INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial hyperresponsiveness (BHR) and airway inflammation are both characteristic features of asthma. Histamine and methacholine, stimuli that mainly act directly on airway smooth muscle, are generally used to measure the presence and severity of BHR. Although the presence and severity of BHR are related to smooth muscle contraction, airway inflammation also contributes to BHR. Therefore indirect stimuli that exert their effect on inflammatory cells, subsequently leading to smooth muscle contraction and edema, may provide additional information. Adenosine 5'-monophosphate (AMP) is such an indirect stimulus, as it has little effect on airway smooth muscle contraction in vitro (1). A major action of AMP appears to involve the release of histamine and other preformed mediators from "primed" mast cells, as AMP-induced bronchoconstriction is associated with a rise of histamine in plasma and bronchoalveolar lavage fluid (2, 3). Moreover, the concentration of AMP causing FEV₁ to drop by 20% (PC₂₀) is inhibited up to 80% by pretreatment with antihistamines (4). AMP may have some additional actions on neural pathways as the airway response is partially attenuated by atropine and ipratropium bromide (5, 6).

Previous studies have shown that PC₂₀ AMP improves to a larger extent with the use of inhaled corticosteroids than PC₂₀ methacholine (7). Furthermore, in a study by Aalbers and colleagues the improvement of BHR after a stay of 1 mo in a hypoallergenic environment in Switzerland could be detected with AMP but not with methacholine (10). These findings suggest that PC₂₀ AMP is more closely associated with airway inflammation in patients with asthma than PC₂₀ methacholine. However, no formal study has investigated this hypothesis so far in the sense of actually collecting data on inflammation in sputum or airway wall biopsy.

The aim of the present study was to investigate whether PC₂₀ AMP is more closely associated with airway inflammation compared with PC₂₀ methacholine. To this end, we tapered down inhaled corticosteroids in a large group of 120 subjects with asthma. Thereafter, we assessed the relationship between PC₂₀ methacholine as well as PC₂₀ AMP with FEV₁ %predicted (pred) and inflammatory markers in sputum, blood, and exhaled air.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Patients with a diagnosis of asthma, 18-65 yr old, were included if they met the following criteria: PC₂₀ methacholine  8 mg/ml, at least one positive skin prick test out of 17 common aeroallergens, reversibility to ₂-agonist  9% of the predicted FEV₁, and ability to expectorate sputum after hypertonic saline inhalation.

Study Design

Inhaled corticosteroids were tapered and when possible discontinued completely in 3 wk. Patients were asked to visit the hospital on two consecutive days when they had discontinued their inhaled corticosteroids completely for 3 wk, or earlier, if they experienced (subjective) symptoms of a pending asthma exacerbation for which they felt that treatment with corticosteroids was desirable. On the first day lung function, exhaled nitric oxide (NO), blood sampling, and sputum induction were performed. On the second day bronchial hyperresponsiveness was measured by methacholine challenge and after 1 h a second challenge with AMP was done. The study protocol was approved by the local medical ethics committee and all participants gave their written informed consent.

Exhaled Nitric Oxide

Exhaled nitric oxide (NO) was measured by tidal breathing as previously described by Alving and coworkers (11).

Lung Function

FEV₁ was measured with a calibrated water-sealed spirometer according to standardized guidelines as described previously (7). PC₂₀ methacholine followed by PC₂₀ AMP 1 h later were performed with a 2-min tidal breathing method. Adenosine and methacholine were prepared in 0.9% saline to produce a range of concentrations from 0.04 to 320 mg /ml for adenosine and 0.038 to 8 mg/ml for methacholine.

Sputum Induction and Sputum Processing

Sputum was induced by inhalation of hypertonic saline aerosols as previously described (7). Fifteen minutes after salbutamol (200 µg) inhalation, hypertonic saline (3%, 4%, and 5%) was nebulized for each concentration during 7 min. Whole samples were processed according to the method of Fahy and coworkers with some modifications (7, 12).

Biochemical Assays

The concentrations of eosinophilic cationic protein (ECP) in serum and sputum were measured using a fluoroenzyme assay, the ImmunoCAP ECP (provided by Pharmacia, Uppsala, Sweden). The concentration of interleukin 8 (IL-8) in sputum was measured using an available ELISA kit (CLB, Amsterdam).

