INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Structural alterations present in airway walls of severe and fatal asthma have been classically described (1, 2). The presence of a chronic inflammation resulting in healing and repair leads ultimately to airway remodeling. The failure to adequately repair the inflammatory damage possibly contributes to the airway hyperresponsiveness and changes in airway mechanics observed in this disease (3). The decline of the ventilatory function over time in some asthmatic patients may be a consequence of airway remodeling (4).

The relationship between structure and function in asthma has been extensively studied. All the compartments of the airway have been shown to have some structural alteration contributing to functional defects (5). Morphometric analysis of asthmatic airways suggests that thickening of the airway wall internal to the muscle layer could increase the effect of a given amount of smooth muscle shortening on airway narrowing (6).

The major extracellular matrix components of the lungs include collagen, elastin, proteoglycans (7), and fibronectin (8). Changes in the extracellular matrix components have already been described in asthma (5, 9, 10). Altered amount or composition of extracellular matrix components, individually or as a network, may contribute to changes in airway mechanics through geometric effects, direct changes in tissue biomechanics, and by altered modulation of the process of inflammation itself (5).

Subepithelial deposition of collagen types I, III, and V and fibronectin in the lamina reticularis beneath the bronchial epithelium has been observed, even in cases of mild asthma (9). Thickening underneath the true basement membrane caused by collagen deposition has also been described in asthmatic children (11). Although there are a number of studies describing collagen changes in asthmatic airways, few reports concern the distribution and configuration of the fibers of the elastic system in this disease.

The elastic system has three components, oxytalan, elaunin, and elastic fibers, defined according to crescent amounts of elastin and fibril orientation (12, 13). Oxytalan fibers are devoid of elastin and, in the organs where they have been described, oxytalan fibers are always anchoring basement membranes and have a mechanical rather than an elastic role (12). In the bronchial lamina propria FES are found as a superficial network underlying the basement membrane, connected to a loose meshwork of coarse fibers close to the smooth muscle (deeper network or lamina elastica) (14). These fibers are connected to the airway adventitia through thin fibers that traverse the airway smooth muscle (15).

Fibers of the elastic system (FES) provide support for airway patency and lung elastic recoil (8). However, their potential role in modulating bronchoconstriction has not yet been described. Because of their mechanical properties and configuration in the airway, FES may act as structures limiting airway constriction and may also provide tension to restore airways to their preconstricted configuration after the contractile stimulus has ceased (15).

There is no consensus in the literature about elastic changes in asthmatic airways. In a descriptive study, Bousquet and colleagues (10) reported fragmentation and paucity of fibers in the subepithelial layer in bronchial biopsies of asthmatics of different severity, with a picture similar to the solar elastolysis of the skin. On the other hand, Godfrey and coworkers (16) could not find differences in the total content of FES in airways in cases of mild, severe, and fatal asthma when compared with control subjects.

This controversy and the potential role of these fibers in contributing to the physiopathology of asthma led us to quantify the FES in asthmatic airways, focusing both total mucosa and superficial layer fiber content.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Thirty-one patients dying of status asthmaticus were selected from our autopsy service. This service is characterized by draining patients from all over the city, including emergency services in the outskirts of the city of São Paulo and patients dying at home. The pathologists work with very little clinical information. Clinical charts or detailed medical history are, in many instances, absent. Information concerning main diseases such as asthma, diabetes, heart disease is obtained from a brief questionnaire that is applied to relatives prior to autopsy. All patients in this study were known to be asthmatics and died during a severe crisis of breathlessness. The diagnosis of fatal asthma was confirmed by macroscopic examination (lungs overinflated) and histologic criteria (mucous plugs, mucosal foldings, epithelium desquamation, mucosal inflammation rich in eosinophils, and hyaline thickening of the basement membrane) (Figure 1A). None of the specimens showed chronic bronchitis, emphysema, or pneumonia. Ten control subjects were selected, dying from causes other than pulmonary disease. None of the control subjects were smokers. All control cases had a normal gross and microscopic examination of the lungs (Figure 1B). Patients younger than 14 and older than 60 yr of age were not included in either group.

View larger version (160K): [in this window] [in a new window]   Figure 1.   Central asthmatic airway (A). Note mucus plugging (m), hyaline thickening of basement membrane (bm), and inflammatory infiltrate containing eosinophils (arrow); c = cartilage (H&E). Central airway of a control lung, c = cartilage (B). There is no airway constriction or inflammation (H&E). Elastic fiber distribution in asthmatic (C ) and control (D) central airways. Control central airway shows two distinct layers of FES. One (thin arrow) lies perpendicular to basement membrane (bm) and the other (thick arrow) lies parallel to airway smooth muscle (asterisk). In central asthmatic airway, there is elastosis and fibers are approximated with a condensed aspect (arrow). Weigert's resorcin-fuchsin with oxidation staining. FES are decreased in the subepithelial layer (SL) of the central asthmatic airway (E ), when compared with control (F ). Weigert's resorcin-fuchsin with oxidation staining. Scale bar = 100 µm.

