INTRODUCTION

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Keywords: airways; asthma; chronic bronchitis; COPD; emphysema; lung; remodeling

Asthma and chronic obstructive lung disease (COPD) are relatively nonspecific clinical terms used to describe two differing patterns of airflow obstruction with respect to reversibility, spontaneously or in response to treatment. In reality asthma and COPD are not single entities; each has a spectrum of reversibility and there is "overlap," most likely associated with the varying extent and the "mix" of both structural and inflammatory changes and the predominant anatomic site within the lung at which these occur (Figure 1). The premise herein is that the clinically observed distinctions between asthma and COPD, at least in individuals selected from polar ends of the clinical spectrum, should be reflected in contrasting patterns of inflammatory cells and cytokines and by differences in the magnitude, anatomic site, and nature of the remodeling process (1). The present synopsis briefly introduces the differences with respect to inflammation and then focuses on the structural changes, the last collectively referred to as "remodeling."

View larger version (24K): [in this window] [in a new window]   Figure 1.   Venn diagram illustrating the overlap between the chronic inflammatory conditions of asthma and COPD.

	DESCRIPTION AND DEFINITIONS

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Inflammation

Asthma and COPD are both chronic inflammatory conditions of the conducting airways and lung parenchyma. However, immunohistology of bronchial biopsies demonstrates differences in the predominant inflammatory cells: biopsies from nonsmoking mild allergic asthmatics have increased numbers of activated CD4⁺ (T helper) lymphocytes and eosinophils and variably reported increases of mast cells. In contrast, the chronic inflammation of smokers with COPD is predominantly of tissue CD8⁺ (T suppressor/cytotoxic) lymphocytes, macrophages, and, variably, neutrophils (2). While inflammation appears to be present at all sites within the lung in both conditions, the predominant anatomic site at which the structural changes occur differs. In COPD it is mainly destruction (or failure to repair) of the lung parenchyma leading to emphysema (Figure 2) and there are also important structural alterations that occur in varying degree to small bronchi and membranous bronchioli (i.e., airways < 2 mm in diameter). The relative contributions that the structural alterations at these two sites make to the airflow obstruction in each patient will vary. In asthma, large airways and small airways are altered structurally but, in the asthmatic nonsmoker, there is no parenchymal destruction. However, the walls of the conducting airways in asthma are thickened by between 50 and 300% of normal and there is lumenal narrowing, which is further compromised by excessive mucus admixed with an inflammatory exudate (Figure 3). In cases of fatal asthma, the longer the duration of asthma, the thicker the airway wall. However, it has been suggested that airway wall thickness per se is not a requirement for asphyxic fatality as a group of relatively young asthmatics (i.e., with a relatively short history of asthma) had an airway wall thickness not significantly different from that of nonasthma controls. It is surmised that lumenal secretions and plugging were the greater contribution to asthmatic death in these young cases of fatal asthma (3). Each tissue structural component, as well as inflammatory cell infiltration and edema, can contribute to the observed thickening; however, in the last-mentioned study it was thickening of the adventitial layers that was most pronounced in the older group with the longest duration of disease.

View larger version (95K): [in this window] [in a new window]   Figure 2.   Gross appearance of the cut surface of a lung, illustrating the characteristic distribution of centracinar emphysema, restricted to the upper aspects of each lobe scale. (Printed by permission from Professor B. Heard.)

View larger version (43K): [in this window] [in a new window]   Figure 3.   Representation of airways in the normal lung (left) and in asthma (right), demonstrating the thickening of the airway wall in asthma due to injury, chronic inflammation, and remodeling of epithelium and subepithelial tissue. The result is encroachment of the airway lumen and a marked increase in resistance to airflow.

Inflammation and Relationship to Remodeling

Acute inflammation is the response of vascularized tissue to injury: the inflammatory reaction is designed to protect the host and to restore tissue and its function to normal. One generally accepted proposal is that the accelerated decline in forced expiratory flow over time in COPD, and that which occurs also in an important subset of asthmatics, is the direct result of a switch from acute, episodic, to chronic inflammation and to consequent parenchymal and airway wall remodeling, respectively (4). The proposal is attractive but, as yet, there is no convincing evidence that the remodeling process is dependent on the prior development of chronic inflammation. It is equally plausible that the processes responsible for the development of chronic inflammation are distinct from those responsible for remodeling (Figure 4). The last consideration has important implications for the design of disease-modifying therapy: those agents that are effective anti-inflamatory compounds will not necessarily prevent or attenuate the process of remodeling.

