Published ahead of print on May 19, 2004, doi:10.1164/rccm.200403-305OC
© 2004 American Thoracic Society doi: 10.1164/rccm.200403-305OC Abnormal Alveolar Attachments with Decreased Elastic Fiber Content in Distal Lung in Fatal AsthmaDepartment of Pathology, Medical School, São Paulo University, São Paulo, Brazil Correspondence and requests for reprints should be addressed to Thais Mauad, M.D., Ph.D., Department of Pathology, São Paulo University Medical School, Avenida Dr. Arnaldo, 455 CEP 01246-903, São Paulo SP, Brazil. E-mail: tmauad{at}usp.br
Small airway disease is thought to contribute significantly to functional impairment caused by asthma. Functional evidence of airway-parenchyma uncoupling in asthma, such as loss of deep breath bronchodilator effect in bronchoconstrictive episodes and enhanced airway closure, has been previously demonstrated. Elastic fibers are essential to maintain adequate elastic recoil of the lungs. In this study, we hypothesized that alveolar attachments could be abnormal and that elastic fibers could be damaged in the distal lungs of patients with fatal asthma. For this purpose, we measured the number of abnormal alveolar attachments and quantified the content of elastic fibers in the adventitial layer of small airways and in the peribronchial and distal alveolar septa of 15 patients who died of asthma (FA) and 9 control subjects (CTRL). Our data (geometric mean [range]) showed an increased proportion of abnormal alveolar attachments per centimeter of basement membrane perimeter in fatal asthma (FA, 0.18 [0.03–4.00]; CTRL, 0.00 [0.00–0.12]; p < 0.001) and decreased elastic fiber content in the small airway adventitial layer (FA, 4.08 [2.22–11.46] µm; CTRL, 6.79 [5.62–10.0] µm; p = 0.01) and in the peribronchial alveoli (FA, 1.08 [0.46–1.91] µm; CTRL, 1.81 [1.22–1.74] µm; p = 0.003), but not in the distal alveoli. We propose that structural alterations at the peribronchiolar level might contribute to the pathogenesis of some functional abnormalities observed in patients with severe asthma.
Key Words: airway–parenchyma uncoupling • autopsy • remodeling • small airways Evidence suggests that the small airways may importantly contribute to functional impairment in asthma, especially in patients with the most severe asthma (1, 2). Previous autopsy and lung resection studies have shown that the small airways are inflamed and thickened in asthma (3–5). The same studies have demonstrated that, unlike in large airways, small airway inflammation predominates in the outer airway wall layers (4–6), that is, in the adventitial layer, and may spread out to the surrounding alveolar walls, the site of alveolar attachments (7). It is known that the elastic load provided by the lung parenchyma is transmitted to the airways through the alveolar attachments, resulting in a mechanical interdependence between airways and parenchyma (8). Alterations at this level could partially explain some of the functional abnormalities observed in patients with asthma, such as the loss of deep breath bronchodilator effect in spontaneous bronchoconstrictive episodes (9), enhanced airway closure (2), and, to a lesser extent, loss of elastic recoil (10). So far, no studies have specifically addressed structural alterations of the small airways and surrounding alveolar walls in asthma. Saetta and coworkers have demonstrated that in small airways of smokers without emphysema, alveolar attachments are decreased in number and histologically abnormal, compared with nonsmokers (11). These findings were correlated with the degree of airway inflammation in the small airways and elastic recoil pressure. Those authors suggested that loss of alveolar attachments represents an early stage of destruction of lung parenchyma, and it could be partially responsible for the loss of elastic recoil measured in their patients (11). It is tempting to imagine that a lesion similar to that described by Saetta and coworkers could occur at the site of alveolar attachments in fatal asthma. The findings of Gelb and coworkers, showing loss of elastic recoil in patients with severe chronic asthma without detectable emphysema (10), reinforce this suspicion. Of the extracellular matrix components, elastic fibers have a pivotal role in the maintenance of lung elastic recoil. We have previously shown that the elastic fibers are fragmented in the subepithelial layer of large airways of patients with fatal asthma and suggested that this alteration could contribute to enhanced bronchoconstriction by decreasing airway smooth muscle postload (12). Integrity of the elastic fibers is also crucial for the maintenance of airway–parenchyma interdependence. In this study, we hypothesized that alveolar attachments are damaged in fatal asthma, leading to uncoupling of the airways from the parenchyma. For this purpose, we measured the number of abnormal alveolar attachments and quantified the content of elastic fibers in the adventitial layer of small airways and in the peribronchial and distal alveolar septa of patients who died of asthma.
