Airway Responsiveness to Indirect Challenges in COPD

ABSTRACT

Patients with chronic obstructive pulmonary disease (COPD) demonstrate airway hyperresponsiveness to a number of indirect stimuli. Hyperresponsiveness to cold air hyperventilation, exercise, and drugs like propranalol and methoxamine seem to be able to distinguish patients with COPD from those with asthma, whereas hyperresponsiveness to stimuli like adenosine 5-monophosphate (AMP) and hypertonic saline seem unable to do so. The relationship of airway responsiveness to indirect stimuli and airway inflammation has received little study. The clinical relevance of hyperresponsiveness to an indirect challenge, including the impact on the natural history, relation to types of bronchitis, baseline airway calibre, and response to treatment need to be studied.

Keywords: :

• COPD
• Indirect challenge
• Airway hyperresponsiveness

INTRODUCTION

Non-specific airway responsiveness refers to the ease with which airways narrow in response to different non-allergic or non-sensitizing stimuli. Airway hyperresponsiveness represents an important pathophysiological feature of asthma and it also exists to some extent in COPD (1). Stimuli such as methacholine and histamine act directly on the smooth muscle, whereas other stimuli like isocapnic hyperventilation, exercise, propranalol, methoxamine, hypertonic saline and adenosine monophosphate act by indirect mechanisms (2). Among the two different categories of airway challenges, the indirect one has the capacity to elicit multiple cellular responses that could provide a better reflection of airway inflammation and a useful tool to evaluate response to treatment. It might also be better in distinguishing asthma from COPD.

The severity of airway responsiveness to a direct stimulus like methacholine is a risk factor for progressive airflow obstruction in smokers (3) and a predictor of a rapid decline in FEV₁ in epidemiologic studies (4). Patients with COPD who have increased airway hyperresponsiveness to histamine (defined as a provocative concentration of histamine less than 8 mg/ml to cause a 10% fall in FEV₁) are more likely to develop chronic respiratory symptoms of cough, sputum and wheeze when compared to patients who do not have airway hyperresponsiveness (5), particularly when associated with peripheral blood eosinophilia (6).

The relevance of airway hyperresponsiveness to indirect stimuli is not well known. Only a few studies have investigated the response to indirect challenges in COPD. The interpretation of their results is confounded by the imprecise definitions of diseases, heterogeneity of the pathologic processes that lead to COPD and by the fact that markers of airway inflammation (which is so important in their pathogenesis) have usually not been measured. In this review, we will discuss the mechanisms of indirect airway challenges, the role of indirect challenges in discriminating between COPD and asthma, what has been learned from the application of indirect challenge tests in evaluating the response to treatment of COPD and indicate what needs to be done in the future to clarify their place in the diagnosis, prognosis and monitoring of the effects of treatment.

Mechanisms of indirect airway challenges

Indirect stimuli act through intermediate mechanisms that activate several cells to release a wide array of mediators and neurotransmitters that induce smooth muscle contraction. These cells include inflammatory cells, epithelial cells, nerve cells and vascular smooth muscle cells (Figure 1).

Figure 1.  Mechanisms of indirect and direct airway challenge.

AMP

Inhalation of AMP can elicit bronchoconstriction in susceptible subjects after conversion to adenosine by the enzyme 5-nucleotidase. Adenosine is a nucleotide that exerts its action through interaction with specific adenosine receptors with subsequent release of inflammatory mediators from mast cells (7). Four different adenosine receptors; A₁, A_2A, A_2B, A₃ are known (8). Human lung mast cells have been shown to express A_2A, and A_2B. The effect of adenosine on mediator release seems to be independent of cAMP protein kinase, because treatment of mast cells with KT5720, which blocks protein kinase activity, does not inhibit this pathway. It has been suggested that activation of phospholipase C that increases intracellular calcium levels may mediate the effects of adenosine (9). Adenosine can also act on A₁, A_2A and A₃ receptors on neutrophils, eosinophils and macrophages that produce different pro- and anti-inflammatory actions (7).

