Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Chest. Author manuscript; available in PMC 2010 April 26.
Published in final edited form as:
PMCID: PMC2859729

Airway remodeling measured by multidetector computed tomography is increased in severe asthma and correlates with pathology

Ravi S. Aysola, M.D.,1 Eric A. Hoffman, Ph.D.,2 David Gierada, M.D.,3 Sally Wenzel, M.D.,4 Janice Cook-Granroth,2 Jaime Tarsi, R.N., M.P.H.,1 Jie Zheng,5 Kenneth B. Schechtman, Ph.D.,5 Thiruvamoor P. Ramkumar, Ph.D.,1 Rebecca Cochran,1 Xueping E, M.D., Ph.D.,1 Chandrika Christie,1 John Newell, M.D.,6 Sean Fain, Ph.D.,7 Talissa A. Altes, M.D.,8 Mario Castro, M.D., M.P.H.,1 and NHLBI Severe Asthma Research Program (SARP)



To prospectively apply an automated, quantitative 3-D approach to imaging and airway analysis to assess airway remodeling in asthma.


Using the Pulmonary Workstation (VIDA Diagnostics) that enables quantitative airway segment measurements of low-dose, thin section (0.625-1.25 mm) multidetector-row CT (MDCT), we compared airway wall thickness (WT) and area (WA) in 123 subjects participating in a prospective multicenter cohort study, NIH Severe Asthma Research Program (SARP): severe asthma (n=63), mild-moderate asthma (n=35), and normal (n=25). A subset of these subjects underwent fiberoptic bronchoscopy and endobronchial biopsies (n=32). WT and WA were corrected for total airway diameter and area - WT%, WA%.


Subjects with severe asthma had significantly greater WT% than mild-moderate asthma and normals [17.2±1.5 v 16.5±1.6, p=0.014 and 16.3±1.2, p=0.031, respectively] and greater WA% compared to mild-moderate asthma and normal [56.6±2.9 v 54.7±3.3, p=.005 and 54.6±2.4, p=0.003, respectively]. Both WT% and WA% were inversely correlated with baseline FEV1% (r=-0.39, p<0.0001 and -0.40, p<0.0001, respectively) and positively correlated with response to bronchodilator (r=0.28, p=0.002 and r=0.35, p<0.0001, respectively). Airway epithelial thickness on biopsy correlated with WT% (r=0.47, p=0.007) and WA% (r=0.52, p=0.003). In the same individual, there is considerable regional heterogeneity in airway wall thickness.


Severe asthmatics have thicker airway walls on MDCT than mild asthmatics or normals, which correlates with pathologic measures of remodeling and the degree of airflow obstruction. MDCT may be a useful technique for assessing airway remodeling in asthma, but overlap among groups limits the diagnostic value in individual subjects.

Keywords: Asthma, airway remodeling, chest CT


Studies of airways of patients who die from asthma demonstrate thickened airway walls, due to increases in smooth muscle mass, infiltration with inflammatory cells, deposition of connective tissue, vascular changes and mucous gland hyperplasia, termed airway remodeling 1-4. Airway remodeling may be a feature of milder and even asymptomatic asthma 5-7. Remodeling of airways can result in worsening of airway narrowing, airflow obstruction and disease progression 8,9.

Airway remodeling in asthma has been studied in vivo by performing endobronchial biopsies that can then be evaluated for structural and inflammatory changes 10-12. Multi-detector computed tomography (MDCT) studies have recently been used to evaluate the extent of airway wall thickening as a non-invasive, highly reproducible method for studying individual airways. Several studies using MDCT have demonstrated that the airway walls of asthmatics are thicker than normals and that airway wall thickness is related to severity of disease and airflow obstruction 7,13-15. This thickening may be partially reversible with inhaled corticosteroid (ICS) treatment in steroid naïve patients and may increase in the absence of ICS. The airway lumen area of stable patients with asthma is not narrowed compared with that of healthy controls and may even be dilated in those with more severe disease 16. These studies are often limited by the number of airways studied, small subject numbers or by the use of semi-quantitative techniques. Thus, the purpose of our study was to apply an automated airway analysis software comparing airway wall measurements among severe and mild-moderate asthmatics and normals and to correlate this with remodeling measurements on biopsies of matched airway segments.


