Abstract
Chronic inflammation and extracellular remodeling of the airway wall characterize asthma. The purpose of this study was to examine whether these features cause a change in airway mechanical properties. We examined 14 healthy and 10 young adults with long-lasting asthma, the latter treated with inhaled bronchodilators and corticosteroids. To obtain area-versus-transmural pressure (A–Ptm) curves during forced expiration (Pedersen, O. F., et al. J. Appl. Physiol. 1982;52:357–369), we used an esophageal balloon and a Pitot static probe positioned at five locations between the right lower lobe and midtrachea. Cross-sectional area (A), airway compliance (Caw = dA/dPtm), and specific airway compliance (sCaw = Caw/A) were obtained from the A–Ptm curves. Results showed that: (1) A was larger in males than in females; (2) Caw and sCaw decreased with a more downstream position; and (3) Caw and sCaw were significantly lower in the patients with asthma, with the differences between the asthmatic patients and the healthy subjects becoming smaller toward the trachea. The lower Caw and sCaw in the patients with long-lasting asthma support the concept that chronic inflammation and remodeling of the airway wall may result in stiffer dynamic elastic properties of the asthmatic airway.
Chronic airway inflammation plays a central role in the pathophysiology of asthma. Mucosal biopsies show that in adult patients with different grades of asthma (1, 2), even after 10 yr of treatment with inhaled corticosteroids (3), inflammatory changes are present in the airways. Studies of asthmatic children indicate that airway inflammation is present in childhood as well (4, 5).
Simultaneously with these effects, extracellular remodeling associated with fibrosis and elastolysis takes place in the airway wall (6), probably as a result of a repair process following acute allergic inflammation and driven by mediator release after chronic allergic inflammation (7). Extensive thickening of the basement membrane, due to subepithelial fibrosis, is commonly observed even in stable asthma (5, 8, 9) at an early stage (9, 10), as well as in childhood asthma (5). The deposition of collagen in the subepithelial matrix is increased (11), and the lamina propria is thickened as a result of increased microvascular permeability and plasma exudation in combination with an increase in the number of blood vessels (12). The entire airway wall is thickened (13), and the percentage cross-sectional area (A) of the airway occupied by airway wall tissue is increased by 1.5- to 2-fold (14).
One may expect that these extracellular structural changes within the airway wall alter the mechanical properties of the airways in asthma through geometric effects (7), by changes in tissue biomechanics (11), and a change in the interaction between airways and the surrounding lung parenchyma (15). However, the net airway mechanical effect of these changes remains unclear. Both increased thickening of the airway wall and increased collagen fibril density are important determinants of airway collapsibility, and may stiffen the subepithelial matrix (11) and increase the ability of the airways to withstand bronchoconstriction (16). On the other hand, reorganization and degradation of elastin and cartilage could result in a decrease in airway wall stiffness (17), thereby decreasing the load on airway smooth muscle (7).
The present study was done to test the hypothesis that structural remodeling, together with chronic inflammation of the airway wall, results in altered mechanical airway properties in long-lasting asthma. It is the first in vivo study comparing differences in airway compliance in asthmatic patients and healthy controls.
The study was done with 24 nonsmoking young adults, consisting of eight healthy female and six healthy male subjects and three female and seven male patients with asthma meeting American Thoracic Society (ATS) criteria (18). The subjects' mean ages and anthropometric data are given in Table 1.
| n | Age (yr) | Height (cm) | TLC (% pred )* | RV/TLC | FVC (% pred )* | FEV1(% pred )* | FEV1/FVC | PEF (% pred )* | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 14 | 26.1 ± 5.1 | 179 ± 9 | 104 ± 9 | 0.26 ± 0.05 | 111 ± 8 | 107 ± 11 | 0.83 ± 0.07 | 114 ± 13 | |||||||||
| Asthma | 10 | 22.2 ± 3.9 | 177 ± 10 | 101 ± 16 | 0.27 ± 0.05 | 97 ± 16 | 85 ± 7 | 0.76 ± 0.09 | 90 ± 14 | |||||||||
| p Value | 0.06 | 0.56 | 0.50 | 0.89 | 0.01 | < 0.001 | 0.05 | < 0.001 |
The patients had a history of moderate to severe asthma (19) beginning in early childhood, and had used maintenance treatment with inhaled corticosteroids for at least 3 yr. They exhibited bronchial hyperresponsiveness (BHR) and were atopic, as defined by total IgE > 100 IU/ml and a positive radioallergosorbent test for at least one inhaled allergen. All of the asthma patients were in stable clinical condition. If an exacerbation or a respiratory tract infection had occurred within 1 mo before scheduled measurements, the measurements were postponed until at least 2 wk after recovery. In an attempt to minimize bronchial obstruction caused by edema or hypersecretion at the time of measurements, all of the asthma patients were pretreated with a 10-d tapering course of prednisolone in addition to their regular treatment (45 mg on Day 1, 25 mg on study Day 7). Within 1.5 h before the Pitot static (PS) probe experiment, all of the asthmatic subjects inhaled the dose of terbutaline (generally 1 mg) that had resulted in maximal bronchodilatation during recording of a dose–response curve 1 wk before the prednisolone course. The healthy subjects had no pretreatment with a β2-agonist, since they showed no significant change from baseline after inhalation of 1 mg terbutaline.
The medical ethics committee of the University Hospital Rotterdam approved the study. Informed written consent was obtained from all subjects.
Baseline lung function tests consisted of maximal expiratory flow–volume (MEFV) maneuvers, whole-body plethysmography, and quasistatic pressure–volume (PV) measurements (Table 1).
MEFV curves were generated with a pneumotachometer (Jaeger, Würzburg, Germany) with a heated Lilly head. Both the pneumotachometer and the volume-constant plethysmograph were part of a Jaeger Masterlab system. Maximal bronchodilatation was determined during generation of a dose–response curve with terbutaline, the need for which was considered if an extra dose of 500 μg of terbutaline resulted in less than a 5% increase in FEV1, expressed as a percent of baseline. PV measurements were made according to the method of Zapletal and colleagues (20) and the European Community for Coal and Steel Standard (21). A latex balloon 8.5 cm long and with a 2-cm perimeter (International Medical Products bv., Zutphen, The Netherlands) was introduced into the esophagus in such a way that the length of the catheter from the nares to the balloon tip was one-fifth of the subject's body height (in cm) plus 9 cm (20). At the corresponding esophageal location, the pressure proved to be most negative during maximal inspiration. The balloon was filled with 1.5 ml of air, remained in situ throughout the subsequent PS probe experiment, and was subsequently checked for leakage.
