INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchial hyperresponsiveness is a key feature of the asthma syndrome, reflecting the underlying disease process (1). Clinically and for research purposes it is measured by bronchial challenge, usually with methacholine or histamine (2). Methacholine acts directly at the level of bronchial smooth muscle, while adenosine monophosphate acts indirectly, causing primed mast cell degranulation and the release of proinflammatory mediators, such as histamine and leukotrienes (3). As such adenosine monophosphate is considered a better surrogate of airway inflammation and more like other indirect stimuli such as allergen or cold air (4). Indeed, studies in small numbers of patients have found adenosine monophosphate to be more sensitive than methacholine in detecting protection against bronchial hyperresponsiveness with inhaled corticosteroid therapy (5, 6). However, to our knowledge, these two types of bronchial challenge have never been directly compared as a screening test in a large group of patients presenting with a diagnosis of asthma.

Genetic polymorphisms of the ₂-adrenoceptor have been implicated in the asthma phenotype (7). The most common are at positions 16 and 27 on the ₂-adrenoceptor gene, located on chromosome 5q32. For example, the homozygous glycine genotype at codon 16 has been associated with nocturnal asthma (8), bronchial hyperresponsiveness (9), bronchodilator desensitization (10, 11), and increased inhaled corticosteroid dependence (12). The homozygous glutamic acid polymorphism at codon 27 has been linked to decreased airways responsiveness (13, 14), a decreased prevalence of childhood asthma (15), and elevated immunoglobulin E (IgE) in families with asthma (16).

We therefore retrospectively evaluated our database of patients presenting with a known diagnosis of asthma with the aim of comparing methacholine and adenosine monophosphate challenge as tools for evaluation of bronchial hyperresponsiveness, and also to investigate their relationship to markers of asthma severity and ₂-adrenoceptor polymorphisms.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The Database

The Tayside asthma and allergy research group recruits patients from the community with a known diagnosis of asthma for further characterization to facilitate clinical trial recruitment. Such patients have had a diagnosis of asthma from their primary or secondary care physician, usually based on symptoms, and have been prescribed at least one antiasthma drug. During their assessment they are characterized according to asthma therapy, spirometry, skin prick allergy testing, methacholine and adenosine monophosphate challenge, as well as blood genotyping for the ₂-adrenoceptor polymorphisms.

Spirometry

Spirometry was performed according to American Thoracic Society guidelines (17) using a Vitalograph Compact spirometer (Vitalograph, Buckinghamshire, UK). Prior to attending the laboratory for spirometry and bronchial challenge, subjects had been asked not to use their short acting ₂-agonists for 6 h, anticholinergics for 12 h, and long acting ₂-agonists, theophyllines, and leukotriene receptor antagonists for 48 h.

Skin Prick Testing

Skin reactivity to common allergens (grasses, trees, weeds, house dust mite, aspergillus, feathers, dog and cat) was determined with skin prick tests on the volar aspect of the forearm, using a standard puncture technique (18). Saline solution (0.9% wt/vol) and histamine (1 mg/ ml) were used as negative and positive controls, respectively. Wheal size was determined 15 min after administration of allergens and a positive reaction was taken as a wheal 2 mm or larger than negative control. For the purposes of the analysis, atopy was defined as having one or more positive skin prick tests.

Methacholine Bronchial Challenge

The methacholine bronchial challenge was performed by using a standardized computer-assisted dosimetric method as previously described (19). In brief, the methacholine was administered in doubling cumulative doses from 3.125 to 1,600 µg given at 5-min intervals until a 20% fall in the forced expiratory volume in 1 s (FEV₁) was recorded. The provocative dose of methacholine producing a fall in FEV₁ of 20% (PD₂₀) was then calculated. A PD₂₀ value of 500 µg equates to a PC₂₀ value of 5 mg/ml. This value for PD₂₀ is used in our laboratory as a cut-off for defining a biologically significant degree of bronchial hyperresponsiveness to methacholine. We are aware that other groups have used a stricter cut-off of 200 µg (2 mg/ml), and therefore we also analyzed our data using these criteria.

