INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Chronic inflammation is recognized as an important aspect of the pathogenesis of asthma. As in other inflammatory conditions which progress to fibrotic changes, certain components of this chronic inflammation have been suggested to lead to architectural remodeling of the end-organ involved, i.e., lung. Presently, this remodeling is poorly defined, but could manifest in a variety of ways, including excessive collagen deposition or altered distribution, mucous gland hypertrophy, elastin deposition/destruction, neovascularization, and/or smooth muscle hyperplasia/hypertrophy (1). Elements of this remodeling are increasingly suggested to contribute to long-term irreversible changes in airway obstruction and severity of the asthmatic disease.

Collagen deposition as a potential marker of this remodeling process has been previously analyzed in large airway endobronchial biopsies of asthmatics of moderately different severity. These studies, using numerous staining methods to evaluate collagen (2), have focused primarily on the thickness of the reticular "basement membrane" or subepithelial "basement membrane" (SBM). The majority of these studies have demonstrated a thickened SBM over a relatively small clinical range of asthma severity compared with normal airways. Although the relationship between the degree of SBM thickness and asthma symptoms/severity remains uncertain, a recent study from Chetta and coworkers suggested that "severe asthmatics" had thicker SBM than mild and moderate asthmatics and the thickness of the SBM was significantly correlated with clinical severity of asthma (11). Despite this individual study, a comparison across all studies has shown widely different ranges for SBM thickness in asthmatics and even among normal control subjects (2).

Whereas collagen has been measured in the SBM of bronchial biopsies in most studies, only a few studies have investigated collagen deposition in the deeper submucosa (that which lies between the external side of the SBM to the internal limit of bronchial smooth muscles and/or mucous glands). Evaluation of this deeper submucosal collagen deposition may also be important, as this portion of the airway has a larger area and is anatomically closer to smooth muscles than the SBM. Recently, Wilson and Li (12) showed increased collagen type III and V in the deeper submucosa of asthmatics compared with normal control subjects. However, studies from other groups have failed to show any difference in collagen deposition in this region when comparing asthmatics and normal subjects (2, 14). There are no biopsy studies to date which have evaluated deeper submucosal collagen deposition in asthmatics over a broad range of severity.

These different results and conclusions make the relative importance of collagen deposition to asthma pathology and physiology unclear. To further evaluate the importance of the fibrotic process in asthmatic airways, it was hypothesized that if fibrosis was a contributing factor to the chronic airway obstruction of severe asthma, then the population of long-term steroid-dependent asthmatics seen at National Jewish Medical and Research Center would manifest the greatest degree of collagen deposition in their airways compared with control groups. Therefore, comparisons of the thickness of the SBM, in conjunction with deeper submucosal collagen deposition, using three different collagen stains, in asthmatics from a wide range of severity and normal control subjects were performed. These results were then evalauted in the light of the level of airway inflammation (as measured by eosinophil numbers), the levels of a fibrogenic cytokine, transforming growth factor- (TGF-), and asthma severity, as indicated by degree of airway obstruction and medication usage.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subject Characteristics

Four groups of nonsmoking subjects were recruited for this study: Group 1: severe asthmatics (n = 17); Group 2: moderate asthmatics (n = 9); Group 3: mild asthmatics (n = 7); Group 4: normal control subjects (n = 8). All asthmatics conformed to the American Thoracic Society definition of asthma (15). Degree of asthma severity was determined by asthma symptoms, pulmonary functions, and medications required by the patients. Severe asthmatics were patients referred to National Jewish for severe, oral-steroid-dependent asthma with frequent hospitalization and emergency room visits. Mild and moderate asthmatics were asthmatics recruited from the community for clinical and research protocols. Mild asthmatics were defined as patients with an FEV₁ of > 80% predicted on -agonist treatment alone, as needed. Moderate asthmatics had an FEV₁ of < 80% predicted, were on -agonists and/or inhaled steroids or theophylline, and did not have a history of emergent health care utilization or oral steroid use. Normal control subjects had no history of respiratory diseases and were on no medications. All subjects signed the informed consent and the protocol was approved by the institutional review board.

