| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Departments of Radiology (C.B.A., P.A.G.), Lung Function Investigation, UPRES 2397 (C.S., M.H.B., M.Z.), and Pneumology (T.S.), and the National Institute of Health and Medical Research (U 494) (P.A.G.), Hôpital Pitié-Salpêtrière, 47-83 Boulevard de l’Hôpital, 75651 Paris 13, France; Assistance Publique-Hôpitaux de Paris, Université Pierre et Marie Curie, Paris, France; and Faculté de Médecine Paris XI, Centre Chirurgical Marie-Lannelongue, Le Plessis-Robinson, France (A.C.). Received April 16, 2001; revision requested May 14; revision received July 31; accepted September 17. Supported by Programme Hôspitalier de Recherche Clinique PCRMG94813. Address correspondence to P.A.G. (e-mail: philippe.grenier @psl.ap-hop-paris.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twelve patients with mild intermittent asthma who were nonsmokers (National Institutes of Health staging) and six nonsmoking healthy volunteers, age and sex ratio–matched, were examined by using helical thin-collimation CT at the level of basal bronchi at 65% of total lung capacity. Three sets of acquisitions were obtained: at baseline and after inhalation of methacholine and then salbutamol. Cross-sectional areas of bronchi greater than 4 mm2 were segmented and calculated from CT images. Lung attenuation was measured in the anterior, lateral, and posterior areas of the right lung parenchyma. Gas trapping was evaluated by using thin-section CT at residual volume in six of the patients with asthma. Statistical analysis included two factors repeated-measurement analysis of variance and Mann-Whitney and Kruskal-Wallis nonparametric tests.
RESULTS: Bronchial cross-sectional areas and lung attenuation did not vary significantly compared with baseline values following bronchial challenge in healthy volunteers or patients with asthma. However, in patients with asthma, bronchial cross-sectional areas were significantly smaller than in healthy volunteers, except after inhalation of salbutamol. Lung attenuation and anteroposterior attenuation gradient were significantly higher in patients with asthma than in healthy patients (P < .001). Air-trapping scores were significantly higher after methacholine challenge.
CONCLUSION: Helical thin-collimation CT at controlled lung volume and at full expiration associated with bronchial challenge may help evaluate bronchoreactivity and inflammation in mild intermittent asthma.
© RSNA, 2002
Index terms: Asthma, 68.754 • Lung, CT, 68.1211 • Lung, diseases, 68.754 • Lung, function, 68.754
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Although the exact mechanism and pulmonary locations of the disease remain partly speculative, physiologic and pathologic observations suggest that small airways and lung parenchyma contribute substantially to asthma pathogenesis (2). The site and type of abnormalities are, of course, of importance to the understanding of the pathophysiologic characteristics of the disease and for optimal target delivery of specific drugs. Thin-section computed tomography (CT) can be used to visualize airways with an internal diameter larger than 2 mm in patients with asthma with much greater detail than can conventional radiography and has made possible preliminary investigations of the site, magnitude, and distribution of airway narrowing in vivo. Several studies (3–6) have brought to light specific CT features in patients with asthma at full inspiration and expiration. CT has also been used to assess bronchial size changes and air trapping after nonspecific bronchial challenge (7). Most of those studies (4–7) have been performed at functional residual capacity and residual volume after bronchial challenge in subjects with asthma (7). Unfortunately, functional residual capacity and residual volume change after bronchial challenge, when compared with prechallenge (baseline) values in patients with asthma, make the comparison of bronchial sections difficult due to the possible changes in mechanical constraints induced by pressure-volume changes. Because inflammation possibly coexists with bronchial hyperreactivity in asthma, assessment of inflammation should ideally be performed simultaneously. So far, to our knowledge, nitric oxide measurement in exhaled gas and inflammatory markers in induced sputum are the only methods available for assessing airway inflammation but remain technically debatable (8,9). CT lung attenuation could constitute a potential index for that purpose. This study evaluates, by using thin-section CT, changes in bronchial cross-sectional area and lung attenuation induced by bronchial stimulation in patients with mild intermittent asthma, at a given lung volume monitored with pneumotachography (1). In addition, this study aims to determine if helical thin-collimation CT can help localize and evaluate bronchoreactivity and inflammation in mild intermittent asthma to provide an integrative approach for monitoring current and new therapies.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Healthy volunteers.—Six nonsmoking healthy volunteers, four women and two men age (mean age, 37 years ± 15 [mean ± SD]; mean weight, 62 kg ± 3; and mean height, 169 cm ± 9) served as healthy subjects. These subjects were age- and sex ratio–matched with the patients with asthma. They were selected at random from a national database of healthy volunteers. All volunteers underwent a complete physical examination and PFTs. None of them had any record of past or current major disease, in particular, heart or lung disease. None of them underwent any sort of medical treatment. Pregnancy was an exclusion criterion.
