Abstract
Airway obstruction and airway hyperresponsiveness are important features of asthma and chronic obstructive pulmonary disease (COPD). Both diseases are characterized by airway wall and lung tissue inflammation, and in asthma there exists a relationship between the inflammatory state of the airways and the severity of hyperresponsiveness. However, the type and cause of this inflammation, as well as the extent and consequences of the inflammatory process, are different in asthma and COPD. Inflammatory processes affecting the airway wall both in peripheral and central areas of the lung appear to be important, the former one dominating in COPD and the latter in asthma. However, it is not clear which structural changes are open for therapy and which are not. Therefore, a better understanding of the consequence of inflammation for lung tissue and airway wall changes in asthma and COPD has to evolve before a full understanding of airway hyperresponsiveness will emanate. Postma DS, Kerstjens HAM. Characteristics of airway hyperresponsiveness in asthma and chronic obstructive pulmonary disease.
Airway obstruction and airway hyperresponsiveness are important features of asthma and chronic obstructive pulmonary disease (COPD). Airway hyperresponsiveness to nonspecific stimuli such as histamine, methacholine, and cold air is a hallmark of asthma, whereas many patients with COPD also have abnormal values in tests of hyperresponsiveness. Airway hyperresponsiveness is defined by an exaggerated response of the airways to nonspecific stimuli, which results in airway obstruction. It is yet unknown which factors within the airways of an individual are responsible for this exaggerated airway narrowing. Therefore, one could question why hyperresponsiveness is currently being assessed in research when this response is so nonspecific.
Airway hyperresponsiveness has recently been shown to be associated both with the development and the remission of respiratory symptoms in an adult general population (Figure 1) (1). Individuals with increased responsiveness who were asymptomatic were more likely to develop symptoms in the ensuing 3 yr than those who were less or not responsive (1). Moreover, when looking into the groups of individuals who were symptomatic, those with less severe responsiveness were more likely to lose their symptoms at follow-up. Increased airway responsiveness usually precedes the development of symptoms in children and young adults as well.
Fig. 1. Development and remission of six respiratory symptoms over 3 yr in an epidemiologic study on 2,684 participants. (Reprinted from Reference 1 by permission.)
[More] [Minimize]In subjects with established asthma and COPD, airway hyperresponsiveness has also been shown to be an important risk factor. In asthma, more severe hyperresponsiveness is associated with more symptoms and a steeper fall in FEV1 (2). The combination of wheeze and hyperresponsiveness discriminates especially the asthmatic group with significant ongoing respiratory impairment. Hyperresponsiveness persists in children with persisting symptoms, but it generally improves in young asthmatics in their teens. A reason behind improvement in hyperresponsiveness is possibly related to growing airway diameter. However, this cannot be the sole solution, as this also occurs in those with persistent hyperresponsiveness. Finally, the severity of airway hyperresponsiveness determined the need of therapy, and the more severe the hyperresponsiveness is, the smaller the response in FEV1 when inhaled corticosteroids are given (3).
Many individuals with COPD demonstrate increased responsiveness as well. In the Lung Health Study, approximately two-thirds of patients with mild or early COPD had demonstrable airway hyperresponsiveness (4). The severity of airway hyperresponsiveness predicted subsequent decline in COPD. Individuals with more severe hyperresponsiveness had a more rapid decline in FEV1, also after adjusting for baseline FEV1, age, baseline smoking history, and changes in smoking status (5). This effect was apparent in both males and females. Furthermore, it was shown that there was a strong interaction between hyperresponsiveness and smoking. Individuals with hyperresponsiveness who continued to smoke were most prone to a rapid decline. Those who smoked intermittently had a less rapid decline than those who continued smoking, and in the latter group the decline increased with more severe hyperresponsiveness.
The interpretation of the test in patients with asthma and COPD is hampered by the dependence of the degree of hyperresponsiveness on the baseline FEV1: the same bronchoconstrictor reaction to a given (nonspecific) stimulus will result in a lower (more abnormal) provocative concentration of, for instance, methacholine causing a 20% reduction in FEV1 (PC20) in a subject with severe obstruction than in a subject with less or no obstruction. The dependency of PC20 on baseline FEV1 is numerically the same in asthma and COPD (6). Since patients with COPD have obstruction that is frequently quite severe, it has been put forward that hyperresponsiveness is a surrogate marker of airway obstruction in COPD, whereas it is something of pathophysiologic importance in asthma. However, in the Lung Health Study approximately two-thirds of patients with mild or early COPD, and hence with only mild obstruction, have demonstrable airway hyperresponsiveness (5). Thus, even though there exists a significant association with FEV1 in individuals with advanced COPD, these data may suggest a more pathophysiologic relevance of hyperresponsiveness in COPD as well.
