Background: Lidocaine inhalation attenuates histamine-induced bronchoconstriction, as well as bronchoconstriction elicited by mechanical irritation. This effect could be mediated by direct effects on smooth muscle or by reflex attenuation. Therefore, we evaluated whether lidocaine attenuated the bronchial response of direct smooth muscle stimulation with methacholine. Methods: In 15 volunteers with bronchial hyperreactivity, a methacholine challenge was performed following the inhalation of lidocaine, dyclonine (which does not attenuate bronchial reactivity) or saline on three different days in a randomized, double-blind fashion. Lung function, response to methacholine, and lidocaine and dyclonine plasma concentrations were measured. Results: The inhaled methacholine concentration (PC20) necessary for a 20% decrease in the forced expiratory volume in 1 s (FEV1) was 8.8 ± 6.1 mg/ml at the screening evaluation. The sensitivity to methacholine challenge (PC20) remained unchanged regardless of which solution was inhaled (9.1 ± 7.5 mg/ml for lidocaine, 10.2 ± 9.0 mg/ml for dyclonine and 9.8 ± 8.3 mg/ml for saline; P= 0.58, means ± standard deviation). Furthermore, the inhalation of all three solutions caused a significant decrease in FEV1 from baseline (P= 0.0007), with a significantly larger effect for dyclonine than lidocaine (P= 0.0153). Conclusions: Although both inhaled and intravenous lidocaine attenuates histamine-evoked bronchoconstriction, it does not alter the response to methacholine. Therefore, the attenuation of bronchial reactivity by lidocaine appears to be related solely to neurally mediated reflex attenuation, rather than to the attenuation of direct constriction of airway smooth muscle.