| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Anesthesiology and Critical Care Medicine, 2Department of Environmental Health Sciences, Division of Physiology, Department of Medicine, 3Division of Pulmonary and Critical Care Medicine and 4Division of Allergy and Clinical Immunology, The Johns Hopkins Medical Institutions, Baltimore, Maryland
Submitted 19 September 2005 ; accepted in final form 1 February 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT GlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
computerized tomography; residual volume; total lung capacity; forced vital capacity
To test this hypothesis, we assessed the contribution of airway structural alterations in asthma to AHR, with the assumption that AHR can be estimated by the effect on pulmonary function of the amount of airways smooth muscle activation associated with the presence of baseline smooth muscle tone (BSMT). To accomplish this, we measured airway structure by high-resolution computed tomography (HRCT) and pulmonary function in 21 asthmatic volunteers before and after maximal bronchodilation with inhaled albuterol. We evaluated luminal diameters and wall thicknesses of airways for three airway sizes. We used the difference in pulmonary function before and after albuterol to quantify the magnitude of the airways response to BSMT and determined whether there was a relationship between the dimensions of the airways and the magnitude of the pre- to postalbuterol differences in pulmonary function.
We found that airway structure played a major role in the magnitude of AHR through different effects on airway closure [residual volume (RV)], on the one hand, and the magnitude of the increase in total lung capacity (TLC), on the other, with the summation of these effects determining the magnitude of the decrease in forced vital capacity (FVC). We propose that some of the major features of AHR in asthma and its absence in health are the result of the relationship between airways structure and the changes in TLC and RV that occur with increasing smooth muscle tone.
|
Glossary |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
METHODS |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
Pulmonary function measurements. Spirometry measurements were performed in triplicate with the subjects in a seated position using a portable Koko spirometer (PDS, Louisville, CO). Immediately after the spirometry measurements, lung volumes were measured in a body plethysmograph (MedGraphics, St. Paul, MN) with the subjects in a seated position. The volunteers were instructed to pant against the closed shutter at a frequency of 1–3 Hz. Most of the volunteers could not tolerate more than three sets of maneuvers in the box. The best quality set of measurements from a single maneuver was utilized. The volunteers underwent computed tomography (CT) scanning within 3–5 min of plethysmography.
We operationally defined the RV as the volume that remains in the lung at the end of a FVC. Therefore, the RV was considered to be the difference between the TLC and the FVC. There was no significant difference between the FVC and the slow vital capacity determined during the lung volume measurements, before or after the inhalation of albuterol.
Albuterol treatment.
After recording baseline spirometric and plethysmographic values, each volunteer received nebulized albuterol (2.5 ml of 0.083%) over 
20 min, after which spirometric and thoracic gas volume measurements were repeated. The nebulized albuterol treatment was repeated until either the increase in FEV1 between treatments was <5% (plateau effect) or the heart rate increased above 140 beats/min. All volunteers reached a plateau before a heart rate of 140 beats/min was reached.
HRCT image acquisitions.
All scans were performed using spiral CT (Somatom Plus 4, Siemens) with settings of 120 kVp, 170 mA, 2-mm slice thickness, rotation feed of 2 mm/s, and a reconstruction interval of 1 mm (total 61 scans per set) during a single breath hold for 24 s at FRC and at TLC (see below) (4). Scanning began 6 cm above the top of the dome of the diaphragm, at FRC, and moved caudally. A reference scan was acquired before each spiral CT scan set to ensure reproducible image location in the lung. The images were reconstructed as a 16-bit 512 x 512 matrix using a field of view of 200 mm. Images were reconstructed with the use of a high-spatial frequency (resolution) algorithm that enhanced edge detection, at a window level of –450 Hounsfield units and a window width of 1,350 Hounsfield units. All airways visualized approximately perpendicular to the scan plane (long-to-short axis ratio <1.5:1) were measured. For repeated airway measurements in a given subject, within each experimental protocol, adjacent anatomic landmarks, such as airway or vascular branching points, were defined on the baseline HRCT images and again identified on the TLC scans and on the scans after the administration of albuterol (7).