Although asthma has been defined by a significant reversible component, there can be an element of fixed obstruction, especially in patients who have longstanding asthma. In the latter group hyperresponsiveness may be determined to some extent by baseline FEV1 as well. In less severe asthma, FEV1 does not importantly contribute to the severity of airway hyperresponsiveness as measured with histamine or methacholine.
Asthma and COPD differ with respect to maximal airway narrowing. There exists a limit to the response in COPD, i.e., a plateau in bronchoconstriction: no further narrowing occurs whatever the dose given. This plateau is lacking in more severe asthma. The most direct evidence of the relationship between excessive airway narrowing and airway inflammation comes from a study showing a correlation between the maximal response to methacholine and the eosinophil counts in bronchial biopsies in asthma. How far the above mechanisms contribute to this excessive narrowing has been studied to some extent but is not resolved as yet.
Asthma and COPD resemble each other in that they both may show variable severity in airway hyperresponsiveness. Both diseases are also characterized by airway and lung tissue inflammation, and in asthma there exists a relationship between the inflammatory state of the airways and the severity of hyperresponsiveness. However, the type and cause of this inflammation, as well as the extent and consequences of the inflammatory process, are different in asthma and COPD. Whether a genetic component drives the presence of this inflammatory process and hyperresponsiveness in both diseases remains to be established. The question arises of what constitutes the pathophysiology of hyperresponsiveness in these two entities.
Inflammation in central and peripheral airways can be assessed with sputum induction, bronchoalveolar lavage, airway wall biopsies, and transbronchial biopsies. They all reflect the inflammatory process, yet give different information. In this article we will discuss the role of inflammation as obtained with biopsies.
One aspect of inflammation is the number and activation of inflammatory cells in the airways. Table 1 shows the different cells involved in the airway wall inflammation in asthma and COPD. In asthma, the number of eosinophils and T lymphocytes, with predominance of CD4 subtype, have been reported to be increased in the subepithelial layer (7). Both the eosinophils and lymphocytes are activated. Moreover, mast cell numbers are increased and activated. With exacerbations, the number of eosinophils may increase next to neutrophils, and it has been shown that the number of neutrophils may be increased in long-standing asthma despite adequate treatment.
| Asthma | COPD | |
|---|---|---|
| T lymphocytes, CD4 | T lymphocytes, CD8 | |
| CD25 | CD25, VLA-1, HLA-Dr | |
| Eosinophilia | Mild eosinophilia | |
| Activated eosinophils | Nonactivated eosinophils? | |
| Mast cells | Mast cells | |
| Neutrophils | Neutrophils | |
| Macrophages |
Although neutrophils have been found to be increased in bronchoalveolar lavage fluid of patients with COPD, they are not abundantly present in the airway wall (7, 8). Recently, however, Saetta (8) reported neutrophils to be preferentially present in the epithelial layer and bronchial mucous glands in COPD. Significant increases are reported in the numbers of total leukocytes, T lymphocytes, activated T lymphocytes (as determined by their positivity to CD25 and VLA-1), and macrophages. Eosinophils have been reported to be present in chronic bronchitis, in particular during exacerbations of the disease. It has been suggested that, in contrast to asthma, the tissue eosinophils found in COPD do not degranulate and are not associated with an increased expression of interleukin (IL)-5 (7, 8). As airflow limitation progressively worsens, T lymphocytes and macrophages increase in the subepithelium. A recent study reported that specifically the number of CD8 T-lymphocyte subsets increases, and a higher number of these cells are associated with a lower level of airflow limitation (9). This report contrasts the inflammation of COPD with the inflammatory process in asthma and its predominance of the CD4 subset.