Histochemistry

Lung tissue was fixed with 4% paraformaldehyde, embedded in paraffin, and processed routinely. Slices 5 µm thick were cut and stained with hematoxylin-eosin and Weigert's resorcin-fuchsin with oxidation (17). Oxidation was performed using oxone as previously described (18). This histochemical method allows for identification of all the three components of the elastic fiber system, i.e., elastic, elaunin, and oxytalan fibers.

Morphometrical Procedure

Airways were divided into central (CA) or peripheral (PA) areas according to the presence or absence of cartilage, respectively. To estimate airway dimensions we assessed bronchial diameters by image analysis in a sample of 20 central and 20 peripheral transversally cut airways by measuring basement membrane perimeter. Airways were defined as transversally cut when the relation between the maximal diameter and the diameter at the widest point perpendicular to this axis was > 0.6.

We measured total and subepithelial elastic fiber content in the airway mucosa These measures were assessed in 63 central asthmatic and 27 control airways. Sixty-four peripheral airways were measured in asthmatics and 28 in control subjects. The technique of measuring elastic fibers by image analysis has been validated mainly in skin studies (19). Mean values were calculated for each patient.

The total content of the fibers of the elastic system in the mucosa was assessed in a zone lying between the epithelial basement membrane and the internal limit of the airway smooth muscle (20). Using a magnification ×200, quantification was carried out with the aid of a digital analysis system, using specific Software (Bioscan-Optimas; Bioscan Incorporated, Edmond, WA). The images were generated by a microscope (Zeiss Axioplan; Zeiss, Oberkochen, Germany) connected to a camera (Sony Triniton CCD; Sony, Tokyo, Japan) and fed into a computer through a frame grabber (Oculus TCX; Coreco, St. Laurent, PQ, Canada) for off-line processing. We analyzed 15 fields in each CA and the total circumference in PA (five to six fields per airway), with a mean of three CA and three PA per case. Mean values were calculated for each airway, and subsequently a mean value was obtained for each patient. The elastic content was assessed for each field and the corresponding length of the basement membrane was measured. In order to avoid the effects of acute bronchoconstriction in the measures, the total elastic content was expressed as a ratio between area of elastic tissue and basement membrane length, results being expressed in micrometers. According to James and colleagues (6) airway internal perimeter remains constant with different degrees of airway smooth muscle shortening, despite substantial changes in luminal area.

In central airways we also quantified the fibers lying in the subepithelial zone, i.e., the zone immediately underneath the basement membrane (10 µm deep). For this procedure, conventional morphometric analysis was adopted. A semicircular coherent grid (Merz grid), used for anisotropic systems, was attached to a video monitor, where fields had their images captured. Using magnification ×1,000, 40 fields per CA were assessed, one to three bronchi per patient. The amount of elastic fibers was expressed as the number of fibers intercepting the grid per field.

In peripheral airways, we could not identify two distinct layers, and therefore only total elastic content was measured.

Data Analysis

An unpaired two-tailed t test was used to compare the content of FES in the mucosa of asthmatics and control subjects. Results were considered statistically significant at a probability level of 5%.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Asthmatic patients ranged from 14 to 58 yr of age (mean ± SD: 38.87 ± 12.51), including 17 female and 14 male patients. The mean ± SD age of control subjects was 37.4 ± 11.9 yr, ranging from 20 to 54 yr, including eight female and two male patients. The causes of the death of the control subjects are shown in Table 1.

Light Microscopy

In central airways two layers of FES were discernible: one deeper layer composed of thick fibers lying parallel to the airway smooth muscle and another one, perpendicular, found closer to the basement membrane. This superficial perpendicular layer was neatly organized in short bundles (Figure 1D). Weigert's resorcin-fuchsin with oxidation staining allowed for further identification of thin fibers anchoring the basement membrane, connected to the superficial layer (Figure 1F). These slender fibers have already been described by Böck and Stockinger (14) and correspond to oxytalan fibers.