Airway measurements with HRCT. Airway dimensions were derived from airway luminal area measurements. The airway luminal area measurement methodology has previously been described (8) and validated (1). Briefly, the HRCT images were transferred to a UNIX-based workstation and analyzed using the airway analysis module of the Volumetric Image and Display Analysis software package (Department of Radiology, Division of Physiologic Imaging, University of Iowa, Iowa City, IA). To measure airway luminal areas, the operator drew a rough isocontour estimate of the lumen of the airway. The software program automatically located a precise isocontour perimeter of the airway lumen by sending out rays in a spoke-wheel fashion to a predesignated pixel intensity level that defined the luminal edge of the airway wall. The length of the rays was set at 6 pixels. The software program used an algorithm for edge detection based on the "full-width-half-maximum" principle. The edge of the wall was defined by the points along the lines where the pixel intensity changed to one-half its maximum through the wall. All full and partial pixels (full pixel size equals 0.1537 mm2 with our settings) within the adjusted isocontour were counted and represented the airway luminal area. The values for the airway wall diameter were obtained by calculating the diameter from the area measurement, assuming a circular structure, and were converted to millimeters by multiplying by the pixel dimension in millimeters.
To measure airway wall thickness (Wth), at least three lines were randomly drawn through the airway wall. The program automatically displayed a histogram of the pixel intensity along that line. The inflection points of increased intensity along the line that represents the inner and outer edges of the airway wall were selected, and the program then automatically measured the distance in pixels between the two points. The values for the Wth were converted to millimeters by multiplying by the pixel dimension in millimeters and averaged.
We measured between 20 and 41 airways in each volunteer. For nomenclature of the HRCT measurements of the airways, we divided the airways by their diameter at FRC at baseline (BSMT) into small (<5 mm in diameter), medium (5–10 mm in diameter), and large (>10 mm in diameter). We measured small and medium airways in all 21 volunteers, but we were able to measure a sufficient number of large airways in only 15 of the volunteers. We determined the diameter in millimeters after maximum relaxation with albuterol at TLC: DiTLCalb(small), DiTLCalb(medium), and DiTLCalb(large) for the small, medium, and large airways, respectively. We also determined the diameters at TLC under baseline conditions: DiTLCbsmt(small), DiTLCbsmt(medium), and DiTLCbsmt(large) for the small, medium, and large airways, respectively. Wth was measured at FRC before and after albuterol for each of the three airway sizes. Wall thickness was not significantly affected by albuterol (P > 0.05). Therefore, we only utilized the postalbuterol measurements designated WthFRCalb(small), WthFRCalb(medium), and WthFRCalb(large). We also measured the difference in the airway diameters at TLC between the albuterol and baseline conditions for the three airway sizes. We focused on a total of 12 airway measurements for each volunteer: three diameters at TLC under albuterol and baseline conditions for a total of six, three differences between the albuterol and baseline conditions, and three wall thicknesses at FRC with albuterol. We have no measurements of the wall thicknesses at TLC, because our initial focus was on the FRC measurement. It was not until the data were analyzed that the significance of the TLC measurements became apparent.
We consider the airway diameters after maximum relaxation with albuterol to be a function principally of the intrinsic structural characteristics of airways, because the effects of smooth muscle tone have been eliminated. We also had measurements of the diameters at FRC, but we inferred that the change in the diameter at TLC with changes in lung volume would be minimized due to the stiffness of the airways at TLC. Therefore, we considered the TLC measurements after albuterol to be the best reflection of intrinsic structure (22).
Protocols. HRCT scans were acquired at baseline FRC and TLC (i.e., with BSMT). On a separate day, after the plateau of the airway smooth muscle relaxing effect of nebulized albuterol treatment was obtained, HRCT scans were again acquired at FRC and TLC (relaxed airways).