Tissue Processing and Staining

Bronchoscopies were performed as previously described (16). Two to four endobronchial biopsies were taken from the first subcarinae of the right or left lower lobes. Endobronchial biopsies were processed as previously reported (17). Briefly, specimens were fixed in acetone at 20° C overnight and then embedded in glycol methacrylate resin. Tissue blocks were stored at 20° C. Serial 2-µm sections were cut from well-oriented tissue blocks with a Reichert Ultracut E ultramicrotome (Leica Inc., Deerfield, IL) by the same researcher (H.W.C.).

Tissue sections were stained using antibodies against collagen type I (Ic) and type III (IIIc) (Biodesign International, Kennebunkport, ME), activated eosinophils (EG2; Pharmacia, Uppsala, Sweden), and transforming growth factor (TGF-1,2,3; Genzyme, Cambridge, MA). A three-step indirect immunostaining method was used. Sections were treated with 0.3% hydrogen peroxide in 0.05 M Tris buffered saline (TBS, pH 7.6) for 30 min to inhibit endogenous peroxidase, followed by incubation with 1% normal horse or goat serum for 30 min to block potential nonspecific binding sites. The slides were then incubated with primary antibodies mentioned previously for 2 h at room temperature, followed by incubation with biotinylated horse anti-mouse IgG or goat anti-rabbit IgG for 1 h at room temperature. Thereafter, avidin-biotin-peroxidase complex (Vector Lab., Burlingame, CA) was added to the slides for 45 min at room temperature. After rinsing the slides in TBS, 0.03% aminoethylcarbazole (AEC) in 0.03% hydrogen peroxide was used as a substrate to develop a peroxide-dependent red color reaction. Slides stained with EG2 and TGF- were counterstained with Mayer's hematoxylin and covered with Crystalmount (Biomeda Corp., Foster City, CA). Slides stained with Ic and IIIc were directly covered with Crystalmount without counterstaining with hematoxylin. Appropriate control slides were similarly treated but with the primary antibodies replaced by nonimmune serum or TBS.

Collective collagen in the tissue was demonstrated using Sirius Red staining method (18). Sections were incubated overnight at 37° C with a drop of solution containing 1% Sirius Red F3BA and 1% Fast Green FCF (Polysciences Inc., Warrington, PA). The slides were then rinsed with water, dehydrated, and mounted with permount.

One slide from each tissue block was stained with hematoxylin- eosin for routine histology.

Analyses of Collagen, Eosinophils, and TGF--Positive Cells

A National Institutes of Health (NIH) scion image analysis program was used to measure the thickness of SBM and the amount of collagen deposition in the submucosa. One to three good quality pieces of tissue were included for the analyses. Black-and-white pictures were taken and transferred to the computer via video camera. The stained area for collagen was black and high in intensity compared with the nonstained area (white).

To measure the SBM thickness, the multiple point-to-point measurement method was used at 50-µm intervals for a total of 35 to 60 measurements. The thickness (µm) of the SBM was vertically measured at ×200 magnification in areas of good epithelial orientation.

Collagen deposition was also measured in total and deeper submucosa. Total submucosal area included the SBM area and deeper submucosal area. The deeper submucosa was defined as the area between the external side of the SBM and the internal side of the smooth muscles and/or mucous glands. Blood vessels, smooth muscles, and mucous glands were erased from the picture and thereby excluded from any measurement. The "density slice function" of the scion image program automatically and selectively identified the collagen staining at predetermined density. The submucosal area (collagen area + noncollagen area) was outlined manually or by the automatic "density slice function." The absolute area of collagen and the corresponding submucosal area were then measured and saved in the computer for each measurement. Collagen deposition in the total and deeper submucosa was expressed as total submucosal collagen area/total submucosal area and deeper submucosal collagen area/deeper submucosal area, both expressed as a percentage (12).

Nucleated cells positive for EG2 (eosinophils) and TGF- were counted at ×400 magnification in the submucosa (measured area, 0.5 [range 0.3 to 0.7] mm²), excluding blood vessels, mucous glands, and smooth muscle cells (19). The cell count was expressed as number of cells per mm² of submucosa. Eosinophils in the bronchial epithelium were also counted and expressed as number of cells per millimeter of basement membrane.

The coefficient of variation for 2 to 3 repeated measurements by the same observer (either H.W.C. or J.L.H.) or between two different observers (H.W.C. and J.L.H.) was < 7%. The observers were blinded to the clinical data of the subjects.