Patients with asthma.—Twelve consecutive patients with mild intermittent asthma who did not smoke, 80 women and four men (mean age 28 years ± 10; mean weight, 65 kg ± 13; and mean height, 169 cm ± 10), were included in the study. All were in healthy condition, apart from having asthma, as assessed by using clinical history and examination; none were currently taking any medication apart from occasional sprays of salbutamol for relief of asthma exacerbation, and, in particular, none were undergoing corticosteroid treatment. Pregnancy was an exclusion criterion. Mild intermittent asthma was diagnosed and graded according to the following National Institutes of Health guidelines (1): symptoms occurring less than twice a week with asymptomatic and normal peak expiratory flow between exacerbations, brief exacerbations (from a few hours to a few days), nighttime symptoms occurring less than twice a month, and forced expiratory volume in 1 second (FEV1) greater than 80% of predicted value. Patients had not suffered from an infectious disease or asthma exacerbation for at least 8 days prior to the study. Patients were not allowed to take any medications during the 48 hours preceding the PFT and CT examination.
Pulmonary Function Tests
Pulmonary function was evaluated 5 days before CT was performed. These tests consisted of measurement of vital capacity, functional residual capacity, residual volume, total lung capacity (TLC), and FEV1. Volume measurements were obtained with the subject in a sitting position by using plethysmography and helium dilution. The latter was also performed with the subject in the supine position to match CT conditions. Airway response to bronchial challenge was assessed by using methacholine according to the American Thoracic Society guidelines (10). Methacholine (Pharmacie Centrale des Hôpitaux, Paris, France) is a nonspecific bronchial challenge drug that induces bronchial hyperreactivity and consequently bronchoconstriction in patients with asthma, following a dose response curve. Patients not responding to a cumulated dose of at least 1,600 µg of methacholine were believed to be nonhyperreactive. Methacholine bronchial challenge is the current standardized test used to mimic asthma exacerbations in a controlled setting. Methacholine was inhaled during inspiration by using an automatic nebulizing device (FCC 88; Mediprom, Paris, France). The cumulated dose of methacholine that resulted in an FEV1 decline of more than 20% compared with the FEV1 reference value for the subject was retained for challenge when thin-section CT was performed. Two breaths of salbutamol (Ventoline, Glaxo Wellcome, Marly-le-Roy, France; 400 µg total) were administered to all subjects at the end of the test and FEV1 was measured 15 minutes later to verify return to baseline. Salbutamol is a ß-sympathetic agonist capable of reversing the bronchoconstrictive action of methacholine, so it is routinely used for reversibility testing of bronchial hyperactivity (10).
CT Acquisition
The subjects breathed through a nonmetallic pneumotachograph (V2000; Sensormedics, Yorba Linda, Calif) during the CT procedure. After a few stable normal breaths, the subjects were asked to perform a maximal inspiration immediately followed by a slow expiration. A computer-driven balloon shutter, incorporated into the expiratory output of the pneumotachograph, was activated at the preset percentage of the patient’s TLC determined during the PFT session. All acquisitions were made after interruption of the slow expiration phase at 65% of TLC. This volume has been found to be a good compromise between 100% TLC, which is difficult to hold for several seconds, and functional residual capacity, which is modified during bronchial challenge in patients with asthma (11). Because TLC varies very little during bronchial challenge in patients with mild intermittent asthma, it was used as the reference volume (12). CT was gated by using the spirometer computer when the chosen volume was reached. Because of the CT scanning response time, acquisition started 2 seconds after gating. Mouth flow and pressure, and volume changes computed from the integrated flow data were displayed on a computer screen to insure that neither leak nor respiratory effort had occurred and that lung volume had remained stable during the imaging procedure. Apnea lasted a maximum of 15 seconds and was found acceptable by all subjects. All subjects had one training session before the acquisition to learn these maneuvers to avoid learning-curve bias.
