INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Little attention has been paid to the differences between bronchial cartilage of healthy and that of diseased airways. The cartilage is primarily responsible for maintaining the stability of the airways, and some investigators have examined the mechanical properties of cartilage in normal airways (1, 2). Meanwhile, recent studies have indicated that chondrocytes in other tissues secrete various biologic substances, including cytokines, nitric oxide (NO), and prostaglandins (PGs) (3). It is well known that all these substances, including both NO and PGs, regulate various functions of airways (i.e., smooth-muscle contraction [6], ion transport across epithelium [7], glandular secretion [8], and ciliary beat [9]), and they have also been shown to play important roles in the pathophysiology of airway diseases, including chronic bronchitis and bronchial asthma (10).

The issue of atrophy of bronchial cartilage or diminished cartilage in chronic obstructive pulmonary disease (COPD) is controversial. Wright (11) and Thurlbeck and colleagues (12) reported that the atrophy of bronchial cartilage was most marked in emphysematous lungs. A quantitative study by Tandon and Campbell (13) also showed atrophy of bronchial cartilage in COPD, but they found that cartilage atrophy was strikingly related to chronic bronchitis. In contrast, two other quantitative studies (14, 15) showed no alteration of cartilage in the bronchi of patients with chronic bronchitis or emphysema, and a recent study also showed no reduction of bronchial cartilage volume in COPD patients (16). Further, there have been two reports of no significant alterations in the amount of bronchial cartilage in bronchial asthma (17, 18), and a recent study by Carroll and associates (19) showed no difference in amounts of bronchial cartilage in the airways of patients who died from a rapid fatal attack of asthma and those who died from an asthma attack of longer duration. All of these investigators described the amount of bronchial cartilage in COPD (chronic bronchitis and/or emphysema) or bronchial asthma, but did not report the changes themselves in the bronchial cartilage.

Therefore, to investigate the morphological changes in bronchial cartilage in airway diseases, which are important for understanding the role of bronchial cartilage (or chondrocytes) in such diseases, we performed morphometric analyses of airways from autopsied lungs of patients with COPD and bronchial asthma, with special reference to the bronchial cartilage.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

As shown in Table 1, we used autopsied lungs from 16 patients for the present study.

The diagnoses of chronic bronchitis, pulmonary emphysema, and bronchial asthma were made from clinical findings, pulmonary function tests, selective bronchoalveolography (SAB) (20), and computed tomography (CT) of the chest (21), according to the definitions of the American Thoracic Society (22). The patients with pulmonary emphysema who did not show any significant prolonged or persistent airway hypersecretion (sputum volume of 10 ml/d or less) were chosen for the present study in order to exclude as far as possible combined emphysema and chronic bronchitis.

Autopsied lungs were obtained from five patients with chronic bronchitis (Group CB), four with pulmonary emphysema (Group PE), and three with bronchial asthma (Group BA), and from four control patients without any respiratory diseases (Group CN) (Table 1). Left lungs were fixed within 24 h after death by immersion in 10% formalin solution for 4 wk or longer. Right lungs were fixed by intrabronchial inflation with 10% formalin at a constant transpulmonary pressure of 25 cm H₂O for a week, and then by immersion in the fixative for 3 wk or more.

All patients in Groups CB and PE had died of chronic respiratory failure. Two patients in Group BA had died of status asthmaticus within 2 d after an asthmatic attack, and the other patient in this group had died of chronic respiratory failure after 2 mo of hospital treatment. Patients in Group CN had died of nonrespiratory diseases; two had died of sudden coronary artery disease, and the others had died of cancer of the digestive tract. All patients in this group were clinically free of respiratory disease. Pulmonary function tests had been performed within 3 yr before death. All patients in Group BA had received corticosteroids orally and/or intravenously for 2 to 5 yr (mean: 3.3 yr) before death, and all patients in Group CB had received corticosteroids exclusively by intravenous infusion before death (2 wk to 3 mo; mean: 1.4 mo) (Table 1). However, there was no significant difference in the dosages of corticosteroid given to Groups BA and CB during the last month before death, as seen in Table 1. Patients in Group PE received few or no corticosteroids before their death (Table 1).