Propranolol

The mechanism of β-blocker-induced airflow limitation in susceptible patients is not fully understood but may be in part due to B receptors blockade. Airflow limitation is induced by the L-isomer of propranolol, in contrast to the D-isomer, which does not have significant β -receptor blocking activity (10). Propranolol inhalation results in the blockade of β-adrenoceptors in the smooth muscle thus removing a sympathetic bronchodilator drive to the airways (11). There is no clear evidence that mast cells are involved in propranolol induced bronchoconstriction (12, 13). Cysteinyl leukotrienes also have no role, as pre-treatment with cys LT1 receptor antagonist did not affect propranolol-induced bronchoconstriction (14).

Methoxamine

Methoxamine is an α-adrenoceptor stimulus that produce airway bronchoconstriction in asthmatic subjects by stimulation of α-adrenceptors. This has been supported by the finding that its bronchoconstrictor effect can be inhibited by administration of Prazosin prior to methoxamine challenge (15). Bronchoconstriction induced in asthmatic subjects by methoxamine can also be inhibited by disodium cromoglycate (16), suggesting that α adrenoceptors may be present on parasympathetic ganglia, stimulation of which leads to increased cholinergic traffic and bronchoconstrictor response. It is also possible that α adrenoceptors may be situated on mast cells, and their stimulation leads to mediators release (16).

Acetaldehyde

The mechanisms by which acetaldehyde cause bronchoconstriction are not well identified. It has been suggested that mast cells or basophils play a role to induce bronchoconstriction in asthmatic subjects by histamine release (17).

Hypertonic saline and mannitol

Hypertonic saline challenge has been used to identify subjects with exercise-induced bronchoconstriction and subjects with active asthma. It acts by increasing airway surface osmolarity, triggering the release of inflammatory mediators from inflammatory cells causing smooth muscle contraction (18). Hypertonic saline challenge has effectiveness similar to exercise and eucapnic hyperventilation. One advantage is that the bronchoconstriction occurs during the challenge in contrast to exercise and eucapnic hyperventilation where it occurs after the test has ceased (19,20). It has an additional advantage as it enables induced sputum collection to identify inflammatory cellular markers.

Mannitol is a dry powder given by inhalation. It acts indirectly by stimulating airway smooth muscle contraction subsequent to release of mast cells mediators such as leukotriene E4 and the prostaglandin D2 (21). Airway hyperresponsiveness to mannitol is more reflective to airway inflammation in asthmatic subjects than methacholine challenge. It is correlated with airway inflammatory subtype in asthmatic subjects (22). The mannitol challenge test has several advantages over other indirect challenge tests. First, inhaling mannitol overcame the problem of need to perform vigorous exercise breathing dry air. Second, there is no need for solutions with different concentrations and nebulizers as it is available as a convenient and standardized test kit with prefilled dry powder capsules delivered in progressively increasing doses using a simple available and single-use dry powder inhaler device.

Exercise and eucapneic hyperventilation

Exercise-induced bronchoconstriction (EIB) is the transient airway narrowing associated with exercise (23). It is the consequence of evaporative water loss caused by conditioning of the inspired air. The surface of the small airways becomes hyperosmolar as a result of water loss (24). Subsequent release of mediators such as prostaglandins, histamine and leukotrienes provokes airway smooth muscle constriction (25).

Airway inflammation could also play a role in exercise-induced bronchoconstriction. Several recent studies have correlated exercise induced bronchoconstriction (EIB) with eosinophils (26) and eosinophils cationic protein (ECP). It also has been proposed that epithelial injury and increase in airway microvascular permeability may play a key role in EIB (27, 28).

The eucapnic voluntary hyperventilation provocation test is a challenge based on the hypothesis that an increased ventilation rate can cause dehydration of the smaller airways which results in hyperosmolarity resulting in release of inflammatory mediators, similar to what occurs during exercise (29). The general procedure involves breathing a mixture of dry compressed gas (5.0% CO₂, 21.0% O₂, balance N₂) at a predetermined rate of 85% of maximum voluntary ventilation (MVV) per minute (calculated as 30 times baseline FEV₁, which approximates 85% MVV) for 6 minutes. Spirometry is performed at 3,5,10, 15, and 20 minutes post-EVH challenge. A decrease in the FEV1 of at least 10% after 6 min hyperpnoea in dry air is considered as a positive test (30).