Statistical Analyses

As the data were not normally distributed, comparisons of data were performed using a nonparametric Kruskal-Wallis analysis of variance (JMP software program; SAS Institute, Cary, NC). If the overall p value was less than 0.05, the Tukey's test was then used to confirm differences in individual groups. Correlations between various groups of data were carried out with Spearman's rank correlation coefficients. A p value of  0.05 was regarded as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of Subjects

The characteristics of the subjects are given in Table 1. Asthmatics had significantly lower FEV₁ values than the normal control subjects. Mild asthmatics showed a higher FEV₁ than moderate and severe asthmatics. Moderate and severe asthmatics were not significantly different in their baseline FEV₁. Duration of asthma was similar among the three groups of asthmatics. There were more females than males in each group, except the moderate group where they were nearly 1:1. Five of nine moderate asthmatics were on inhaled corticosteroids at < 1,000 µg/d. All the severe asthmatics were using both oral prednisone (40 mg/d [range, 15 to 40 mg/d]) and inhaled corticosteroids in addition to ₂-agonists and other medications. No patients were using methotrexate, cyclosporin, or gold.

SBM Thickness

Initial analyses compared the SBM thickness among the three groups of asthmatics and normal control subjects (Table 2 and Figure 1). When analyzed in this fashion, there were no significant differences among the four groups in the SBM thickness independent of staining method used. As these results were unexpected, a post hoc analysis was used to determine whether differences existed when evaluating the asthmatics as a single group and comparing them with normal control subjects. As a group, asthmatics (n = 33) had a thicker SBM than normal control subjects (median and interquartile range: 7.5 µm [6.0 to 10.1] versus 5.5 µm [5.0 to 7.5], p = 0.04) when the SBM was measured by IIIc staining method. However, asthmatics and normal control subjects were not significantly different in the thickness of the SBM evaluated by Ic (p = 0.09) or collective collagen (Sirius Red staining) (p = 0.19). Figures 2a-2h show examples of collective collagen Sirius Red staining and IIIc staining for airway tissue from the four groups.

View larger version (15K): [in this window] [in a new window]   Figure 1.   No significant differences were found among the four groups of subjects in the thickness of SBM as measured by stainings for collective collagen (a), collagen type I (b) and III (c). CT = normal control subjects; MIA = mild asthmatics; MA = moderate asthmatics; SA = severe asthmatics. Cross bars represent the medians.

View larger version (138K): [in this window] [in a new window]   View larger version (138K): [in this window] [in a new window]   View larger version (138K): [in this window] [in a new window]   View larger version (139K): [in this window] [in a new window]   View larger version (145K): [in this window] [in a new window]   View larger version (134K): [in this window] [in a new window]   View larger version (144K): [in this window] [in a new window]   View larger version (138K): [in this window] [in a new window]   View larger version (142K): [in this window] [in a new window]   View larger version (144K): [in this window] [in a new window]   Figure 2.   Representative photomicrographs of different staining techniques on endobronchial biopsy samples. (a-d ) Sirius Red staining (collective collagen) in normal control subject, mild, moderate, and severe asthma, respectively. (e-h) Collagen type III staining in normal control subject, mild, moderate, and severe asthma, respectively. (i ) Negative control section without primary antibody incubation (counterstained with hematoxylin). ( j ) Eosinophil EG2 immunostaining in a subject with mild asthma. Original magnification, ×400.

Variation of the SBM thickness was consistently observed in both single-biopsy samples and between different pieces of tissues from the same subject. This variation was reflected in the large coefficients of variation (CV) for the SBM thickness measurement within an individual biopsy sample (36.9% [22.5 to 59.6%] [median, interquartile range], 36.4% [22.9 to 58.5%] and 35.3% [24.4 to 53.3%]) for collective collagen (Sirius Red staining), collagen type I and III, respectively. The CV for the SBM thickness between different pieces of biopsy samples from the same subject in all the groups were 12.7% (3.2 to 26.8%), 13.5% (5.2 to 44.4%), and 11.8% (0.2 to 32.5%) for collective collagen, collagen type I and III, respectively. The absolute differences in SBM thickness were 1.06 µm (0.8 to 1.46), 0.9 µm (0.62 to 1.8), and 1.0 µm (0.2 to 1.25) between different pieces from the same subject for collective collagen, collagen type I and III, respectively. Three different staining techniques correlated well with each other (Ic versus IIIc, r = 0.96, p < 0.0001; Ic versus Sirius Red staining, r = 0.93, p < 0.0001; and IIIc versus Sirius-Red staining, r = 0.93, p < 0.0001).

