INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: asthma; exercise-induced; adrenergic -agonists; tolerance

Inhaled ₂-agonist bronchodilators are widely used to treat asthma symptoms and have an important role in the treatment of acute severe asthma. Short-acting -agonists are usually prescribed "as-required," but many patients use them on a frequent or regular basis. Long-acting inhaled -agonists are now available that provide sustained bronchodilation and are used as additional maintenance treatment to inhaled corticosteroids to achieve asthma control. A concern arising from this frequent use of -agonists is that patients may become tolerant to their effects and that they may be less effective in the treatment of acute severe asthma.

Until recently, it was thought that clinically significant tolerance to the bronchodilator effects of -agonists did not occur because very few studies were able to demonstrate this. By contrast, loss of the bronchoprotective effect of -agonists has been demonstrated in numerous studies (1). This means that when used regularly -agonists become less effective at preventing bronchospasm induced by natural or artificial stimuli such as exercise, methacholine, or histamine. Some studies have found that tolerance to the bronchodilator effects can also be demonstrated after use of the long-acting -agonists (2, 3). The importance of the loss of bronchoprotective effects and a small reduction in bronchodilator effect is not clear in the light of a sustained improvement in asthma symptoms using long-acting -agonists (4, 5). However, we have recently shown that bronchodilator tolerance can be easily demonstrated after regular use of both short- and long-acting -agonists if bronchoconstriction has first been induced with methacholine (6, 7). This finding indicates that although -agonists appear to remain effective in stable asthma, they may be less effective in acute severe asthma in patients who have been using the drugs frequently.

So far, bronchodilator tolerance in the setting of acute bronchoconstriction has been demonstrated after prior bronchoconstriction using methacholine. This is an artificial stimulus that acts directly on airway smooth muscle. To demonstrate the clinical relevance of this finding it is important to confirm that tolerance also occurs in the setting of bronchoconstriction induced by natural stimuli. We therefore examined whether the regular use of an inhaled -agonist leads to a reduction in the bronchodilator response to -agonist after exercise-induced bronchoconstriction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Volunteers aged 18-50 years with a history of exercise-induced bronchoconstriction were recruited at St Joseph's Hospital and McMaster University Medical Centre. Subjects with recent unstable asthma or respiratory tract infection, requiring oral or high-dose inhaled corticosteroids (> 1,500 µg/day beclomethasone or equivalent), or with significant cardiovascular disorders were excluded. The study was conducted outside the usual allergen season and subjects were not aware of recent allergen-induced exacerbations. Each subject gave written informed consent. The study was approved by the research ethics boards of both institutions.

Long-acting -agonists were withheld for 36 hours and other inhaled bronchodilators for 8 hours before an incremental exercise challenge. After measurement of baseline spirometry (ATS criteria), exercise was commenced at 100 kilopond meters per minute (kpm/min) on a cycle ergometer. The work rate was increased by 100 kpm/min every minute until the subject was unable to continue. Abbreviated spirograms to measure the forced expiratory volume in 1 second (FEV₁) were performed 1, 3, 5, 10, 15, 20, 25, and 30 minutes after exercise. At each time interval the highest of two measurements taken 30 seconds apart was recorded. Subjects who had a fall in FEV₁ were invited to return for a dry-air exercise challenge on another day.

The screening dry-air exercise challenge was undertaken at 80% of the maximum work rate achieved in the incremental exercise challenge. After measurement of baseline spirometry, subjects cycled at this constant work rate for 7 minutes. They inhaled dry air during exercise and for 1 minute after exercise. The FEV₁ was measured 1, 3, 5, 10, 15, 20, 25, and 30 minutes after exercise. Subjects who demonstrated a greater than or equal to 15% fall in FEV₁ which was sustained at greater than or equal to 10% for 5 minutes were entered into the study. Subjects who did not have a sufficient fall in FEV₁ were eligible to be re-screened using an increased workrate.

Subjects received either salbutamol 100 µg (Airomir, 3M, Minnesota), two puffs four times daily, or matching placebo in a random-order, double-blind, crossover manner. Subjects used a spacer device if necessary (Aerochamber, Trudell Medical, London, ON, Canada). All other -agonists were withheld. Ipratropium bromide 20 µg/puff (Boehringer Ingelheim, Germany) was provided for as-required relief of asthma symptoms. Any other medication was continued at a constant dose.

After 1 week (6-10 days) of each treatment, the study treatment and ipratropium were withheld for 8 and 12 hours, respectively, before attending the laboratory. Dry-air exercise challenges were performed using the same work rate and protocol as the screening challenge. After exercise the FEV₁ was measured at 1, 3, 5, 10, 15, 20, and 25 minutes. Salbutamol 100, 100, and 200 µg (Ventolin, Glaxo-Wellcome, Greenford, UK) was administered via an Aerochamber after the 5-, 10-, and 15-minute measurements, respectively. There was no washout between treatments. Both challenges were performed at the same time of day.

Analysis

The FEV₁ values after exercise (1, 3, and 5 minutes) and after administration of salbutamol (10, 15, 20, and 25 minutes) were compared between treatments by analysis of variance (ANOVA) (SPSS version 10.0). The pre-exercise FEV₁ was used as a covariate. p values less than 0.05 were regarded as statistically significant.

Sample size calculations were based on previous studies using methacholine (6, 7). Anticipating a similar reduction in response after exercise, eight subjects provided 80% power to achieve statistical significance at  = 0.05. It was hoped to recruit 10 subjects.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Of 22 subjects screened for exercise-induced bronchoconstriction, nine (8 female) were randomized to the study. Their mean age was 26 years (range 18-44), and the mean (SD) FEV₁/VC ratio 83 (10)%. Only one was using inhaled corticosteroid, and none was using other anti-inflammatory medications, theophylline, or anticholinergics. One subject was withdrawn during the first treatment period due to an exacerbation of her asthma. The remaining eight completed the study without incident.