View larger version (20K): [in this window] [in a new window]   Figure 4.   Representation of the proposed hypothesis for two distinct pathways of response to chronic injury: chronic inflammation or remodeling. Effective treatment for each of these distinct pathways will likely need to be different.

Definition

The concept of "remodeling" implies that a process of "modeling" must have preceded it. We have seen in the article by Warburton and colleagues in this issue (pp. S59-S62) that the lung, in utero, undergoes extensive modeling and remodeling yet these processes are entirely appropriate to the normal process of lung development. Many of the cytokines and growth factors thought to be proinflammatory in asthma and in COPD are also expressed normally without detriment to the developing lung. These include members of the fibroblast growth factor family, the transforming growth factor family, epithelial-derived growth factor, granulocyte-macrophage colony-stimulating factor, platelet-derived growth factor, vascular endothelial growth factor, and hepatocyte growth factor (4, 5) (see also the article by Warburton and colleagues). Accordingly the working definition of remodeling proposed herein recognizes that the process of remodeling per se is not of necessity abnormal. It is an alteration in size, mass, or number of tissue structural components that occurs during growth or in response to injury and/or inflammation. It may be appropriate, as in normal lung development or that which occurs during acute reaction to injury, or "inappropriate" when it is chronic and associated with abnormally altered tissue structure and function as, for example, in asthma, COPD, or fibrosing alveolitis.

In wound healing (in the skin) there is an appropriate response: clot formation, swelling/edema, rapid restitution of the denuded areas by epithelial dedifferentiation, and proliferation and migration from the margins of the wound. This is normally associated with an inflammatory reaction: that is, early infiltration of the injured tissue by neutrophils and later by lymphocytes and macrophages. Reticulin is deposited within days and this may mature to form interstitial collagen, a scar, within 2-3 wk. In addition, healing may involve contraction of the surrounding tissue in the case of an open wound, by myofibroblasts that may proliferate transiently in relatively large numbers (6). Vasodilatation, congestion, and mucosal edema are also cardinal signs of acute inflammation and the angiogenesis of the granulation tissue is an integral part of the reparative response (7). Thus, normal tissue architecture and function are restored consequent to an entirely appropriate inflammatory and remodeling process. Each of these stages in normal wound healing and many of the inflammatory cell types and cytokines involved (e.g., transforming factor ) appear also in asthma and in COPD, but in asthma both the inflammation and remodeling persist and the consequences are inappropriate to the maintenance of normal (airway) function. The reasons for the persistence are unknown but may be the result of repeated inhalation of allergen or exposure to high concentrations of allergen, or infection, or a genetically influenced abnormal host inflammatory response or defect in the repair process. In COPD the injury is presumed to be due to repeated exposure to cigarette smoke, environmental and/or occupational pollutants, chronic (or latent) infection, or an interaction of these. The host responses in asthma and COPD may of course differ due to differences in genetic makeup or consequent to developmental abnormality or to the effects of perinatal environmental exposure: these may be associated later with differing patterns and sites of chronic inflammation and inappropriate remodeling.

	ALTERATIONS TO LARGE AIRWAYS

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Epithelial Injury

Histologically, damage and shedding of the airway surface epithelium are reported in asthma, postmortem, but this change is highly variable: some airways have intact surface epithelium even in the presence of marked inflammation and other structural change (8). Loss of epithelium induces a healing process that results in either complete restoration of the columnar/ cuboidal epithelium with normal proportions of goblet and ciliated cells or, if the injury is repeated, squamous cell metaplasia and/or goblet cell hyperplasia. Histologically, damage and shedding of airway surface epithelium are an often-reported feature of asthma: there are clusters of sloughed epithelial cells (referred to as Creola bodies) in asthmatic sputa, increased numbers of epithelial cells in bronchoalveolar lavage (BAL) fluid, and loss of the surface epithelium in biopsy specimens (9). Aggregations of platelets together with fibrillary material, thought to be fibrin, have been observed in association with the damaged surface (11). In asthmatics with varying severity and symptoms the greater the loss of surface epithelium in biopsy specimens the greater the degree of airway responsiveness (11). However, the loss of epithelium observed in biopsies of mild asthmatics is highly variable and an unreliable end point for determination of extent of injury or the response to treatment (12). The variability between specimens and laboratories is likely due to several factors, including inherent fragility of the epithelium in asthmatics, mechanical stresses imposed during bronchoscopy, and differences between the types of forceps used, their sharpness and operator methodology and experience. In contrast, epithelial loss is a less often reported feature of bronchial biopsies taken from smokers with bronchitis or COPD when goblet cell hyperplasia and squamous metaplasia are often seen (Figure 5A and 5B) (10, 11, 13).