No specific staining was seen in negative control slides (Figure 2i). In this case, slides were counterstained with hematoxylin.

Deeper and Total Submucosal Collagen Deposition

Correlations among deeper or total submucosal collagen deposition evaluated by three differing staining methods were performed. Ic staining correlated well with IIIc staining (r = 0.76, p < 0.0001 for deeper submucosal collagen and r = 0.78, p < 0.0001 for total submucosal collagen, respectively). However, correlations between Sirius Red staining and Ic or IIIc staining for deeper or total submucosal collagen deposition were poor (all r values less than 0.5, p > 0.05). Sirius Red staining showed less deeper or total submucosal collagen deposition than Ic or IIIc staining (p < 0.05 for all four comparisons). No significant differences were found among the four subject groups in the amount of deeper or total submucosal collagen deposition (Table 2, Figures 3 and 4) as demonstrated by all the staining methods. Similar to the analysis of the SBM, the data were also analyzed as total asthmatics (n = 33) versus normal control subjects. However, unlike the SBM, there were no significant differences in the amount of deeper or total submucosal collagen deposition between asthmatics and normal control subjects (deeper submucosal collagen, 29.6% [22.1 to 38.6%] versus 24.1% [22.4 to 27.1%], p = 0.16; for collective collagen, 35.6% [26.8 to 45.5%] versus 36.6% [29.8 to 40.4%] for Ic, p = 0.78; and 39.0% [26.2 to 47.1%] versus 37.8% [31.2 to 51.2%] for IIIc, p = 0.72; total submucosal collagen, 36.1% [27.8 to 43.4%] versus 28.9% [28.3 to 33.7%] for collective collagen, p = 0.07; 41.6% [31.9 to 51.1%] versus 39.5% [34.2 to 43.8%] for Ic, p = 0.56; and 45.9% [33.7 to 52.3%] versus 43.6% [35.7 to 55.3%] for IIIc, p = 0.87).

View larger version (16K): [in this window] [in a new window]   Figure 3.   No significant differences were found among the four groups of subjects in the amount of deeper submucosal collagen deposition as measured by stainings for collective collagen (a), collagen type I (b) and III (c). Cross bars represent the medians. Definition of abbreviations as in Figure 1.

View larger version (16K): [in this window] [in a new window]   Figure 4.   No significant differences were found among the four groups of subjects in the amount of total submucosal collagen deposition as measured by stainings for collective collagen (a), collagen type I (b) and III (c). Cross bars represent the medians. Definition of abbreviations as in Figure 1.

Airway Eosinophils and TGF- Protein Expression

Cell counts for submucosal eosinophils and TGF- expressing cells are shown in Figure 5. An example of eosinophil EG2 staining from a mild asthmatic is shown in Figure 2j. No significant differences were found in the number of eosinophils and TGF- expressing cells in the submucosa among the four groups. Eosinophil counts in the epithelium were also not significantly different (p = 0.50) among the severe (0 [0-0]/mm basement membrane), moderate (0 [0-0.8]), or mild (0 [0-0]) asthmatics, and normal controls (0 [0-0]). As a group, asthmatics (n = 33) were not significantly different from the normal control subjects in the number of eosinophils in either submucosa (p = 0.37) or epithelium (p = 0.61). The number of TGF--positive cells in the submucosa tended to be higher in asthmatics (n = 33) than in normal control subjects (12.5 [6.4 to 25.9] versus 6.6 [4.7 to 10.2], p = 0.06).

View larger version (14K): [in this window] [in a new window]   Figure 5.   No significant differences were found among the four groups of subjects in the number of eosinophils (a) or TGF--positive cells (b) in the submucosa. Cross bars represent the medians. Definition of abbreviations as in Figure 1.

Correlations between Collagen Deposition, Airway Inflammation, Pulmonary Functions, and Duration of Asthma

Table 3 shows correlation analysis data in all the subjects (n = 41). No significant correlations were found between the SBM thickness, deeper submucosal collagen deposition, and total submucosal collagen deposition.

Correlations were also performed between collagen deposition and airway inflammation as measured by the number of eosinophils in the submucosa and the epithelium (Table 3). No significant correlations were seen between collagen deposition and eosinophil infiltration in either submucosa or epithelium in all the subjects. The number of TGF--positive cells did not correlate with collagen deposition.