Study Design

As part of the NIH Severe Asthma Research Program (SARP), a prospective cohort of subjects, following informed consent, underwent a detailed testing and MDCT – using a standardized protocol developed by the SARP17. MDCT data was analyzed and compared on a total of 123 subjects: severe asthma (n=63), mild-moderate asthma (n=35), normals (n=25). In a subset whom underwent airway biopsies (n=32), we correlated segmental airway WT and area (WA) from MDCT with measures of airway remodeling from biopsies, including epithelial and lamina reticularis (LR) thickness, from the corresponding segment. All subjects at the Washington University SARP site were given the option of participating in the airway biopsy substudy. The study was approved by each site's Institutional Review Board and monitored by an independent Data and Safety Monitoring Board.

Human Subjects

The inclusion criteria by group were: Normals: 18 – 60 years of age, in good overall health, no smoking within past 5 years and <5 pack years smoking, and PC20 >16 mg/ml. Mild to moderate asthma18: 18 - 60 years of age, physician diagnosis of asthma, on asthma therapy >12 months, daytime asthma symptoms > 2 times per week but less than continual, and/or nocturnal asthma symptoms > 2 times per month but less than nightly, no smoking within past 5 years and <5 pack years smoking, no concurrent lung disease, and PC20 8mg/ml or 15% improvement in FEV1 post bronchodilator. Severe asthma: 18 - 60 years of age subjects with one or both of two major criteria and two of the minor criteria (American Thoracic Society workshop 19). The major criteria include (in order to achieve control to a level of mild-moderate persistent asthma): (1) Treatment with continuous or near continuous (≥50% of the previous year) oral corticosteroids; or (2) Treatment with high dose inhaled corticosteroids: beclomethasone (>1,260 μg/d), budesonide (>1,200 μg/d), flunisolide (>2,000 μg/d), fluticasone (>880 μg/d), or triamcinolone (>2,000 μg/d). The minor criteria include: (1) daily treatment with a long term controller medication in addition to inhaled corticosteroids (e.g. long-acting -agonist, theophylline, or leukotriene antagonist), (2) asthma symptoms requiring short-acting -agonist use on a daily or near daily basis, (3) persistent airflow obstruction (FEV1 <80% predicted; diurnal peak flow variability >20%), (4) one or more urgent care visits for asthma per year, (5) three or more oral corticosteroid “bursts” per year, (6) prompt deterioration with 25% reduction in oral or inhaled corticosteroid dose; and (7) near fatal asthma event in the past. In these subjects with severe asthma, conditions other than asthma were excluded, exacerbating factors were treated, and the subject did not have a history of poor adherence.

CT technique

All subjects underwent MDCT chest with 16 or 64 detector rows (GE Light Speed Ultra 16 or Siemens Volume Zoom, Sensation 16 or 64) after maximal bronchodilation with albuterol (540-720 mcg) to minimize the effect of acute bronchoconstriction on airway dimensions. Subjects were administered increasing doses of albuterol till the FEV1% difference was 5% or a maximal dose of albuterol (8 puffs or 720 mcg) was reached. Suspended full inspiratory measurements were obtained at the following settings: GE: 1.675-1.75 pitch, 0.6 sec rotation time, 120 kV, 0.625-1.25 mm reconstructed slice thickness, medium smooth “standard” reconstruction algorithm; Siemens: 1.5 pitch, 0.5 sec rotation time, 120 kV, 1mm reconstructed slice thickness, a medium smooth reconstruction algorithm, effective mAs was 30-57. To obtain isotropic voxels, the slice reconstruction interval was set to equal the inplane spatial resolution (field of view in mm/512 pixels).