Indirect challenges to discriminate COPD and asthma

Airway responsiveness is determined by a number of factors including the magnitude of airflow limitation and the degree of inflammation in the airways. Therefore, it is not surprising that airway hyperresponsivness also occurs in about two thirds of smokers with COPD (31). Table 1 shows prevalence of airway hyperresponsiveness to direct and indirect stimuli in asthma and COPD. Part of the problem in discriminating between COPD and asthma relates to the definition of these conditions. COPD and asthma are both a part of airway diseases, which are composed of chronic airflow limitation, variable airflow limitation, bronchitis (airway inflammation) and airway hyperresponsiveness (Figure 2) (32). They can occur in different combinations that are not always correlated together. Asthma, by definition, is reversible and variable airflow limitation. COPD is rather an imprecise term that includes chronic bronchitis, chronic bronchiolitis and emphysema all of which lead to fixed airflow limitation and slow forced emptying of the lung. Chronic airflow limitation is the physiological component and bronchitis is the inflammatory component of the disease. Airway inflammation is an important component that has various causes, patterns and responses to treatment. It causes symptoms, variable and chronic airflow limitation by releasing of various mediators and stimulating remodelling (32). The chronic airflow limitation in COPD may be a result of different aetiologies and pathologies. For example, it may be due to smoker's bronchitis, bronchiectasis, or bronchiolitis. The presence or absence of atopy may be another relevant factor. Although, in contrast to asthma, atopy is not frequently present in patients with COPD, it may well be present. Although not formally tested, atopy may affect the severity of airway responsiveness similar to patients with asthma (33). This heterogeneity of pathology (Table 2) may be relevant in the interpretation of results of studies of airway responsiveness.

Figure 2.  Components of airway diseases and their relationship to one another.

Table 1.  Prevalence of hyperresponsiveness to direct and indirect stimuli in 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%

*90% in smokers, 39% in nonsmokers.

Postma DS, Kerstjens HAM. Characteristics of airway hyperresponsiveness in asthma and chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998;158:S187–S192.

Table 2.  Some studies which investigated indirect AHR in different disease heterogeneity

	Study	Clinical characteristics	Nature of CAL	N	Indirect challenge
1	Ramsdale et al., Thorax 1984	Age 26–79, six atopic	Smoker's bronchitis	27	Methacholine, Isocapnic hyperventilation of cold air
2	Du Toit et al., ARRD 1986	Age 40–81, atopy not described	Smoker's bronchitis	18	Histamine, methacholine, methoxamine
3	Godfrey et al., Thorax 1991	52 – pediatric asthma	Cystic fibrosis, bronchiolitis obliterans, primary ciliary dyskinesia, bronchiectasis	93	Methacholine, exercise
		22 – pediatric COPD
		19 – control children (age 6–25)
4	Woolcock et al., ARRD 1991	7 CAL nonatopic	Smoker's bronchitis	30	Propranalol
		13 Asthma atopic
		10 normal nonatopic
5	Woolcock et al., ARRD 1991	Nonatopic	Smoker's bronchitis	14	4.5% NaCl
6	Oosterhoff et al., ARRD, 1993	30 COPD: > 40 yrs, nonatopic	19 smokers with COPD	58	Methacholine and AMP
		28 Asthma 18–45 yr, atopic	11 non- or ex smokers with COPD
7	Avital et al., Thorax 1995	51 – pediatric asthma	Cystic fibrosis, bronchiolitis obliterans, primary ciliary dyskinesia, bronchiectasis	91	Methacholine, adenosine, exercise
		21– pediatric COPD
		19 – control children (age 6–25)
8	Rutgers et al., AJRCCM 1998	Age 45–75, nonatopic	Smoker's bronchitis	44	Methacholine, AMP
9	Rutgers et al., Clin Exp Allergy 1999	Age 45–75, nonatopic	Smoker's bronchitis	44	Methacholine, AMP
10	Rutgers et al., Clin Exp Allergy 2000	18 COPD Age > 50 yr, nonatopic	Smoker's bronchitis	29	AMP
		11 normal, nonatopic
11	Leuppi et al., Pulm Pharmacol Ther 2005.	COPD, Age 44–83	Smoker's bronchitis, emphysema	30	Mannitol