Correlations between FEV₁ or duration of asthma and collagen deposition in all the subjects were also not significant (Table 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Collagen deposition in large airway biopsies of mild, moderate, and severe asthmatics, as well as normal control subjects, was evaluated to determine if the amount of collagen deposition increased among asthma patients of increasing severity. Although these results confirmed the thickening of the SBM in asthmatics compared with normal subjects, perhaps surprisingly, no significant differences in collagen deposition between very severe and milder forms of asthma existed, suggesting that, at the large airway level, the amount of collagen deposition may not predict the clinical severity of disease.

The current study confirmed that the thickening of the SBM existed, at least as measured by collagen type III staining, in asthmatics (2, 4, 5, 7). However, unlike what has been reported by other groups (10, 11), the thickness of the SBM did not appear to vary among asthmatics of differing severity. The thickness of the SBM was not predictive of or associated with the levels of severity of disease. These results may raise questions regarding the use of the SBM thickness as a "marker" of the irreversible fibrotic component of more severe groups of asthma.

The current study was also one of only a few to examine the collagen deposition in the asthmatic airway deeper submucosa. Deeper submucosal collagen deposition was evaluated because inflammatory cells are prominent there and its volume is considerably larger than the SBM. Thus, any significant change of collagen in this area may have the potential to influence airway pathophysiology. The present study did not show any significant differences between severe and milder asthmatics or between asthmatics in general and normal control subjects in the amount of deeper or total submucosal collagen deposition. These results support some previous reports (2, 14), but do not agree with other recent studies that demonstrated increased deeper submucosal collagen deposition in asthmatics which correlated with the severity of disease (12, 20). Given the lack of differences in collagen deposition among asthmatics with a broad range of severity observed in this study, it was not surprising that no relationships existed between subepithelial collagen deposition and the clinical data of asthmatics, such as duration of asthma or FEV₁ (2).

These differing results in both SBM thickness and submucosal collagen deposition may be due to factors such as the number of measurements taken, the tissue processing methods, and the study subjects. Considerable limitations inherent to the technique of endobronchial sampling of the airways may also play an important role in causing the disparate results from the different studies. In the present study, it was clear that the SBM was not uniformly distributed. For this reason, measurements of the SBM thickness were taken from areas of good epithelial orientation at 50-µm intervals for a total of 35 to 60 measurements to minimize the variation at different points. Many other recent studies (9, 21) have used fewer than 10 measurements. Interestingly, Sullivan and coworkers (22) recently suggested that the number of measurements required to achieve a precision of approximately 10% is 96 for normal control subjects and 68 for asthmatics at 20-µm intervals, suggesting that a large number of measurements may be necessary for a better evaluation of the SBM thickness. Tissue processing variations, including different techniques of fixation, the thickness of sections for the staining, and staining methods, could affect collagen quantification in the tissue. These may be part of the reasons for differing results from different studies. For example, SBM thickness in normal control subjects in some studies is more than that from asthmatics in other studies (2, 3, 6, 7). In the current study, collective collagen (evaluated by Sirius Red staining) and the two most common types of collagen in the tissue, type I and type III, were examined. When submucosal collagen was evaluated, the nuclei of cells on the tissue sections were not stained so that collagen quantification was not interfered with by the nuclear staining. We found that Sirius Red staining showed less deeper or total submucosal collagen deposition than collagen type I or type III staining, suggesting a reduced sensitivity of Sirius Red staining. This may be due to the fact that Sirius Red staining was observed to pick up the larger bundles of collagen, whereas collagen type I and III staining picked up both larger and smaller bundles of collagen. Possibilities for other explanations await further study. These observations reiterate the importance of mutiple staining methods for evaluating airway collagen deposition.

Previous studies have generally evaluated only mild asthmatics (2) with few studies looking at collagen deposition in moderate and severe asthmatic airways (10, 11, 20). In the current study, well-defined groups of mild, moderate, and severe asthmatics were included. The range of severity of patients in this study was considerably different from those of Cho and coworkers (10) and Chetta and coworkers (11) which demonstrated the thickened SBM in severe asthmatics compared with milder asthmatics. Their severe asthmatics were not as severe as ours based on symptoms, FEV₁, and steroid usage and would have been described as mild or moderate asthmatics for this study. While differences may exist in the definition of asthma severity among studies, the patients in this study clearly demonstrated the broadest range of severity. Despite this, no differences in collagen deposition were found. Interestingly, a recent study from Vignola and coworkers (13) also failed to find a difference of collagen deposition as measured by the SBM thickness between asthmatics of varying severity, although that study focused on considerably milder asthmatics.