This study was approved by the Review Board for Human Studies of the School of Medicine of the University of São Paulo (São Paulo, Brazil). Fatal asthma (FA) and control lung (CTRL) tissue samples were obtained from autopsied subjects between 1996 and 2000. Fifteen nonsmoking subjects with fatal asthma were included in the study. All individuals had a previous history of asthma (information provided by the next of kin before autopsy) and died during an acute attack. Clinical data were further obtained at a subsequent interview with the family of the deceased. Postmortem inclusion criteria were pathologic changes consistent with asthma: lung hyperinflation, abundant thick mucus in the airways, epithelial desquamation, thickening of the basement membrane and smooth muscle, and mucosal inflammation with or without eosinophils. The control group consisted of nine nonsmoking subjects who died of nonpulmonary causes, with no previous lung disease. Specimens were fixed in 4% paraformaldehyde for 24 hours and embedded in paraffin. Lungs were not fixed under pressure. Five-micron sections, stained with hematoxylin and eosin, were screened to identify suitable small airways for analysis. For identification of elastic fibers, the Weigert resorcin–fuchsin technique with oxidation was used as previously described (12). This technique selectively stains the elastic fibers (elastin and preelastin fibers) (13). We quantified the number of normal and abnormal alveolar attachments and elastic fibers in transverse sections of small airways, defined by a basement perimeter of less than 6 mm (14). Alveolar attachments were defined as alveolar walls that extend radially from the outer wall of the nonrespiratory bronchiole (11). For an airway to be included in the study, no more than 25% of its outer perimeter was shared with the accompanying blood vessel. Abnormal attachments were considered to be those showing any discontinuity or rupture of the radial alveolar walls. Using image analysis, we measured the average number of normal and abnormal alveolar attachments, as well as the corresponding airway basement membrane perimeter, in two or three transversally cut small airways from each patient, at x200 magnification. Values were expressed as the proportion of abnormal to normal attachments per centimeter of basement membrane perimeter. We performed inter- and intraobserver analysis of alveolar attachment measurements in 17 randomly selected slides (26 airways). The content of elastic fibers in small airways and lung parenchyma was determined in three compartments, using x200 magnification: (1) the outer wall area of the bronchioles, or the area between the outer border of the airway smooth muscle and the alveolar attachments (15), results being expressed as elastic fiber content (µm2) per basement membrane perimeter (µm). Ten fields per lung were quantified, usually comprising the whole circumference of 2 or 3 small airways; (2) the alveolar septa of the peribronchiolar region, or the site of alveolar attachment, expressed as elastic fiber content (µm2) per septum length (µm); and (3) the distal parenchyma, defined as alveolar septa distant (at least one x100 field) from the alveolar attachments, also expressed as elastic fiber content (µm2) per septum length (µm). For (2) and (3), 20 randomly selected alveolar septa were measured per lung. Areas of local hyperinflation secondary to airway closure could lead to thinner alveolar septa in the peribronchiolar areas, affecting elastic fiber content in a given alveolar segment. We have tested this hypothesis by measuring the mean linear intercept (Lm) in five fatal asthma and five control cases. Using a magnification of x200, the Lm was assessed in 10 random noncoincident microscopic fields (16) in peribronchiolar areas. The Lm was determined by counting the number of intercepts between the eyepiece lines and the alveolar septa of each microscopic field, and expressed as the relation between total line length and the number of intercepts per field. A mean value (µm) was calculated for each lung. Image-Pro Plus 4.1 for Windows (MediaCybernetics, Silver Spring, MD), running on a microcomputer, connected to a digital camera (JVC TK-C1380 color video camera; Victor Company of Japan, Yokohama, Japan) coupled to an optical microscope (Leica DMR; Leica Microsystems, Bensheim, Germany), was used for the measurements. Airway basement membrane perimeter and Lm values were expressed as means ± SE. Other values were log transformed before statistical analysis and expressed as geometric means and ranges. Paired and unpaired t tests were used to compare means between groups. Correlations were performed with the Pearson or Spearman test. p Values lower than 0.05 were considered significant.