The number and activity of inflammatory cells have been established to be linked with airway hyperresponsiveness. In asthma, many studies have suggested an association of the number of inflammatory cells and the number of activated cells with airway hyperresponsiveness. Although these correlations exist, it is clear from Figure 2 that for a given number of cells along the basement membrane, there is a wide scattter of hyperresponsiveness and vice versa (10). Thus, other factors should contribute to the severity of hyperresponsiveness. These factors are most likely the consequences of the airway wall inflammation, which will be discussed below. In COPD, Mullen and colleagues (11) showed that more inflammation of membranous airways was associated with more severe hyperresponsiveness, but no relationship was found between inflammation in the cartilaginous airways. Further studies in COPD are presently lacking.
Fig. 2. Relationship between inflammatory cell numbers in the airway wall and hyperresponsiveness in patients with asthma treated with inhaled corticosteroids. Note the wide scatter of cell numbers for a given degree of hyperresponsiveness. (Reprinted from Reference 10 by permission.)
[More] [Minimize]In 1968, Hogg and coworkers (12) demonstrated that the major site of increased resistance in smokers with airflow limitation was the peripheral airways, suggesting that the increase in resistance was due to inflammatory and structural changes in the peripheral zone. Peripheral airways include small bronchi and bronchioles of less than 2-mm diameter. In COPD, immunohistochemical characterization of the inflammatory cells shows an infiltration of B lymphocytes in the adventitia and of CD8 T lymphocytes in the airway wall (7). An increased smooth muscle thickness in peripheral airways of these subjects has also been observed.
In asthma, eosinophils have been shown to predominate, and especially so in asthmatics with nocturnal symptoms, probably reflecting more severe disease. Hamid and associates (13) found that eosinophils predominated in both central and peripheral airways, but that T lymphocytes were only present in large numbers (above the numbers in healthy controls) in central airways. The presence of enhanced inflammation in the peripheral airways is compatible with the measured increase in peripheral airway resistance in patients with asthma and with the prediction that this is also the site of increased airway responsiveness in patients with asthma (14). The degree to which inflammatory processes in peripheral airways contribute to airway hyperresponsiveness has, however, not formally been established. One can anticipate that the consequences of inflammation are not only present in the central airways.
The changes that occur in the airways in asthma are caused by an influx of inflammatory cells with release of their mediators. This episodic and ongoing allergic inflammation may cause structural changes that affect the severity of airway obstruction after inhaling a bronchoconstrictor stimulus. Results of the inflammatory process in asthma are airway wall edema, deposition and remodeling of connective tissue components, hypertrophy and hyperplasia of tissue cells (for example, smooth muscles), and new vessel formation in the bronchial vasculature. These changes result in airway wall thickening, which may have a profound effect on airway function. This includes changes in thickness of airway wall areas, but also in stiffness of these areas due to biochemical changes of tissue. Structural changes in the airway wall, including extracellular airway remodeling, are prominent features of asthma. For instance, collagen deposition in the subepithelial matrix and hyaluronan and versican deposition around and internal to the smooth muscle can be present in asthma. These depositions can be expected to oppose the effect of smooth muscle contraction.
In COPD, the expiratory flow is influenced by the elastic properties of lung tissue, since elastolytic destruction of lung parenchyma decreases the elastic lung recoil and therefore the driving pressure for flow. Moreover, remodeling of airway wall tissue and airway smooth muscle contributes to the mechanical properties of the airway wall. Luminal secretions may increase resistance in the airways together with increased mucus gland size and goblet cell number and, finally, airway wall inflammation may change airway collapsibility as well. The increased airway collapsibility and loss of lung elasticity are irreversible, whereas the acute inflammatory changes may be open for therapeutic intervention. Acute deposition of proteoglycans and edema formation in submucosal and adventitial tissue may occur in COPD. This is potentially reversible. Chronic inflammation in COPD includes progressive collagen deposition and possibly changes in the balance of different proteoglycans and fibrosis of the airway wall. With the increase of airway wall thickness, mechanical properties change.
The wall of the airway can be divided into three areas (15): (1) the inner wall, consisting of epithelium, basement membrane, lamina reticularis, and loose connective tissue between the lamina reticularis and the airway smooth muscle; (2) the smooth muscle layer in between; and (3) the outer wall, with loose connective tissue between the muscle layer and the surrounding parenchyma (the adventitia).