For a detailed description of the equipment used to measure airway pressures, we refer to the studies by Pedersen and colleagues based on similar PS probe experiments (22, 23). The PS probe was a device comparable to the one used by Macklem and Mead (24) (Figure 1). The tubes connected to the PS probe for measurement of end-hole pressure (Ptot) and lateral pressure (Plat), and the tube attached to the esophageal balloon for indirect measurement of pleural pressure (Ppl), were connected to three identical pressure transducers (EMT34; Elema Schönander, Stockholm, Sweden), and via EMT 311 amplifiers to an electronic subtractor. Thus, both the pressure loss due to convective acceleration (Pca) (= Ptot − Plat) and the transmural pressure (Ptm) (= Plat − Ppl) were calculated on-line.
Fig. 1. Pitot static probe. O.D. = 3 mm; hole at tip to record Ptot; 1.3 cm from tip: six lateral holes with 0.5 mm diameter to record Plat. (From Reference 22, with permission.)
[More] [Minimize]Mouth flow (V′m) was measured with a nonheated Fleisch No. 3.5 pneumotachograph (Fleisch, Lausanne, Switzerland). Mouth flow and the pressure signals for Pca, Ptm, and Ppl were visible on-line. The electronic subtractor, flow amplifier, and PS probe were fabricated for this purpose at the University of Aarhus. The method used for tuning of the catheters among themselves and in combination with the pneumotachometer, and for testing of the PS probe, is described elsewhere (23). During in vitro tests, the accuracy of measurement of the A of different rigid tubes was in the range of ± 10% with Pca > 0.5 kPa, except for values of A > 2 cm2 (23).
All subjects were premedicated with 0.5 mg atropine intramuscularly at 0.5 h before introduction of the PS probe. The mouth, pharynx, vocal cords, and bronchial tree were anesthetized with topical anesthetic. A cuffless endotracheal tube (ETT) was passed between the vocal cords into the trachea after introduction of the bronchoscope. The bronchoscope was pulled back, and after placing the PS probe with its two catheters into the trachea through the ETT, the tube was removed. Through use of the reintroduced bronchoscope, the tip of the PS probe was positioned just above the entrance to the right lower lobe (in one subject the left lower lobe). This was the most peripheral position (Position 0) used for measurements. The PS probe was then pulled back under bronchoscopic view until the four other intended locations were reached. These were, respectively, the entrance to the middle lobe (Position 1), the middle of the mainstem bronchus (Position 2), 1 cm above the main carina (Position 3), and the midtrachea (Position 4) (Figure 2). In the following text, the term “position” indicates the intrabronchial position of the PS probe unless used in another context. The distance between two intended intrabronchial locations was determined individually by measuring at the subject's mouth the distance across which the catheter was pulled back between these locations. Next, the PS probe was repositioned at Position 0, the bronchoscope was carefully withdrawn, and the two catheters were pushed through and secured in two tightly fitting sideholes in a specially designed mouthpiece. The free ends of the PS probe catheters and the esophageal balloon catheter were then connected to the pressure transducers, and the mouthpiece was connected to the pneumotachograph. For all measurements subjects were in the sitting position in a chair, and were wearing a nose clip. Before each individual forced expiratory maneuver, the Ptot and Plat catheters were flushed forcefully with air to remove secretions from the end and side holes in the PS probe. If the on-line visible pressure signals indicated plugging or blockage of the endhole or sideholes, flushing was repeated and/or the PS probe was carefully turned or moved upstream and downstream for a maximum of 1 cm. Starting at Position 0, all subjects were asked to perform several MEFV maneuvers with the PS probe at each position. In addition, most of the subjects were asked to produce sigh and huff maneuvers, for the purposes of another study (24). At each position the procedure was repeated until acceptable results were obtained or four or five maneuvers had been performed. A total of up to 30 MEFV maneuvers were performed. A complete PS probe experiment took about 1 to 1.5 h.
Data selection. Initial assessment of the quality of the experimental maneuvers was done by eye, using the on-line visible curves on the computer screen. The V′m, Pca, Ptm, and Ppl signals from accepted maneuvers were saved digitally for subsequent calculations. Curves with obvious errors (nonmaximal effort or blockage of the endhole or sideholes in the PS probe) were discarded.
Asyst software (Asyst Software Technologies Inc., Rochester, NY) and the Statistical Package for Social Sciences (SPSS Inc., Chicago, IL) were applied for data acquisition and calculation of additional variables. Initially, the results of V′m, Ppl, Pca, and Ptm, and their derived variables at 13 different “levels” of each MEFV maneuver with the PS probe, were selected for further analysis. Each level was characterized by a lung volume equal to 80, 70, 60, 50, or 40% TLC, or 75, 50, and 25% FVC, or by a V′m equal to 100, 80, 60, 40, or 20% of peak expiratory flow (PEF). Each of these volume or flow levels counted as one “case” in the subsequent analysis. Ideally, a complete data set from a single PS probe MEFV maneuver therefore consisted of 13 cases. Subsequently, cases with a lung volume within 20% to 0% of FVC and cases with Ppl, Pca, airway compliance (Caw), or upstream pressure loss (Pfr) values of zero or less were discarded. In order to reduce problems of differences in the number of measurements and/or missing values among subjects and/or between positions, all data per variable per position from a single subject were later pooled within the following four lung volume ranges: 100% to 80%, 79% to 60%, 59% to 40%, and 39% to 20% of FVC.
Calculations. During each individual MEFV maneuver, with the PS probe in situ, data on V′m, Ppl, Pca, and Ptm were collected simultaneously. The curves for these variables served as the basis for the calculation of additional parameters. Figure 3 gives an overview of the different intra- and peribronchial pressures during forced expiration in terms of a “pressure walk” along the airway, as suggested by Mead (25).
Fig. 3. Pressures and pressure differences measured in an airway during expiration. V′= expiratory flow; Palv = alveolar pressure; Ppl = pleural pressure; Ptot = impaction pressure; Plat = lateral pressure; Pel = elastic recoil pressure; Ptm = transmural pressure; Pfr = upstream pressure loss; Pca = pressure loss due to convective acceleration; J = pressure head. (From Reference 23, with permission.)
[More] [Minimize]Using the Bernoulli equation, and assuming a blunt velocity profile and an incompressible medium, gives:
| Equation 1 |
where ρ is the density in g/cm3, A is cross-sectional area in cm2, P is pressure in kPa, and V′ is flow in L/s.