In all patients with asthma the events surrounding death and macro- and microscopic pathologic features were consistent with fatal asthma. Histologic examination showed lung hyperinflation but no signs of emphysema. Characteristics of the patients with asthma are shown in Table 1. Patients' median age was 38 years, ranging from 11 to 68 years, and 10 of them were female. The available clinical data include previous treatment, duration of disease, medical follow-up, and previous hospitalizations. As depicted in Table 1, all patients had been using inhaled ß-agonists. Five patients were being treated with steroids: two of them received beclomethasone regularly plus oral steroids during the exacerbations, one patient received inhaled and oral steroids continuously, one patient received only beclomethasone regularly, and the other received oral steroids during the exacerbations.
Five patients with asthma (the ones whose detailed clinical history we were able to retrieve were nonsmokers, and yielded suitable material for analysis) were part of our previous study (12). All control patients had normal lungs at gross and microscopic examination. Table 2 shows ages, sex, and cause of death of the control patients. Seven of them were female; their median age was 47 years, ranging from 20 to 74 years.
We analyzed 73 airways in fatal asthma and 43 in control subjects. The mean (± SE) airway perimeter was 1.45 ± 0.12 mm in fatal asthma and 1.59 ± 0.12 mm in control subjects (p = 0.41), corresponding to terminal bronchioles (17). There was an increased number of abnormal alveolar attachments in asthma compared with control subjects (Figures 1A–1D). The proportion of abnormal to normal alveolar attachments per centimeter of airway basement membrane perimeter was greater in fatal asthma than in control subjects (p < 0.001). Data are shown in Table 3. The inter- and intraobserver correlation coefficients for the number of abnormal attachments were r = 0.88, p < 0.001 and r = 0.89, p < 0.001; and r = 0.88, p < 0.001 and r = 0.91, p < 0.001 for the proportion of abnormal to normal alveolar attachments. No significant differences were found within and between observers by using the paired t test.
There was decreased elastic fiber content in the outer wall of small airways of FA subjects compared with CTRL subjects. Fiber fragmentation was not evident. Elastic fiber content was 4.08 (2.22–11.46) µm in subjects with fatal asthma and 6.79 (5.62–10.00) µm in control subjects (p = 0.01). Similarly, the elastic fiber content in peribronchiolar septa (site of alveolar attachments) was less in fatal asthma (1.08 [0.46–1.91]) µm than in control subjects (1.81 [1.22–1.74]) µm (p = 0.003) (Figures 1E and 1F). There was no statistically significant difference in elastic fiber content in the distal alveoli between fatal asthma and control subjects (p = 0.89). Interestingly, elastic fiber content was significantly higher in the peribronchial than in the distal alveoli of control subjects (p < 0.001), but not in fatal asthma. Data are presented in Figures 2 and 3. There was no significant difference in Lm at peribronchiolar areas between fatal asthma and control patients (asthma, 70.55 ± 3.17 µm; control subjects, 66.45 ± 1.87 µm; p = 0.27), indicating that elastic fiber content in peribronchiolar areas of patients with asthma was not affected by hyperdistension at this level.
There were significant correlations between number of normal alveolar attachments and content of elastic fiber in the outer wall of the small airways (r = 0.52, p = 0.009) and in the proximal septa (r = 0.52, p = 0.009) (Figure 4). The content of elastic fiber in the outer wall of small airways significantly correlated with the content of elastic fiber in the proximal alveolar septa (r = 0.753, p < 0.001), but not with the distal alveolar septa (p = 0.09). The content of elastic fiber in the distal septa correlated with the content of elastic fiber in the proximal septa (r = 0.5, p = 0.01). Furthermore there were significant negative correlations between the proportion of damaged abnormal to normal alveolar attachments and the content of elastic fibers in the proximal septa (r = –0.50, p = 0.015) and the content of elastic fiber in outer wall of small airways (r = –0.48, p = 0.02). There were no significant correlations between morphologic parameters and age, use of steroids, or the available clinical data.