MDCT airway evaluation software

MDCT scans were analyzed using automated, quantitative software designed to reliably label and segment the first five to six airway generations and allow accurate measurement of airway wall and lumen diameters obtained perpendicular to the long axis of each airway (Pulmonary Workstation, Version 0.139, VIDA Diagnostics) 20-23. Previous studies have validated the lung and airway segmentation methods when compared to manual measurements 24,25. Airway measurements for each segment were made at each centerline voxel and averaged over the middle third of the segment. The specific MDCT measurements used included airway WT, percent WT (WT%), WA, percent WA (WA%), luminal area (LA) and percent LA (LA%). The calculations are as follows: WT = Average Outer Diameter – Average Inner Diameter, WT% = (WT/Average Outer Diameter) x 100, WA = Total Area (TA) – LA, WA% = (WA/TA) × 100, LA% = (LA/TA) × 100, (Figure 1).

Figure 1
Airway measurements by MDCT

In the primary analysis, we averaged third generation airway wall measurements for all automatically segmented and labeled airways in each subject. An average of 18 third generation airways per subject were measured (9 for each lung). In the secondary analysis of subjects who underwent biopsies, we obtained cross-sectional CT measurements of each biopsied airway in a plane perpendicular to the airway long-axis at a distance 30% of the segment length distal to the origin of the airway segment. Two independent readers trained in the use of Pulmonary Workstation were blinded to the subject status or the results of the biopsy measurements. Inter-rater reliability on a random sample of 50 subjects between these two readers was excellent (intraclass correlation (ICC) = 0.98).

Pulmonary Function Tests

Spirometry, methacholine bronchoprovocation and plethysmographic lung volumes were performed within 1-2 days of the MDCT scan among the SARP sites in accordance with standardized ATS criteria 26,27. Subjects were asked to abstain from long-acting 2 agonists for 12 hours and short acting β2 agonists for 4 hours prior to assessment. Spirometry was performed before and after shorting-acting β2 agonist (4 puffs/360 mcg of albuterol) was delivered by metered dose inhaler and spacer.

Endobronchial Biopsies

A subset of 32 subjects (15 severe asthma, 9 mild-moderate asthma, and 8 normals) underwent bronchoscopy and 12-16 endobronchial biopsies were obtained from the third generation segmental carinas of the upper lobes. Three readers independently measured the areas of epithelium and LR of at least 3 biopsies with intact airway epithelium using the ImageJ software for morphometric analysis. The areas of the epithelium and LR were normalized for the length of the basement membrane yielding an epithelial and LR ratio 28. The average of these measurements made in triplicate was then used for subsequent analysis.

Statistical methods

All data were analyzed using analysis of variance and chi square tests to compare continuous and categorical variables across groups (SAS v9). Stepwise multiple linear regression identified variables that had independent significant associations with outcome measures that included FEV1%pred, WA%, and WT%. Variables were retained in these models if they had a significant (p < 0.05) or borderline significant (p < 0.1) association with the outcome measure.



The characteristics of the subjects in our study (Table 1) demonstrated that those with severe persistent asthma were on average: older, reported more frequent symptoms of allergies, had elevated levels of IgE, increased airway hyperresponsiveness and decreased baseline FEV1%pred compared to mild-moderate asthmatics and normals. Subjects in the biopsy subset were representative of the overall cohort (Table 1) with the exception that mild-moderate asthmatics had earlier onset of asthma (11.6 v 13 yrs, p=0.01) and greater ED visits within the last 12 months (50% v 5.6%, p=0.02).

Table 1
Group characteristics


There was no significant difference in average WT among groups. To account for differences in airway size, we calculated the WT% (Figure 1) for the labeled airways in each subject. Severe asthmatics had significantly greater WT% than those with mild-moderate asthma and normals (Figure 2, Table 2). There was no significant difference between mild-moderate asthmatics and normals for WT%. The increase in WT% inversely correlated with baseline FEV1%pred (r = -0.39, p < 0.0001), and positively correlated with change in FEV1% postbronchodilator (r = 0.28, p = 0.002). This finding was primarily due to the relationship between WT% and FEV1%pred (r = -0.47, p = 0.0003) in severe asthma. There was no significant relationship between WT% and FEV1%pred in mild-moderate asthmatics and normals. There was no significant correlation between WT% with FEV1 PC20.