Adenosine Monophosphate Bronchial Challenge

Adenosine monophosphate bronchial challenge was performed as previously described (20). The test was continued until the FEV₁ had dropped by more than 20% from the baseline level or the maximum concentration of 400 mg/ml had been given. The provocative concentration of adenosine monophosphate producing a 20% fall in FEV₁ (PC₂₀) was calculated using a computer-assisted curve-fitting package. A cut-off value for PC₂₀ of 200 mg/ml is used in our laboratory for defining biologically significant bronchial hyperresponsiveness to adenosine monophosphate. Again, we also analyzed our data using the stricter cut-off of 100 mg/ml, which other groups have used previously.

Identification of the ₂-Adrenoceptor Polymorphisms

₂-Adrenoceptor polymorphisms were identified as previously described (14). In brief, genomic DNA was extracted from whole blood, and a 234 base-pair fragment generated by PCR that spanned the regions of interest. The primers used were 5': CCCAGCCAGTGCCTTACCT and 3': CCGTCTGCAGACGCTCGAAC. The genotype was identified by allele-specific oligonucleotide hybridization using 19-mer probes homologous for the Arg-16, Gly-16, Gln-27, or Glu-27 forms of the receptor. A random selection of PCR fragments was also directly sequenced to confirm the specificity of genotype identification by allele-specific digonucleotide.

Statistical Analysis

Data for methacholine PD₂₀ and adenosine monophosphate PC₂₀ were log transformed to normalize their distribution prior to analysis. Least-squares regression analysis was used to devise correlation coefficients using SPSS for Windows (Statistical Products and Service Solutions Inc., Chicago, IL). The statistical analysis between genotype and methacholine PD₂₀ or adenosine monophosphate PC₂₀ was performed by analysis of variance followed by multiple range testing, using Statgraphics (STSC Software Publishing Group; Rockville, MD).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Four hundred and eighty-seven subjects, presenting with a diagnosis of asthma over a period of 3 yr, had been entered into the database. Demographic data are shown in Table 1. Most patients had asthma of mild to moderate severity and 95% were atopic. Of these, 258 had a methacholine challenge, 259 an adenosine monophosphate challenge, and 185 both. A summary of the 185 subjects having both is shown in Table 2. The sensitivity, therefore, of methacholine challenge in an unselected group of patients presenting with a diagnosis of asthma is 140/185 (76%) using a cut-off of 500 µg, and 119/185 (64%) using 200 µg. The sensitivity of adenosine monophosphate challenge is 92/185 (50%) using 200 mg/ml and 69/185 (37%) using 100 mg/ml. Methacholine had a sensitivity of 46% (37% using the stricter cut-off), and adenosine monophosphate (AMP) 31% (26% using the stricter cut-off), in the small steroid-naive subgroup (n = 35).

FEV₁ and the forced mid expiratory flow rate (FEF_25-75) expressed as % predicted for age and height were significantly lower in patients who were hyperresponsive to either methacholine or adenosine monophosphate (see Table 3). Daily- inhaled corticosteroid dose was significantly higher only for those hyperresponsive to methacholine. Also, for patients hyperresponsive to methacholine (PD₂₀ < 500 µg only), PD₂₀ was weakly correlated to FEV₁ and FEF_25-75, but not to daily inhaled corticosteroid dose (Table 4). This was not the case for patients hyperresponsive to adenosine monophosphate, except in the small subgroup (n = 21) that was steroid naive, where PC₂₀ was correlated to FEV₁ (R = 0.46, p < 0.05) and FEF_25-75 (R = 0.45, p < 0.05). All the above correlations held in the large subgroup that was atopic.

Allelic and genotype frequencies for our subjects are shown in Table 5. Using a cut-off of 500 µg, subjects with allelic substitution of glycine at codon 16 had a lower methacholine PD₂₀ than those who were homozygous for arginine at codon 16, whereas using the stricter cut-off of 200 µg, only those patients with the homozygous mutation had a significantly lower PD₂₀ than those with the Arg-Arg genotype (Table 6). No such relationship existed for the adenosine monophosphate-responsive subjects at the 200 mg/ml cut-off, but again using the 100 mg/ml level, patients with the homozygous mutation had a significantly lower PD₂₀ than those with the Arg-Arg genotype (Table 7).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our study has examined the potential for methacholine and adenosine monophosphate challenge to be used as screening tools for bronchial hyperresponsiveness in an unselected population presenting with a diagnosis of asthma. We analyzed our data using two cut-offs for each challenge to represent alternative definitions of bronchial hyperresponsiveness, although any cut-off below the lower limit for normal bronchial responsiveness (e.g., PD₂₀ > 1,600 µg for methacholine [21]) is arbitrary, especially when used to investigate the relationship between the PD₂₀ and markers of asthma severity. There are no population-based data for AMP PC₂₀, but a value of < 200 mg/ml is discriminatory in our database. In the group that had both challenges, methacholine challenge had a sensitivity of 76% (64% using the stricter definition) and adenosine monophosphate challenge had a sensitivity of 50% (34% using the stricter definition). Furthermore, although only 11% (17% using the stricter definition) of methacholine-negative patients were adenosine monophosphate positive, 57% (53% using the stricter definition) of adenosine monophosphate-negative patients were methacholine positive.