View larger version (134K): [in this window] [in a new window]   View larger version (146K): [in this window] [in a new window]   Figure 5.   The airway mucosa as it appears by light microscopy (LM) of bronchial biopsies from (A) an atopic asthmatic showing loss of surface epithelium and early thickening and hyaline appearance of the reticular basement membrane (RBM) and (B) a heavy smoker with COPD (FEV1 = 40% of predicted), demonstrating epithelial squamous metaplasia and a thinner RBM of normal thickness. Stained with hematoxylin-eosin (H&E).

Injury to superficial epithelium is normally accompanied by mitotic activity in the remaining cells and rapid restoration of the denuded surface (14). There are, however, newly emerging data indicating that the mitotic response in asthma, unlike COPD, may be abnormally suppressed: the proposal in asthma is that there may be an abnormal repair response of the epithelium to injury (15). One hypothesis is that chronic injury or defective repair of the surface epithelium results in its persistent activation and the chronic secretion of a variety of proinflammatory epithelial-derived cytokines and growth factors that drive both the subsequent chronic inflammatory and remodeling responses seen in subepithelial compartments. These factors include epithelial-derived growth factor and granulocyte-macrophage colony stimulating factor and would induce alterations to the epithelial reticular basement membrane, via activation of adjacent fibroblasts/myofibroblasts, and deeper structures including bronchial smooth muscle, mucus-secreting glands, and wall vessels. The release of these and other molecules including interleukin 8 (IL-8), eotaxin, and RANTES (regulated on activation, normal T-cell expressed and secreted) would also provide a chemoattractant gradient to both inflammatory and phenotypically altered structural cells (see below).

Reticular Basement Membrane Thickness

Thickening of the lamina reticularis or subepithelial reticular basement membrane (RBM) is a characteristic change in asthma (See Figure 5A) albeit subtle differences between atopic and nonatopic forms of asthma have been reported (16, 17). This example of remodeling, when homogeneously thickened and hyaline in appearance, is highly characteristic and usually pathognomonic of asthma. The thickening is not found in smokers with chronic bronchitis or COPD (compare Figure 5A and 5B), providing there is no evidence of reversible airways obstruction (18). The RBM is not present in the fetus (at least up to 18 wk of gestation) (19) but develops later even in normal, healthy individuals, presumably during childhood. Relative to the norm, RBM thickening occurs early in the asthma process (20), even before asthma is diagnosed (21). In a collaboration with A. Bush and D. Payne (London, UK) we have had the opportunity to examine bronchial biopsies from asthmatic children (n = 19) between the ages of 6 and 16 yr who had remained symptomatic despite relatively high doses of inhaled corticosteroid. We have compared the RBM thickness with age-matched nonasthmatic children (n = 7) undergoing bronchoscopy for other clinical indications and found that the RBM of the asthmatics was thickened even at this early age (20). In a further collaboration with K. Guntapalli (Baylor, Texas) we have had the opportunity to extend the study to include 10 steroid-naive adult asthmatics (mean age, 28.5 yr), 6 adult asthmatics intubated for life-threatening asthma (mean age, 34.5 yr), and 8 healthy adult nonasthmatic controls (mean age, 33.5 yr) (22). The RBM thickness was significantly increased in the asthmatic children compared with the healthy adult controls but the thickness in the asthmatic children and the adults with mild and severe asthma was similar. There was no correlation between RBM thickness and either symptom duration for asthmatic children or the age or severity for all the asthmatics.

These results highlight how early and maximal is this characteristic aspect of remodeling and indicates that it is unrelated to the duration of chronic inflammation and severity of the disease. Also, the thickening remains even when asthma is mild and well controlled by antiasthma treatment (23) and it is present postmortem in patients with a long history of asthma but who have not died of their asthma (24) (Figure 6). Table 1 shows the data for the comparison of RBM thickness in normal healthy nonsmokers, smokers with bronchitis, smokers with chronic bronchitis and COPD, and asthmatics matched for age and disease severity (23). Apart from the asthmatics receiving inhaled steroids, none of the others received such treatment. However, despite their steroid treatment the asthmatics still had a significantly thicker RBM at either of the two airway levels studied (23). The RBM thickness in the smokers with chronic bronchitis alone or with COPD was within the normal range. One caveat to this apparently clear distinction between asthma and COPD is that a proportion of patients with COPD defined clinically by their irreversibility to inhaled -agonist can be shown to be partially reversible after a (2-wk) course of oral corticosteroid. Interestingly, this group shows raised levels of eosinophil cationic protein in their bronchoalveolar lavage and a significantly thickened RBM in their bronchial biopsies compared with COPD patients, who have a airflow obstruction irreversible to oral corticosteroids (18).