We examined the mean FEV₁ values in relation to each exercise challenge (Figure 1). The baseline FEV₁ after the salbutamol and placebo treatment periods was not significantly different. After exercise, there was a significantly greater fall in FEV₁ in the salbutamol arm at 1, 3, and 5 minutes (p < 0.001); at 5 minutes the fall was 90% greater in the salbutamol arm. The FEV₁ remained significantly lower in the salbutamol arm throughout the salbutamol dose-response curve (p < 0.001). The differences in the FEV₁ measurements during the dose- response curves was largely explained by the fall in FEV₁ post-exercise. If this was taken into account as a covariate, the FEV₁ values during the dose-response curve were not significantly different.

View larger version (17K): [in this window] [in a new window]   Figure 1.   FEV1 changes before and after exercise and during the dose- response to salbutamol. Error bars represent 95% confidence intervals. For comparison the FEV1 changes from the prerandomization screening challenge are shown to illustrate the spontaneous changes in FEV1 following exercise.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study has shown that regular salbutamol treatment leads to increased bronchoconstriction and a suboptimal response to salbutamol after exercise. Although the absolute magnitude of the response to salbutamol was similar in both treatment arms, the greater fall in the salbutamol arm meant that the FEV₁ remained lower after salbutamol treatment. This difference persisted after salbutamol 400 µg and for 25 minutes after exercise.

We have previously shown that regular salbutamol treatment leads to increased exercise-induced bronchoconstriction (8). In that study, there was also a difference in baseline FEV₁ values on the placebo and salbutamol days that may have partially explained the differences in the post-exercise fall. In the present study the baseline FEV₁ values were almost identical, and the demonstration of increased bronchoconstriction after salbutamol confirms our earlier observations. The bronchodilator response to salbutamol once exercise-induced bronchoconstriction has occurred has not previously been studied and the finding that the FEV₁ remains low despite salbutamol is a new observation.

These findings indicate that patients with exercise asthma will have worsening exercise symptoms and a reduced response to their -agonist reliever inhaler if they are treated with regular or frequent -agonist. We chose to study the effects of regular treatment with short-acting rather than long-acting -agonists. The reason for this was that although not recommended in recent guidelines, regular or frequent self-medication with short-acting drugs remains a common practice. Clearly, a similar study should be performed using long-acting -agonists. However, there is no reason to believe that these effects would not be observed with long-acting -agonists. Both short and long-acting -agonists lose some of their effectiveness in preventing exercise-induced bronchoconstriction if they are used regularly (8, 9), and a similar degree of bronchodilator tolerance is observed after regular dosing with short- and long-acting -agonists in the context of methacholine induced bronchoconstriction (6, 7). How frequently or at what dose -agonists need to be taken for these effects to develop and how soon after starting treatment they occur is not yet known.

The mechanism behind our findings is likely to be ₂-receptor downregulation due to regular -agonist treatment. During exercise it is likely that circulating catecholamines help to maintain airway patency by functional antagonism of airway smooth muscle contraction induced by mediators released during exercise. This may in part account for the increase in FEV₁, which is sometimes observed immediately after exercise. In the present study, on the prerandomization and placebo days, we observed little change in the mean FEV₁ at 1 minute (Figure 1); however on the salbutamol day there was an immediate fall. From this it may be inferred that the "bronchoprotective" effect of endogenous catecholamines was reduced. Similarly, reduced functional antagonism against bronchoconstricting mediators may explain the subsequent suboptimal response to treatment with salbutamol.

An alternative explanation could be that regular salbutamol treatment caused increased airway inflammation leading to increased mediator release during exercise and hence both greater exercise-induced bronchospasm and a reduced response to -agonist. In support of this possibility there is evidence that regular short-acting -agonist treatment leads to an increase in sputum eosinophils (10, 11). However, -agonist tolerance after methacholine-induced bronchoconstriction does not appear to be affected by inhaled corticosteroid treatment suggesting that the mechanism is independent of airway inflammation (6). Possibly a combination of receptor downregulation and proinflammatory mechanisms are involved.

The development of bronchodilator tolerance to -agonists is potentially a major clinical problem. The consequences may be far greater when asthma is more severely uncontrolled than the effects on exercise-induced bronchoconstriction, which we have demonstrated here. -Agonists are the most important rescue treatment for acute asthma. Although asthma deaths are uncommon, most occur outside hospital despite self-administration of inhaled -agonist and thus represent a failure to respond to this emergency treatment (12). It is plausible that prior overuse of -agonist contributes to this failure to respond. Bronchodilator tolerance has previously been demonstrated in the context of methacholine-induced bronchoconstriction (6, 7). We have now shown that regular -agonist treatment not only leads to increased exercise-induced bronchoconstriction (a natural asthma stimulus), but also a failure to fully bronchodilate in response to inhaled salbutamol. Similar effects have been noted using an allergen-challenge model (13) and it is probable that the same effect will occur regardless of the cause of bronchoconstriction. It has recently been shown that the degree of -agonist tolerance observed increases with increasing bronchoconstriction (14). Thus, the effect of -agonist resistance may be profound in patients with acute severe asthma.

This study lends weight to current clinical opinion that regular short-acting -agonist treatment should be avoided in asthma. The implications for long-acting -agonist treatment used in conjunction with inhaled corticosteroids is less certain because this combination has been shown to improve asthma control (4, 5). However, we have demonstrated that regular -agonists have adverse effects on exercise-induced bronchoconstriction. We suggest that they are prescribed with care, particularly in patients with prominent exercise symptoms.