Statistical Methods

All calculations of PC₂₀ were performed with the base-2 logarithm (log₂) as this reflects doubling concentrations and normalizes the distribution. Patients already responding to saline were assigned a PC₂₀ value half of the lowest concentration applied (13). Patients not responding to the highest concentration of methacholine or AMP were assigned a value twice the highest concentration applied. Normality of distributions was assessed using the Kolmogorov-Smirnov test. If this test resulted in a p value < 0.05, normalization by logarithmic transformation was attempted. Correlations between variables were calculated by Pearson's correlation test in case of normal distribution or by Spearman's correlations test otherwise. To determine independent prognostic factors of PC₂₀ methacholine and PC₂₀ AMP multiple regression analysis was employed, in the stepwise algorithm (SPSS PC⁺ 9.0). In this analysis, the use of corticosteroids (yes/no), and the number of days corticosteroids had been stopped at the time of measurement were entered as covariates.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Characteristics

One hundred and twenty patients with mild to moderately severe asthma were enrolled in the study. Baseline characteristics are presented in Table 1. In this study, it was aimed for that patients completely discontinued their inhaled corticosteroids for at least 3 wk. During the steroid tapering period, 16 patients returned to the hospital earlier due to symptoms of a pending asthma exacerbation. From these 16 patients, 6 patients still used inhaled corticosteroids (3 patients 400 µg/d budesonide or beclomethasone, 2 patients 250 µg/d fluticasone, and 1 patient 800 µg/d budesonide). The remaining 10 patients had not used inhaled corticosteroids for a median period of 12 d (range 2-21 d). It was not possible to measure the PC₂₀ AMP in all patients, due to asthma symptoms and low FEV₁ (the lowest FEV₁ was 15 %pred). Thus, we were able to measure PC₂₀ AMP in 114 (95%) patients, respectively. All patients were, by design, responsive to methacholine (PC₂₀  8 mg /ml), whereas 102 of the 114 (89%) patients were responsive to AMP (PC₂₀  320 mg/ml). Of the 16 patients who returned to the hospital before they had completely discontinued their inhaled corticosteroids for at least 3 wk, the PC₂₀ AMP could not be measured in 2 patients, 13 patients were responsive to AMP, and 1 patient was not responsive to AMP.

Individual Correlations of PC₂₀ Methacholine and PC₂₀ AMP with Clinical and Inflammatory Parameters

A high correlation for both PC₂₀ methacholine and PC₂₀ AMP was found with FEV₁ %pred, correlations being very comparable (r = 0.45, p < 0.01, and r = 0.43, p < 0.01 respectively) (Table 2 and Figure 1). There was an inverse correlation between percentage of sputum eosinophils and PC₂₀ methacholine as well as PC₂₀ AMP, the correlation being stronger with PC₂₀ AMP ( = 0.49, p < 0.01) than with PC₂₀ methacholine ( = 0.29, p < 0.01) (Figure 2). Furthermore PC₂₀ methacholine was significantly correlated with the percentage of lymphocytes and ECP per eosinophil in sputum, whereas PC₂₀ AMP significantly correlated with ECP per eosinophil in sputum, ECP in sputum, number of eosinophils in blood, and ECP in blood.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Unadjusted relationship between PC20 methacholine and PC20 AMP with the level of FEV1, %pred.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Unadjusted relationship between PC20 methacholine and PC20 AMP with the percentage of sputum eosinophils.

Independent Correlates of PC₂₀ Methacholine and PC₂₀ AMP in a Multiple Linear Regression Model

In a multiple linear stepwise regression model, PC₂₀ methacholine was predominantly predicted by FEV₁ %pred, whereas the number of peripheral blood monocytes was an independent additional predictor (Table 3). In contrast, PC₂₀ AMP was predominantly predicted by the percentage eosinophils in sputum, whereas FEV₁ %pred was an additional independent predictor for the level of PC₂₀ AMP. The total percentage explained variance was larger for PC₂₀ AMP (36%) than for PC₂₀ methacholine (23%). When the use of corticosteroids (yes/no) and the number of days corticosteroids had been stopped at the time of measurement were entered as covariates, their regression coefficients were not significant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study in a large group of patients with asthma demonstrates that the levels of PC₂₀ methacholine and PC₂₀ AMP are both associated with baseline level of FEV₁ %pred. PC₂₀ AMP provides a better reflection of airway inflammation than PC₂₀ methacholine, as the percentage of sputum eosinophils does explain 25% of the variance in PC₂₀ AMP but is not a significant independent predictor for PC₂₀ methacholine. These results were independent of whether patients were still using inhaled corticosteroids at the time of measurement.

A positive association between the severity of bronchial hyperresponsiveness and the level of FEV₁ has been observed before (14). It can be explained by the fact that a given stimulus will result in a larger bronchoconstrictor response in a subject with more severe airway obstruction than in a subject with less severe obstruction, resulting in a lower provocative concentration of the stimulus under study causing a 20% reduction in FEV₁ (15). Thus, it was expected that both the severity of PC₂₀ methacholine and PC₂₀ AMP would be associated with the level of FEV₁ %pred.

The observed association between more severe bronchial hyperresponsiveness to both PC₂₀ methacholine and PC₂₀ AMP and more extensive airway inflammation, that is, the percentage of eosinophils in sputum, was to be expected. Worsening of asthma following viral infection or antigen exposure is associated with an increase in bronchial hyperresponsiveness and in case of experimental rhinovirus accompanied by an increase in sputum eosinophils and ECP (18, 19). Furthermore, bronchial hyperresponsiveness and sputum eosinophilia improve after therapy with inhaled corticosteroids (7). Activation of different inflammatory cells leads to airway wall edema with a concomitant increase in airway wall thickness. In addition, smooth muscle cells become more sensitive to contracting stimuli, all contributing to an increase in bronchial responsiveness (20, 21).

