Regular use of short-acting β-agonists may decrease control of asthma and increase airway responsiveness to bronchoconstrictor stimuli. The aim of this study was to determine the effects of regular treatment with the long-acting β-agonist, salmeterol, on the methacholine dose-response curve (DRC) in mild-tomoderate asthmatics.Changes in methacholine airway responsiveness were measured in 14 stable adult asthmatics, randomized in a double-blind, three-way cross-over design to receive salmeterol 50 µg, salbutamol 200 µg or placebo, each twice daily for 4 days. Two baseline methacholine DRC, were performed, one without premedication and one following a single dose of 200 µg salbutamol. Following 4 days of regular treatment, methacholine DRC to plateau were carried out commencing 15 min after the final dose of trial medication.There were no significant differences in mean baseline forced expiratory volume in one second (FEV1) between treatments. Four days treatment with salmeterol and salbutamol shifted the DRC to the right, but salmeterol provided less protection than salbutamol. The point of inflection of the curve from baseline moved 1.9 and 3.2 doubling doses, respectively, compared to placebo (p≤0.001), and the provocative concentration of methacholine required to produce a 20% fall in FEV1 (PC20) increased 1.6 and 3.1 doubling doses, respectively (p≤0.001). The slope of the DRC was increased slightly by both β-agonists compared to placebo (log slope 3.11, 3.06 and 2.77 for salmeterol, salbutamol and placebo, respectively). This effect of regular salmeterol on slope was more marked in subjects with lower baseline FEV1. Maximal response plateaus did not differ between the three treatments.These results suggest that regular use either of short-or long-acting β-agonists could increase the risk of a more precipitous asthma episode associated with "breakthrough" bronchoconstrictor responses, particularly in those with more severe initial airflow obstruction, if subjects are exposed to a sufficiently potent stimulus.