Inhalation of a direct stimulus such as histamine or methacholine is generally used to measure bronchial hyperresponsiveness (BHR). Provocation with adenosine 5'-monophosphate (AMP), an indirect airway challenge, has been suggested to be a better marker of airway inflammation than direct challenges. However, so far little information on this subject is available. The aim of our study was to assess whether the concentration of AMP causing the FEV(1) to drop by 20% (PC(20)) is more closely associated with inflammatory parameters in asthma than PC(20) methacholine. In 120 patients with atopic asthma (median FEV(1) 81% predicted [pred], median age 27 yr), PC(20) methacholine and PC(20) AMP as well as sputum induction, blood sampling, and measurement of nitric oxide in exhaled air were performed. PC(20) methacholine was predominantly predicted by FEV(1) %pred (explained variance [ev] = 18%) with the percentage of peripheral blood monocytes being a weak additional independent predictor (total ev = 23%). By contrast, PC(20) AMP was predominantly predicted by the percentage of eosinophils in sputum (ev = 25%), while FEV(1) %pred was only an additional independent predictor (total ev = 36%). PC(20) AMP reflects more closely the extent of airway inflammation due to asthma than PC(20) methacholine.