Data analysis.
Pulmonary function measurements in this study are expressed as either percent or fraction predicted values (http://www.hopkinsmedicine.org/pftlab/pfpred.htm). In a few instances, only absolute values are reported. For both lung function and structural measurements, the effect of BSMT was expressed as the percent change as determined by
 | (1) |
|
RESULTS |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
) in FVC was the pulmonary function measurement that was the most closely correlated with both %FEV1 (r = 0.86, P < 0.0001) (Fig. 2A) and the baseline FEV1 fraction predicted (r = 0.77, P < 0.0001) (Fig. 2B).
|
|
The change in FRC with BSMT played a dominant role in the control of the change in TLC relative to the change in RV. The strongest correlation between the change in the TLC and the change in any other pulmonary function measurement was with FRC (r = 0.93, P < 0.0001; Fig. 3A). The strongest correlation between the change in RV and the change in any other pulmonary function measurement was also with the change in FRC (r = 0.88, P < 0.0001; Fig. 3B). Thus the change in FRC with BSMT closely tracked the changes in RV (Fig. 3B) and TLC (Fig. 3A).
|
 | (2) |
 | (3) |
FVC. Figure 4 shows the correlation between FRCratio and TLCratio to be highly significant (r = 0.84, P < 0.0001). Moreover, the %FVC significantly correlated with both the FRCratio (r = 0.57, P = 0.007) (Table 3) and TLCratio (r = 0.65, P = 0.001).
|
|
FEV1 that occurred from BSMT (r = 0.62, P = 0.018; Fig. 5), suggesting that the mechanism(s) that produced the BSMT acted in an analogous way to the inhaled methacholine on the FEV1. Furthermore, log10PC20 was also significantly correlated with the FRCratio (r = 0.66, P = 0.0108, Table 3, Fig. 6). In addition to %FVC, the FRCratio was significantly correlated with other effects of the BSMT on pulmonary function: %FEV1 and %RV (r = 0.50 and –0.74, respectively; P = 0.020 and 0.0001, respectively; Table 3).
|
|
FVC and %RV (r = –0.51, P = 0.018), but this degree of correlation was less than that between %FVC and the FRCratio (r = 0.58, P = 0.0065, Table 3). Furthermore, with stepwise regression of both %RV and the FRCratio as independent variables against %FVC as the dependent variable, the %RV became nonsignificant, leaving only the FRCratio as a significant determinant of %FVC. The implication of these relationships is that, whatever the mechanism of the increase in RV, and we shall later explore what that mechanism is, the effect of the BSMT on the FVC was more through how much the FRC increased relative to the increase in RV than through an independent effect of the increase in RV itself. Because of the very strong correlation between FRC and TLC (see above), we could hypothesize that the effect of BSMT on FVC was also determined by the TLCratio. Indeed, the TLCratio was a significant determinant of %FVC (r = 0.65, P = 0.001).
Thus the increase in RV was necessary but not sufficient for the decrease in FVC with BSMT. For the FVC to decrease with BSMT, the TLC must not have increased enough from an increase in FRC to prevent the decrease in FVC. Although the log10PC20 was significantly correlated with both %FEV1 (Fig. 5) and the FRCratio (Fig. 6, Table 3), multiple stepwise regression showed that only the FRCratio was a significant determinant of log10PC20, suggesting that the magnitude of the response to both the BSMT and inhaled methacholine was significantly determined by how much the FRC (and, therefore, TLC) increased relative to the increase in RV.
The relationship between the FRCratio and airway dimensions.