A at the PS probe can be calculated if Pca and V′ at the PS probe (V′PS) are known. The change in A with change in Ptm is a measure of Caw at the PS probe (Caw = dA/dPtm).
For the calculation of V′PS and of other parameters used in the study, and for the correction of exhaled volume for intrathoracic pressure (VPpl), we refer to the article by Pedersen and coworkers (23). A values calculated from Pca values below 0.5 kPa were disregarded in the subsequent analysis, since they proved to be less reliable during testing of the PS probe in rigid tubes (23). V′m was not corrected to body temperature–pressure–saturation conditions, and no attempt was made to correct for the difference in composition of air and alveolar gas. This will introduce a small but systematic error in both asthmatic and the healthy subjects, which we considered to be of no significance in the present study (26).
Figure 4A shows an example of an MEFV maneuver by a healthy subject with the PS probe at the lower end of the trachea (Position 3). V′PS and different pressures at the PS probe are plotted against the change in VPpl. For each individual PS probe experiment, A at the PS probe was plotted against local Ptm. This generally resulted in a curve with a regular and an irregular region (Figure 4B). A 2nd- to 5th- degree polynomial was fitted through the regular part of the A-versus-Ptm curve, and its agreement with the actual curve was assessed visually. The polynomial was saved for subsequent calculation of Caw values (provided by the first-order differentiation of the polynomial), together with the lower and upper Ptm values of the fitted part of the A–Ptm curve. For further analysis, A and Caw derived from the polynomial were used. Specific airway compliance (sCaw) was calculated as Caw/A.
Fig. 4. (A) MEFV maneuver in a healthy subject; local flows and pressures with the Pitot static probe at the lower end of the trachea. V′PS = local flow; VPpl = exhaled volume; Ptm = transmural pressure; Pca = pressure drop due to convective acceleration; J = pressure head at tip of probe; Ppl = pleural pressure. (B) On-line calculated curve of cross-sectional area A versus Ptm (solid line) and fitted polynomial (dashed line).
[More] [Minimize]Lung function results for asthmatic and healthy subjects were compared through a two-sample t test. All initial values for A, Ptm, Caw, and sCaw were logarithmically transformed.
When two or more values existed for a variable from a particular subject and specific position within a single volume range, the values were averaged. The aim was to ensure that for each subject, all 20 combinations of the five PS probe positions and four volume ranges used in the study contained a single mean logarithmic value of Caw, A, sCaw, and Ptm. Nevertheless, values for some positions and/or volume ranges were still missing for some subjects.
The outcome variables A, Caw, sCaw, and Ptm were then analyzed, using release 7 of the BMDP software system, module 5V (BMDP Statistical Software Inc., Los Angeles, CA), by applying repeated measurements analysis of variance (rmANOVA).
In this analysis, the following independent factors were used: (1) two intersubject factors, each with two levels: sex (male/female) and asthma (yes/no); and (2) two intrasubject factors: volume (four ordinal levels) and position (five ordinal levels [pos]). The assumed variance or covariance structure of the residuals was compound symmetry. The analysis began with a full model containing the four explanatory factors just described (volume and position as categorical factors) and all of their first-order interactions.
The initially used full model was as follows: Predicted variable = dis + sex + pos + volume + dis · sex + dis · pos + dis · volume + sex · pos + sex · volume + pos · volume, where dis = disease and pos = airway position. A backward elimination method was then applied, using likelihood ratio tests (embedding tests), in which in each step the factor with the highest p value above 0.05 was eliminated, in hierarchial fashion starting with the first-order interaction (i.e., main effects were not eliminated as long as they had a significant interaction with one of the other factors). If the restricted model thus selected still contained at least one of the categorical (ordinal) factors of volume and/or position, then these categorical factors were replaced (one at a time) by their continuous (numeric) versions in order to test whether an embedded model with a linear trend gave a better goodness of fit, considering the lesser number of degrees of freedom required for the further restricted model.
As a consequence of the logarithmic transformation, model predictions were back-transformed to geometric means, and effects must be interpreted in a multiplicative way.
When in the following text A, Ptm, Caw, or sCaw are discussed, their predicted values according to the final model are meant. Otherwise, the suffix “measured” is used. A two-sided level of significance of p < 0.05 was used.
Because of the invasive nature of the measurements, no complete 13-case data set from each MEFV maneuver with the PS probe could be obtained for each subject at each airway position. For the healthy subjects, an average of 2.9 volume ranges and for the asthmatic subjects an average of three volume ranges with data from each position per subject could be obtained. A data set with 1,477 cases from a total of 218 accepted MEFV maneuvers with the PS probe was used for the final analysis of A, Ptm, Caw, and sCaw. Tables 2 and 3 give the geometric means of the measured values of A, Ptm, Caw, and sCaw for the healthy and the asthmatic subjects, respectively.