In our study, we found a dichotomy in the factors explaining the severity of PC₂₀ methacholine and PC₂₀ AMP. The level of FEV₁ %pred was the most important explanatory factor for the severity of PC₂₀ methacholine. Although the level of FEV₁ %pred was also a significant factor in explaining the severity of PC₂₀ AMP, greater contribution was derived from the percentage of sputum eosinophils. Therefore PC₂₀ AMP, which acts indirectly via the release of mediators from primed mast cells, reflects to a larger extent the cellular activation state in asthma than PC₂₀ methacholine. There are two possible explanations for the finding that PC₂₀ AMP is more closely associated with eosinophilic airway inflammation than PC₂₀ methacholine. First, an increased number of eosinophils may reflect an overall increased inflammatory activity resulting in an increased production of cytokines that stimulate mast cell chemotaxis, maturation, or activation directly or indirectly (22). Second, it can be speculated that the exposure to aeroallergens in subjects with allergy activates mast cells through immunoglobulin E (IgE)-dependent pathways resulting in the release of mediators that are responsible for an influx of eosinophils in sputum. We hypothesize that if tryptase or another marker of mast cell activity was to be measurable in sputum or blood, a closer association might be demonstrable for PC₂₀ AMP with sputum tryptase than for PC₂₀ AMP with the percentage of sputum eosinophils.

Polosa and colleagues recently reported an association between the number of eosinophils in sputum and PC₂₀ AMP, but not PC₂₀ methacholine in 12 subjects with allergic rhinitis in whom the diagnosis of asthma was specifically excluded (23). Although asthma and allergic rhinitis may be different diseases, patients with allergic rhinitis without clinical evidence of asthma have been shown to exhibit airway inflammation as reflected by an increase of eosinophils in sputum (24). Thus, the data of Polosa and coworkers (23) are compatible with ours. We extended their observation in two ways. First, we measured sputum ECP and showed in a simple regression that activation of eosinophils, as measured by ECP, was associated with more severe responsiveness to AMP but not to methacholine. This did not, however, contribute significantly in the multiple regression analysis. Multiple regression was not performed by Polosa and coworkers. The current analysis for the first time clearly showed that PC₂₀ AMP is more closely associated with eosinophils in sputum than PC₂₀ methacholine.

The observation that a lower number of peripheral blood monocytes is independently associated with more severe responsiveness to methacholine was not expected. In a previous study, Sont and coworkers also found an association between peripheral blood monocytes and bronchial hyperresponsiveness to hypertonic saline (25). It has been suggested that peripheral blood monocytes may play a role in immune responses (26). To date, however, relatively little is known about the modulatory role of peripheral blood monocytes in airway inflammation. Therefore this finding is of interest and merits further investigation.

In our study, PC₂₀ AMP but not PC₂₀ methacholine was associated equally strongly with the number of peripheral blood eosinophils and with the concentration of ECP in serum. Peripheral blood eosinophils have been shown to be associated with the severity of symptoms, the level of FEV₁, and bronchial hyperresponsiveness to methacholine and histamine (27- 30). Furthermore, serum ECP has been shown to reflect the degree of eosinophil activation (31). Thus, it has been suggested that both peripheral blood eosinophils and serum ECP may be indirect markers of airway inflammation in asthma (24). However, both the number of peripheral blood eosinophils and the concentration of ECP in serum did not contribute to a better prediction of PC₂₀ AMP in multiple regression analysis. The probable explanation for this finding is that the percentage of eosinophils in sputum, the number of peripheral blood eosinophils, and serum ECP provide overlapping information about the inflammatory activation state in asthma.

Finally, we did not find exhaled NO to be related to either PC₂₀ methacholine or PC₂₀ AMP. It has been suggested that the concentration of exhaled NO reflects airway inflammation. Patients with asthma have a higher concentration of exhaled NO compared with normal individuals, which is reduced after treatment with corticosteroids. However, the direct association between the concentration of exhaled NO and airway inflammation as reflected by the number of eosinophils in bronchial biopsies or sputum has been a matter of controversy (32, 33). Therefore, the absence of an association between the concentration of exhaled NO and PC₂₀ AMP does not exclude PC₂₀ AMP as a marker of airway inflammation.

In conclusion, the results of this study show for the first time that PC₂₀ AMP is more closely associated with airway inflammation than PC₂₀ methacholine. Therefore, PC₂₀ AMP provides both clinicians and researchers with a noninvasive marker of disease activity. Further studies have to assess whether reduction of airway wall inflammation in asthma (e.g., by corticosteroids) is also more closely associated with improvement in PC₂₀ AMP than PC₂₀ methacholine.