Results

Pretreatment with salmeterol produced a significant increase in FEV₁ compared with placebo (mean (SD) ΔFEV₁10 (5)% after salmeterol, 4 (5)% after placebo, p<0.05) and an inhibition of both EAR (median (range) ΔFEV₁ –2% (–28 to +2) after salmeterol, –41% (–57 to –27) after placebo, p<0.01) and LAR (median (range) ΔFEV₁ –5% (–37 to +6) after salmeterol, –45% (–65 to –20) after placebo, p<0.01; fig 1). After pretreatment with salmeterol EAR persisted in one subject and LAR in three.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Mean (SE) values of FEV₁ before and up to seven hours after allergen challenge in subjects pretreated with placebo (open circles) or salmeterol (closed circles). **p<0.01 between placebo and salmeterol.

Sputum was successfully obtained in all subjects at baseline evaluation and 24 hours after both allergen challenges. When placebo was inhaled before allergen challenge the sputum eosinophil percentage increased significantly (median 34%, p<0.05 versus baseline). Compared with placebo, pretreatment with 50 μg salmeterol resulted in a significant inhibition of the increase in sputum eosinophil percentage at 24 hours after allergen (15%, p<0.05 with respect to placebo; table 1).

Sputum ECP concentrations also increased at 24 hours after allergen inhalation in subjects pretreated with placebo (geometric mean 581 μg/ml, p<0.01 versus baseline) as well as after pretreatment with salmeterol (410 μg/ml, p<0.01 versus baseline) with no significant difference between placebo and salmeterol treatment (table1).

Blood eosinophil percentages did not change with respect to the baseline value at 24 hours after allergen inhalation with either placebo or salmeterol. Compared with baseline values, serum ECP concentrations increased at 24 hours after allergen with placebo pretreatment (9.2 versus 6.0 μg/l, respectively, p = 0.03) and salmeterol pretreatment (9.4 versus 6.0 μg/l, respectively, p = 0.02; table 1).

Discussion

This study shows that salmeterol inhibits the increase in sputum eosinophils induced by allergen challenge in sensitised asthmatic subjects. Moreover, our data confirm that pretreatment with salmeterol prevents both EAR and LAR to allergen inhalation.

Eosinophil recruitment into the airway can be prevented by anti-inflammatory drugs such as inhaled corticosteroids.10It is debatable whether long acting β₂ agonists also have some anti-inflammatory properties in vivo. In a group of subjects with mild asthma salmeterol significantly reduced serum ECP levels by approximately 50%.11 Moreover, salmeterol significantly reduced the increase in plasma proteins in nasal lavage fluid of subjects with allergic rhinitis eight hours after nasal allergen challenge12 and inhibited the recruitment of eosinophils in bronchial lavage fluid 24–48 hours after segmental allergen challenge.13 Pizzichini et al found that the late increase in sputum eosinophils after allergen inhalation was not prevented by pretreatment with salmeterol or with beclomethasone.4 The disagreement between our results and the results of Pizzichini et almay be explained by the different study design. In the study by Pizzichini et al each subject repeated five allergen challenges and four hypertonic saline sputum inductions for each allergen challenge. This could have resulted in a progressive increase in airway inflammation in each subject during the progression of the study, leading to a more persistent eosinophilic inflammation and consequently to the low repeatability reported by these authors in sputum eosinophil percentages measured before each allergen challenge. In fact, a small change in markers of airway inflammation can be induced by repeated hypertonic saline challenges14 and allergen inhalation increases non-specific bronchial reactivity for many days.15

Although airway eosinophilic recruitment induced by allergen challenge was inhibited by salmeterol, ECP levels in induced sputum obtained after pretreatment with salmeterol were no different from those obtained after placebo pretreatment. While it has been shown that salmeterol can inhibit diapedesis of inflammatory cells into the tissue at bronchodilator doses,16 higher concentrations of drug are required to affect ECP release from eosinophils.17 On the other hand, ECP levels were more variable than eosinophil percentages and there is therefore considerable potential for a type II error in failing to detect a modest effect of salmeterol on ECP levels.

Tolerance to some anti-inflammatory effects of salmeterol has been reported as a possible explanation for the loss of the protective effect of salmeterol on allergen challenge after repeated administrations of the drug.18 This fact could produce a discrepancy between the acute and chronic effects of salmeterol on airway inflammation in asthma.

In conclusion, we have shown that a single dose of salmeterol reduces the recruitment of eosinophils in the airways after allergen challenge, in addition to the prevention of the early and late airway responses.