Figure 3 shows the effect of inner airway wall thickening (14). If the inner airway wall thickens, e.g., occupying 40% of the area below the smooth muscle instead of 20%, this will increase the airway resistance to some extent, in this example, 1.8-fold. However, smooth muscle shortening will affect airway resistance to a larger extent when the inner airway wall is thicker than without airway wall thickening (increase in resistance of 80-fold in this example). This increased airway resistance can be further enhanced when intraluminal airway mucous secretions are present. Figure 4 shows the effect of outer airway wall adventitial thickening. The three layers of the airway wall are depicted. The smooth muscle preload is represented by the springs attached to the outer surface area. It is clear that these springs are loosened when the outer airway wall is thicker and airway smooth muscle contracts (compare left side of lower panel with that of upper panel). The ultimate airway narrowing that occurs is the result of the balance between airway smooth muscle shortening and the opposing force (dilating force) of the surrounding tissue. This elastic load prevents excessive airway narrowing, as can be seen in the upper panel, where resistance increases from 1 to 8. However, when the springs are unloaded (lower panel), this allows a larger smooth muscle contraction before the surrounding parenchyma (represented by the springs) is deformed. The larger airway smooth muscle shortening results in larger airway narrowing, and resistance increases up to 50 times above baseline in this example. The third component is airway smooth muscle thickness. When airway smooth muscle area increases, this will allow the muscle to shorten more against the elastic loads surrounding it, and thus larger airway narrowing occurs.
Fig. 3. The effect of inner wall thickening. In this example, the normal airways have an inner wall area of 20%. When 30% airway smooth muscle (ASM) shortening occurs, the airway resistance (R) increases from an arbitrary value of 1 to 7.6. An increase in wall area to 40% produces a small increase in resistance to 1.8× baseline, but with the thicker airway wall, 30% ASM shortening increases resistance to 80× baseline. (Reprinted from Reference 14 by permission.)
[More] [Minimize]
Fig. 4. The effect of outer airway wall thickening. In this example, the normal airway has a resistance (R) of 1. When airway smooth muscle (ASM) contraction is stimulated, shortening occurs until the stress in the surrounding lung parenchyma (represented by the springs) prevents further shortening and narrowing at a resistance of 8. With adventitial thickening the stress in the surrounding parenchyma is decreased, and greater smooth muscle shortening is possible before the parenchymal recoil prevents further airway narrowing. (Reprinted from Reference 14 by permission.)
[More] [Minimize]Kuwano and coworkers (14) performed morphometric studies on lung tissue of patients with asthma (n = 15), patients with COPD (n = 15), and control subjects (n = 15). They compared the airway size, as measured by the inner airway wall perimeter by assessing the length of the basement membrane (Pbm) and the airway wall compartment areas (WA). WA consists of inner airway area, smooth muscle area, and adventitial area. They found that the relationship between the square root of thickness of the airway wall and Pbm was different in asthma and COPD. Whereas patients with asthma had a significant relationship of Pbm with all three areas of the airway wall, i.e., adventitia (r = 0.137), smooth muscle area (r = 0.063), and submucosa (r = 0.088), in COPD there existed only a significant relationship with airway smooth muscle area (r = 0.035). This does not imply that the other compartments do not affect changes in airway obstruction in COPD. As mentioned above, the changes in the quality of the connective tissue may change the effect of an obstructive stimulus as well.
Hamid and associates (13) were the first to assess inflammation in the inner and outer airway wall in asthmatic individuals. They compared the results with those of nonasthmatic control subjects. There were a larger number of eosinophils in the outer airway wall area than in the inner area in patients with asthma. However, findings on activated eosinophils were opposite, suggesting that tissue damage will occur largely in the inner submucosal airway wall area.
Hyperresponsiveness can be assessed with a stimulus directly acting on the smooth muscle and a stimulus that induces smooth muscle contraction in an indirect way. The latter stimulus may activate either inflammatory cells or neural pathways, thereby enhancing airway obstruction. Methacholine and probably histamine are stimuli that directly act on the smooth muscle; cold air and adenosine-5-monophosphate (AMP) are indirectly acting stimuli. Table 2 shows the prevalence of hyperresponsiveness in asthma and COPD with the different stimuli used, as derived from several publications (16-18). In asthma and COPD many patients are hyperresponsive to inhalation of histamine and methacholine. Furthermore, patients with COPD are more responsive to histamine as compared with methacholine, whereas patients with asthma are comparably responsive to these two stimuli. Only 10% of the population with COPD responds to hyperventilation of cold air compared with 96% of the population with asthma. The correlation of the response to methacholine and hyperventilation of cold air is very good in the group with asthma but nonexistent in COPD. The prevalence of hyperresponsiveness to methacholine, histamine, propranolol, SO2, and isocapnic hyperventilation of cold air is larger in asthma than in COPD, whereas increased responsiveness to acetylcholine and fog predominate in COPD. These observations make it difficult to postulate a single mechanism responsible for the hyperresponsive phenomenon. The above discussion of the pathologic findings in asthma and COPD corroborate this. Even though different mechanisms are likely playing a role in hyperresponsiveness in asthma and COPD, it is unknown to what extent the different stimuli are related to the different aspects of the pathology of asthma and COPD.