Figure 2
Pulmonary Workstation MDCT images and bronchial biopsy from normal and severe asthma subjects
Table 2
MDCT airway measurements by group

There was no statistically significant difference in WA among groups. The analysis of WA% showed that severe asthmatics had greater WA% compared to mild-moderate asthmatics and normals. There was no significant difference in WA% between mild-moderate asthmatics and normals (Figure 2, Table 2). WA% was inversely correlated with baseline FEV1%pred (r = -0.4, p < 0.0001), and positively correlated with change in FEV1% postbronchodilator (r = 0.35, p < 0.0001). The correlation between WA% and baseline FEV1%pred was due to the relationship in severe asthmatics (r = -0.49, p = 0.0001). There was a significant inverse correlation between WA% with FEV1 PC20 (r = -0.29 in all asthmatic subjects, p = 0.02, and r = -0.48 in those with severe asthma, p = 0.01).

LA was not significantly different between groups. Severe asthmatics did have smaller LA% compared to mild-moderate asthmatics and normals. There was no significant difference in LA% between mild-moderate asthmatics and normals (p = ns).

Segmental MDCT Comparisons

Individual airways segments were compared among the three groups in regards to WT% and WA%. WT% in a few airways, and WA% in most airways were significantly greater in severe asthmatics compared to mild-moderate asthma and normals (Table 3). Because segmental WT measurements are not independent of each other, we calculated a slope of airway WT% and WA% from the apex to the base of the lung in each individual subject 21. The slopes for WT% and WA% were not significantly different between groups.

The range (minimum-maximum measurement) in airway thickness across segments per subject for WT% was 9.7-29.9 (normals), 9.7-29.6 (mild-moderate asthma), and 9.8-30.1 (severe asthma) and for WA% was 38.6-65.5 (normals), 38.9-64.7 (mild-moderate asthma) and 39.0-67.3 (severe asthma). The range of WA% across segments was significantly different between groups, with the severe asthmatics (28.3±3.4) having more variability than the mild-moderate asthmatics (25.8±3.4) or normals (26.9±4.0, p = 0.0035) but not for WT% (p = 0.92). The WT% and WA% range in an individual's segmental airways is highly variable and is dependent on the underlying disease status. However, if one focuses on the RUL apical segment (previously used in other studies 5,15), there was a statistically significant correlation between the RUL apical segment WA% (Pearson correlation coefficient r=0.75, p<0.0001) and WT% (r=0.52, p<0.0001) with all other segments (up to 19).

Multivariate analysis of WA/WT

The variables that distinguished severe asthmatics from mild-moderate asthmatics and normals included: age, baseline FEV1%pred and FVC%pred, log IgE, and change in FEV1 post-bronchodilator. We also found that WT% and WA% were significantly greater in severe asthmatics versus subjects with mild-moderate asthma and normals. Stepwise multiple regression models used FEV1%pred, WA% and WT% as dependent variables. The only significant independent predictors of FEV1%pred were WT% (P = 0.0004) and group (p < 0.0001) (R2 = 0.48). When WT% was the dependent variable in the stepwise regression, the significant independent correlates were FEV1%pred (p < 0.0001) and history of intubation (p = 0.04, R2 = 0.415). The only significant independent correlate of WA% was FEV1%pred (p < 0.0001, R2 = 0.331).

Correlation between MDCT airway indices and remodeling

Epithelial thickness ratio was positively correlated with both WT% and WA% (r = 0.47, p = 0.007 and r = 0.52, p = 0.003 respectively) (Figure 3). The relationship between LR thickness ratio and WT% and WA% demonstrated a similar trend (r = 0.33, p = 0.07 and r = 0.33, p = 0.06 respectively). The sum of epithelial and LR ratios was also positively correlated with WT% and WA% (r = 0.46, p = 0.008 and r = 0.49, p = 0.004 respectively).