| Healthy | 100%–80% FVC | 79%–60% FVC | 59%–40% FVC | 39%–20% FVC | ||||
|---|---|---|---|---|---|---|---|---|
| Position 0 A | 1.31 (0.69/2.27) | 1.17 (0.80/1.88) | 0.72 (0.44/1.02) | 0.65 (0.37/1.17) | ||||
| Ptm | −0.67 (−0.18/−2.10) | −1.04 (−0.15/−2.66) | −1.80 (−0.43/−7.16) | −2.72 (−0.96/−8.05) | ||||
| Caw | 0.71 (0.20/1.82) | 0.60 (0.21/1.41) | 0.40 (0.11/0.88) | 0.26 (0.10/0.51) | ||||
| sCaw | 0.54 (0.21/1.34) | 0.51 (0.19/1.32) | 0.55 (0.21/1.19) | 0.41 (0.13/1.07) | ||||
| n | 9 | 10 | 9 | 7 | ||||
| Position 1 A | 1.42 (0.87/2.27) | 1.18 (0.66/1.75) | 0.87 (0.47/1.44) | 0.64 (0.08/1.40) | ||||
| Ptm | −0.87 (−0.41/−1.51)* | −0.95 (−0.08/−3.90) | −1.83 (−0.36/−5.71) | −2.35 (−0.93/−8.08) | ||||
| Caw | 0.47 (0.18/3.84) | 0.39 (0.17/2.28) | 0.31 (0.08/1.49) | 0.23 (0.04/0.84) | ||||
| sCaw | 0.33 (0.16/1.02) | 0.33 (0.16/1.36) | 0.35 (0.08/1.49) | 0.36 (0.07/1.26) | ||||
| n | 10 | 12 | 13 | 11 | ||||
| Position 2 A | 1.22 (0.59/2.41) | 1.07 (0.58/1.69) | 0.85 (0.52/1.37) | 0.85 (0.53/1.32) | ||||
| Ptm | −1.37 (−0.82/−4.41) | −1.61 (−0.40/−3.40) | −2.63 (−0.94/−5.28) | −2.83 (−1.67/−4.80) | ||||
| Caw | 0.22 (0.03/1.14) | 0.27 (0.09/0.82) | 0.19 (0.04/0.59) | 0.23 (0.06/0.44) | ||||
| sCaw | 0.18 (0.05/0.64) | 0.25 (0.08/0.57) | 0.22 (0.07/0.43) | 0.27 (0.11/0.50) | ||||
| n | 7 | 10 | 8 | 8 | ||||
| Position 3 A | 1.05 (0.61/1.77) | 0.84 (0.40/2.04) | 0.72 (0.33/1.29) | 0.80 (0.34/1.40) | ||||
| Ptm | −2.15 (−0.51/−5.60) | −2.82 (−0.55/−11.53) | −3.26 (−1.04/−11 .17) | −2.58 (−1.17/−7.35) | ||||
| Caw | 0.24 (0.08/0.64) | 0.12 (0.02/0.40) | 0.12 (0.03/0.81) | 0.11 (0.02/0.34) | ||||
| sCaw | 0.23 (0.12/0.59) | 0.15 (0.03/0.45) | 0.16 (0.05/0.60) | 0.14 (0.05/0.38) | ||||
| n | 9 | 13 | 12 | 11 | ||||
| Position 4 A | 0.98 (0.41/1.79) | 0.90 (0.30/1.58) | 0.86 (0.42/1.67) | 1.02 (0.53/1.78) | ||||
| Ptm | −2.69 (−0.90/−5.77) | −3.62 (−1.01/−10.18) | −4.34 (−1.68/−10.10) | −2.43 (−1.03/−3.96) | ||||
| Caw | 0.14 (0.04/0.68) | 0.12 (0.01/1.41) | 0.15 (0.04/1.36) | 0.15 (0.06/0.49) | ||||
| sCaw | 0.14 (0.04/0.69) | 0.13 (0.02/0.89) | 0.17 (0.07/0.81) | 0.15 (0.04/0.55) | ||||
| n | 10 | 13 | 10 | 10 |
| Asthma | 100%–80% FVC | 79%–60% FVC | 59%–40% FVC | 39%–20% FVC | ||||
|---|---|---|---|---|---|---|---|---|
| Position 0 A | 0.78 (0.21/1.49) | 0.63 (0.20/1.44) | 0.82 (0.57/1.32) | 0.97 (0.50/1.84) | ||||
| Ptm | −2.08 (−1.22/−2.62) | −3.33 (−0.56/−8.77) | −2.14 (−0.22/−7.18)* | −2.53 (−1.17/−4.38) | ||||
| Caw | 0.11 (0.04/0.22) | 0.17 (0.07/0.85) | 0.21 (0.11/0.33) | 0.16 (0.04/0.28) | ||||
| sCaw | 0.15 (0.04/0.43) | 0.27 (0.09/1.48) | 0.25 (0.11/0.41) | 0.16 (0.04/0.35) | ||||
| n | 6 | 8 | 8 | 7 | ||||
| Position 1 A | 1.38 (0.67/2.84) | 1.04 (0.54/2.41) | 1.06 (0.46/2.62) | 1.14 (0.69/2.42) | ||||
| Ptm | −1.36 (−0.11/−5.23) | −2.68 (−0.51/−6.96) | −2.96 (−1.12/−8.40) | −2.09 (−0.85/−5.25) | ||||
| Caw | 0.19 (0.11/0.28) | 0.22 (0.07/0.57) | 0.27 (0.15/0.77) | 0.21 (0.07/0.43) | ||||
| sCaw | 0.14 (0.04/0.33) | 0.21 (0.06/0.66) | 0.26 (0.08/0.74) | 0.19 (0.08/0.39) | ||||
| n | 7 | 9 | 8 | 6 | ||||
| Position 2 A | 1.41 (0.88/2.45) | 1.22 (0.62/2.58) | 1.22 (0.74/2.32) | 1.37 (0.83/3.26) | ||||
| Ptm | −1.51 (−0.47/−2.94) | −2.05 (−0.43/−4.77) | −2.21 (−0.47/−5.79) | −2.48 (−0.73/−7.23) | ||||
| Caw | 0.20 (0.04/1.03) | 0.15 (0.06/0.24) | 0.12 (0.05/0.27) | 0.14 (0.02/0.43) | ||||
| sCaw | 0.14 (0.03/0.42) | 0.13 (0.09/0.31) | 0.10 (0.03/0.15) | 0.10 (0.01/0.17) | ||||
| n | 7 | 9 | 8 | 7 | ||||
| Position 3 A | 1.02 (0.86/1.41) | 1.01 (0.63/2.13) | 0.98 (0.67/1.89) | 0.97 (0.67/1.87) | ||||
| Ptm | −1.89 (−0.53/−4.65) | −3.14 (−1.43/−7.38) | −3.39 (−1.39/−7.99) | −3.11 (−1.29/−7.43) | ||||
| Caw | 0.15 (0.10/0.30) | 0.14 (0.07/0.24) | 0.10 (0.07/0.17) | 0.12 (0.07/0.18) | ||||
| sCaw | 0.14 (0.09/0.35) | 0.14 (0.08/0.22) | 0.11 (0.06/0.21) | 0.13 (0.07/0.20) | ||||
| n | 6 | 8 | 8 | 7 | ||||
| Position 4 A | 1.13 (0.83/1.34) | 0.91 (0.43/1.35) | 0.93 (0.70/1.52) | 1.10 (0.86/1.57) | ||||
| Ptm | −3.52 (−1.41/−7.03) | −4.35 (−0.75/−12.86) | −4.17 (−0.72/−12.36) | −3.41 (−0.89/−9.54) | ||||
| Caw | 0.17 (0.08/0.72) | 0.10 (0.04/0.26) | 0.10 (0.05/0.22) | 0.12 (0.07/0.34) | ||||
| sCaw | 0.15 (0.07/0.60) | 0.11 (0.05/0.19) | 0.11 (0.07/0.20) | 0.11 (0.06/0.22) | ||||
| n | 7 | 8 | 8 | 8 |
A was significantly determined by sex, airway position, volume, the interactions of disease with airway position, and volume level, and of sex with volume (Table 4). In general, A became smaller with decreasing volume, with the greatest decline in A occurring between the volume ranges of 100% to 80% and 79% to 60% FVC. A was smaller in the female subjects at all volume ranges: at 100% to 80% FVC, Afemale was 65% of Amale (p < 0.001); at 79% to 60% FVC it was 63% of Amale (p < 0.001); at 59% to 40% FVC it was 77% of Amale (p < 0.05); and at 39% to 20% FVC it was 83% of Amale (p > 0.1). The difference in A between males and females did not depend on the healthy or asthmatic status of the subjects or on the position of the PS probe.