CT scans were obtained with one of two (model 7000 SR [n = 6] or Twin Flash [n = 12]; Philips, Eindhoven, the Netherlands) scanners. Three helical CT acquisitions (10-mm thickness) were obtained at the level of the basal bronchi with 1-mm collimation, 1.5 pitch, and 1-mm reconstruction interval. Effective section thickness was 1.1 mm. We used 120 kV, 100–165 mAs settings, a 512 x 512 pixel matrix, and a spatial resolution filter. The first set of acquisitions was obtained at baseline. The second set was performed less than 3 minutes after methacholine inhalation in accordance with American Thoracic Society guidelines, applying the procedure used during PFT. Each subject with asthma inhaled the methacholine dose that had been found to result in a significant decrease (>20%) of FEV1 during PFT. The healthy volunteers inhaled 1,600 µg of methacholine. The third CT acquisition was performed 15 minutes after inhalation of 400 µg of salbutamol. Healthy volunteers received methacholine and salbutamol to insure that both groups received aerosols.
Three regularly spaced 1-mm-thick scans were also obtained at full expiration (or residual volume), at the end of each set of acquisitions, in the last six patients with asthma, above the aortic arch, at the level of the tracheal carina, and above the diaphragm, to depict air trapping in both lungs (see Statistical Analysis).
Measurements
Images obtained at 65% TLC.—Images were displayed by using standardized lung windows (level, -600 HU; width, 1,600 HU). Magnified fields of view on the right lung were reconstructed from raw data. We used anatomic landmarks (airways or vascular branching points) to match the sections obtained during the successive sets of measurements in each patient. We took into account, for further analysis, all airways that were visualized approximately perpendicular to the scanning plane and had long-to-short section–axis ratios of less than 1.5.
Two observers (C.B.A., A.C.) selected regions of interest away from vessels in the right lung–matched sections. Regions of interest were measured (HU) in the anterior (nondependent), lateral and posterior (dependent) areas. These measurements were confined to regions consisting mainly of alveolar units and small conducting airways (13). These selected regions of interest consisted of squares measuring 12–15 x 12–15 mm.
The areas of bronchial cross-sections included in the selected sections were segmented and calculated by using software previously developed and validated by Préteux et al (14). Bronchi with large to small axis ratios greater than 1.5 were automatically excluded by the procedure, as were bronchi with incomplete cross-sectional outlines on the section. Bronchi with cross-sectional areas less than 4 mm2 were also excluded due to the inaccuracy of the computation method for those values (14). For each patient and set, we computed the total sum of all cross-sectional areas of the selected bronchi on the right side.
Images obtained at residual volume.—Air trapping was assessed on the entire cross-sectional area of both lungs on expiratory scans obtained in six of the patients with asthma. We used a scoring system, with consensus obtained between the two observers, that is based on the percentages of lung parenchyma with air trapping on expiratory CT scans, as described by Ng et al (15). The extent of air trapping visible was estimated at each of the three levels and for each lung by using the following five-point scale: 0, no air trapping visible; 1, 1%–24% of the cross-sectional area of the lung affected; 2, 25%–49% affected; 3, 50%–74% affected; 4, 75%–100% affected. The maximum possible score for one lung was 12 (three levels times four points at each level) and 24 for both lungs.
Statistical Analysis
The number of healthy volunteers was kept to a minimum (n = 6) to avoid unnecessary exposure of healthy subjects to radiation. The number of subjects with asthma incorporated into the study was planned to be larger because of the potentially higher predictable data variability expected in this group, with a total of 12 patients with asthma. Preliminary results appeared to be statistically homogeneous. Imaging data obtained in the first six subjects with asthma indicated no objective changes of bronchial diameter for bronchi of 2 mm or more of cross-sectional area after methacholine challenge; it was important to assess the impact of the challenge on smaller airways by evaluating air trapping at residual volume. To obtain a statistically relevant group, air trapping was measured in the six additional subjects with asthma. There was no statistical difference, as tested by using analysis of variance (ANOVA), between the two groups of six patients with asthma in terms of morphologic characteristics, age, sex ratio, PFTs, and imaging data. All patients were pooled for further statistical analysis. A power analysis was performed to ensure that the small sample size did not impair the level of significance of the statistical analysis. Type 2 errors (not rejecting the null hypothesis when it is in fact false) were computed from ANOVA and t tests and were always less than the acceptable value of 0.1.
PFTs and CT lung attenuation are expressed as mean ± SD. PFT data at baseline were compared between groups with a nonpaired Student t test. PFT data in each group were compared by using repeated measurement ANOVA. Lung attenuations in healthy volunteers and subjects with asthma were compared among acquisitions with a two-factor repeated-measurement ANOVA. The air-trapping scores for patients with asthma were analyzed with one-way ANOVA for repeated measurements.
