Recent work has demonstrated the ability of continuous positive airway pressure (CPAP) to relieve dyspnea during exercise in patients with severe chronic obstructive pulmonary disease (COPD). The present study examined the effects of CPAP (7.5-10 cmH2O) on the pattern of respiratory muscle activation and its relationship to dyspnea during constant work load submaximal bicycle exercise [20 +/- 4.8 (SE) W] in eight COPD patients (forced expiratory volume in 1 s = 25 +/- 3% predicted). Tidal volume, respiratory rate, minute ventilation, and end-expiratory lung volume increased with exercise as expected. There was no change in breathing pattern, end-expiratory lung volume, or pulmonary compliance and resistance with the addition of CPAP. CPAP reduced inspiratory muscle effort, as indicated by the pressure-time integral of transdiaphragmatic (integral of Pdi.dt) and esophageal pressure (integral of Pes.dt, P less than 0.01 and P less than 0.05, respectively). In contrast, the pressure-time integral of gastric pressure (integral of Pga.dt), used as an index of abdominal muscle recruitment during expiration, increased (P less than 0.01). Dyspnea improved with CPAP in five of the eight patients. The amelioration of dyspnea was directly related to reductions in integral of Pes.dt (P less than 0.001) but inversely related to increases in integral of Pga.dt (P less than 0.01). In conclusion, CPAP reduces inspiratory muscle effort during exercise in COPD patients. However, the expected improvement in dyspnea is not seen in all patients and may be explained by more marked increases in expiratory muscle effort in some individuals.