We examined the relationships between the FRCratio and the airway measurements for the three sizes of airways, with BSMT and after albuterol: 1) the luminal diameter at TLC, 2) the wall thickness at FRC, 3) the % in the luminal diameter between BSMT and albuterol [18 total airway measurements for determination of the correlation coefficients (r) with the FRCratio]. The only significant correlations were between the luminal diameters at TLC of the large airways, both with BSMT and albuterol [DiTLCbsmt(large), r = 0.60, P = 0.0190 and DiTLCalb(large), r = 0.70, P = 0.0036]. With stepwise multiple regression, DiTLCbsmt(large) was not significant, leaving only the correlation between the FRCratio and DiTLCalb(large) as significant. DiTLCbsmt(large) was not significant, because its effect was largely redundant through its tight correlation with DiTLCalb(large) (r = 0.92, P < 0.0001).
At all values of DiTLCalb(large) that were <13 mm, FRCratio was <1 (the increase in FRC was less than the increase in RV with the BSMT); at all values of DiTLCalb(large) that were >13 mm, the FRCratio was 
1 (Fig. 7). The significant correlation between FRCratio and DiTLCalb(large) occurred only in the range where DiTLCalb(large) was <13 mm and the FRCratio was <1 (r = 0.71, P < 0.05). In the range where DiTLCalb(large) was >13 mm and the FRCratio was 1, there was no apparent correlation between the FRCratio and DiTLCalb(large) (r = 0.14, P = 0.8). Thus the ability of the BSMT to decrease the FVC and FEV1 through a low FRCratio appeared to require a critical narrowing of the large airways in the absence of tone. When the luminal diameter of the large airways without tone was above this critical diameter (critical Diam), BSMT had little effect on the FVC and FEV1.
|
FVC (r = 0.73, P = 0.0002), FRCratio (r = 0.54, P = 0.01), and TLCratio (r = 0.64, P = 0.002). There was no significant correlation between either DiTLCalb(small) or DiTLCalb(medium) and (FEV1/FVC)alb (P = 0.54 and 0.57, respectively). The correlates of the response to a methacholine challenge. The observation that the log10PC20 of a methacholine challenge was significantly correlated with the magnitude of the reduction in FEV1 from the BSMT lends support to the idea that the reduction in FEV1 from either BSMT or a methacholine challenge involves similar mechanisms. The observation that both the log10PC20 and the reduction in FEV1 from BSMT were significantly correlated with the FRCratio (Table 3) suggests that the magnitude of the response to either BSMT or a methacholine challenge does not depend simply on how the smooth muscle tone affects the FEV1 directly, but also how much the FRC and therefore the TLC increases relative to the increase in RV with an increase in smooth muscle tone. The question that remains to be answered is why the FRCratio is positively correlated with the diameter of the large airways [DiTLCalb(large)].
The effects of a diminished diameter of the large airways in the absence of smooth muscle tone [DiTLCalb(large)], as reflected by a diameter <13 mm (critical Diam), was an increase in FRCalb%, TLCalb%, and FVCalb%. This can be seen by partitioning these parameters by the critical Diam. One-way ANOVA revealed that FRCalb%, TLCalb%, and FVCalb% were significantly increased in volunteers with a critical Diam of <13 mm (P = 0.0004, 0.02, and 0.01, respectively). Narrowing of the large airways was also associated with a decrease in FEV1 alb/FVCalb that is compatible with an increased expiratory resistance. The mechanism of the association of increased lung volumes and narrowing of the large airways in the absence of smooth muscle tone is not clear, but dynamic hyperinflation or intrinsic positive end-expiratory pressure is likely playing a role.
The greater the FRCalb%, TLCalb%, and FVCalb% before the BSMT, the greater was the responsiveness to the methacholine challenge (Table 4). One can infer from the relationships between the FRCratio and log10PC20 (Fig. 6) and between the FRCratio and TLCratio (Fig. 4) that the significant correlations between the lung volumes and the log10PC20 (Table 4) indicate that the greater the lung volumes were before the addition of smooth muscle tone, the less the TLC was able to increase for a given increase in RV from the challenge.
|
The relationship between the %RV and the airway dimensions.