| p Values | In A | In −Ptm | In Caw | In sCaw | ||||
|---|---|---|---|---|---|---|---|---|
| Disease | 0.793 | 0.654 | < 0.001 | < 0.001 | ||||
| Sex | 0.002 | 0.003 | — | 0.018 | ||||
| Position (categorical) | < 0.001 | < 0.001 | — | — | ||||
| Position (continuous) | — | — | < 0.001 | < 0.001 | ||||
| Volume (categorical) | < 0.001 | < 0.001 | — | — | ||||
| Volume (continuous) | — | — | < 0.001 | — | ||||
| Dis·pos | 0.002 | 0.025 | < 0.001 | 0.003 | ||||
| Dis·vol | < 0.001 | < 0.001 | — | — | ||||
| Sex·pos | — | 0.039 | — | 0.013 | ||||
| Sex·vol | 0.019 | 0.017 | — | — |
Ptm became more negative with decreasing volume during forced expiration in the individual subjects. Overall, disease status did not contribute significantly to Ptm. However, there was a significant interaction between disease status and airway position and volume range that contributed in a non–log-linear way to Ptm. In the asthmatic subjects, Ptm was on average more negative in the upstream positions at comparable high lung volumes.
Caw was log-linearly related to disease status, airway position, and volume level, and to the interaction of disease and airway position. Caw at the entrance to the lower lobe (Position 0) in the asthmatic subjects was 0.52 of local compliance in the healthy subjects. This difference decreased toward the trachea and became minimal at the trachea. In both the asthmatic and the healthy subjects, Caw was lower at a more central position of the PS probe (from Position 0 to Position 4). With each downstream step in position, Caw decreased in the healthy subjects by 29.5% and in the asthmatic subjects by 10.5%. In the healthy subjects, Caw at the lower lobe (Position 0) was 4.04 times larger than at the midtrachea (Position 4), whereas in the asthmatic subjects this factor was 1.56.
Disease status, sex, airway position, and the interaction of airway position with sex and disease contributed to sCaw in a log-linear way. sCaw was significantly lower in the asthmatic than in the normal subjects (Figure 5: measured sCaw). In contrast to Caw, sCaw was not significantly determined by volume. As could be expected from the dependence of A on sex, sCaw was also determined by sex, and was smaller in the male subjects for a given disease status (Table 5). The difference in sCaw between males and females became smaller with a more downstream position. There was no interaction between sex and disease status. With each subsequent downstream position, sCaw decreased by 33% in the healthy females, 24% in the healthy males, 23% in the asthmatic females, and 13% in the asthmatic males
Fig. 5. Geometric mean values of measured sCaw (1/kPa) versus Pitot static probe position for lung volume range of 79% to 60% FVC in healthy and asthmatic subjects.
[More] [Minimize]| sCaw Asthmatic: Healthy Subjects | sCaw Male:Female | |||
|---|---|---|---|---|
| At lower lobe | 0.55 | 0.70 | ||
| At middle lobe | 0.62 | 0.78 | ||
| Mid-mainstem bronchus | 0.72 | 0.88 | ||
| End trachea | 0.84 | 0.99 | ||
| Midtrachea | 0.96 | 1.12 |
In summary, the foregoing results show that: (1) Central airways of asthmatic subjects behave more stiffly during forced expiration than those of healthy subjects. The difference between healthy and asthmatic subjects became smaller at the level of the trachea. (2) Caw and sCaw decrease with a more downstream position in the tracheobronchial tree. (3) A in the central airways is larger in males than in females. The difference between the sexes became smaller with decreasing lung volume. (4) sCaw is volume-independent.
This is the first in vivo study in long-lasting asthma of functional airway change as a possible result of chronic inflammation and structural remodeling of the airway wall. The main purpose of the study was to examine the compliance of central airways during forced expiration in young adult subjects with long-lasting asthma. The results indicate that central airways behave more stiffly in asthma patients than in healthy subjects.
With regard to the technical aspects of the study and their limitations, we refer to related studies by Pedersen and colleagues (22, 23) done with the same method and calculations.
The invasive nature and time-consuming complexity of the experiments conducted in the study limited the inclusion of more volunteers. A complete matching of the two study groups could therefore not be obtained (Table 1). However, because we found no significant differences in TLC or measures of airway patency, we assumed that differences between the two study groups in lung-volume changes caused by air compression could be neglected.
For ethical reasons, only the asthmatic subjects were pretreated with systemic corticosteroids and a β2-agonist. The β2-agonist may have reduced bronchial tone and therefore increased Caw in the asthmatic as compared with the healthy subjects (27). The difference found in sCaw may therefore have been even larger before bronchodilator treatment.
Despite instructions and encouragement, not all subjects could repeatedly exhale across the full range of FVC with the PS probe in situ. Typical problems were fits of coughing during the experiment, increased mucus production with a need to swallow, and a premature cessation of local anesthetic effect. At some PS probe positions in some subjects, no reliable data could be obtained because of wedging of the probe, especially at low lung volumes, or because of temporary mucus obstruction of one or more holes in the probe.
Accepted maneuvers could still have contained artifacts over certain volume ranges. These ranges were omitted in the final analysis.
Although FVC was lower in the asthmatic subjects, FVC was chosen as the basis for distribution of data over volume ranges because both FVC and PS probe data were obtained during comparable forced expiratory maneuvers. It was expected that the mechanism(s) responsible for a reduced FVC in the asthmatic subjects also occurred during the MEVF maneuvers with the PS probe in these subjects.
For each individual PS probe measurement, only the volume range corresponding specifically to the Ptm range for which the polynomial through the A–Ptm curve was fitted could be used for analysis. Because this volume range was generally smaller than FVC, this often resulted in less than the intended total of 13 cases per measurement.
