Short-term trials of bronchodilator drugs are widely used to assess patients with stable chronic obstructive pulmonary disease (COPD), but there is an uncertainty about the equivalence of the FEV1 response to beta-agonists and anticholinergic drugs, their relative ability to identify patients likely to improve with corticosteroids, the most appropriate way to express the results of these tests, and whether age or allergic status affects the beta-agonist and anticholinergic response differently. We studied 100 consecutive patients with stable COPD (mean FEV1, 0.96 +/- 0.48 L; mean age, 62 +/- 8 yr). Spirometry was measured before and after either 5 mg of nebulized salbutamol or 500 micrograms of nebulized ipratropium bromide and repeated after 2 wk of 30 mg of oral prednisolone daily. Total IgE, specific RAST, and skin prick testing values were recorded. Using modified American Thoracic Society response criteria, 33 patients failed to bronchodilate after the acute trials, 16 responded only to nebulized salbutamol, 17 to nebulized ipratropium, and 34 to both drugs. Twenty-two patients improved after corticosteroids. This was usually detected by a positive acute trial response (salbutamol 90% specific; ipratropium 84% specific). Baseline FEV1 differed between days, and in those who responded on only 1 day, this variation correlating with the response to ipratropium (r = 0.66). Expressing the response criterion as a percentage change in the available bronchodilatation increased the numbers responding with a high baseline FEV1, and vice versa. Neither age nor allergic status was related to the change in FEV1 after either drug in these patients. In COPD patients, testing with high-dose nebulized bronchodilators identifies a substantial number of partially reversible patients whatever age it is employed.(ABSTRACT TRUNCATED AT 250 WORDS)