Morphometry

The lungs were cut into sagittal slices (1 to 2 cm thick), and 20 stratified, randomly selected tissue blocks were taken from the middle sagittal slices; 10 blocks from each upper or lower lobe of right lungs that were macroscopically free of pneumonia were chosen at random for paraffin embedding and sectioning. Histologic sections were cut 4 µm thick and stained with hematoxylin and eosin (H&E) and elastica- Goldner stains. With the latter staining, perichondrial fibrosis could be observed more clearly, and it was easy to distinguish this fibrosis from other connective tissue. Further, to access the changes in the extracellular matrix of the bronchial cartilage, histologic sections cut 4 µm thick from each group were stained with safranin-O (23). Preliminary examination revealed that the right lungs, fixed by intrabronchial inflation with the fixative, had a larger number of bronchial profiles cut vertically than did the left lungs fixed by immersion in the fixative. We therefore used the right lungs for the ensuing morphometric procedures. Twenty or more bronchi containing cartilage (3 to 8 mm in diameter) from each patient, cut vertically in the cross-section profile, were chosen for morphometric study. Bronchi showing ectasis in tissue sections were excluded from morphometry.

Proportions of the areas of total cartilage to bronchial wall, degenerated cartilage to total area of cartilage (Deg%), and perichondrial fibrosis (Fib%) were measured with a digitizing tablet coupled to a computer, using the image on a television monitor obtained directly from the microscope (PC 9801; NEC Corporation, and LA-500; PIAS Corporation, Tokyo, Japan), as previously described (24, 25). Using the same bronchial specimens described earlier, we determined the area proportion of glands (Gland%) by measuring the area of the bronchial wall excluding the cartilage-containing area. Gland% was determined as the ratio of summed gland areas to the corresponding wall area in each subject. Simultaneously, the area proportion of mucus-containing goblet cells to the total epithelial layer (Goblet%) was measured in the bronchial wall. For the goblet-cell morphometry, 20 or more visual fields in each airway were chosen at random. The epithelial area obtained in this manner constituted 20% or more of the total length of the epithelial layer in the airways. Because some epithelial-cell layers were seen to be detached, airways with 75% or more of the epithelial layer intact were chosen for morphometry, and such airways represented 90% or more of total airways observed in the present study. The diameter of airways was calculated by measuring the length of the basement membrane. For the measurement of basement membrane thickness (BMT), 64 or more fields were randomly chosen in each airway. Various cell types in the submucosal layer of the bronchi were identified and counted randomly in an area of submucosal tissue 0.25 mm in arc length extending between the epithelium and the cartilage, at a magnification of ×400. Each cell type counted was expressed as the number of cells per square millimeter in the paraffin sections.

All morphometric measurements were made by two observers (M.H. and S.S.) without knowledge of the group from which the paraffin sections had come, and the mean value was used for the data analysis. However, the morphometric values recorded by one observer were significantly correlated with those recorded by the other, suggesting the reproducibility of the morphometric findings. For example, Deg% values reported by the two observers were significantly correlated (r = 0.87, p < 0.001).

Statistical Analysis

Data are expressed as mean ± SEM. For multiple comparison of means, one-way analysis of variance (ANOVA) and Duncan's multiple range test were used. For direct comparisons of two means, Student's two-tailed paired or unpaired t test was used, and the Cochran- Cox t test was used when Bartlett's test showed the variance to be uniform. The regression coefficient was also used for statistical analysis. Significance was accepted at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although degenerated cartilage was absent or scarce in airways from Group CN (Figure 1A), bronchial cartilages in airways from Groups CB, PE and BA, showed loss of cellular or pericellular metachromasia (territorial matrix), and chondrocytes were pyknotic or absent, yielding a vacuolelike appearance with empty lacunae (Figures 1B and 1C). These degenerative changes in bronchial cartilage were most dominantly observed in airways from Group CB. Further, in some bronchial cartilages of Group CB, the cartilage plates lacked their normal round or oval contour (Figure 1A), and instead had irregular, serrated contours (Figure 1C). Such degenerative change was accompanied by a decrease in the intensity of safranin-O staining (Figure 2). For the morphometry of degenerated cartilage (Deg%), we defined the area of absence of or vacuolelike changes in chondrocytes as the degenerated area in bronchial cartilage. We also observed increased perichondrial fibrosis in airways from Groups CB and BA, as shown in Figure 3. The perichondrial fibrosis was observed not only at the ends of the cartilage plates, but also uniformly around the plates (Figure 3B). The increased perichondrial fibrosis in Group BA seemed to be more dominant than in Group CB.