In asthmatic airways the most striking abnormalities were observed in the superficial network of CA. At this level the fibers were fragmented, disorganized, and sparse, immersed in an edematous stroma (Figure 1E). Fibers of the deep layer were tangled and seemed to be approximated in areas of bronchoconstriction, with a condensed aspect (Figure 1C). In peripheral airways only one layer of elastic fibers was discernible at optical level. The abnormalities in the FES were not so remarkable as observed in CA. The distribution of FES in normal airways and the alterations observed in asthmatics are schematically shown in Figure 2.

View larger version (43K): [in this window] [in a new window]   Figure 2.   Schematic drawing showing the distribution of FES in normal airways and the altered FES pattern observed in asthmatic airways. Adapted from Reference 14.

Morphometry

Airway diameters ranged from 0.65 to 0.21 cm in central airways and from 0.17 to 0.065 cm in peripheral airways, which correspond to Generation 3 to 9 for central airways and to Generation 10 to 13 for peripheral airways (21).

The mean values (µm) of total mucosal elastic content in CA in asthmatics and control subjects are shown in Tables 2 and 3. The mean values (µm) of total mucosal elastic content in PA in asthmatics and control subjects are shown in Tables 4 and 5. Total mucosal elastic content in asthmatics and control subjects for CA and PA is shown in Figure 3. There was a statistically significant increase in FES content in asthmatic CA. There was no difference in FES content in PA. Total elastic content in CA was 12.16 ± 1.65 µm in asthmatics and 7.95 ± 0.88 µm in control subjects (p = 0.03). In PA, total elastic content was 3.96 ± 0.33 µm in asthmatics and 3.01 ± 0.32 µm in control subjects (p = 0.09). The distribution of FES mucosal content according to patient's age in CA and PA, respectively, is shown in Figures 4 and 5. Note that CA elastosis tended to occur in patients older than 40 yr of age. For PA, FES content in asthmatics and control subjects did not vary with age.

View larger version (35K): [in this window] [in a new window]   Figure 3.   Total mucosal content of FES, expressed in micrometers, in peripheral (PA) and central (CA) airways of asthmatic and control lungs. There was a statistically significant increase in FES content in asthmatics CA (p = 0.03). There was no statistical difference in FES content in PA (p = 0.09). Values are expressed as mean ± SEM.

View larger version (20K): [in this window] [in a new window]   Figure 4.   Distribution of FES total mucosal content in CA according to patients' ages (asthma and control). Data are expressed as mean ± SEM (µm).

View larger version (23K): [in this window] [in a new window]   Figure 5.   Distribution of FES total mucosal content in PA according to patients' ages (asthma and control), expressed as mean ± SEM (µm).

The mean values (number of intercepts) of superficial FES content in asthmatics and control subjects are shown in Tables 6 and 7, respectively. Morphometric assessment of FES content, expressed as the number of fibers intercepting the grid, in the subepithelial region in central airways of asthmatic and control patients is shown in Figure 6. There was a statistically significant decrease in FES content in asthmatic airways. Subepithelial elastic content was 5.30 ± 0.31 in asthmatics and 9.85 ± 0.46 in control subjects (p < 0.0001). The distribution of FES subepithelial superficial content in CA according to the patient's age is shown in Figure 7. Note that asthmatic patients of any age had a low fiber content when compared with control subjects.

View larger version (43K): [in this window] [in a new window]   Figure 6.   FES content in the superficial network of central airways in asthmatic and control lungs, expressed as numbers of fiber intercepts. There was a statistically significant decrease in the content of FES in asthmatics (p < 0.0001). Values are expressed as mean ± SEM.

View larger version (21K): [in this window] [in a new window]   Figure 7.   Distribution of FES subepithelial content in CA according to patients' ages (asthma and control). Data are expressed as number of intercepts.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Fibers of the elastic system, because of their natural properties (FES act returning an organ to its original conformation after release of a deforming force), are natural candidates to contribute to bronchoconstriction modulation. Impairment of FES in the airway wall is likely to contribute to airway hyperresponsiveness and also to prevent the airway from returning to its preconstricted status. Leick-Maldonado and coworkers (15) described, when studying the distribution of FES in bronchoconstricted intraparenchymatous guinea-pig airways, how there is a mechanism of delamination of the fibers within the subepithelial layer, part of them following the invaginations, attached to the basement membrane, the other remaining attached to the airway smooth muscle. This configuration is similar to what we observed in asthmatic bronchoconstricted airways and is compatible with the theory that FES participate in bronchoconstriction modulation.

It has already been demonstrated that interaction between parenchyma airway is important in the regulation of bronchoconstriction, as demonstrated in the experiments of Ding and colleagues (22). These investigators showed that the effects of bronchoconstrictive agents might be reversed by large inflation of the lungs. Less attention, however, has been devoted to the interactions occurring between airway smooth muscle and epithelial basement membrane in the constrictor event, and how the transmission of forces generated by airway smooth muscle are regulated by elements of the extracellular matrix.

