Inhalers containing corticosteroids and long-acting beta2-agonists are becoming increasingly important in asthma management. A rapid effect is important to patients, particularly during exacerbations. We compared the onset of bronchodilation and patient-perceived relief from dyspnoea following single-inhaler budesonide/formoterol or salmeterol/fluticasone in a model of acute bronchoconstriction. A randomised, double-blind, double-dummy, single-dose, crossover study included 27 outpatients with asthma (mean age 35 years; mean FEV1 90% predicted normal). Immediately following methacholine-induced bronchoconstriction (fall in FEV1 > or = 30%), patients inhaled budesonide/formoterol (160/4.5 microg, 1 or 2 inhalations; Symbicort Turbuhaler), salmeterol/fluticasone (50/250 microg; Seretide Diskus) or placebo on 4 study days. Lung function and Borg score were assessed for 30 min. During methacholine-induced provocation (final mean FEV1 62.5% of baseline), mean Borg score increased 10-fold (from 0.3 to 3.0 units). Hereafter, mean FEV1 at 3 min improved significantly more after budesonide/formoterol 1 and 2 inhalations (37 and 38%, respectively) than after salmeterol/fluticasone (23%; P < 0.001) or placebo (10%; P < 0.001). Median recovery times to 85% of baseline FEV1 were shorter for budesonide/formoterol (1 or 2 inhalations: 3.3 and 2.8 min, respectively) than salmeterol/fluticasone (8.9 min; P < 0.001) and placebo (> 30 min). One min after budesonide/formoterol, dyspnoea was significantly reduced (Borg score -0.86 units, both doses) compared with salmeterol/fluticasone (-0.55 units; P < 0.05) and placebo (-0.23 units; P < 0.001). Budesonide/formoterol provides immediate bronchodilation, faster than salmeterol/fluticasone, which patients can feel during acute methacholine-induced bronchoconstriction.