Figure 3
MDCT wall thickness % and wall area % are correlated with airway remodeling


Airway remodeling describes structural changes, that in the asthmatic airway collectively result in thickening of the airway wall 2,3,8,9. CT imaging of the airways is being developed as a technique to study airway remodeling in vivo. Multiple studies have documented increased airway wall thickness in asthma using CT. Most of these studies 5,13-15 employed manual tracing methods, with either manual or digital measurement of airway dimensions. These previous studies using manual tracing methods have demonstrated significant differences in comparing asthmatics to normals though these differences may be related to selection bias. Kasahara et al. found that the radiographic measurement of WT and WA were: increased in asthmatics, correlated with LR, and inversely related to FEV1 13. A pediatric study found similar, although less robust results 29. To improve objectivity, reproducibility, and efficiency, automated airway segmentation and analysis methods have been implemented30. However, automated assessments previously have been restricted to those airways that are nearly round in transaxial images, so that the long axis is roughly perpendicular to the scan plane. Consequently, quantitative measurements were obtained only on (nearly) round airways in a limited number of slices, which may lead to potential selection bias as our data demonstrates significant regional heterogeneity in segmental airway remodeling. The automated technique we used eliminated selection bias by segmenting, labeling and measuring all of the proximal airways and allowing comparison of those airways that may have been excluded using previous methods.

In this study, we used an automated, quantitative software program to analyze and measure the differences in airway WT between severe asthmatics versus milder disease. We found that when using MDCT indices of airway WT that account for TA, specifically WT% and WA%, that severe asthmatics have on average slightly more thickened airway walls compared to mild-moderate asthma and normals. Interestingly, there was no significant difference when comparing airway WT% or WA% in mild-moderate asthmatics with normals. Previous studies normalized airway measurements to the total airway diameter/area or body surface area 13,30. Our data demonstrates that there is substantial variability in airway diameter and lumen between airways in the same subject and between subjects and that this must be taken into consideration when averaging data. Our measurements were normalized for segmental total thickness or area which accounts for inter-airway variability.

Furthermore, we found that the MDCT indices WT% and WA% correlated with physiologic measures of airflow obstruction across subjects. Similarly, Kasahara et al. also found that WT, WA, and LR were all inversely correlated with post-bronchodilator FEV1 13. Increased segmental airway thickness may correlate with more distal, small airway narrowing 31. Gono et al. compared expiratory and inspiratory (E/I) lung density measurements to assess air trapping as a measure of small airways disease. Asthmatics with irreversible airflow obstruction had significantly higher E/I ratios than asthmatics whose expiratory flows normalized after bronchodilation 5. Hasegawa et al. in a recent study using 3-D measurements of airway dimensions in chronic obstructive pulmonary disease focused their analysis on two bronchi (the right upper lobe and right lower lobe) and found statistically significant correlations between lumen area, WA% and FEV1 (% predicted). They also found a stronger correlation when distal smaller airways (up to 6th generation) were analyzed 32. These findings suggest that airway remodeling in more proximal airways reflects similar changes occurring in more distal airways, resulting in measurable airflow limitation and air trapping.

A current limitation of MDCT measures of airway thickness is the inability to be more specific about which component of the airway has truly changed. In our study we are unable to discern if the MDCT findings of increased WT% and WA% in severe asthma truly reflects increases in epithelial and LR changes or some other feature of airway remodeling such as increased smooth muscle mass 1,3. This particular aspect of airway remodeling has not been well characterized due to the limited sample of smooth muscle available with an endobronchial approach. Future longitudinal studies are needed to evaluate temporal changes in airway wall thickness within individual patients and to evaluate the effects of different treatments.

There are limitations to our ability to measure airway remodeling and accurately obtain airway biopsies from the same segment measured by MDCT. The biopsy sites were obtained from the upper lobes only and therefore we are not able to generalize our remodeling measures to other lung segments. Accurate histologic measurement of the epithelial and LR layers requires that the biopsies have an intact epithelium 28, therefore not all biopsies were included in our analysis.

The natural history of airway remodeling in asthma, its rate of progression and its response to treatment are questions that remain unanswered. Non-invasive measures of airway remodeling using MDCT would allow us to longitudinally monitor the effects of various stimuli and treatments on remodeling. Eventually, this technique may help identify individuals with asthma who are likely to develop severe disease and who may benefit from early targeted, aggressive therapy.