These factors explain why almost all subjects had volume ranges with missing data for A, Ptm, Caw, and sCaw. However, no clear pattern was noticeable in the distribution of these “empty” volume ranges among the 10 asthmatic and 14 healthy subjects that could have significantly influenced the results of the multiple regression modeling. With regard to the differences in A, ln A, Ptm, ln Ptm, Caw, ln Caw, sCaw, and ln sCaw between the healthy and the asthmatic subjects, there were no statistically significant differences between the data set used for final analysis and the data set containing the finally excluded cases. This was analyzed for the pooled data for all five PS probe positions and all four volume ranges. Therefore, no bias with regard to the differences found for asthmatic versus healthy subjects and caused by the exclusion of these cases is likely to have occurred. Because data from only 24 subjects were analyzed, we assumed a compound symmetry structure for the variance and covariance matrices of the residuals in the statistical model used in the study, in order to reduce the number of estimated parameters.
The choice for the logarithmic transformation was based on the theory that the physically defined relationships between the various outcome measures (A, Ptm, Caw, and sCaw) imply that values for these variables are generated by multiplicative rather than by additive processes.
The values of A found in our study are comparable to those described by Macklem and Wilson (28), who used the same method that we used. Brooks, using the acoustic reflection technique, found a mean tracheal area of 1.87 ± 0.1 cm2 in 11 adolescent subjects (29). This is somewhat larger than our results at midtrachea. However, our findings were obtained during forced expiration with a compressive negative transmural pressure. The value of A throughout the central airways was significantly smaller in the females in our study at all volume ranges, independent of whether they had asthma or not. This is consistent with the difference in A between sexes as found by Martin and colleagues with the acoustic reflection technique during tidal breathing (30).
The decrease in A during forced expiration can be explained by dynamic compression causing a decrease in the dilating effect of Ptm. As shown in our study, Ptm decreases with volume. A may also decrease in direct relation to volume because of a reduction in axial tension and possibly because of a decrease in tethering by surrounding tissue. This is in agreement with findings by others. Hoffstein and associates found that the area of tracheal and bronchial segments increases with an increase in lung volume and transpulmonary pressure (31). Hughes and coworkers found decreasing airway diameters with a history of volume deflation (32).
At all positions we found that on average, measured Ptm (= Plat − Ppl) became more negative during forced expiration with decreasing lung volume. This can predominantly be explained as follows: Ppl and therefore peribronchial pressure are maintained during an MEFV maneuver. In contrast, elastic recoil pressure and therefore outward-acting intrabronchial pressure (Plat) decrease progressively when lung volume becomes smaller. The end result is a negative Ptm, compressing the airway.
Narrowing of the airways may increase the pressure loss caused by friction and (possibly) by convective acceleration during forced expiration. This leads to a further reduction of intrabronchial pressure and therefore of A. Extra narrowing of the airways stretches the airway–parenchyma attachments and increases parenchymal support. This may lead to a decrease in dynamic airway compliance (Cdyn) (33-35).
At comparable (high) lung volumes, Ptm was on average more negative at the more peripheral airway positions in the asthmatic subjects. A more leftward (negative) position on the A–Ptm curve of an airway will coincide with a lower A and with a less steep slope. A more negative Ptm at comparable lung volume may therefore (partly) explain lower Caw values in our study, and may hamper comparison of data on compliance for the asthmatic as compared with the healthy subjects for evaluation of differences in the two groups' airway structures.
The (nonlinear) A–Ptm relationship was volume dependent, mainly at high lung volumes, with Ptm becoming more negative and A and Caw decreasing with decreasing lung volume. This was also concluded from similar PS probe experiments in humans during repetitive huff maneuvers by Pedersen and associates (23). This means that the volume at which Caw is measured has to be taken into account when comparing compliance data.
The decrease in Caw with decreasing volume may be caused by cartilage rings moving toward each other when length tension is reduced during exhalation, resulting in stiffer central airways, as reported for calves' tracheas by Suki and coworkers (36). This is in agreement with findings by Olsen and colleagues (27). They observe that muscular constriction of central airways may lead to a decrease in compressibility by covering the soft parts of the airways with sleeves of fibrocartilage (27).
The larger Caw at more upstream airways in both groups of subjects in our study is in agreement with findings in dogs (37) and humans (31).
The dependency of Caw on volume and Ptm complicates the interpretation of differences in Caw in healthy and asthmatic subjects. When Caw was corrected for A (Caw/A = sCaw), the volume dependency of Caw disappeared. This indicates that the decrease in Caw with decreasing lung volume is mainly related to a decrease in A. We also made this finding during maximal-effort huff maneuvers in our study of the speed index at peak flow (23). When correlated with the dependence of A on sex, sCaw was sex-dependent: a lower sCaw was found in the central bronchi for male subjects. The difference between males and females in sCaw did not depend on disease status, and became smaller toward the trachea.
Sasaki and colleagues (35), Takishima and coworkers (34), and Nakamura and coworkers (37) deduced from experiments with excised dog lungs that sCaw is independent of Ptm (see Appendix). The use of sCaw as a measure of airway elastic properties therefore circumvents the problem of differences in Ptm between asthmatic and healthy subjects at comparable lung volumes. In the studies just cited, the slope of the change in bronchial volume plotted against the change in the bronchial pressure (dlog Vbr/d Pbr) became steeper with a lower distending pressure at a lower elastic recoil. These findings indicate that comparison of elastic behavior of bronchi should preferably be made with sCaw as the most relevant parameter, and at fixed elastic recoil values. This latter condition was met in our study because sCaw was determined over equal ranges of lung volume with comparable elastic recoil values.
In our study, sCaw was significantly lower in the asthmatic than in the healthy subjects, especially at more peripheral positions in the airway. This strengthens the conclusion that in asthmatic individuals the airway wall behaves more stiffly during forced expiration.
The decrease in sCaw found in our asthmatic subjects may be due to structural remodeling in a chronically inflamed airway wall. However, the net effect on airway mechanics of each of the many different structural changes that occur in the asthmatic airway wall is unclear.