Ramsdale and co-workers examined the effects of methacholine and hyperventilation of cold dry air in 27 smokers with chronic bronchitis (with and without COPD) (34) and in 27 subjects with asthma with similar FEV₁ values (35). The degree of responsiveness to methacholine correlated inversely with the FEV₁ whether expressed as% predicted or the% VC in both the chronic bronchitic and asthmatic groups. However, hyperventilation caused bronchoconstriction in all asthmatics except one, but none in those with chronic bronchitis except for three. Of these exceptions, it is possible that the one asthmatic had associated COPD and that one of the three with chronic bronchitis also had atopy and asthma. These observations suggest that the increased responsiveness to methacholine is primarily due to the airflow obstruction and that the lack of response to hyperventilation reflects the absence of true airway hyperresponsiveness.

Godfrey and co-workers (36) compared the airway responsiveness to methacholine and exercise in children and younger adults who were healthy, or had asthma, or had chronic lung disease. The latter group had a mixture of conditions which included cystic fibrosis, bronchiectasis, immotile cilia and recurrent pneumonia, which may have been associated with COPD. When the mean minus two standard deviation of methacholine responsiveness in the healthy subjects was taken as the lower limit of normal, most of the asthmatic subjects and most of those with chronic lung disease were hyperresponsive to methacholine. In contrast, when the mean plus two standard deviation of the exercise response (% fall in FEV₁) was taken as the upper limit of normal, only those with asthma but not of those with chronic lung disease were hyperresponsive. In a similar group of subjects, Avital and colleagues demonstrated that a 5% fall in FEV₁ following an exercise study had 85% sensitivity and specificity in distinguishing asthmatic subjects from those with COPD (37). It is quite possible that a non-smoking child with COPD may have a different patho-physiological process in the airway compared to an older smoking adult.

Woolcock and co-workers (38) examined the effect of propranolol in healthy subjects and in those with asthma or COPD. Most of the asthmatic subjects were hyperresponsive to propranalol while only 2 of the 7 with COPD were hyperresponsive, supporting better specificity of this test for asthma.

Du Toit and co-workers (39) examined the effects of histamine, methacholine and methoxamine in subjects with COPD. While the subjects were hyperresponsive to histamine and methacholine, only 1 of 18 was hyperresponsive to methoxamine.

Acetaldehyde challenge could be used to differentiate between asthma and COPD. SaÂnchez-Toril et al. reported that 92% of asthmatic subjects developed bronchoconstriction in response to acetaldehyde compared to 5% of subjects with chronic bronchitis. bronchoprovocation with acetaldehyde was more (95% specificity) than bronchoprovocation with methacholine (24% specificity) (40).

These observations suggest that hyperresponsiveness to cold air hyperventilation, exercise, propranalol, acetaldehyde and methoxamine are more commonly associated with asthma than with COPD.

One indirect stimulus that does not seem to discriminate between asthma and COPD is hypertonic saline. Woolcock and co-workers (38) studied responsiveness to hypertonic saline (4.5% NaCl) in 18 subjects with chronic airflow limitation due to smoker's bronchitis and another 18 with asthma who had the same prechallenge values for FEV₁% predicted. The majority of the subjects in both groups responded to the stimulus and it was not possible to distinguish asthmatics from subjects with COPD on the basis of their PD₂₀ 4.5% NaCl values. However, when the dose-response curves of the subjects with COPD and a PD₂₀ 4.5% NaCl of 5 ml or less were compared with those of asthmatics, they had much lower slopes. Although patients with COPD responded with a reduction in FEV₁ with each dose, the change was gradual. For the asthmatic subjects a threshold was apparent above which the airways rapidly narrowed. In this particular group of asthmatic subjects, the median value for PD₂₀ was 3.0 ml, while it was 3.7 ml in the subjects with COPD. However, when an analysis of a larger number of asthmatic subjects was carried out, the median PD₂₀ value for 4.5% NaCl was 2 ml compared with a median value of 11 ml in the 25 subjects with COPD. It is therefore possible that a lower cut point of PD₂₀ may be able to distinguish those patients who have additional asthma. There was no significant relationship between pre-challenge FEV₁% predicted and PD₂₀ 4.5% NaCl.