| Stimulus | Asthma | COPD | ||
|---|---|---|---|---|
| Acetylcholine | 75 | 64 | ||
| Methacholine | 80 | 70 | ||
| Histamine | 82 | 36 | ||
| Propranolol | 67 | 21 | ||
| SO2 | 95 | 30 | ||
| AMP | 90 | 90/39* | ||
| Hyperventilation† | 96 | 11 | ||
| Fog | 30 | 81 |
One stimulus in Table 2 deserves a little more attention. AMP-induced bronchoconstriction can be found in both individuals with asthma and COPD (17, 18). Figure 5 shows results of AMP provocation tests in groups of patients with asthma and COPD with comparable levels of methacholine responsiveness (17, 18). Despite the same response to methacholine, clear differences exist in response to AMP. Patients with COPD who smoke are more responsive to AMP than nonsmokers. This opens speculations as to the mechanism of AMP-induced bronchoconstriction. It is clear that inhaled corticosteroids dampen the inflammatory process in asthma. This improvement reflects itself in a beneficial effect on airway hyperresponsiveness. Some studies have shown that effects are larger when the stimulus is an indirect one, like AMP, than with methacholine (19). AMP exerts its effects mainly via stimulation of mast cells. This has been supported by the observation that the selective H1 receptor antagonist terfenadine inhibits AMP-induced bronchoconstriction up to 80% in asthma. Furthermore, histamine levels in bronchoalveolar lavage fluid increase upon adenosine provocation. The role of the mast cell in the inflammatory process and direct antigen recognition in asthma has now been well established. Therefore, the enhanced effect of inhaled corticosteroids upon adenosine challenge versus methacholine challenge indicates involvement of these cells in airway hyperresponsiveness.
Fig. 5. Levels of AMP responsiveness in smokers and nonsmokers with COPD and subjects with asthma. The groups had similar methacholine responsiveness. The control group was unresponsive to methacholine and did not have airflow limitation. (Modified from References 17 and 18 by permission.)
[More] [Minimize]Studies of COPD have shown that short-term use of inhaled corticosteroids does not affect hyperresponsiveness as assessed with histamine or methacholine. Our group has investigated the effect of inhaled corticosteroids (1,600 μg budesonide/d for 6 wk) in nonallergic individuals with COPD. There was neither an effect on methacholine nor on AMP (20). Furthermore, terfenadine protected AMP challenge significantly. This suggests either that mast cells are not very much activated in COPD even though hyperresponsiveness to AMP is increased, which might explain the lack of response to corticosteroids, or that mast cell activation is not affected by inhaled corticosteroids in COPD, whereas it is in asthma.
There is a need for better understanding of the pathophysiology of hyperresponsiveness, since it reflects the risk for development of asthma and COPD as well as the progression of these diseases. Inflammatory processes affecting the airway wall both in peripheral and central areas of the lung appear to be important. However, it is not clear which structural changes are open for therapy and which are not. Therefore, a better understanding of the consequences of inflammation for lung tissue and airway wall changes in asthma and COPD has to evolve before a full understanding of airway hyperresponsiveness will emanate.
