We investigated the constituents of bronchoalveolar lavage (BAL) regarding cell profiles and released eosinophilic cationic protein (ECP) in 11 patients treated occasionally with inhaled bronchodilators (Group A) and 11 patients treated regularly with inhaled corticosteroids (Group B). A normal, healthy control group of 12 subjects was also recruited. Compared with Group A, Group B had a reduced recovery percentage of infused volume (p less than 0.05) and total cell number (p less than 0.01). Compared with the control group, there was a significant increase in the percentage of eosinophils (p less than 0.05) in both groups of asthmatics. In Group A there was also a significant increase in mast cells (p less than 0.05), serum-ECP (p less than 0.05), and BAL-ECP (p less than 0.001). No correlations between any of the cell variables and the level of airway responsiveness measured as PC20 histamine were found in any group. Group A patients were investigated twice--before and after 4 wk of randomly allocated treatment with either a regular beta-2-receptor agonist (terbutaline 250 micrograms, two puffs four times a day) or a regularly inhaled corticosteroid (budesonide 200 micrograms twice a day). The BAL differential cell counts were similar and not significantly affected by either treatment. However, BAL-ECP levels were decreased by budesonide treatment (p less than 0.05). ECP levels in serum and BAL were significantly correlated (p less than 0.05 to 0.001). The eosinophilic cell involvement in asthma is further emphasized by this study but the increase in numbers of eosinophils seems less important than their activity, here measured as release of one degranulation product, ECP. To suppress disease activity, repeated long-term treatment is important, but clear preference for either treatment cannot be given on the basis of our present results.