Stephens and Jiang (23) have speculated that in central airways parenchymal tissue should not exert an important load to the airway smooth muscle constriction insofar as at this level the parenchyma does not attach to the wall directly but via a compliant connective tissue sheath. It is only at the tenth or eleventh generation that alveolar septa seem to insert directly on the wall of the airway. However, even at this level the intervening structures may possess considerable compliance. Using computer imaging, these investigators have shown that during bronchoconstriction there is no significant variation in the external perimeter of the airway, whereas there is a significant narrowing of the lumen, suggesting that some mechanism of recoil occurs within the airway wall itself. Furthermore, Bramley and coworkers (24), comparing airway strip mechanical properties, concluded that a reduced tissue elastance of nonmuscle elements within airway walls of asthmatic patients results in a greater muscle shortening by reducing the load limiting bronchoconstriction.

In the present study we quantified both superficial and deeper layer of FES in airways of patients dying of fatal asthma. Our results indicate for the first time the presence of elastosis in asthmatic central airways when the entire thickness of the mucosal layer is considered. The elastosis could be a result of repair and contribute to airway remodeling after inflammation and fiber fragmentation, as already suggested by Bousquet and colleagues (25).

In the subepithelial region (superficial network) of central airways we showed a paucity of elastic fibers. Fiber disrupture at this level could determine loss of attachments between submucosa and epithelium. We have speculated that since the elastic fiber system is a continuum network of fibers connecting the airway epithelium to the lung parenchyma, transmission of elastic forces within the airway wall would have a "weak point" at this region. Impairment of the elastic fiber network could contribute to airway hyperconstriction and to the impaired mechanism of airway recoil.

There is some evidence that airway wall remodeling may occur in both large and small airways, as demonstrated in some autopsy studies (26). We do not have a clear explanation for the lack of elastosis or significant morphologic alterations in the FES content and distribution in the peripheral airways. One possible explanation is that there is very little elastic tissue in smaller airways, and maybe our methods were not sensitive enough to bring up a statistically significant difference between asthmatics and control subjects (although there was a trend towards elastosis, as shown in Figure 3). On the other hand, there are few studies that suggest that airway inflammation in fatal asthma is less important in peripheral airways (27). If this is true, one could expect less elastic damage within peripheral airways because of chronic inflammation.

The chronic inflammatory infiltrate present in asthma produces proteases, including elastases and collagenases (28) released from various inflammatory cells such as macrophages (29), neutrophils (30), and even eosinophils (31). Recent work has shown increased levels of elastase and ₁-antitrypsin in the sputum of asthmatic patients (32). This chronic insult probably alters the FES and the other components of the extracellular matrix. Experimental work has shown that inhaled elastase can cause bronchoconstriction in guinea pigs (33). Furthermore, the edema and stretching of the airway observed in cases of fatal asthma could contribute to the fragmentation of the fibers (10). It is important to emphasize that our results are derived from fatal status asthmaticus, and some of the observed changes could be the result of the final attack. Although the presence of elastosis probably denotes a chronic repair of FES within the airway wall, the large amount of edema observed underneath the basement membrane may represent an acute additional damage to these fibers during the fatal attack.

Our observation that elastosis in central airways tended to occur in patients older than 40 yr of age (Figure 4) supports the idea that elastosis could follow repair after chronic, prolonged inflammation characteristic of the disease.

The lack of clinical data is the limiting factor of this autopsy study. In the present work all patients had a clinical history of asthma, but for some of the patients we do not have further information concerning cigarette smoking, cause and duration of asthma, and treatment. In all these cases death was ascribed as status asthmaticus. The lungs had a gross aspect compatible with fatal asthma: they were hyperinflated and contained tenacious mucous plugs in the bronchi. We also used histologic criteria to confirm the diagnosis: bronchoconstriction, (pseudo) hyaline thickening of the basement membrane, eosinophilic infiltrate, and mucous plugging. Each lung was included in our study only when most of these features were present. Indeed, we believe that we might have excluded some asthmatic patients from the study because of the rigid histologic criteria. Also, when signs of emphysema or "chronic bronchitis" were present we did not use the tissue. We have not used patients older than 60 and younger than 14 yr of age. Especially in the older population it may be very difficult to distinguish asthma from chronic obstructive pulmonary disease.

In summary, the present data show that both elastosis and elastic fiber fragmentation are present in asthmatic airway walls, indicating that FES may participate in airway remodeling. Our results also reinforce the idea that FES can contribute to the mechanism of bronchoconstriction.