As in the analysis of the FRCratio and airway dimensions, we analyzed the correlation coefficients between the same 18 airway dimensions and %RV. There were five significant correlation coefficients: DiTLCalb(medium) r = –0.46, P = 0.035; %DiTLC(medium) r = 0.44, P = 0.0488; DiTLCalb(large) r = –0.56, P = 0.0312; WthFRCbsmt(large) r = 0.65, P = 0.0083; and WthFRCalb(large) r = 0.64, P = 0.0100. With backward stepwise regression, there were only two significant determinants of %RV, accounting for nearly 70% of the variance: r2 = 0.68, P = 0.0011; DiTLCalb(medium), F ratio = 9.89, P = 0.0085; WthFRCalb(large), F ratio = 18.26, P = 0.0011.
Since the luminal diameter of the large airways [DiTLCalb(large)] was the determinant of the FRCratio that quantified how much the FRC increased in relation to the increase in RV with BSMT, the addition of DiTLCalb(large) to the two determinants of the increase in RV [DiTLCalb(medium) and WthFRCalb(large)] should be a major determinant of the increase in FRC (%FRC). Indeed, the addition of DiTLCalb(large) to the determinants of RV accounted for nearly 90% of the variance on the %FRC: r2 = 0.88, P < 0.0001; DiTLCalb(medium), F ratio = 59.21, P < 0.0001; WthFRCalb(large), F ratio = 37.88, P < 0.0001; DiTLCalb(large), F ratio = 14.55, P = 0.0029.
The BSMT produced a significant decrease in the luminal diameter at TLC in all three sizes of airways (Table 5). There were no significant changes in the WthFRC of any sized airway with BSMT. There was no significant correlation between the decrease in the diameter of any sized airway and the increase in the RV with the BSMT.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This study evaluated pulmonary function and airway structure on 2 separate days: on 1 day, under baseline conditions, and on another day, after a maximum response to albuterol. We made the assumption that the difference in pulmonary function between the 2 days was caused by the BSMT that had been abolished by albuterol. We assumed that the change in pulmonary function caused by albuterol is due only to the abolition of the bronchial smooth muscle tone (32). We defined the magnitude of the difference in pulmonary function between the 2 days as the response to BSMT. We used the magnitude of the response to BSMT in relation to airway dimensions determined by HRCT to make inferences concerning the relationship between airways structure (remodeling) and AHR in volunteers with moderate to severe asthma.
To understand the relationship between airway structure and the severity of asthma, we need to understand how airway structure was related to 1) the mechanism of the increase in RV with BSMT, and 2) the mechanism of the increase in TLC with BSMT.
Before proceeding further, several potential limitations to these measurements should be considered. First, the spirometry and lung volume measurements were determined in the upright position, whereas HRCT was performed supine. There is an 25% decrease in FRC in the supine position (25), associated with an increase in airway resistance (31). We have shown this to occur in previous HRCT studies from our laboratory (4). Given that this phenomenon occurred in all volunteers in the present study, we do not think it affected the validity of our comparisons between patients. A second potential limitation of our study was the inability of HRCT to visualize and measure very small airways below the resolution of the CT scanner. While we have difficulty visualizing and measuring airways <1 mm in diameter with our current CT technology, recent data from Nakano et al. (30) demonstrate the utility of conducting airways to reflect the changes in the respiratory airways. Finally, we assumed that we could eliminate smooth muscle tone with albuterol, despite the fact that airway smooth muscle contraction in asthma is mediated by a variety of mechanisms. Thus it is possible that complete smooth muscle relaxation was not achieved with high-dose albuterol alone. Despite this potential limitation, we were able to observe significant changes in airway luminal diameter with albuterol (Table 5) and thus consider it reasonable to assume that the airway structure under the albuterol condition was dominated by changes due to remodeling.