View larger version (108K): [in this window] [in a new window]   Figure 6.   Scanning electron micrographs demonstrating the airway mucosa in (A) a nonasthmatic subject with epithelium attached to reticular basement membrane (RBM) of normal thickness, beneath which there is interstitial collagen, and (B) a subject with a 25-yr history of asthma but who died of nonrespiratory causes, demonstrating thickened RBM and damaged epithelium.

Because of the immunopositivity of the RBM for epitopes recognized in collagen (e.g., types I, III, and V), its thickening in asthma has been referred to as "subepithelial fibrosis" (25). In the author's opinion this is an unfortunate application of the term "fibrosis" as reticulin, a major component of the RBM, is ultrastructurally different from the "banded" (65-nm periodicity) and relatively thicker collagen fibers that lie deeper in the interstitium of the airway wall. In contrast to interstitial collagen, reticulin fibers are linked to glycosaminoglycans, fibronectin, and a tenascin-rich matrix (26) that entraps a variety of molecules such as heparin sulphate and growth factors. These entrapped molecules may modulate the state of differentiation, integrity, and function of the overlying surface epithelium. The author proposes that the additional entrapped and adsorbed molecules may also provide an increased osmotic force and gradient that encourages thickening of the RBM by swelling due to uptake of water. Against the "fibrotic" theory, the "osmotic" proposal is supported by the early maximal thickening, which is usually only of the order of 2-3 µm and, at most double, its norm and the thickening is not progressive with time. Unlike fibrotic tissue, the RBM can also relatively quickly change its reticular state and be "lost" after allergen challenge, this associated with the subepithelial accumulation of fibromyocytes (see below). Also, the thickened RBM does not behave as a barrier to the transmigration of inflammatory cells, which by the release of enzymes (e.g., metalloproteases) or via predetermined pores may pass through it with apparent ease (27). Thickening of the RMB has been reported to show a positive correlation with airway hyperresponsiveness, the frequency of asthma attacks, and the numbers of fibroblasts and "myofibroblasts" that lie external and adjacent to it (28).

The data with respect to increases in the amounts of airway wall interstitial collagen are controversial. These data show that with increasing severity of asthma there are increases in the area of the mucosa staining for collagen (31) but in contrast there are other studies finding no such relationship (32). Our own findings in mild asthma show no differences with respect to collagen or elastic tissue (33). In contrast, airway wall fibrosis is generally, but not always, considered a feature of the airways in smokers who develop COPD, albeit these studies have focused on small rather than large airways (34).

Enlargement of Bronchial Smooth Muscle Mass

Bronchial smooth muscle is arranged in a geodesic pattern encircling the airway as two apposing spirals: when the muscle shortens it not only constricts but also shortens the airway (Figure 7). This may be important as any factor that stiffens the airway, such as increased RBM thickness, increased collagen deposition, vascular congestion, or edema, will result in resistance against airway shortening (37). It is hypothesized that reduction of the capacity to shorten an elastic airway, due to stiffening, will result in the consequent redirection of the tension developed by the geodesic/oblique anatomic arrangement of airway smooth muscle such that more of the tension than normal will resolve in favor of airway constriction than of airway shortening.

View larger version (78K): [in this window] [in a new window]   Figure 7.   A drawing illustrating the geodesic pattern of airway smooth muscle as it spirals around the airway. (Reprinted by permission from WS Miller. The Lung. Springfield, IL: Charles C. Thomas; 1937.)