Bronchial cross-sectional areas are provided as medians and ranges because of their non-Gaussian distribution (Fig 1). Consequently, a nonparametric Mann-Whitney U test was used to compare groups. The three measurement times within groups were compared by using a Kruskal-Wallis nonparametric test. Differences were considered significant when the probability of a type 1 error was 5% or less.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
In this study, at 65% TLC the bronchial cross-sectional areas at baseline were significantly smaller in patients with mild intermittent asthma compared with healthy volunteers. Unfortunately, these data could not be compared with those of Goldin et al (7), as their results were expressed not as absolute surface areas but as percentages of baseline values. The lower bronchial cross-sectional areas at baseline in patients with asthma compared with those of healthy volunteers may be explained by using the following three mechanisms: baseline level of bronchoconstriction, inflammation resulting in bronchial wall thickening and narrower lumen, and intrinsic impairment of the ability of inspiration to stretch airway smooth muscles in asthma (17,18). Despite the demonstration of airway wall thickening in patients with asthma in postmortem CT studies (19), it does not seem contributory to bronchi of cross-sectional area of more than 4 mm2 in patients with mild intermittent or mild persistent asthma. Obviously, this observation does not exclude the possibility of inflammatory processes in bronchi smaller than 4 mm2, which could not be detected by using our direct measurement technique, as previously reported (20). Clearly, some degree of baseline bronchoconstriction could reasonably be expected in the patients with intermittent asthma as they had stopped all treatments 48 hours prior to the study. This was confirmed by the fact that inhalation of salbutamol after methacholine challenge brought their bronchi not only back to cross-sectional areas comparable with those of the healthy volunteer group but also to higher than their own baseline values. Because imaging was performed after a full inspiration at a fixed volume during expiration, conditions for impairment of airway smooth muscle stretching in patients with asthma were met (21,22). Cross-sectional area was not influenced more by methacholine challenge in patients with asthma than that in healthy volunteers. Our findings do not provide argument in favor of the hypothesis of baseline chronic impaired stretching in bronchi of more than 4 mm2. Such a hypothesis as an alternative explanation for the reduced cross-sectional area at baseline in the patients with asthma, rather than baseline bronchoconstriction, as proposed earlier remains speculative. The presence of air trapping (previously described [5,6] in patients with asthma) induced in our study by methacholine challenge, and partly reversible after salbutamol inhalation at residual volume in patients with asthma, confirmed that the methacholine-induced bronchoconstriction involved mainly or exclusively the smallest airways (23,24).
Because attenuation can be quantified in a reproducible way with CT, and values for normal subjects lie within a narrow range, Wollmer et al (13) used lung attenuation to assess various stages of inflammation in smokers. We applied the same approach in this study to try to characterize further inflammation in the more distal lung of patients with intermittent asthma by using CT, in particular at the level of bronchi cross-sectional area of less than 4 mm2, which could not otherwise be quantified with our method. As could be expected, the anterior, lateral, and posterior lung attenuation values calculated for our healthy volunteers at 65% TLC were about 25 HU greater than the corresponding attenuation values for the nonsmokers studied by Wollmer et al (13) at 100% TLC. This difference can be explained by lung volume difference. The overall dependent-nondependent gradients were similar (30 HU) in both studies. The attenuation values found in our patients with asthma were not only always greater than our normal values but were also greater than the values reported for the smokers studied by Wollmer et al (13), even after taking into account the different pulmonary volumes. Furthermore, the overall dependent-nondependent gradient was significantly higher for patients with asthma than for smokers (125 vs 75 HU, respectively). The relationship between anterior, lateral, and posterior attenuation values was linear. We hypothesize that the well-known peribronchial and small-airway inflammation occurring in asthma explains the attenuation and gradient increases and that it is limited to the parenchyma, which includes the smaller bronchi and small airways in the patients with mild intermittent asthma included in our study (23–25).
The question of the potential role of the constriction of small bronchi and small airways in increased lung attenuation in patients with asthma at baseline cannot be raised. Indeed, such a constriction should have induced extended air trapping at full expiration, which was not observed; air trapping was not observed in patients before methacholine inhalation. We could also hypothesize that lung perfusion may contribute to lung attenuation differences between patients with asthma and healthy volunteers. However, hypoxic vasoconstriction is expected in areas of hypoventilation occurring distal to airway constriction. This should normally result in a decrease in lung attenuation rather than an increase. Thus, we can only speculate that increased lung attenuation observed at baseline in our patients with asthma was related to inflammatory changes or to increased lung perfusion related to inflammation. Actually, inflammation consisting in cellular infiltration and cellular and extracellular lung water accumulation may increase lung attenuation at CT.
