another criterion for disease evaluation, particularly in cases in which the severity of the symptoms and the findings on lung function tests are discordant. Discrepancies such as this one may be caused by differences in the information provided on HRCT and on lung function tests.
Unlike the results obtained in adults with asthma [27] and in a Turkish study of pediatric patients [28] and despite the fact that we studied a population of children with severe asthma, we found no mucoid impaction, emphysema, areas of hyperlucency, bronchiectasis, or line shadows corresponding to sequelae. These abnormalities have been correlated with disease severity in adults [12], and some of these abnormalities are thought to be irreversible (i.e., bronchiectasis, emphysema, areas of hyperlucency, and line shadows [1]). The population in the Turkish pediatric study was different from ours, and the imaging findings described probably resulted from infections (major bronchiectasis, hilar lymphadenopathies, or subsegmental or segmental atelectasis). We may not have observed these abnormalities in children because of the shorter duration of the disease (mean duration, 7.9 years). We hope that treating inflammation at an early stage of the disease may prevent such sequelae. However, the absence of visible sequelae does not exclude the possibility of their occurrence during childhood.
No correlation was found between areas of hyperlucency on expiratory scans and clinical severity or bronchial wall thickening scores on HRCT. These findings are consistent with those of Park et al. [17]. We will therefore no longer obtain expiratory scans.
In conclusion, asthma is a chronic inflammatory disease that is common in pediatric patients. The initially reversible inflammation can progress as a result of remodeling and the formation of sequellar lesions. These stages are indirectly visible on HRCT scans.
We established a simple, objective, reproducible quantitative criterion for assessing bronchial wall thickening: counting all the visible bronchial sections in the right lung at three predefined levels. The symptomatic patients receiving treatment had bronchial wall thickening scores that were significantly higher than those of the control subjects.
In children with difficult-to-treat asthma, HRCT did not show any of the typical findings reported in adults, such as emphysema and bronchiectasis. This finding may be because the sequelae appear only after the disease has progressed for many years.
HRCT could be used in association with clinical and lung function data to assess the severity of asthma in patients with persistent asthma and perhaps to identify patients with a higher risk of remodeling. Furthermore, HRCT could be of use in decision-making to optimize the treatment of inflammation. Additional studies are required to investigate the links among clinical data, lung function test results, and scores for bronchial wall thickening on HRCT. Indications for HRCT during the follow-up period of children with difficult-to-treat asthma should be investigated.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Bergeron, M. K. Tulic, and Q. Hamid Tools used to measure airway remodelling in research Eur. Respir. J., March 1, 2007; 29(3): 596 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Aukland, T. Halvorsen, K. R. Fosse, A. K. Daltveit, and K. Rosendahl High-Resolution CT of the Chest in Children and Young Adults Who Were Born Prematurely: Findings in a Population-Based Study Am. J. Roentgenol., October 1, 2006; 187(4): 1012 - 1018. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Brody, H. A. W. M. Tiddens, R. G. Castile, H. O. Coxson, P. A. de Jong, J. Goldin, W. Huda, F. R. Long, M. McNitt-Gray, M. Rock, et al. Computed Tomography in the Evaluation of Cystic Fibrosis Lung Disease Am. J. Respir. Crit. Care Med., November 15, 2005; 172(10): 1246 - 1252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U G Rossi and C M Owens The radiology of chronic lung disease in children Arch. Dis. Child., June 1, 2005; 90(6): 601 - 607. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