The percentage of the bronchial wall occupied by bronchial smooth muscle is increased in fatal asthma (38) (Figure 8A). Using a "point-counting" morphometric technique applied to tissue sections of conducting airways, Dunnill and coworkers showed that approximately 12% of the airway wall in segmental bronchi obtained from cases of fatal asthma was composed of muscle compared with about 5% in normal subjects. Hogg and colleagues (39, 40) have confirmed this trend in airways larger than 2 mm in diameter and demonstrated a 2- to 4-fold increase over normal in the area of the wall occupied by bronchial smooth muscle. The increase is 4-fold in older (with a longer duration of disease) and 2-fold in younger cases of fatal asthma (3). The absolute increase in muscle mass is reported to be particularly striking in large intrapulmonary bronchi of lungs obtained after a fatal attack as compared with that seen in asthmatic subjects dying of other causes (8) (compare Figures 8B and 8C). Occasionally, the author (in collaboration with K. Guntapalli and colleagues) observed that large areas (> 85%) of a bronchial biopsy may be occupied by subepithelial muscle bundles in severe (intubated) cases of asthma (unpublished, 2001). The increase in muscle mass may also begin relatively early in the child. In an ongoing collaboration (with V. Macek and coworkers) we have identified, in bronchial biopsies, airway smooth muscle in relatively large amounts, close to the epithelium even in children with asthma (unpublished observations).

View larger version (187K): [in this window] [in a new window]   View larger version (132K): [in this window] [in a new window]   Figure 8.   Increased bronchial smooth muscle. (A) A histological section of the airway wall of a case of fatal asthma stained with H&E to show enlarged smooth muscle blocks lying relatively close to the surface epithelium. (B) Scanning electron microscopy of part of the mucosa in a nonasthmatic and (C) fatal asthma, demonstrating the three-dimensional appearance of the enlarged blocks of bronchial smooth muscle and dilatation of bronchial vessels, which both contribute to thickening of the airway wall. There is loss of bronchial epithelium.

Few studies of COPD have focused attention on the larger (cartilaginous) airways. One systematic study has described changes in large airway dimensions in relation to the lung function of patients with COPD (41). These authors tested the hypothesis that airflow obstruction and estimates of small airway inflammation correlate with large airway wall thickness and muscle mass. They found that the wall area internal to the muscle was significantly thickened over the entire range of cartilaginous airways measured and that this was associated with a reduction in FEV/FVC. However, alterations in large airway smooth muscle mass were not observed and there was no correlation between muscle mass and airflow limitation. Interestingly, there was a positive association between peripheral airway inflammation and large airway inner wall area and the authors argued that their findings and those of others favor inflammation as the cause of the increasing inner airway wall thickness that occurs in both large and small airways in COPD.

There are several proposals for the mechanisms responsible for the increased smooth muscle mass in asthma: each may contribute in varying degree and the predominant pathway may differ from airway generation to generation. Muscle fiber hyperplasia (42), hypertrophy (43, 44), or other mechanisms may each contribute. Smooth muscle cell (myocyte) proliferation resulting in an increase in myocyte number is likely to be a major contributor to the increase in smooth muscle mass. Other novel mechanisms may also be responsible. We have observed that solitary contractile cells, originally referred to as "myofibroblasts," appear in substantial numbers after the late-phase reaction (i.e., in biopsies taken 24 h postchallenge) in response to experimental allergen challenge (45). The ultrastructure of these large, irregularly shaped cells (compare Figures 9A and 9B) demonstrates not only the intracellular presence of secretory organelles (i.e., rough endoplasmic reticulum and Golgi apparatus) but also elongate bundles of filaments with electron-dense condensations that are identical to the contractile apparatus present in bronchial smooth muscle myocytes. With respect to their cell size and filamentous content these cells are distinct from the myofibroblasts described by Roche and colleagues (25). Our original report considered that the fibroblast was a likely origin of the allergen-induced myofibroblast and there is in vitro work that supports this (46). However, 2 yr ago, M. J. Gizycki made the interesting observation that myocytes of the intact airway smooth muscle bundles showed ultrastructural evidence of dedifferentiation to a "secretory" phenotype. Outside the normally occurring smooth muscle bundles there were solitary contractile cells that we have previously called myofibroblasts. These solitary contractile cells were identified at varying positions in the mucosa including a zone immediately adjacent to the surface epithelium. He speculated that, with repeated exposure to allergen, these contractile cells, which he considered should be called "fibromyocytes," represent dedifferentiated bronchial smooth muscle that was in the process of migrating toward the surface epithelium. He proposed that in response to allergen challenge in asthma, dedifferentiation of existing smooth muscle myocytes occurs and they migrate, in the form of a fibromyocyte phenotype, to a subepithelial site where they may form new muscle of abnormal phenotype and abnormal function (45). In this respect, the epithelial injury and smooth muscle response in asthma parallels the findings of the response to endothelial injury and the associated changes of vascular smooth muscle in atheroma (47). In atheroma there is transformation of vascular myocytes to a "synthetic" phenotype and their subsequent migration to the neointima. The capacity of bronchial smooth muscle myocytes to migrate has been demonstrated (48). These findings are interesting in the light of previous comparisons of the inflammation of atheroma and asthma (47). The author proposes herein that the remodeling processes associated with injury, inflammation, and smooth muscle migration and increase in asthma and atheroma are similar. Figure 10 shows the schema we propose for two novel pathways to explain the smooth muscle increase in asthma. A further possibility is that pericytes, immature contractile cells associated with vessels, may also act as a source of newly emerging myofibroblasts. There is new experimental evidence in vivo for the response of airway wall vessels to inhalational allergen challenge and for a similar dedifferentiation and migration of vascular myocytes to repeated allergen challenge as that described above in airway smooth muscle (49). It would appear that the contribution to inflammation and plasticity of structural cells has been hugely underestimated both in asthma and in COPD.