A number of arguments favor the hypothesis that inflammatory remodeling results in a decrease in Caw. Thus, all layers of the airway wall have been found to thicken as a result of extensive formation of granulation tissue (11). The end result may be scarlike tissue with an increased deposition of densely packed subepithelial collagen below an essentially normal epithelial basal lamina (38), with fragments of elastic fibers and with inactive-appearing fibroblasts. The mechanical consequences of this probably depend on the chemistry of the different collagen types involved and on the architecture of the collagen deposits. It can be imagined that an increased collagen fibril density and thickened subepithelial matrix will increase the tensile stiffness and resistance to deformation of the airway wall (7, 11, 15). Indeed, Wilson and colleagues found a reduced increase in airway dead space volume during inspiration in asthmatic as compared with control subjects (39). They suggested that the addition of scar-type collagen to the airway wall contributes to a reduction in airway distensibility without reducing the airway lumen. Work by Colebatch and associates supports the notion that airways of asthmatic individuals are less distensible than those of normals (40). However, one may question whether distensibility can be compared with compressibility of airways as a measure of Caw. Mitchell and coworkers conclude, on the basis of a recent model study of sensitized pig bronchi, that lung inflammation increases airway stiffness and, by implication, airway wall load on the smooth muscle (41).
Despite factors that may decrease Caw, airway remodeling in chronic asthma will involve degradation of matrix components as a result of the chronic destruction, healing, and repair that occur in the disease. Bousquet and colleagues reported that an abnormal, fragmented, superficial network of elastic fibers was present in most asthmatic subjects, together with patchy, tangled, and thickened elastic fibers of the deeper layers of the airway wall (6, 17). Electron microscopic studies suggested a severe elastolysis with fragmented elastin fibers in the asthmatic subjects' large airways, possibly as a consequence of chronic inflammation, and/or mechanical stretching caused by hyperinflation or airway edema in asthma or both (6, 17). An intact network of elastic fibers is crucial for the maintenance of structure and mechanical properties of the lung. A consequence of elastic fiber degradation could be an increased compressibility of central airways in long-lasting asthma, coinciding with decreased elastic retraction in the lung. This was supported by the finding of Bramley and associates that a single asthmatic airway preparation showed less passive tension and a threefold greater shortening than did six nonasthmatic lobar airway preparations, without any difference in the amount of smooth muscle in the two types of preparation (42). Bramley and associates suggested that the release of protease enzymes (such as elastases and/or collagenases) from inflammatory or resident cells degrades elements of the extracellular matrix and reduces airway tissue elastance.
It remains unclear whether these pathophysiologic findings in vitro may be extrapolated to the in vivo situation, in which many other, different factors interact and determine the final mechanical behavior of the airways. Paré and coworkers reported cartilage proteoglycan degradation and cartilage remodeling in each of six fatal cases of asthma—changes that may decrease the force needed to deform the airways (7, 11). Recent computational modeling by Lambert and associates (16) predicted that the Ptm resulting in airway collapse depends on the number of folds in the airway mucosa. The degree of mucosal buckling is, according to the computational-model studies by Wiggs and colleagues, mainly determined by the compressive stiffness of the airway wall (43). The sum of the elastic moduli of the different layers in the airway wall is therefore an important determinant of airway compressibility.
We found stiffer airway behavior during forced expiration in stable asthmatic patients than in healthy subjects. This supports the hypothesis that chronic airway inflammation and/or remodeling in asthmatic individuals may lead to a decreased Cdyn of the airways. Since our observations were limited to the central airways, we cannot predict whether a decreased compliance can also be found in the more peripheral, intrapulmonary airways. It is unclear how chronic inflammation with subsequent remodeling of the airway wall changes the mechanical properties of the airways. The only way to obtain more insight into this process seems to be by means of in vivo measurements. However, the exact role of airway inflammation as a factor contributing to decreased airway compliance can only be assessed when subepithelial structural changes (e.g., collagen deposition) and airway mechanical properties are studied simultaneously.
Increased stiffness of the airway wall will impose a greater load on bronchial smooth muscle during bronchoconstriction, and therefore has, to a certain degree, a protective effect. However, it does not rule out the appearance of excessive bronchoconstriction related to BHR, because many factors causing thickened airways also contribute to excessive airway narrowing (14).
The finding in relatively young asthmatic subjects of functional abnormalities of the airways as a possible result of long-lasting inflammation stresses the importance of preventing and treating the airway inflammation in asthma as early and optimally as possible. It has to be kept in mind, however, that the possible long-term effects of these functional abnormalities remain unknown.
The authors thank T. F. Pedersen for Asyst programming and J. Aerts for assistance during the PS probe experiments.
Supported by grant 89.61 from the Netherlands Asthma Foundation.
| 1. | Beasly C. R. W., Roche W. R., Roberts J. A., Holgate S. T.Cellular events in the bronchi in mild asthma and after bronchial provocation. Am. Rev. Respir. Dis.1391989806817 |
| 2. | Foresi A., Bertorelli G., Pesci A., Chetta A., Olivieri D.Inflammatory markers in bronchoalveolar lavage and in bronchial biopsy in asthma during remission. Chest981990528535 |
| 3. | Lundgren R., Soderberg M., Horstedt P., Stenlin R.Morphological studies of bronchial mucosa from asthmatics before and after ten years of treatment with inhaled steroids. Eur. Respir. J.11988883889 |
