INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inhaled ₂-agonists are the most effective drugs available for the relief of acute asthma symptoms (1). More potent, long-acting ₂-agonists have recently been introduced, which provide nocturnal bronchodilatation and also protect against bronchoconstrictor stimuli for more than 12 h. While short-acting ₂-adrenoceptor agonists are recommended for use only as a relief medication, long-acting ₂-agonists are recommended for regular use in patients with asthma not controlled adequately on inhaled corticosteroids (2).

Concerns have been raised regarding the regular use of ₂-adrenoceptor agonists in asthma. Studies have linked short-acting ₂-agonists with loss of asthma control (3) and an increased risk of asthma death (3, 4). With the prolonged receptor occupancy of long-acting ₂-agonists, any effects may be greater than with the short-acting drugs. The development of tolerance to the nonbronchodilator effects of these drugs, resulting in an increase in airway inflammation, has been suggested as a mechanism for any adverse effects (5).

Several recent studies have shown that the degree of protection conferred by short-acting ₂-agonists against indirectly acting bronchial provocation challenge agents, such as adenosine monophosphate (AMP) and allergen, is significantly reduced after regular therapy with short-acting ₂-agonists (5- 9). This also occurs, but to a lesser degree, with directly acting bronchoconstrictors such as methacholine (MCh) (7, 8), suggesting greater downregulation of mast cell than airway smooth muscle ₂-receptors. Whether this differential effect also occurs with long-acting ₂-agonists is not known. Regular therapy with salmeterol and formoterol may, however, result in diminution of bronchoprotection to MCh challenge (6, 7), although single doses of salmeterol do not protect more against AMP than against MCh challenge in mild asthma (10).

In clinical practice, patients with asthma are maintained on regular long-acting ₂-agonist treatment and use short-acting ₂-agonists for symptom control. An interaction between short- and long-acting ₂-agonists could occur due to shared ₂-adrenoceptor stimulation. The effect of salmeterol on the protection conferred by albuterol, measured directly and indirectly, has not been examined. We therefore studied the effect of salmeterol on bronchial hyperreactivity after albuterol in mild asthmatic subjects, using MCh and AMP challenge to assess airway smooth muscle and mast cell ₂-adrenoceptors, respectively.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Sixteen nonsmoking subjects meeting the American Thoracic Society's diagnostic criteria for asthma (11) (9 males, mean age [SE] 29.5 [2.4] yr (Table 1) were studied. All gave written informed consent to participate in the study, which was approved by the Royal Brompton and National Heart & Lung Hospital Ethics Committee. All subjects had occasional symptoms of variable wheeze and dyspnea, controlled by ₂-agonists alone. None had suffered an asthma exacerbation or respiratory tract infection within the 6 wk preceding the study. Baseline FEV₁ for all subjects was  70% predicted. All subjects were sensitive to MCh and AMP, showing a PC₂₀ of < 8 mg/ml and < 50 mg/ ml, respectively, and a PC₂₀, measured 15 min after a single inhalation of 200 µg of albuterol, of < 32 mg/ml MCh and < 200 mg/ml AMP. Subjects were excluded if they had smoked within 6 mo before study entry or had used any glucocorticoids within 4 mo. Inhaled bronchodilators and caffeine-containing drinks were withheld for > 12 h before each study visit.

Study Design

The study was double-blind, randomized, placebo-controlled, and crossover. After screening, patients entered a run-in period of 14 d, at which time inhaled ipratropium bromide MDI was substituted for ₂-agonist as rescue medication for the study duration. Either active treatment (salmeterol 50 µg b.i.d. via dry powder inhaler) or identical placebo were both inhaled regularly over a period of 14 d, crossing over to the alternative treatment after a washout period of  14 d. MCh and AMP challenges were performed on consecutive days in randomized order (8), with identical order of challenge for each patient throughout the study. PC₂₀ was measured (15 min after a single dose of albuterol, 200 µg via metered-dose inhaler and spacer) at baseline, prior to starting treatment, and exactly 12 ± 0.5 h after cessation of salmeterol treatment, at an identical time of day. Patients inhaled study medication the evening before the first challenge, attended the laboratory exactly 12 h later, then continued medication during that day, attending for the alternative challenge exactly 12 h after the last dose. Twice daily peak flows, measured before any bronchodilator, symptom scores, and rescue inhaler usage were measured during the study period.