View larger version (124K): [in this window] [in a new window]   View larger version (144K): [in this window] [in a new window]   Figure 9.   Transmission electron micrograph of (A) a subepithelial fibroblast of normal appearance and (B) a myofibroblast or "fibromyocyte" in the bronchial mucosa of a mild atopic asthmatic 24 h after experimental inhalational allergen challenge. The number of these cells increases markedly compared with the patients challenged with diluent. (Reprinted by permission from Reference 45.)

View larger version (18K): [in this window] [in a new window]   Figure 10.   Schema summarizing the two likely pathways that generate allergen-induced myofibroblast and fibromyocyte forms. (Reprinted by permission from Reference 45.)

Lumenal Content, Epithelial Goblet Cells, and Submucosal Glands

Many asthmatics suffer from excessive production of mucus, which, admixed with the inflammatory exudate, forms highly tenacious plugs that block the airways and are exceedingly difficult to clear by cough (50) (Figure 11). There is clearly a component of mucus to these secretions but the especially high additional contribution by inflammatory cells and their secretory products in asthma may explain the sticky nature of these secretions as compared with the intralumenal mucus of chronic bronchitis.

View larger version (129K): [in this window] [in a new window]   Figure 11.   Gross appearance of the cut surface of the lung in a case of fatal asthma. The airway lumena are completely blocked by sticky secretions composed of mucus admixed with inflammatory exudates.

Epithelial goblet cells and mucus-secreting submucosal glands are the major sources of lumenal mucus. Goblet cell hyperplasia is a feature of the large airways in both asthma (51, 52) and chronic bronchitis (53). Submucosal gland hypertrophy is also seen to about the same extent in both asthma and COPD (38). There are serous and mucous secretory units (acini) in the glands: it is reported that, while the normal proportions of serous to mucous acini are retained in asthma there is a disproportionate increase in mucous acini and loss of serous acini in chronic bronchitis (54). As the serous acini contribute a variety of antibacterial substances including lysozyme, lactoferrin, and the secretory component of secretory IgA, their reduction during the remodeling process in chronic bronchitis may be relevant to the ease with which the respiratory tract becomes chronically colonized by bacteria.

	ALTERATIONS TO VASCULATURE

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Dilatation of bronchial mucosal blood vessels, congestion, and wall edema are consistently reported features of fatal asthma and these can account for considerable swelling and stiffening of the airway wall (Figure 12) (37, 55, 56). There are indications that the increased proportion of the wall occupied by vessel may be due in part to a proliferation of bronchial vessels (angiogenesis) (57). While angiogenesis has been reported in mild asthma (58) it is particularly marked in severe corticosteroid-dependent asthma (59). Whether these changes are the consequence of chronic allergic inflammation or due to the response to chronic (or latent) viral, mycoplasm, or bacterial infection is not known. While proliferation of the bronchial vasculature is a feature of bronchiectasis and occurs in response to infection, changes to the bronchial vasculature have not been reported as a particular feature of COPD (55). However, patients with moderate to severe COPD do have elevated pulmonary vascular pressures during exercise and there are structural changes in the pulmonary arteries consistent with endothelial dysfunction and pulmonary hypertension when compared with patients with minimal or no disease. Small (< 500 µm) pulmonary vessels in airway-obstructed smokers show intimal thickening as compared with those of nonobstructed nonsmokers: in severely obstructed smokers, there is medial hypertrophy also (60). Such structural changes likely contribute to the narrower lumens and vascular obstruction of these vessels. Interestingly, there is infiltration of the pulmonary arterial wall by T lymphocytes. The CD8⁺ T cell phenotype is increased in both nonobstructed smokers and smokers with COPD compared with nonsmokers and the intensity of the inflammatory infiltrate has been shown to correlate with both endothelium-dependent relaxation and intimal thickness (63).