| 4. | Jöbsis Q., Raatgreep H. C., Hermans P. W. M., de Jongste J. C.Hydrogen peroxide in exhaled air is increased in stable asthmatic children. Eur. Respir. J.101997519521 |
| 5. | Cutz E., Levison H., Cooper D. M.Ultrastructure of airways in children with asthma. Histopathology21978407421 |
| 6. | Bousquet J., Chanez P., Lacoste J. Y., White R., Vic P., Godard P., Michel F. B.Asthma: a disease remodelling the airways. Allergy471992311 |
| 7. | Paré P. D., Roberts C. R., Bai T. R., Wiggs B. J.The functional consequences of airway remodelling in asthma. Monaldi Arch. Chest Dis.521997589596 |
| 8. | Jeffery P. K., Wardlaw A. J., Nelson F. C., Collins J. V., Kay A. B.Bronchial biopsies in asthma: an ultrastuctural quantitative study and correlation with hyperreactivity. Am. Rev. Respir. Dis.140198917451753 |
| 9. | Laitinen, L. A., A. Laitinen, A. Altraja, I. Virtanen, M. Kampe, B. G. Simonsson, S. E. Karlsson, L. Hakansson, P. Venge, and H. Sillastu. 1996. Bronchial biopsy findings in intermittent or ‘early' asthma. J. Allergy Clin. Immunol. 98(Pt. 2):S3–S6 (discussion: S33–S40). |
| 10. | Chetta A., Foresi A., Del Donno M., Bertorelli G., Pesci A., Olivieri D.Airways remodelling is a distinctive feature of asthma and it is related to the severity of the disease. Chest1111997852857 |
| 11. | Roberts C. R.Is asthma a fibrotic disease? Chest1071995111s117s |
| 12. | Li X., Wilson J. W.Increased vascularity of the bronchial mucosa in mild asthma. Am. J. Respir. Crit. Care Med.1561997229233 |
| 13. | Okazawa M., Müller N., McNamara A. E., Child S., Verburgt J., Paré P. D.Human airway narrowing measured using high definition computed tomography. Am. J. Respir. Crit. Care Med.154199615051510 |
| 14. | James A. L., Paré P. D., Hogg J. C.The mechanics of airway narrowing in asthma. Am. Rev. Respir. Dis.1391989242246 |
| 15. | Paré P. D., Bai T. R.The consequences of chronic allergic inflammation. Thorax.501995328332 |
| 16. | Lambert R. K., Codd S. L., Alley M. R., Pack R. J.Physical determinants of bronchial mucosal folding. J. Appl. Physiol.77199412061216 |
| 17. | Bousquet J., Lacoste J. Y., Chanez P., Vic P., Godard P., Michel F. B.Bronchial elastic fibers in normal subjects and asthmatic patients. Am. J. Respir. Crit. Care Med.153199616481654 |
| 18. | American Thoracic SocietyStandards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am. Rev. Respir. Dis.143198714331437 |
| 19. | National Heart, Lung, and Blood Institute, National Institutes of Health. 1992. International asthma management project: international consensus report on diagnosis and treatment of asthma. Clin. Exp. Allergy 22(Suppl. 1):1–72. |
| 20. | Zapletal A., Paul T., Samánek M.Pulmonary elasticity in children and adolescents. J. Appl. Physiol.401976953961 |
| 21. | Quanjer, P. H. 1983. Standardized lung function testing. Bull. Eur. Physiopathol. Respir. 19(Suppl. 5):22s–27s. |
| 22. | Pedersen O. F., Thiessen B., Lyager S.Airway compliance and flow limitation during forced expiration in dogs. J. Appl. Physiol.521982357369 |
| 23. | Pedersen O. F., Brackel H. J. L., Bogaard J. M., Kerrebijn K. F.Wave-speed determined flow limitation at peak flow in normal and asthmatic subjects. J. Appl. Physiol.83199717211732 |
| 24. | Macklem P. T., Mead J.Factors determining maximum expiratory flow in dogs. J. Appl. Physiol.251968159169 |
| 25. | Mead J.Expiratory flow limitation: a physiologist's point of view. Fed. Proc.39198027712775 |
| 26. | Quanjer P. H., Tammeling G. J., Cotes J. E., Pedersen O. F., Peslin R., Yernault J. C.Lung volumes and forced ventilatory flows: report working party standardization of lung function tests, European Community for Steel and Coal. Eur. Respir. J.61993540 |
| 27. | Olsen C. R., Stevens A. E., Pride N. B., Staub N. C.Structural basis for decreased compressibility of constricted trachea and bronchi. J. Appl. Physiol.2319673539 |
| 28. | Macklem P. T., Wilson N. J.Measurement of intrabronchial pressure in man. J. Appl. Physiol.201965653663 |
| 29. | Brooks L. A.Tracheal size and distensibility in patients with cystic fibrosis. Am. Rev. Respir. Dis.1411990513516 |
| 30. | Martin T. R., Castile R. G., Fredberg J. J., Wohl M. E., Mead J.Airway size is related to sex but not lung size in normal adults. J. Appl. Physiol.63198720422047 |
| 31. | Hoffstein V., Castile R. G., O'Donnell C. R., Glass G. M., Strieder D. J., Wohl M. E., Fredberg J. J.In vivo estimation of tracheal distensibilitly and hysteresis in normal adults. J. Appl. Physiol.63198724822489 |
| 32. | Hughes J. M. B., Hoppin F. G., Mead J.Effect of lung inflation on bronchial length and diameter in excised lungs. J. Appl. Physiol.3219722535 |
| 33. | Hughes J. M. B., Jones H. A., Wilson A. G., Grant B. J. B., Pride N. B.Stability of intrapulmonary bronchial dimensions during expiratory flow in excised lungs. J. Appl. Physiol.371974684694 |
| 34. | Takishima T., Sasaki H., Sasaki T.Influence of lung parenchyma on collapsibility of dog bronchi. J. Appl. Physiol.381975875881 |
| 35. | Sasaki H., Takishima T., Sasaki T.Influence of lung parenchyma on dynamic bronchial collapsibility of excised dog lungs. J. Appl. Physiol.421977699705 |
| 36. | Suki B., Habib R. H., Jackson A. C.Wave propagation, input impedance, and wall mechanics of the calf trachea from 16 to 1,600 Hz. J. Appl. Physiol.75199327552766 |
| 37. | Nakamura M., Sasaki H., Sekizawa K., Ishii M., Takishima T., Hoppin F. G. J.Series distribution of airway collapsibility in dogs. J. Appl. Physiol.501981325333 |
| 38. | Roche W. R., Williams J. H., Beasly R., Holgate S. T.Subepithelial fibrosis in the bronchi of asthmatics. Lancet11989520524 |
| 39. | Wilson J. W., Li X., Pain M. C. F.The lack of distensibility of asthmatic airways. Am. Rev. Respir. Dis.1481993806809 |
| 40. | Colebatch, H. J. H., I. A. Greaves, and C. K. Y. Ng. 1979. Pulmonary mechanics in diagnosis. In M. A. de Kock, J. A. Nadel, and C. M. Lewis, editors. Mechanics of Airway Obstruction in Human Respiratory Disease. A. A. Balkema, Cape Town. 25–48. |
| 41. | Mitchell H. W., Turner D. J., Gray P. R., McFawn P. K.Compliance and stability of the bronchial wall in a model of allergen- induced lung inflammation. J. Appl. Physiol.861999932937 |
| 42. | Bramley A. M., Thomson R. J., Roberts C. R., Schellenberg R. R.Hypothesis: excessive bronchoconstriction in asthma is due to decreased airway elastance. Eur. Respir. J.71994337341 |
| 43. | Wiggs B. R., Hrousis C. A., Drazen J. M., Kamm R. D.On the mechanism of mucosal folding in normal and asthmatic airways. J. Appl. Physiol.83199718141821 |
