To compare the sensitivity and validity of mannitol, histamine, and cold air challenges to demonstrate the effect of inhaled corticosteroids (ICS) in asthma. A prospective study. Seventeen patients with recently diagnosed, steroid-naive asthma who fulfilled the diagnostic criteria of Finnish Social Insurance Institution and were hyperresponsive to both mannitol and histamine. The following procedures were carried out at baseline and after 3 months and 6 months of treatment with inhaled budesonide, 800 microg/d: symptom assessment with a questionnaire, ambulatory peak expiratory flow (PEF) measurements twice daily for 2 weeks, and bronchial challenges with mannitol, histamine, and cold air. Budesonide decreased the sum symptom score, daily use of bronchodilating drugs, and diurnal PEF variation, but did not change FEV(1) percentage of predicted significantly. In addition, budesonide significantly decreased mannitol (p = 0.005) and histamine (p = 0.002) response dose ratios. The magnitude of the budesonide-induced change in responsiveness to these two challenges did not differ significantly. The effect of budesonide on cold air responsiveness did not reach statistical significance (p = 0.064). Change in mannitol responsiveness correlated significantly with the changes in sum symptom score and in FEV(1). Change in cold air responsiveness correlated with the changes in sum symptom score and in diurnal PEF variation. Change in histamine responsiveness correlated only with change in FEV(1). Mannitol challenge is both a sensitive and valid test to demonstrate the effects of ICS in asthma. Histamine challenge is equally sensitive for this purpose, but its validity may be lower than that of mannitol challenge. Cold air challenge seems to be a valid test to demonstrate the effects of ICS, but its sensitivity may be lower than that of mannitol and histamine challenges.