Grant Support

National Institutes of Health HL69149, HL64368, HL69349, HL69170, HL69155, HL69174, HL69130, HL69167, HL69116, HL69174-05


Analysis of variance
American Thoracic Society
Computed tomography
Data Safety and Monitoring Board
Emergency Department
Forced expiratory volume in one second
Percent predicted of Forced expiratory volume in one second
High resolution computed tomography
Ig E
Immunoglobulin E
Intraclass correlation
Inhaled corticosteroids
Lamina reticularis
Luminal area
Luminal area percent
Multidetector-row computed tomography
National Asthma Education and Prevention Program
National Institutes of Health
Not significant
Provocative concentration of methacholine to cause a 20% decline in FEV1
Severe Asthma Research Program
Standard deviation
Total area
Wall area
Wall area percent
Wall thickness
Wall thickness percent


Disclosures/Potential conflict of interest: None declared by authors.

Contributor Information

Ravi S. Aysola, ten.knilhtrae@alosyasr.

Eric A. Hoffman, ude.awoiu@namffoh-cire.

David Gierada, ude.ltsuw.rim@DadareiG.

Sally Wenzel, ude.cmpu@esleznew.

Janice Cook-Granroth, ude.awoiu@htornarG-kooC-ecinaJ.

Jaime Tarsi, ude.ltsuw.mi@isratj.

Jie Zheng, ude.ltsuw.soibuw@enaj.

Kenneth B. Schechtman, ude.ltsuw.soibuw@nek.

Thiruvamoor P. Ramkumar, ude.ltsuw@umar.

Rebecca Cochran, ten.retrahc@25ceb.

Xueping E, ude.ltsuw.grusn@XE.

Chandrika Christie, ude.ltsuw.mi@eitsirhcc.

John Newell, GRO.CJN@JlleweN..

Sean Fain, ude.csiW@niaFS.

Talissa A. Altes, ude.pohc.liame@setla.

Mario Castro, ude.ltsuw@mortsac..