Bronchial Provocation Challenge

Bronchial provocation challenge was performed according to our standardized technique, as previously reported (8). Solutions of MCh or AMP (Sigma, Poole, UK), were dissolved in 0.9% saline in doubling dilutions (0.06-32 mg/ml and 0.39-800 mg/ml, respectively) and were administered from a nebulizer attached to a breath-activated dosimeter (Mefar, Brescia, Italy). After resting quietly for 15 min, baseline spirometry was assessed by three forced expiratory maneuvers using a dry-wedge spirometer (Vitalograph, Buckingham, UK). Subjects then inhaled five breaths of saline, and FEV₁ was measured 2 min after the last inhalation. Incremental doses were administered at 3-min intervals. Challenges were terminated when a 20% decrease in FEV₁ from the postsaline value was reached. PC₂₀ was determined by linear interpolation of the concentration-FEV₁ response curve.

Statistical Methods

PC₂₀ values were log10 transformed for analysis and reported here in milligrams per milliliter as geometric mean and geometric standard error of the mean (GSEM). Log values were compared by analysis of variance (ANOVA), taking into account period and patient effects. Thus, the effect of treatment on responses to MCh and AMP challenge was examined by comparing log PC₂₀ after the administration of placebo with that after salmeterol. The geometric mean represents the anti-log of logarithmic mean, and the GSEM, the anti-log of the logarithmic standard error. In addition, the number of doubling dilutions were calculated according to the formula (8):

Doubling dilution changes were compared by ANOVA in an identical manner to log PC₂₀ values. FEV₁ data were analyzed as absolute values and compared by ANOVA, taking into account period and patient effects, and reported as means ± standard error. Significance was taken as p < 0.05. The power of the study was calculated on the basis of variation of the individual PC₂₀ (8). With an  value of 5% and power of 80%, 15 patients were required to detect a twofold difference in PC₂₀.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Fifteen of the sixteen patients successfully completed the study. One subject withdrew during the placebo treatment period, due to an upper respiratory tract infection. Her results until withdrawal were included on an intention-to-treat basis. Subject characteristics are shown in Table 1.

Bronchial Responsiveness

Geometric mean PC₂₀ (after albuterol) at the start of both treatment periods showed no significant difference (2.9 ± 1.3 mg/ml [salmeterol] versus 2.2 ± 1.3 mg/ml [placebo], respectively). PC₂₀ with placebo treatment did not change significantly at any time during the study, either with MCh or with AMP. With MCh, mean PC₂₀ decreased significantly with salmeterol treatment from 2.24 ± 1.3 mg/ml to 1.08 ± 1.2 mg/ml (p < 0.03), while placebo decreased from 2.9 ± 1.3 mg/ml to 2.6 ± 1.3 mg/ml (p = NS) (Figure 1A). With AMP, a slight decrease was observed after salmeterol treatment (27.5 ± 1.5 to 9.5 ± 1.4 mg/ml; p = NS), while a small decrease was also seen with placebo (43.6 ± 1.5 to 21.4 ± 1.4 mg/ml; p = NS (Figure 1B).

View larger version (14K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 1.   (A) Effect of regular salmeterol (50 µg b.i.d.) or identical placebo treatment for 2 wk on the acute protection provided by 200 µg of albuterol on MCh PC20, measured 12 h after cessation of salmeterol. (B) Effect of regular salmeterol (50 µg b.i.d.) or identical placebo treatment over 2 wk on the acute protection provided by 200 µg of albuterol on AMP PC20, measured 12 h after cessation of salmeterol; n = 15; triangles = placebo; squares = salmeterol. *p Value < 0.05 compared with baseline. Data shown as log mean ± SEM, before and after treatment.

When analyzed in terms of doubling dilutions, PC₂₀ decreased by 1.05 ± 0.4 doubling dilutions with salmeterol and methacholine challenge, compared with 0.18 ± 0.4 doubling dilutions with placebo (p = NS) (Figure 1A). For AMP challenge, PC₂₀ decreased by 1.41 ± 0.72 doubling dilutions with salmeterol, and 1.0 ± 0.44 with placebo (p = NS) (Figure 1B).