View larger version (125K): [in this window] [in a new window]   View larger version (117K): [in this window] [in a new window]   View larger version (115K): [in this window] [in a new window]   Figure 1.   (A) Micrograph from a patient with acute myocardial infarction (male, age 53 yr), showing bronchial cartilage with normal chondrocytes in the bronchial wall. (H&E stain; original magnification: ×100). (B) Micrograph from a patient with chronic bronchitis (male, age 61 yr), showing loss of cellular or pericellular metachromasia (territorial matrix) and pyknotic or absent, vacuolelike appearance with empty lacunae of chondrocytes in the bronchial wall. (H&E stain; original magnification: ×100). (C ) Micrograph from a patient with chronic bronchitis (male, age 61 yr), showing some defects of cartilage, with marked degeneration of chondrocytes at the end. (H&E stain; original magnification: ×100).

View larger version (156K): [in this window] [in a new window]   View larger version (156K): [in this window] [in a new window]   Figure 2.   Micrographs of bronchial cartilage stained with 0.1% safranin-O for 30 s. (A) Micrograph from a patient with acute myocardial infarction without any pulmonary disease (male, age 53 yr), showing well-stained extracellular matrix with normal chondrocytes. (Original magnification: ×250.) (B) Micrograph from a patient with chronic bronchitis (male, age 53 yr). Note decreased intensity of extracellular matrix on safranin-O staining, with pyknotic chondrocytes. (Original magnification: ×250.)

View larger version (161K): [in this window] [in a new window]   View larger version (163K): [in this window] [in a new window]   View larger version (159K): [in this window] [in a new window]   Figure 3.   (A) Micrograph from a patient who died of colon cancer without any pulmonary diseases (male, age 69 yr), showing bronchial cartilage with normal perichondrium in the bronchial wall. (Elastica-Goldner stain; original magnification: ×100.) (B) Micrograph from a patient with bronchial asthma (male, age 52 yr). Note markedly increased perichondrial fibrosis around bronchial cartilage. (Elastica-Goldner stain; original magnification: ×100). (C ) High-magnification micrograph from a patient with bronchial asthma (male, age 52 yr), showing increased perichondrial fibrosis around bronchial cartilage. (Elastica-Goldner stain; original magnification: ×250.)

There were no significant differences in the diameter of bronchi in which this was measured (3.48 ± 0.92 mm, 3.15 ± 0.80 mm, 2.72 ± 0.72 mm, and 3.01 ± 0.44 mm in Groups CB, BA, PE, and CN, respectively) or in the area proportion of cartilage to bronchial wall among the four groups (4.47 ± 1.21%, 4.45 ± 0.26%, 3.16 ± 0.81%, and 4.14 ± 0.85% in Groups CB, BA, PE, and CN, respectively). Further, no significant difference in the extent of cartilage expressed as mean area of cartilage (mm²) was found among the four study groups (0.44 ± 0.21 mm², 0.48 ± 0.16 mm², 0.35 ± 0.10 mm², and 0.42 ± 0.11 mm² in groups CB, BA, PE, and CN, respectively). However, there was a higher proportion of bronchial cartilage the larger the bronchial diameter (r = 0.68, p < 0.01).

As shown in Figure 4A, Groups CB, BA, and PE all showed 10-fold higher values of Deg% than did Group CN (p < 0.01 each), and Group CB showed the highest value, which was also significantly higher than that for Group PE (p < 0.05).