The mechanism of the increase in RV from the BSMT.
The magnitude of the increase in RV produced by the BSMT was significantly correlated with the wall thickness of the large airways and the luminal diameter of the medium airways, in the absence of tone. Although the luminal diameters at TLC of all three airway sizes under the baseline conditions were significantly less than the luminal diameters in the absence of the smooth muscle tone, neither the baseline luminal diameters nor the changes in the luminal diameters with BSMT were significant determinants of %RV.
It might seem paradoxical that the effect of the BSMT was only determined by the caliber of the airways in the absence of smooth muscle tone and not by the caliber measured in the presence of tone or the difference in the caliber produced by the smooth muscle tone. Nevertheless, the findings are exactly those that would be predicted from a theoretical analysis of how smooth muscle tension in blood vessels or airways produces a critical closing pressure (9, 34, 35). If the elastic recoil pressure (airway pressure relative to pleural pressure under static conditions) is considered the distending or transmural pressure (Ptm) of the airways, then the critical closing pressure of the airways in the presence of smooth muscle tone is the Ptm at airway closure, or Ptm' (15, 35). An increase in Ptm' causes the RV to increase because of airway closure. The theoretical analysis of how smooth muscle tone causes a critical closing pressure showed that the magnitude of Ptm' was determined by the ratio of the smooth muscle tension to the radius of the unstressed lumen in the absence of smooth muscle tension (9, 34). This phenomenon was recently demonstrated to be the predominant mechanism of increased respiratory impedance in the antigen-challenged, sensitized mouse lung (43).
The fact that only the luminal diameter of the medium airways and the wall thickness of the large airways at TLC in the absence of the smooth muscle tone were determinants of the change in RV produced by smooth muscle tone could explain why there has been so little evidence of an association between the luminal diameter of the airways and pulmonary function measurements (6). The relationships between the responses to smooth muscle tone and airway structure have not been previously studied with the airways at TLC in the absence of smooth muscle tone.
We did not measure the change in the luminal diameters with increasing doses of a muscle spasmogen, but we have evidence from other studies that the greater the activation of the airway smooth muscle, the greater the decrease in the luminal diameter of the airways (4). Since we found no correlation between the magnitude of the change in pulmonary function and the change in the luminal diameter of the airways that was produced by the BSMT in the present study, it became plausible that there was purely a structural basis to the variance of the effect of BSMT on pulmonary function. These findings are compatible with a recent study of Schueller et al. (38), who found that the difference in airway luminal area between baseline and postalbuterol or between baseline and histamine challenge was the same for three groups: a group of healthy volunteers and two groups of asthmatic subjects, those with or without a significant decrease in FEV1 with the histamine challenge.
In the present study, there was no evidence that closure of the measured small airways played a significant role in the effect of BSMT on RV. The results of the present study are compatible with other studies on the role of closure of large airways. Even large, cartilaginous airways can completely close from smooth muscle contraction (5, 28). Moreover, there is evidence that small airways respond less to cholinergic stimulation than large airways (10, 29). Nevertheless, the luminal narrowing in the visualized airways in the present study could also be a marker for functionally important unvisualized changes in the distal lung (30). In addition, modeling studies have suggested that neither small nor large airway closure alone can explain the ventilation defects observed on PET scanning associated with bronchocontriction in asthmatic subjects (42).
The mechanism of the increase in TLC with BSMT. The increase in TLC was closely related to the increase in FRC (Figs. 3A and 4). Without an increase in FRC, there was little increase in TLC. Perhaps an increase in FRC produced by the BSMT is somehow sensed by the central nervous system, leading to a greater neural drive to the inspiratory muscles, resulting in an increased TLC. As lung volume is increased from a voluntary effort, the neural drive to the inspiratory muscles of the thorax continuously increases the outward retractive force of the thorax, acting in opposition to the continuously increasing inward retractive force of the lungs (the elastic recoil of the lungs) until they are balanced at TLC (15, 27). A greater neural drive to the inspiratory muscles would cause the balance between the outward force of the thorax and the inward force of the lungs to be reached at a higher TLC.
