Epidemiologic data suggest that patients with chronic obstructive pulmonary disease (COPD) might be more sensitive than normal persons to the respiratory effects of oxidant pollutant exposure. Our study was designed to determine the response of patients with COPD to ozone. Thirteen white men with nonreversible airways obstruction (mean FEV1/FVC, 58%), of whom 8 were current smokers, were randomly exposed for 2 h to air and to 0.2 ppm ozone on 2 consecutive days using a single-blind crossover design. During either exposure, subjects exercised for 7.5 min every 30 min. Measures of respiratory mechanics obtained pre-exposure and postexposure were not significantly affected by either exposure. Similarly, ventilation and gas exchange measured during exercise showed no difference either between exercise periods or exposure days. However, arterial O2 saturation (SaO2), measured by ear oximetry during the final exercise period each day was lower (94.8%) at the end of O2 exposure, than SaO2 obtained at the end of air exposure (95.3%), the difference (0.48%) being significant (p = 0.008). Because normal subjects undergoing comparable exposures show a threshold for respiratory mechanical effects at about 0.3 ppm ozone, our data suggest that mild to moderate COPD is not associated with increased sensitivity to low ozone concentrations. However, our data do not rule out the possibility that the response of such subjects might be exaggerated at higher ozone concentrations. The consistent (in 11 of 13 subjects), though small, decrease in SaO2 may indicate that indexes of ventilation/perfusion distribution might be more sensitive measures of ozone effect in this compromised patient group than are conventional respiratory mechanics measures.