View larger version (31K): [in this window] [in a new window]   Figure 4.   (A) Deg% values in each group. Groups CB, BA, and PE all showed significantly higher values of Deg% than did Group CN, and the highest value was observed in Group CB. *p < 0.05, **p < 0.01 compared with Group CN. p < 0.05, compared with Group PE. (B) Fib% values in each group. Groups CB and BA both showed significantly higher values of Fig% than did Group CN, and the highest value was observed in Group BA. Group PE also showed a slight but significant increase in the value of Fib%. **p < 0.01, compared with Group CN. p < 0.05, p < 0.01 compared with Group PE.

As shown in Figure 4B, groups CB and BA both showed 1.5-fold higher values of Fib% than did Group CN (p < 0.01 each), and Group PE also showed a slight but significant increase in values of Fib% as compared with Group CN (p < 0.05). Group BA showed the highest value of Fib%, which was significantly higher than that of Group PE (p < 0.01) (Figure 4B).

The neutrophil number in the bronchial wall in Group CB showed a significant increase as compared with that in Group CN (p < 0.01), whereas the neutrophil number in Groups BA and PE did not, as shown in Table 2. Further, the neutrophil number in Group CB was significantly larger than that in Group BA (p < 0.05) or Group PE (p < 0.05). The eosinophil number in Group BA showed the largest value among the four groups, although it did not reach statistical significance because of the small number of subjects and large variation (Table 2). The eosinophil number in Groups CB and BA was slightly but significantly larger than that in Group CN (p < 0.05 each) (Table 2).

As shown in Figure 5, the values of Deg% correlated significantly with the number of neutrophils in the bronchial wall (p < 0.01). The values of Fib% correlated significantly with the number of eosinophils in the bronchial wall (r = 0.51, p < 0.05). Further, the values of Fib% correlated significantly with BMT(r = 77, p < 0.0002); Gland% (r = 56, p < 0.02), and Goblet% (r = 0.55, p < 0.02), and the values of Deg% correlated significantly with those of Fib% (r = 0.64, p < 0.01).

View larger version (13K): [in this window] [in a new window]   Figure 5.   Relationship between number of neutrophils in airway walls and Deg%. Deg% value correlated significantly with number of neutrophils in airway wall in each patient (r = 0.63, p < 0.01). Open circles: Group CB; open triangles: Group PE; closed circles: Group BA; and open squares: Group CN.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the morphometric analysis reported here, bronchi of 3 to 8 mm diameter, containing subsegmental bronchi in which cartilage atrophy was reported to be most dominant, were chosen for examination (12). The larger bronchi (segmental, main bronchi, or trachea) could not be included in the morphometric analysis because it was impossible to obtain with certainty the entire cross section of such large bronchi. The amount of bronchial cartilage in COPD (Groups CB and PE) or bronchial asthma (Group BA) was not significantly different from that in controls. This finding was consistent with those by Restrepo and Heard (14), Greenberg and colleagues (15), Dunnill and coworkers (17), and Takizawa and Thurlbeck (18). In contrast, Wright (11) and Thurlbeck and associates (12) described marked bronchial cartilage atrophy or a decreased amount of bronchial cartilage in emphysematous lungs, and Tandon and Campbell (13) reported that cartilage atrophy was most strikingly related to chronic bronchitis. It is well known that pulmonary emphysema is combined with chronic bronchitis (26). Further, advanced or longstanding chronic bronchitis is often accompanied by diffuse bronchiectasis that shows atrophy and/or degenerative changes in bronchial cartilage (27, 28). A recent study by Ogrinc and colleagues (29) showed that destruction and loss of bronchial cartilage were accompanied by bronchiectasis in cystic fibrosis airways in which neutrophil accumulation occurred, as in airways of chronic bronchitis. In the present morphometric analysis we excluded as much as possible bronchi showing ectasis. This may be a possible explanation for the lack of a difference in the amount of total bronchial cartilage in controls and COPD (chronic bronchitis and emphysema) patients. In addition, our morphometric analysis may not have been very sensitive to small changes in the cartilage content of bronchi both because of the different diameters (3 to 8 mm) of the bronchi that were examined and the small numbers of samples taken, although mean bronchial diameters were not significantly different among the four groups. In contrast, the two parameters of Deg% and Fib% are not influenced by these factors.