It has been suggested that adenosine monophosphate challenge is a better marker of airway inflammation in asthma (3), and seems to be more responsive to treatment with inhaled corticosteroids (5, 6). As the proportion of patients taking inhaled corticosteroids was similar for methacholine and adenosine monophosphate responders, it is possible that there were relatively more adenosine monophosphate nonresponders due to the effects of steroids. However, in our steroid-naive subgroup (which was admittedly small, due to adherence to British Thoracic Society Guidelines) the relative sensitivities of the two challenges were the same, with methacholine having a sensitivity of 46% (37% with the stricter criteria) and AMP 31% (26% with the stricter criteria). The diagnosis of asthma is usually based on symptoms, lung function, and bronchial hyperresponsiveness. All of our subjects had a presumed diagnosis of asthma, mostly from primary care, based on symptoms with or without peak flow. It is therefore not possible for us to comment on the specificity of the challenges. It is possible that those patients who were unresponsive to methacholine may have been misdiagnosed as asthmatic or have asthma that has become quiescent, either spontaneously or as a result of treatment. In looking for a screening tool for bronchial hyperresponsiveness in an unselected group of patients, we would suggest that the greater sensitivity of methacholine challenge makes it the favored option.

We have shown that patients with more severe airway obstruction (as measured by FEV₁ and FEF_25-75) are more likely to have bronchial hyperresponsiveness to methacholine or adenosine monophosphate. Also, for patients responsive to methacholine (with a PD₂₀ < 500 µg only), and steroid-naive patients responsive to AMP, increasing severity of asthma (as measured by FEV₁ and FEF_25-75) was weakly correlated to increasing bronchial responsiveness. In this respect bronchial hyperresponsiveness to histamine has been shown to correlate with early morning peak expiratory flow rate (PEFR) and improvement in PEFR after bronchodilator and diurnal fluctuation of PEFR (2). Moreover, acetylcholine PC₂₀ is associated with epithelial damage (22), and methacholine PC₂₀ to CD4 T-lymphocytes in bronchoalveolar lavage (23), exhaled nitric oxide, sputum eosinophils (24), and airway remodeling (25). As in our study, bronchial hyperresponsiveness has previously been associated with increased steroid requirement in patients with asthma (26), reinforcing the link between asthma severity and bronchial hyperresponsiveness to methacholine.

The allelic frequencies in our study population were similar to those previously reported (27). We found the presence of the glycine allele at codon 16 to be associated with increased bronchial hyperresponsiveness to methacholine, in keeping with other studies (28). We did not find any association between bronchial hyperresponsiveness and genotype at codon 27, in contrast to Ramsey and coworkers (13). The vast majority (92%) of our patients were skin prick positive to at least one allergen, and unsurprisingly analysis of this subgroup did not reveal any different results to the overall analysis.

We recognize the limitations of this retrospective analysis. Our patients are a heterogeneous group unselected other than by their given diagnosis of asthma. As such they do represent the type of patients we see in our clinical practice. We feel that this sample provides an example of how the methacholine and adenosine monophosphate challenges would perform in "real life" screening of patients with asthma for bronchial hyperresponsiveness.

In conclusion, our results suggest that methacholine challenge is a more appropriate initial screening tool for bronchial hyperresponsiveness in asthma as it was more sensitive than adenosine monophosphate and was related to asthma severity. In addition, we have demonstrated an association between the glycine allele at codon 16 and increased bronchial hyperresponsiveness to methacholine.