In this study we have shown that patients who died of asthma present an increased proportion of damaged alveolar attachments when compared with patients without asthma. We have also shown that patients with fatal asthma have less elastic fibers in the adventitial wall of the small airways and peribronchial alveolar septa, but not in the distal alveoli. To our knowledge, this is the first study to describe structural abnormalities of the extracellular matrix in the small airways and alveoli of patients with asthma. Other authors had previously hypothesized that alveolar attachments could be altered in asthma, and our present data confirm this suspicion (18). The functional consequences of the presence of inflammation and structural alterations in the small airways of patients with asthma cannot be underestimated. Small airways have a larger cross-sectional area than large airways, and perhaps more importantly, are directly connected to the alveolar parenchyma, especially those of the tenth generation and beyond (19). Structural alterations in small airways with the potential to cause airway parenchyma uncoupling could, at least partially, explain a known functional abnormality of patients with asthma: the loss of deep inspiration bronchodilator effect, the extent of which seems to be associated with asthma severity and/or the degree of airway hyperresponsiveness (20). Structural alterations in small airways have indeed been implicated as an underlying reason for increased asthma severity. Studies demonstrated that even in asymptomatic patients with asthma with some degree of fixed airflow obstruction, small airway wall thickening could be observed (21). Moreover, Bai and coworkers demonstrated that fatal asthma in older patients is characterized by thicker airways due to adventitial thickening. These authors suggested that loss of airway–parenchyma interdependence might be a more important mechanism of severe attacks in the older group (22). Indeed, in't Veen and coworkers showed that patients with severe asthma with recurrent exacerbations have early airway closure during expiration when compared with patients with severe but stable asthma, and proposed that early airway closure could be a key factor in the pathogenesis of fatal asthma (2). Some well established concepts about airway remodeling in asthma have been challenged. The concept that thickening of the airways may be detrimental to lung function has been drawn into question. Niimi and coworkers (23) showed that thicker airways may have a protective rather than a deleterious role in lung function in asthma, attenuating airway reactivity. Our findings may help explain this apparent contradiction. Pare (24) alleged that not only thickness, but also alterations in the components of the airways, could contribute to the final degree of airway hyperresponsiveness. Although the airways are thicker in fatal asthma (thus increasing airway smooth muscle load), damage to alveolar attachments and a decrease in elastic fibers in the outer wall of small airways and peribronchial alveoli might promote airway–parenchyma uncoupling by decreasing the tethering mechanical forces exerted by the surrounding parenchyma (thus decreasing airway smooth muscle preload), resulting in a tendency to excessive narrowing on stimulation of the airway smooth muscle. Previous studies have demonstrated alveolar inflammation in asthma (25, 26), and we are the first to demonstrate structural changes in the alveolar compartment of patients with asthma, such as damaged alveolar attachments and decreased elastic fiber content in peribronchiolar alveoli. In smokers, loss of alveolar attachments is implicated as an early stage in the destruction of the lung parenchyma and could be responsible for the early loss of elastic recoil (11), probably representing the initial phase of a more progressive and diffuse lesion in terms of alveolar destruction. Although we observed damaged alveolar attachments in the region of airway–parenchyma coupling with decreased elastic fiber content in our patients with asthma, we did not demonstrate alterations in elastic fiber content in the distal alveoli. As parenchymal tissue plays the major role in regulating elastic recoil in the lungs, other factors should certainly contribute to the recoil alterations observed in patients with chronic persistent asthma (27), such as mechanical stress relaxation due to recurrent bronchoconstriction and hyperinflation, stretching of the connective tissue, and altered lung growth (28). Indeed, we cannot totally exclude that part of the elastic fiber alterations observed in this study could reflect abnormal lung growth due to chronic lung changes in asthma. It has been previously demonstrated that adults with childhood-onset asthma present larger lung volumes than patients with adult-onset asthma and control patients (29, 30). One could allege that our findings represent acute rupture due to the final severe attack, what may be partially true. However, considering that patients with severe asthma are prone to several acute attacks during their lifetime, and that mechanical stretching elicits remodeling signaling (31), attachment disruption secondary to previous asthma attacks could have long-term pathologic consequences as well. Furthermore, rupture of alveolar attachments secondary to exacerbated bronchoconstriction only during the final attack could not explain our finding of decreased elastic fiber content in peribronchial alveolar septa. In addition, the significant correlation between the content of elastic fibers in the adventitial layers of small airways and peribronchial alveolar septa with the number of normal alveolar attachments reinforces the idea that elastic fibers contribute to the integrity of the alveolar attachments. We did not find differences in the content of elastic fibers in the distal alveoli between patients with asthma and control subjects. Moreover, control lungs