TGF- and eosinophils have been proposed to be involved in the airway remodeling process (20, 23). The current study showed that, similar to data on the SBM thickness, asthmatics tended to have more TGF- expressing cells than normal control subjects. In contrast, eosinophil numbers were not significantly different between asthmatics and normal control subjects. Asthmatics of varying severity had similar numbers of TGF--positive cells and eosinophils. These differences in eosinophil numbers from previous studies may be due to the steroid use in the moderate and severe patients. As anticipated from these results, no significant correlations were found between TGF- expressing cells or eosinophils and collagen deposition. These data suggest that TGF- and eosinophils may not be important factors in asthmatic airway collagen deposition, although longitudinal and inhibitor studies may be necessary to confirm this. Although these results are similar to those of Aubert and coworkers (24), they differ from two other recent studies (20, 23) which demonstrated correlations between TGF- messenger RNA (mRNA) (20) or protein (23) expression and airway fibrosis. These differences may be partly explained by the withdrawal of corticosteroids from moderate and severe asthmatics in both of those studies (20, 23) and by differing methods to quantitate airway fibrosis compared with the current study.

The current study differs from many previous studies in that it included asthmatics on a wide range of corticosteroid therapy. Results from limited numbers of previous studies have been indeterminate regarding the effects of steroids on airway remodeling. Several groups demonstrated that inhaled corticosteroids reduced airway inflammation, but could not reduce the thickness of the SBM in mild/moderate asthmatics (6, 25). In contrast, two studies showed reduction of the SBM thickness after 3 to 4 mo of inhaled corticosteroid treatment in mild/moderate asthmatics (9, 28). The effects of corticosteroids on the expression of TGF- have also been variable with both upregulation and downregulation demonstrated in asthmatic airways after steroid treatment (29, 30). Most previous studies on collagen deposition have not included patients on any form of steroids at initiation of study. As both severe and moderate asthmatics in this study received inhaled and/or oral corticosteroids, it is possible that the lack of differences in collagen deposition among asthmatic groups was secondary to an effect of corticosteroids to reduce the thickness of the SBM and lessen submucosal collagen deposition in the moderate and severe asthmatics. In spite of this, it must be stated that although the degree of collagen deposition among asthmatic groups was similar (whether due to corticosteroid use or not), severe asthmatics in this study were dramatically different clinically from the other groups, with ongoing symptoms, poor pulmonary functions, and oral corticosteroid requirements. This finding suggests that large airway collagen deposition may not be a key contributor to the symptoms and pathophysiology of asthma.

It should also be emphasized that only one aspect of airway remodeling, collagen deposition in the large airways, was evaluated in this study. We cannot exclude the involvement of other tissue components, such as proteoglycans, elastin, tenascin, and smooth muscles in the airway remodeling process. Although some have shown that the amount of airway elastic fibers did not differentiate asthmatics of differing severity from normal control subjects (14), Bousquet and coworkers (31) demonstrated that asthmatics showed an abnormal distribution of elastic fibers in the submucosa. Tenascin, one of the anti-adhesive glycoproteins which can inhibit the spreading of many different types of cells, including epithelial cells on extracellular matrices (32), has also been demonstrated in increased amounts in asthmatic airway SBM. Inhaled budesonide was effective in reducing the expression of this protein (26). Correlations of tenascin expression with epithelial shedding or with asthma severity and airway hyperresponsiveness were not evaluated. Increased shortening and/or mass of airway smooth muscles have also been proposed to contribute to abnormal airway narrowing in asthmatics (33, 34). Finally, all studies to date have been done at the large airway level. It is certainly conceivable that small airway remodeling is of great importance; something which can only be analyzed at the transbronchial biopsy level. Clearly, the definition of airway remodeling in asthma requires better articulation.

In conclusion, the current study did not show excessive collagen deposition at the large airway level of severe asthmatics as compared with milder subjects. Further, the amount of collagen deposition did not correlate with airway inflammation, pulmonary functions, or duration of disease. A better understanding of airway remodeling remains of critical importance to improve our approach to the treatment of asthma.