1. Carroll N, Elliot J, Morton A, et al. The structure of large and small airways in nonfatal and fatal asthma. Am Rev Respir Dis. 1993;147:405–410. [PubMed]
2. James A, Paré P, Hogg J. The mechanics of airway narrowing in asthma. Am Rev Resp Dis. 1989;139:242–246. [PubMed]
3. Kay AB. Pathology of mild, severe, and fatal asthma. Am J Respir Crit Care Med. 1996;154:S66–69. [PubMed]
4. Kuwano K, Bosken C, Paré P, et al. Small airways dimensions in asthma and in chronic obstructive pulmonary disease. Am Rev Respir Dis. 1993;148:1220–1225. [PubMed]
5. Gono H, Fujimoto K, Kawakami S, et al. Evaluation of airway wall thickness and air trapping by HRCT in asymptomatic asthma. Eur Respir J. 2003;22:965–971. [PubMed]
6. Ketai L, Coutsias C, Williamson S, et al. Thin-section CT evidence of bronchial thickening in children with stable asthma: bronchoconstriction or airway remodeling? Acad Radiol. 2001;8:257–264. [PubMed]
7. Awadh N, Muller NL, Park CS, et al. Airway wall thickness in patients with near fatal asthma and control groups: assessment with high resolution computed tomographic scanning. Thorax. 1998;53:248–253. [PMC free article] [PubMed]
8. Benayoun L, Druilhe A, Dombret M, et al. Airway structural alterations selectively associated with severe asthma. Am J Respir Crit Care Med. 2003;167:1360–1368. [PubMed]
9. Homer RJ, Elias JA. Consequences of long-term inflammation. Airway remodeling. Clin Chest Med. 2000;21:331–343. ix. [PubMed]
10. Adelroth E. How to measure airway inflammation: bronchoalveolar lavage and airway biopsies. Can Respir J. 1998;5(Suppl A):18A–21A. [PubMed]
11. Bousquet J. The use of biopsy to study airway inflammation. Respir Med. 2000;94(Suppl F):S1–2. [PubMed]
12. Chetta A, Foresi A, Del Donno M, et al. Airways remodeling is a distinctive feature of asthma and is related to severity of disease. Chest. 1997;111:852–857. [PubMed]
13. Kasahara K, Shiba K, Ozawa T, et al. Correlation between the bronchial subepithelial layer and whole airway wall thickness in patients with asthma. Thorax. 2002;57:242–246. [PMC free article] [PubMed]
14. Little SA, Sproule MW, Cowan MD, et al. High resolution computed tomographic assessment of airway wall thickness in chronic asthma: reproducibility and relationship with lung function and severity. Thorax. 2002;57:247–253. [PMC free article] [PubMed]
15. Niimi A, Matsumoto H, Amitani R, et al. Airway wall thickness in asthma assessed by computed tomography. Relation to clinical indices. Am J Respir Crit Care Med. 2000;162:1518–1523. [PubMed]
16. Niimi A, Matsumoto H, Takemura M, et al. Clinical assessment of airway remodeling in asthma: utility of computed tomography. Clin Rev Allergy Immunol. 2004;27:45–58. [PubMed]
17. Moore W, Bleecker E, Curran-Everett D, et al. Characterization of the severe asthma phenotype by the National Heart, Lung, and Blood Institute's Severe Asthma Research Program. J Allergy Clin Immunol. 2007;119:405–413. [PMC free article] [PubMed]
18. National Asthma Education and Prevention Program . Expert Panel Report 2: Guidelines for the diagnosis and management of asthma. National Institutes of Health; National Heart, Lung, and Blood Institute; 1997.
19. American Thoracic Society Workshop Proceedings of the ATS workshop on refractory asthma. Current understanding, recommendations, and unanswered questions. Am J Respir Crit Care Med. 2000;162:2341–2351. [PubMed]
20. Palagyi K, Tschirren J, Sonka M. Quantitative analysis of intrathoracic airway trees: methods and validation. Inf Process Med Imaging. 2003;18:222–233. [PubMed]
21. Tschirren J, Hoffman EA, McLennan G, et al. Intrathoracic airway trees: segmentation and airway morphology analysis from low-dose CT scans. IEEE Trans Med Imaging. 2005;24:1529–1539. [PMC free article] [PubMed]
22. Tschirren J, Hoffman EA, McLennan G, et al. Segmentation and quantitative analysis of intrathoracic airway trees from computed tomography images. Proc Am Thorac Soc. 2005;2:484–487. 503–484. [PMC free article] [PubMed]
23. Tschirren J, McLennan G, Palagyi K, et al. Matching and anatomical labeling of human airway tree. IEEE Trans Med Imaging. 2005;24:1540–1547. [PMC free article] [PubMed]
24. Hu S, Hoffman E, Reinhardt J. Automatic lung segmentation for accurate quantitation of volumetric Xray CT images. IEEE Trans Med Imaging. 2001;20:490–498. [PubMed]
25. Reinhardt J, Raab S, D'Souza N, et al. Intrathoracic airway measurement: ex-vivo validation. In: Hoffman E, editor. Medical Imaging 1997. Physiology and function from multidimensional images. Proceedings of SPIE; Newport Beach, CA: 1997.
26. Miller M, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J. 2005;26:319–338. [PubMed]
27. Wanger J, Clausen J, Coates A, et al. Standardisation of the measurement of lung volumes. Eur Respir J. 2005;26:511–522. [PubMed]
28. Cohen L, E X, Horiuchi T, et al. Epithelial cell proliferation contributes to airway remodeling in severe asthma. Am J Res Crit Care Med. 2007;176:138–145. [PMC free article] [PubMed]
29. de Blic J, Tillie-Leblond I, Emond S, et al. High-resolution computed tomography scan and airway remodeling in children with severe asthma. J Allergy Clin Immunol. 2005;116:750–754. [PubMed]
30. Matsumoto H, Niimi A, Takemura M, et al. Relationship of airway wall thickening to an imbalance between matrix metalloproteinase-9 and its inhibitor in asthma. Thorax. 2005;60:277–281. [PMC free article] [PubMed]
31. Nakano Y, Muller NL, King GG, et al. Quantitative assessment of airway remodeling using high-resolution CT. Chest. 2002;122:271S–275S. [PubMed]
32. Hasegawa M, Nasuhara Y, Onodera Y, et al. Airflow limitation and airway dimensions in chronic obstructive pulmonary disease. Am J Resp Crit Care Med. 2006;173:1309–1315. [PubMed]