View larger version (154K): [in this window] [in a new window]   Figure 12.   Histological transverse section of a bronchus from a case of fatal asthma. The lumen is blocked with secretions, there is goblet cell hyperplasia, thickening of the RBM, inflammation, increased smooth muscle, and marked dilatation and congestion of bronchial vessels. Stained with H&E. (Printed by permission from Professor B. Heard.)

	CHANGES TO SMALL AIRWAYS AND LUNG PARENCHYMA

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Inflammation and structural alterations occuring in the small airways and lung parenchyma in COPD are considered the most important contributors to the airflow limitation and to the accelerated decline of FEV₁ in COPD. Similarly, changes in small airways in severe asthma may also contribute to the accelerated decline in a small but significant proportion (~ 10%) of asthmatics and there is currently much interest as to the role of small airway abnormalities in asthma (Figure 13). This may be especially important in smokers with asthma, a group that has been little studied with respect to their inflammation and remodeling processes.

View larger version (148K): [in this window] [in a new window]   Figure 13.   Bronchiole in a case of asthma showing eosinophilic plugging of the small airway, associated with inflammation and thickening of the airway wall. The surrounding alveoli remain intact. (Printed by permission from K. Hamada and colleagues.)

Inflammation

There is now bronchoscopic evidence of an inflamed bronchiolar epithelium in COPD with increased release of proinflammatory cytokines including IL-8 (64). In COPD there is infiltration of the wall of small airways by CD8⁺ T cells and there is an increase in the number of pigmented macrophages in both airway and alveolar lumena (65). This pattern and the extent of inflammation are likely to be associated with thickening of the airway wall, loss of alveolar-bronchiolar attachments, and consequent loss of elastic recoil and lumenal narrowing (Figure 14). There is also a strong relationship between respiratory bronchiolitis (consisting of accumulations of pigmented macrophages), peribronchiolar fibrosis, and smoking: the relationship of these airway changes to septal thickening of surrounding alveolar walls and the development of patchy alveolar wall fibrosis with a peribronchiolar distribution (referred to as "respiratory bronchiolitis-asociated interstitial lung disease") is unclear but also of current interest to the development of both interstitial lung disease and emphysema (66).

View larger version (138K): [in this window] [in a new window]   Figure 14.   Small airway disease in COPD: a transverse section of a small airway showing peribronchiolitis consisting predominantly of lymphocytes. There is damage and loss of alveolar-bronchiolar attachments associated with centriacinar emphysema and consequent collapse of the conducting airway, especially during expiration. Stained with H&E.

While inflammation in young smokers distinguishes smokers from nonsmokers, it is still not clear which, if any, of the several small airway lesions, that is, fibrosis, goblet cell metaplasia, or smooth muscle mass enlargement, contribute most to the stenotic lesions seen in plastic casts of the small airways (Figure 15) or the increased rate of decline in lung function characteristic of COPD. In asthma, eosinophilic and T-helper "Type 2" inflammation has been demonstrated in the small airways and surrounding alveolar walls but there is no associated loss of peribronchiolar attachments or emphysema reported (67).

View larger version (37K): [in this window] [in a new window]   Figure 15.   Plastic cast of the small airways in COPD, showing focal stenoses that are responsible for increased resistance to airflow locally. (Printed by permission from Professor Bignon.)

Goblet Cells

In contrast to large airways, goblet cells are normally absent or sparse in airways less than 2 mm in diameter (i.e., small bronchi and bronchioli) but they appear and increase in number in these peripheral airways in COPD (68), a process referred to as mucous metaplasia. Whether mucous metaplasia occurs also in asthma is debated. Some investigators suggest that the mucus found at this peripheral anatomic site may be aspirated from the larger airways rather than being produced locally. However, the work of Aikawa and colleagues is persuasive and indicates that mucous metaplasia is also a feature of asthma, when goblet cell degranulation, during a fatal attack, may be widespread, leading to the release of copious amounts of mucus that may remain adherent to the goblet cells (51).