The observation that lung attenuation in patients with asthma remained unaffected by methacholine and salbutamol inhalations at 65% TLC further supports the hypothesis that bronchoconstriction played only a small role or no role in lung attenuation changes and that inflammation was a more likely contender.
In conclusion, patients with mild intermittent asthma with no ongoing therapy, normal PFTs, and at least 8 days without asthma exacerbation or acute infection had smaller cross-sectional areas of bronchi of more than 4 mm2 compared with healthy volunteers. That these bronchi did not respond to methacholine challenge but responded to a bronchodilatator may be suggestive of increased baseline tone of the bronchi, although, this finding remains compatible with the theory of impaired stretching (17). The lung-parenchyma attenuation and dependent-nondependent gradient values were significantly higher in patients with asthma compared with those in healthy volunteers and were not affected by any challenge, thereby suggesting distal inflammation. We conclude that helical thin-collimation CT at a controlled lung volume, associated with bronchial challenge, may be a useful tool to help localize and evaluate bronchoreactivity and inflammation in early intermittent asthma. The ability to follow up bronchial reactivity and attenuation by using imaging techniques over time in cohorts of patients receiving different treatments can provide an independent tool to assess and monitor current and new therapies in patients with mild intermittent asthma and also at later stages of the disease, possibly in conjunction with exhaled breath and induced sputum analysis techniques.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, A.C., P.A.G., M.Z.; study concepts and design, A.C., P.A.G., M.Z., T.S.; literature research, C.B.A., A.C., P.A.G.; clinical studies, C.S., M.H.B.; data acquisition, C.B.A., A.C., M.H.B., C.S., M.Z.; data analysis/interpretation, C.B.A., A.C., M.Z.; statistical analysis, A.C.; manuscript preparation, C.B.A., A.C., M.Z.; manuscript definition of intellectual content, C.B.A., A.C., P.A.G., M.Z.; manuscript editing, A.C., M.Z.; manuscript revision/review, C.B.A., A.C., P.A.G., T.S., M.Z.; manuscript final version approval, P.A.G., M.Z.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. A. Kaminsky, C. G. Irvin, L. K. A. Lundblad, J. Thompson-Figueroa, J. Klein, M. J. Sullivan, F. Flynn, S. Lang, L. Bourassa, S. Burns, et al. Heterogeneity of bronchoconstriction does not distinguish mild asthmatic subjects from healthy controls when supine J Appl Physiol, January 1, 2008; 104(1): 10 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Winkler and J. G. Venegas Complex airway behavior and paradoxical responses to bronchoprovocation J Appl Physiol, August 1, 2007; 103(2): 655 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Harris, T. Winkler, N. Tgavalekos, G. Musch, M. F. V. Melo, T. Schroeder, Y. Chang, and J. G. Venegas Regional Pulmonary Perfusion, Inflation, and Ventilation Defects in Bronchoconstricted Patients with Asthma Am. J. Respir. Crit. Care Med., August 1, 2006; 174(3): 245 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. de Jong, N. L. Muller, P. D. Pare, and H. O. Coxson Computed tomographic imaging of the airways: relationship to structure and function Eur. Respir. J., July 1, 2005; 26(1): 140 - 152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Berger, V. Perot, P. Desbarats, J. M. Tunon-de-Lara, R. Marthan, and F. Laurent Airway Wall Thickness in Cigarette Smokers: Quantitative Thin-Section CT Assessment Radiology, June 1, 2005; 235(3): 1055 - 1064. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Orlandi, C. Moroni, G. Camiciottoli, M. Bartolucci, M. Pistolesi, N. Villari, and M. Mascalchi Chronic Obstructive Pulmonary Disease: Thin-Section CT Measurement of Airway Wall Thickness and Lung Attenuation Radiology, February 1, 2005; 234(2): 604 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Matsuoka, Y. Kurihara, Y. Nakajima, H. Niimi, H. Ashida, and K. Kaneoya Serial Change in Airway Lumen and Wall Thickness at Thin-Section CT in Asymptomatic Subjects Radiology, February 1, 2005; 234(2): 595 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. H. Brown and W. Mitzner Invited Review: Understanding airway pathophysiology with computed tomograpy J Appl Physiol, August 1, 2003; 95(2): 854 - 862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. H. Mills, J. M. Wild, B. Eberle, and E. J. R. Van Beek Functional magnetic resonance imaging of the lung Br. J. Anaesth., July 1, 2003; 91(1): 16 - 30. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