The relationship between AMP responsiveness and chronic airflow limitation due to smoker's bronchitis is more complex. There is controversy regards using provocation with AMP to differentiate between asthma and COPD. Oosteroff et al. (41) examined the effect of inhaled methacholine and inhaled AMP in current smokers with COPD, ex- or non-smokers with COPD, a group of asthmatic non-smokers and a control group of smokers without airflow limitation. The three groups with disease had a similar degree of responsiveness to methacholine. The responsiveness to AMP was also significantly increased in the three diseased groups but the COPD group of non-smokers or ex-smokers was significantly less responsive than the COPD smoking group and the asthma non-smoking group in whom the AMP responsiveness was not significantly different.

These results suggest that AMP responsiveness cannot be used to differentiate COPD from asthma. They also indicate that current smoking is a determinant of increased responsiveness to AMP. However, smoking does not seem to increase the sensitivity to AMP in young subjects with asthma (42). In a group of 16 non-smoking asthmatic subjects, 12 smoking atopic asthmatic subjects and 5 non atopic healthy smokers, the same group of investigators demonstrated a significantly higher sensitivity to AMP provocation in nonsmokers with asthma as compared with smokers with asthma. The baseline FEV₁, the sensitivity to methacholine provocation and the atopic status were comparable between the groups.

In contrast to these studies, Avital and colleagues (37) demonstrated that an AMP responsiveness of 25 mg/ml had 90% sensitivity and specificity in differentiating patients with asthma and COPD. This may be a reflection of the selection of the patient population and the causes of chronic airflow limitation, which varied from cystic fibrosis and bronchiectasis to bronchiolitis obliterans. All the patients in this study were children who had never smoked. Atopic status was not reported.

A recent study tried to investigate the effect of smoking cessation on direct and indirect challenge in smoker subjects with allergic rhinitis with or without asthma. They reported more improvement in AHR to AMP than methacholine after 6 months of smoking cessation. However, improvement in AHR to AMP and methacholine were comparable after 12 months of cessation. This suggests that responsiveness of the airways to AMP could be used to detect early changes in AHR during smoking cessation (43).

Although further studies are needed to clearly identify the indications for clinical use of AMP bronchoprovocation and a standardised population based cut off PC20 AMP value needs to be determined, there is an increasing body of evidence to support that it could be used to differentiate between COPD and asthma when clinical diagnosis is uncertain. (44).

Indirect challenge to assess response to treatment

Unlike patients with asthma, who show significant improvement in methacholine airway responsiveness following short-term (45) and long-term treatment with inhaled or ingested corticosteroids (46), patients with COPD show little or no improvement in airway responsiveness to a direct stimulus like histamine (47) or methacholine (48). There is little information on the effects of treatment, particularly with corticosteroids, on responsiveness to indirect stimuli in patients with COPD. Rutgers et al (49) evaluated the effects of budesonide 1600 mcg a day for 6 weeks in 54 patients with COPD, on responsiveness to AMP and methacholine challenges.

Compared to placebo treatment, budesonide treatment did not significantly change PC₂₀ AMP or PC₂₀ methacholine. Therefore it appears that assessment of responsiveness to indirect stimulus does not provide additional information to that provided by responsiveness to histamine or methacholine following short-term treatment with inhaled corticosteroids. Effects of long-term treatment with inhaled corticosteroids on airway responsiveness to AMP or other indirect stimuli have not been studied. Several studies have shown that inhaled corticosteroid could attenuate airway response to AMP to a greater extent than Methacholine in asthma both after 2 hours of single dose and after regular treatment. This important observation could be exploited in discrimination between asthma and COPD (50–52).

In contrast to methacholine and AMP responsiveness, Leuppi and colleagues have reported responsiveness to mannitol as a predictor to response to treatment with inhaled corticosteroids in 30 patients with COPD (53). All patients had post bronchodilator FEV1/forced vital capacity ratio of less than 70% and a reversibility of less than 12% and 200 ml from baseline. After 3 months of treatment with ICS, FEV₁% predicted improved from 67% to 79% in mannitol positive patients; whereas it was unchanged in the mannitol negative patients.