| 1. | Xu X., Rijcken B., Schouten J. P., Weiss S. T.Airways responsiveness and development and remission of chronic respiratory symptoms in adults. Lancet350199714311434 |
| 2. | Peat J. K., Woolcock A. J., Cullen K.Rate of decline of lung function in subjects with asthma. Eur. J. Respir. Dis.701987171179 |
| 3. | Kerstjens H. A. M., Overbeek S. E., Schouten J. P., Brand P. L. P., Postma D. S.the Dutch CNSLD Study GroupAirways hyperresponsiveness, bronchodilator response, allergy and smoking predict improvement in FEV1 during long-term inhaled corticosteroid treatment. Eur. Respir. J.61993868876 |
| 4. | Tashkin D. P., Altose M. D., Bleecker E. R., Connett J. E., Kanner R. E., Lee W. W., Wise R. A.for the Lung Health Study Research GroupThe lung health study: airway responsiveness to inhaled methacholine in smokers with mild to moderate airflow limitation. Am. Rev. Respir. Dis.1451992301310 |
| 5. | Tashkin D. P., Altose M. D., Connett J. E., Kanner R. E., Lee W. W., Wise R. A.for the Lung Health Study Research GroupMethacholine reactivity predicts changes in lung function over time in smokers with early chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med.153199618021811 |
| 6. | Brand P. L. P., Postma D. S., Kerstjens H. A. M., Koëter G. H.the Dutch CNSLD Study GroupRelationship of airway hyperresponsiveness to respiratory symptoms and diurnal peak flow variation in patients with obstructive lung disease. Am. Rev. Respir. Dis.1431991916921 |
| 7. | Jeffery, P. K. 1997. Pathology of asthma and COPD: a synopsis. Eur. Respir. Rev. 7:43, 111–118. |
| 8. | Saetta, M. 1997. Central airways inflammation in the development of COPD. Eur. Respir. Rev. 7:43, 109–110. |
| 9. | O'Shaughnessy T. C., Ansari T. W., Barnes N. C., Jeffery P. K.Inflammation in bronchial biopsies of subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocytes with FEV1. Am. J. Respir. Crit. Care Med.1551997852857 |
| 10. | Sont J. K., Van Krieken H. J. M., Evertse C. E., Hooijer R., Willems L. N. A., Sterk P. J.Relationship between the inflammatory infiltrate in bronchial biopsies and clinical severity of asthma in patients treated with inhaled steroids. Thorax511996496502 |
| 11. | Mullen J. B. M., Wiggs B. R., Wright J. L., Hogg J. C., Pare P. D.Nonspecific airway reactivity in cigarette smokers: relationship to airway pathology and baseline lung function. Am. Rev. Respir. Dis.11986120125 |
| 12. | Hogg J. C., Macklem P. T., Thurlbeck W. M.Site and nature of airway obstruction in chronic obstructive lung disease. N. Engl. J. Med.278196813551360 |
| 13. | Hamid Q., Song Y., Kotsimbos T. C., Minshall E., Bai T. R., Hegele R. G., Hogg J. C.Inflammation of small airways in asthma. J. Allergy Clin. Immunol.10019974451 |
| 14. | Kuwano K., Bosken C., Pare P. D., Bai T., Wiggs B., Hogg J. C.Small airway dimensions in asthma and chronic obstructive pulmonary disease. Am. Rev. Respir. Dis.148199312201225 |
| 15. | Pare, P. D., and T. R. Bai. 1996. Airway wall remodelling in chronic obstructive pulmonary diseases. Eur. Respir. Rev. 6:39, 259–263. |
| 16. | Postma D. S., Koëter G. H., De Vries K.Clinical expression of airway hyperreactivity in adults. Clin. Rev. Allergy71989321343 |
| 17. | Oosterhoff Y., Jansen M. A. M., Postma D. S., Koëter G. H.Airway responsiveness to adenosine-5′ monophosphate in smokers and nonsmokers with atopic asthma. J. Allergy Clin. Immunol.921993773776 |
| 18. | Oosterhoff Y., De Jong J. W., Jansen M. A. M., Koëter G. H., Postma D. S.Airway responsiveness to adenosine-5′ monophosphate in chronic obstructive pulmonary disease is determined by smoking. Am. Rev. Respir. Dis.1471993553558 |
| 19. | Weersink E. J. M., Douma W. R., Postma D. S., Koëter G. H.Fluticasone propionate, salmeterol xinafoate, and their combination in the treatment of nocturnal asthma. Am. J. Respir. Crit. Care Med.155199712411246 |
| 20. | Rutgers S. R., Koëter G. H., Van der Mark Th. W., Postma D. S.Short term treatment with budesonide does not improve hyperresponsiveness to adenosine-5′-monophosphate in COPD. Am. J. Respir. Crit. Care Med.1571998880886 |