Both chronic bronchitis and pulmonary emphysema are characterized by neutrophil infiltration or accumulation in the airways (30). Because the number of neutrophils in the bronchial wall correlated significantly with the Deg%, and because the degenerated bronchial cartilage was most dominant in chronic bronchitis, it is possible that neutrophils in the airways play a role in the degeneration of bronchial cartilage. This may result in the decreased amount of bronchial cartilage or cartilage atrophy reported in previous studies (11), and also in the development of bronchiectatic changes in chronic bronchitis (27, 28).

We can speculate about the mechanism of degeneration of bronchial cartilage in COPD as follows: Slices of rabbit articular cartilage are reported to synthesize large quantities of NO after exposure to human recombinant interleukin-1 (hrIL-1) (3) and lipopolysaccharide (LPS) (4), and endogenously synthesized NO reduces cartilage proteoglycan synthesis (3). In addition to neutrophil chemotactic activity, IL-8 has been shown to have a neutrophil-activating capability in vivo with respect to the release of neutrophil elastase and the induction of IL-1 and IL-1 receptor antagonist (IL-1ra), provoking the release of neutrophil elastase and leading to cartilage destruction when injected into rabbit knee joints (5). Thus, the infiltrating neutrophils are known to be responsible for the cartilage destruction. It is possible that a similar mechanism acts in the degeneration or destruction of bronchial cartilate in diseased airways.

Airways in bronchial asthma are characterized by both eosinophil accumulation and thickened basement membrane (31). The latter is known to be due to the deposition of interstitial collagen and fibronectin on the original basement component, with increased numbers of myofibroblasts in the submucosal layer (33). A similar deposition of collagen may occur around bronchial cartilage. In fact, we found that Fib% correlated significantly with BMT. Further, Fib% correlated significantly with Gland% and Goblet% in our study, and both bronchial gland hyperplasia and goblet-cell metaplasia are also pathologic features of asthmatic airways (34). Although the factors responsible for the increased collagen-deposition around bronchial cartilage remain unknown, the same factors may be related to the hyperplasia of secretory cells, thickened basement membrane, and increased collagen surrounding the cartilage in asthma airways.

The alteration of bronchial cartilage can be expected to influence airway function, since cartilage in airways and other tissues is known to secrete various PGs, cytokines, and NO (3- 5, 35). However, whether bronchial cartilage contributes to the alterations in the pathophysiology of airway diseases is unknown, because almost all experiments have been performed after excluding this cartilage. In addition to study of its mechanical properties, airway cartilage has been shown experimentally to release PGE₂, PGF₂, and PGI₂, and to play a role in the development of tone in the airway smooth muscle of guinea pig, rabbit, and canine lungs (35). However, the role of pathologic alterations in the bronchial cartilage in the pathophysiology of airway diseases remains unresolved.

We used autopsied lungs for the morphometric analysis in our study. All of the patients had undergone various therapies including cardiopulmonary resuscitation and/or treatment with various drugs. These therapies may therefore have influenced our findings (38). For example, before their death, all patients in Groups BA and CB received large amounts of glucocorticoid, whereas the patients in Groups PE and CN did not. However, because glucocorticoid treatment is known to decrease the number of eosinophils infiltrating into airways and to reduce goblet-cell metaplasia (39), it is unlikely that the glucocorticoid treatment altered the main findings in our study.

The present findings indicate that airways in COPD and bronchial asthma have both tissue resulting from cartilage degeneration and increased connective tissue surrounding cartilage. Further, these alterations in bronchial cartilage may differ in chronic bronchitis and bronchial asthma. Because cartilaginous airway lesions as well as small-airway lesions contribute to airflow limitation (40), the present findings are important for understanding the pathophysiology of chronic airway diseases including COPD and bronchial asthma.