Thirty-two patients presenting with acute exacerbations of chronic obstructive pulmonary disease were entered into the following double-blind, crossover study. First (time 0), patients inhaled either ipratropium bromide (54 micrograms) or metaproterenol sulfate (1.95 mg) via a metered dose inhaler (MDI) attached to a device (Inspirease) (phase 1). After 90 minutes, they inhaled whichever of the two medications they had not received in phase 1. This is referred to as phase 2. Pulmonary function (FEV1 and FVC) was measured at time 0, and at 30, 60, and 90 minutes following phase 1 treatment, and at 30, 60, and 90 minutes following phase 2 treatment (120, 150, and 180 minutes from the start of the study). Arterial blood gas samples (n = 20) were obtained at entry into the study and 30 and 90 minutes after phase 1 medication. The groups did not differ in age, degree of airway obstruction, hypoxemia, or theophylline usage at the start of the study. In phase 1, at 90 minutes, pulmonary function in both groups significantly and similarly improved. For ipratropium, FEV1 improved from 0.62 +/- 0.08 L to 0.88 +/- 0.11 L (p less than 0.01) and for metaproterenol FEV1 improved from 0.69 +/- 0.06 to 0.92 +/- 0.09 L (p less than 0.01). There was no further improvement with phase 2 treatment for either group. Thirty minutes after inhaling ipratropium, there was a small but significant rise in PO2 (5.8 +/- 3.0 mm Hg; p less than 0.05) while metaproterenol inhalation resulted in a 6.2 +/- 1.2 mm Hg decline in PO2 (p less than 0.05). These changes were not sustained at 90 minutes. We concluded that for acute exacerbations of COPD, both ipratropium and metaproterenol are effective medications when administered via an MDI attached to a device (Inspirease). However, ipratropium may be a safer choice as it initially did not cause a decline in blood oxygenation.