showed a higher content of elastic fibers in the peribronchial region when compared with the distal septa, what is not observed in FA. The present data suggest that structural alterations of the lung parenchyma are localized in the peribronchial areas. Indeed, alveolar inflammation has been observed in transbronchial biopsies of patients with asthma (25, 26), a procedure that samples the peribronchial alveoli. It is possible that inflammation localized at this site could contribute to the observed elastic fiber changes in the adventitial layer of the small airways and peribronchial alveoli. In fact, we have previously described fragmentation of elastic fibers in the subepithelial layer of large airways, but no elastic fiber alterations in the inner wall of small airways in fatal asthma (12). Now we describe decreased elastic fiber content in the adventitial (outer) layer of small airways. Haley and coworkers (4) have previously demonstrated that in fatal asthma, small airway inflammation predominates in the outer wall layers. The same authors hypothesized that peribronchiolar inflammation of the small airways in asthma seemed to be disease specific, because it was not present in cases of cystic fibrosis. The inner and outer walls of airways have a different blood supply and it has been previously suggested that, in asthma, they might present a different inflammatory/cytokine milieu (32). If we believe that part of the observed elastic fiber changes are a consequence of inflammation, and that inflammation is more prominent in the outer wall, it is plausible that we have found elastic fiber alterations in the outer wall, but not the inner wall, of the small airways. What mediators could be involved in the degradation of the elastic fibers in the small airways of patients with FA? Increased levels of matrix metalloproteinase-9, a metalloproteinase with elastinolytic properties, have been demonstrated in the plasma of patients with severe acute asthma (33). Pham and coworkers (34) have further demonstrated increased levels of matrix metalloproteinase-9 and decreased tissue inhibitor of metalloproteinase-1 in patients with nocturnal asthma and decreased airway elastin density. Increased levels of active elastase have been demonstrated by Vignola and coworkers (35) in the sputum of both young and elderly patients with asthma. Assessment of matrix metalloproteinases in our patients could improve understanding of the mechanisms involved in changes in elastic fiber content. One important limitation of our study is the limited availability of clinical and functional data, so that correlations with functional parameters were not feasible. Only 8 of 15 patients had some medical follow-up. It is striking that, despite signs of asthma severity (previous hospitalization, use of two or more drugs), only five patients were using inhaled steroids. A local study has shown that the sales of inhaled steroids are low in São Paulo, despite the increase in the number of asthma fatalities in the last 10 years (36). Although we might be dealing with a heterogeneous population regarding asthma severity, all of them died during an acute asthmatic attack and most likely share the same physiopathological mechanisms leading to death due to asthma. Because the lungs of these patients were not fixed at pressure, Lm (mean alveolar diameter) measurements were not totally adequate to evaluate the presence of emphysema in our samples. However, there were no differences in alveolar size in peribronchiolar areas or histologic signs of emphysema in the patients with fatal asthma. Therefore, the localized elastic fiber changes in the peribronchiolar area cannot be attributed to areas of peribronchiolar hyperdistension leading to thinner alveolar septa. We believe that, in large airways, elastic fibers are important for the maintenance of the intrinsic airway recoil (12), whereas in the distal lung they are involved mainly in the airway–parenchyma coupling mechanism. Our previous and present data suggest that alterations of the elastic fibers might be involved in different ways in the mechanisms of enhanced bronchoconstriction in fatal asthma. These results contribute to our understanding of the pathophysiology of asthma, but the extent to which these results can be transposed to less severe cases of asthma remains unclear. However, the methodologic limitation of requiring large amounts of distal lung tissue makes this sort of investigation possible only with autopsy or lung excision material. In conclusion, we have shown an increased number of damaged alveolar attachments and decreased elastic fiber content in the adventitial layer of small airways and peribronchial alveoli in fatal asthma. We propose that these alterations can contribute to the pathogenesis of some of the functional abnormalities observed in patients with severe asthma. Increasing our knowledge about the structure–function relationship in small airways is essential to create a better understanding of the mechanisms related to asthma severity.
The authors thank Dr. Arthur F. Gelb and Dr. Noe Zamel for constructive discussions and suggestions during the performance of this study.
Supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and the Laboratório de Investigação Médica LIM-05 do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo. Conflict of Interest Statement: T.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; L.F.F.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.A.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; L.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; F.D.C.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.A.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; P.H.N.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form March 8, 2004; accepted in final form May 13, 2004
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
1-antitrypsin levels in adult asthmatics. Eur Respir J 2003;22:795–801.
