We examined the role of CO2 responsiveness and breathing efficiency in limiting exercise capacity in 15 patients with chronic airway obstruction (FEV1 = 0.88 +/- 0.25 L, mean +/- SD). Responses of minute ventilation and P0.1 (mouth pressure 0.1 s after the onset of occluded inspiration) to hypercapnia (delta VE/delta PCO2, delta P0.1/delta PCO2) were measured by rebreathing, and the ratio of the two (delta VE/delta P0.1) was defined as an index of breathing efficiency during hyperventilation. Exercise capacity was measured as symptom-limited, maximal oxygen consumption (VO2max/BW) in an incremental treadmill test and also as the 12-min walking distance (TMD). All patients discontinued the treadmill test because of dyspnea, and the exercise capacity correlated with the degree of airway obstruction, although there was a wide variability among patients with comparable FEV1. There were no significant correlations between the responses to CO2 and exercise capacity. However, there was a significant correlation between delta VE/delta P0.1 and VO2max/BW (r = 0.87, p less than 0.001) or TMD (r = 0.78, p less than 0.001), and these correlations remained significant even when the relational effects of FEV1 were taken out. These results support the hypothesis that airway obstruction and breathing efficiency are important, but that CO2 responsiveness is not a major factor in determining the exercise capacity of patients with chronic airway obstruction.