Smooth Muscle Mass

The relative contribution of airway wall smooth muscle mass to overall airway wall thickness in small airways is much greater than that in the large airways. Airway smooth muscle increases significantly in the small airways in COPD (36, 65, 68, 69). In a study of small (membranous) airways of 15 patients with COPD compared with the lungs of nonobstructed subjects and a group of asthmatic patients, it was only the airway smooth muscle area that was significantly increased in COPD (55). In asthma, the increase in the wall area occupied by muscle, in absolute terms, is not as striking in small airways as in the large (8). It is considered that the increased muscle mass that occurs at all generations of airway is likely to be the most important abnormality responsible for the increased airflow resistance observed in response to bronchoconstricting stimuli in both asthma and COPD (37). Further studies and a greater understanding of the changes occurring in small airways is required, as is a means of effective delivery of anti- inflammatory and anti-remodeling therapy to this distal anatomic site.

Finally, this synopsis does not allow consideration of the important changes of emphysema in COPD, also likely the consequence of a chronic CD8⁺ cell inflammatory process. The current definition of emphysema excludes the presence of obvious fibrosis, yet it is now known that fibrosis may also occur even in the presence of alveolar wall loss (70, 71). The appearance of the fenestrae of "microscopic emphysema" (Figure 16) and subsequent enlargement of alveolar spaces, distal to the terminal bronchiolus, in COPD may thus represent the consequence of lung injury and a failure of adequate repair rather than of alveolar wall destruction per se. Thus the focal fibrosis that may be identified in some cases of emphysema may represent the remainder of a repair process. Further studies of the mechanisms that balance the production and degradation of collagen that occurs during the reparative and remodeling response to lung injury may yield important findings applicable to the treatment or prevention of the parenchymal lesions so important to COPD.

View larger version (150K): [in this window] [in a new window]   Figure 16.   Human lung alveoli as they appear by scanning electron microscopy in a smoker who has developed microscopic emphysema. The fenestrae are too small to be seen by the naked eye on gross examination of the lung but likely contribute to loss of lung elastic recoil.

	SUMMARY AND CONCLUSIONS

TOP ABSTRACT INTRODUCTION DESCRIPTION AND DEFINITIONS ALTERATIONS TO LARGE AIRWAYS ALTERATIONS TO VASCULATURE CHANGES TO SMALL AIRWAYS... SUMMARY AND CONCLUSIONS REFERENCES

Remodeling of the airway wall occurs in both asthma and COPD, but there are differences in the structures affected and the prime anatomic site at which they occur (see Table 2). Early thickening of the epithelial RBM and infiltration of the wall by eosinophils and T-helper (CD4⁺) lymphocytes is highly characteristic of asthma. The RBM thickening does not change with age, severity, or duration of asthma. The large airway increase in airway smooth muscle is a feature of fatal asthma, as is the lumenal obstruction due to mixtures of inflammatory exudate and mucus. The last severely increases surface tension forces, making the airway difficult to maintain patent in asthma. Small airway mucous metaplasia, increased muscle mass, airway wall fibrosis, emphysema, and infiltration of the airways, pulmonary vessels, and lung parenchyma by (CD8⁺) T-cytotoxic lymphocytes are features of COPD. The predominant anatomic site of the alterations varies in individuals and will determine the extent and rate of decline in the airflow obstruction that develops. The involvement of small airways seems to be an important determinant of fixed airflow obstruction in both COPD and asthma. In asthma, the remodeling usually begins early, in the child. In COPD, the changes begin in later life. The nature of the inflammation and its response to corticosteroid differ: usually responsive in nonsmoking asthmatics and, for the most part, resistant in smokers with COPD. We do not fully understand the relationship between chronic inflammation and remodeling. The relationship between remodeling and accelerated decline in lung function is also unclear. We may learn much about the remodeling process in asthma from the work already done in the field of vascular research. Finally, it is likely that distinct forms of treatment are required to target separately the inflammatory and remodeling processes. Table 3 summarizes what we know about the likely effects of inhaled corticosteroids (P. K. Jeffery, unpublished, 2001) and other classes of agent such as long-acting -agonists and leukotriene receptor antagonists that may also be beneficial (72). Some, but not all, of the remodeling seen in humans can be induced successfully in vivo (in the rat and more recently the monkey). We look forward to further results of such experimental investigations that will help us to unravel some of the many remaining questions that concern inappropriate remodeling in the conducting airways and lung parenchyma in both asthma and COPD.