In some studies, during the recovery from acute asthma or during challenge of asthmatic patients with spasmogens, an increase in TLC has been observed (13, 16), but not invariably (24). Martin et al. (26) suggested that there was increased activity of the inspiratory muscles in proportion to the increase in FRC during inhaled histamine challenge in seven asymptomatic asthmatic subjects. In three of the seven volunteers, there was an increase in TLC by 10, 12, and 22% over the baseline TLC. The authors suggested that increased muscle action was a major determinant of the degree of hyperinflation. The increased neural drive suggested by Martin et al. (26) was directly observed in a study of the electromyographic activity of the diaphragm in asymptomatic, asthmatic patients with an increase in FRC produced by histamine challenge (20). Peress et al. (33) examined respiratory mechanics during an induced attack of acute asthma from exercise in a volunteer who was trained to carry out the respiratory maneuvers required to measure the maximum outward retractive force of the thorax at TLC. There was an increase in RV, FRC, and TLC during the attack, accompanied by a significant increase in the outward retractive force of the thorax at TLC that the authors attributed only partly to an increase in inspiratory muscle contractility. In addition to the increased contractility of the inspiratory muscles, the authors suggested that part of the increase in outward force was caused by a shift of the pressure-volume (PV) curve of the chest wall that they inferred was from a change in the intrinsic elastic properties of the chest wall. Martin et al. (26) found a similar shift in the PV curve of the relaxed chest wall.
In addition to the shift in the PV curve of the thorax (26, 33), shifts in the PV curve of the lungs have been observed where there has been a decrease in the elastic recoil pressure at the same volume and an increase in the expiratory compliance (13, 16, 17). A primary shift in the PV curves of the lungs has been suggested as the cause for the increase in the TLC (19), but the changes in the PV characteristics are exactly opposite to what was found in a study of submaximal neuromuscular blockade in healthy volunteers (14). Thus the change in the static PV characteristics of the lungs and thorax that accompanies the increase in TLC would appear to be more likely the result of the increase in the ability of the inspiratory muscles to expand the thorax. The PV characteristics of excised lungs change in a nearly identical manner to those found during an asthmatic attack, with nothing more than an increase in the end-inspiratory pressure and volume to a new level (18). These changes are caused by the marked effect of volume history on the PV relationship (18, 21).
On the basis of the evidence that increased inspiratory muscle contractility and increased FRC go together in asthma, we infer that the increase in TLC with BSMT or acute asthma is the result of a greater neural drive to the inspiratory muscles that is a response to the increase in FRC.
Comparisons of the responses of RV, TLC, and FVC to the BSMT. The response to the BSMT in terms of the magnitude of the mean absolute increase in RV and the decrease in FVC for the 21 volunteers was 0.91 and 0.27 liters, respectively (Table 2, Fig. 1). The mean increase in TLC was 0.63 liters, exactly equal to the difference between the RV and FVC response (Fig. 1). If there had been no increase in TLC, the decrease in FVC would have been equal to the increase in RV, but the measured mean increase in RV was 3.3 times greater than the decrease in FVC; i.e., the decrease in FVC with BSMT would have been more than threefold greater, if there had been no increase in TLC with the BSMT.