Airway inflammation and responsiveness to indirect challenge

Airway inflammation plays an important part in the pathogenesis of COPD. Bronchial mucosal biopsy studies (54), and more recently measurements in induced sputum (55), have demonstrated an inflammatory pattern with predominantly macrophages, neutrophils and T-lymphocytes. Mast cells have been found in increased numbers in mucous glands and therefore they might have a role in airway inflammation in COPD (56). One third of patients with chronic airflow limitation due to smoker's bronchitis may have an eosinophilic bronchitis (55–57). It is important to know the nature of the airway inflammation because the natural history of the disease and the response to treatment may depend on the predominant cell type. For example, a neutrophilic bronchitis was associated with a more severe airflow limitation in a cross-sectional study (60, 58) and a more rapid decline in lung function in a retrospective cohort study in which sputum was examined at the end of 15 years of follow-up (59). Eosinophilic bronchitis is steroid-responsive while non-eosinophilic bronchitis may not be (55, 60) There is very little information regarding the association between the nature of airway inflammation in COPD and airway responsiveness to either direct or indirect challenge.

Airway hyperresponsiveness to indirect stimuli might represent the extent of the inflammatory process more closely than direct stimuli. In asthma Green et al. have shown that there was 0.4 doubling dose reduction in PC ₂₀ to methacholine with 1.6-fold reduction in sputum eosinophils after treatment with budesonide (800ug /day) for 4 weeks (61). 4 weeks of treatment with 160 ug ciclesonide increased the mean values of PC20 to AMP by 7-fold and that was associated with reduction in eosinophils in induced sputum (62). Van den Berge and colleagues reported that PC20 AMP was closely associated with eosinophils percentage and eosinophilic cationic protein (ECP) in sputum and peripheral blood of asthmatic patients (63).

In COPD, it has been shown that hyperresponsiveness to AMP is correlated with eosinophilic inflammation. Rutgers et al. (64) studied AMP responsiveness in a group of non-atopic patients with COPD who were either ex-smokers or non-smokers and examined the relationship of hyperresponsiveness to sputum, bronchoalveolar lavage and bronchial biopsy inflammatory variables. They observed that most of the 12 subjects with hyperresponsiveness to AMP had an increase in sputum eosinophils and CD-8 positive cells per mm² in biopsies compared with the results in 6 and 5 subjects respectively who had no response to AMP. Such studies may be useful in understanding the differences in responses in patients with asthma and COPD. Perhaps, COPD patients with an eosinophilic bronchitis may show hyperresponsiveness to exercise, cold air hyperventilation and propranalol inhalation. In another study, the number of T lymphocytes in the airway walls was correlated with AHR in subjects with centrilobular emphysema (65).

CLINICAL RELEVANCE

A reliable indirect provocation test is likely to have two important clinically relevant applications. First, since baseline airflow obstruction is likely to determine bronchoconstrictive response to direct stimuli such as methacholine and histamine, it would be helpful to have a provocation test that is not influenced by airway caliber to identify true airway hyperresponsiveness. Second, since indirect provocation tests seem to correlate more with the presence of steroid responsive airway inflammation than direct provocation, these tests are likely to identify COPD patients who could benefit from inhaled corticosteroids.

Summary

Airway hyperresponsiveness to various indirect stimuli is frequently observed in patients with COPD. While some stimuli like cold air hyperventilation, exercise and propranalol are able to discriminate patients with COPD from those with asthma, other stimuli such as AMP and hypertonic saline are less able to do so. This may partly be due to the demographic and clinical characteristics of the patient populations studied which include smoking habits, atopy, nature of the chronic airflow limitation and bronchodilator and corticosteroid reversibility. Since indirect challenge depends on the presence of inflammatory cells in sufficient numbers to release their inflammatory mediators in response to the stimuli, it is considered to be more associated with airway inflammation and gives complementary information to direct tests. A better understanding of the clinical significance of indirect airway hyperresponsiveness requires further studies examining its relationship with measurements of airway inflammation in bronchial biopsies or by less invasive methods like induced sputum.

ACKNOWLEDGEMENT

Dr Nair is supported by a Canada Research Chair in Airway Inflammometry. Dr Hassan is funded by an Egyptian Ministry of Higher Education Fellowship.

Declaration of interests

The authors have no conflicts of interest to disclose. The authors alone are responsible for the content and writing of the paper.