Effect of Salmeterol Treatment on FEV₁

Baseline FEV₁ was 85 ± 4% predicted on screening for entry into the study. Mean FEV₁ before salmeterol treatment was slightly lower than before placebo at the beginning of each treatment period (3.29 ± 0.15 L versus 3.38 ± 0.18 L), but this difference was not statistically significant. Two weeks of salmeterol treatment resulted in a significant increase in pre-challenge FEV₁ compared with placebo (3.47 ± 0.16 L versus 3.33 ± 0.17 L; p < 0.05). The mean change was 0.18 ± 0.08 L with salmeterol and 0.05 ± 0.08 with placebo, and was statistically significant (p < 0.02), reflecting the long duration of bronchodilator effect of salmeterol.

Effect of Regular Salmeterol Treatment on Bronchodilator Response

After the run-in period, mean FEV₁ increased after a single dose of albuterol from 3.29 ± 0.15 to 3.59 ± 0.16 L (p < 0.01), or 0.30 ± 0.15 L. After treatment, there was a significant difference between change in FEV₁ between salmeterol and placebo; mean change was 0.25 ± 0.5 L after salmeterol treatment, and 0.35 ± 0.5 L after placebo (p < 0.01). However, this could be accounted for by the rise in baseline FEV₁ with salmeterol treatment.

Peak Flow

Mean morning peak expiratory flow (PEF) increased with salmeterol treatment relative to placebo. Mean pretreatment morning peak flow was 483 ± 29 with salmeterol and 454 ± 26 with placebo. Evening PEF also increased with salmeterol compared with placebo (488 ± 24 versus 507 ± 29), but these differences did not achieve statistical significance.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our study has shown that regular treatment with salmeterol for 2 wk leads to a small loss of the acute protective effect of albuterol on MCh-induced bronchoconstriction. However, no significant loss of protection to AMP occurred, implying that significant mast cell tolerance did not develop. Our findings imply that the mechanisms underlying bronchoprotective tolerance with short-acting ₂-agonists during regular therapy differ from those with the long-acting ₂-agonist salmeterol.

Our study confirms the work of others (5), and adds new information regarding protection against mast cell-mediated bronchoconstriction by the long-acting ₂-agonist salmeterol. The mechanism underlying bronchoprotective tolerance with short-acting ₂-agonists appears to involve differential tolerance in airway cells, occurring more rapidly on mast cell than on smooth muscle ₂-adrenoceptors, but whether this also applies to long-acting ₂-agonists was not known. One study suggested that, in single dose, salmeterol does not protect more against AMP challenge than against MCh, similar to our results (10).

Given the readiness with which loss of bronchoprotection to AMP was induced previously after only 1 wk of treatment with terbutaline in our laboratory (8), our findings suggest that mast cell ₂-adrenoceptors are not downregulated by regular salmeterol treatment. Our study design excludes the possibility of very rapid downregulation, as we measured PC₂₀ 15 min after a single dose of albuterol both before and after salmeterol treatment, and patients had been taken off ₂-agonists before any treatment. Our results thus suggest that salmeterol's primary mechanism of action is by functional antagonism at the level of airway smooth muscle. Our findings regarding indirect bronchial provocation challenge differ from those of Ramage and colleagues (12), who found loss of bronchoprotection to exercise after regular salmeterol treatment. This discrepancy can perhaps be accounted for by differences in mechanisms between exercise-induced bronchoconstriction and AMP, or perhaps by the use of a longer treatment period.

The loss of bronchoprotection we observed with methacholine was small (1 ± 0.4 doubling dilutions) and fell just short of statistical significance when analyzed as doubling dilutions. A loss of bronchoprotection of similar magnitude has been noted in several other studies (5, 13). With such a small bronchoprotective loss and preservation of significant bronchoprotection, the clinical relevance of our findings is questionable. However, an interaction between albuterol and salmeterol could potentially be of clinical relevance during, for example, an acute asthma exacerbation, when ₂-adrenoceptor function may be impaired.

Our study was conducted on patients not maintained on inhaled glucocorticoids. This is contrary to the recommended use of salmeterol. However, recent studies have suggested that inhaled glucocorticosteroids in therapeutic doses do not prevent adverse effects of short-acting ₂-agonists (3), and the only placebo-controlled study available to date found inhaled glucocorticosteroids to be ineffective in preventing bronchoprotective tolerance (14). Whether loss of bronchoprotection is of any clinical relevance, either in maintenance asthma therapy or in an acute asthma exacerbation, requires further investigation.