The results of the present study are not compatible with the conventional interpretation of the role of TLC in asthma, where it is assumed that increases in TLC occur only infrequently and in the most severe asthmatic attacks. On the contrary, the change in FVC and FEV1 appears to be the tip of the iceberg of a much greater change in RV and TLC in the typical response to an increase in smooth muscle tension, i.e., where there is no luminal narrowing of the large airways. However, when the lumen of the large airways is sufficiently narrowed to lead to dynamic hyperinflation, the increase in smooth muscle tension leads to an increase in RV without a compensatory increase in TLC, thus causing the FVC to fall. In support of this hypothesis, Carroll et al. (11) compared airway remodeling between fatal and nonfatal cases of asthma. They found more remodeling of the large airways in the fatal than nonfatal cases of asthma, but there were no differences between the two groups in the remodeling of the small airways. The relationship between large airway remodeling and asthma severity in this study could be explained, at least in part, by the diminished compensation of an increase in TLC with smooth muscle contraction when there is a narrowing of the lumen of the large airways.
A healthy person has no narrowing of the lumen of the large airways and is able to tolerate a very high dose of a contractile agonist with little decrease in FVC or FEV1. The results of the present study would suggest that this apparent hyporesponsiveness should be accompanied by a considerable increase in RV, but a simultaneous increase in TLC suppresses the change in FVC and FEV1. At this point, this postulate has not been experimentally tested. If the challenge with the contractile agonist were carried out with suppression of deep inspirations in healthy volunteers, as in the study of Skloot et al. (39), it is possible that the expected increase in TLC accompanying the increase in RV could be suppressed. If this is so, an alternative interpretation of the results of the study of Skloot et al. in healthy volunteers could be made: the suppression of deep inspirations during the methacholine challenge markedly increased the response, not because the suppression of deep inspirations caused a greater increase in RV, but, rather, the suppression prevented the compensatory increase in TLC.
The role of (FEV1/FVC)alb in the structural basis of AHR. The results from the present study suggest that the major structural basis of the AHR, as estimated by the effect of BSMT, resulted from a narrowed caliber of the fully relaxed large airways that attenuated the increase in TLC with smooth muscle constriction. A conventional pulmonary function measurement in the absence of smooth muscle tone, (FEV1/FVC)alb can be used as a surrogate for DiTLCalb(large). Indeed, (FEV1/FVC)alb was significantly correlated with the same indexes as the DiTLCalb(large). In support of the special significance of FEV1/FVC, as a reflection of the caliber of the large airways in relation to AHR in asthma, Britton et al. (3) found that FEV1/FVC was a determinant of AHR in asthma that was independent of atopy and even FEV1. Bacharier et al. (2) studied the relationship between asthma severity, judged by symptoms and medication use, and pulmonary function in children. They found that FEV1/FVC, but not FEV1, decreased as asthma severity increased.
The longitudinal study of Rasmussen and his colleagues (36) used (FEV1/FVC)alb as an index of airway remodeling without consideration of its special relationship to remodeling of the large airways. Their study showed that subjects with a reduced (FEV1/FVC)alb at 18 or 26 yr old were more likely to have asthma, AHR, and an increased probability of having had the same characteristics throughout childhood. The link found in the present study between large airway dysfunction, AHR, and severity, similar to the findings in the study by Rasmussen et al. (36), raises the possibility that the stage is set in early childhood for significant AHR being caused by some chronic insult to the large airways that is superimposed on, and independent of, the more typical characteristics of asthma.
In summary, the findings of this study suggest that the magnitude of the hyperresponsiveness in asthma is a function of the intrinsic structure of the airways. Increases in the wall thickness of the large airways or decreases in the luminal diameter of the medium airways were associated with increases in RV. The increase in RV is normally accompanied by an increase in TLC that attenuates the decrease in FVC. Luminal narrowing of the large airways prevented the compensatory increase in TLC that would normally accompany the increase in RV that occurs with BSMT, thus leading to a significant reduction in FVC. This relationship between the changes in RV and TLC is related to the magnitude of the change in FRC relative to the change in RV. The hyperresponsiveness from narrowing of the large airways was related to the degree of dynamic hyperinflation that was present before the increase in smooth muscle tone.
|
GRANTS |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTGlossaryMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
2-agonist. J Appl Physiol 84: 1500–1505, 1998.
P < 0.01 vs. corresponding baseline value.
