To cope with the increased ventilatory demands of exercise, patients with severe expiratory flow limitation adopt strategies that ultimately place greater demands on their inspiratory muscles. Increased inspiratory muscle work may contribute to dyspnea causation and exercise limitation in such patients even before their ventilatory ceiling is attained. In this setting, continuous positive airway pressure (CPAP) should, by favorably affecting inspiratory muscle function and respiratory sensation, improve exercise performance. Six patients with chronic airflow limitation (CAL) (FEV1 +/- SD = 35 +/- 12% predicted) undertook constant-load, submaximal, cycle exercise at 50% of their predetermined maximal oxygen consumption: CPAP of 4 to 5 cm H2O was delivered during one exercise session and bracketed by one or two unassisted control sessions. In four patients, CPAP-assisted (4 to 5 cm H2O) exercise was bracketed by two unassisted control exercise sessions; two remaining patients undertook CPAP-assisted exercise and one unassisted control session. CPAP resulted in a significant increase in exercise endurance time (TLIM) (by 48%: CPAP TLIM (mean +/- SE) = 8.82 +/- 1.90 min; averaged control TLIM = 5.98 +/- 1.23 min (p less than 0.01). CPAP effectively ameliorated exertional dyspnea in the majority of patients; selected dyspnea ratings (Borg scale) during control (final minute) and CPAP at isotime, at comparable levels of ventilation, were (mean +/- SD) 7.83 +/- 2.25 and 5.5 +/- 2.2, respectively (p less than 0.025). Breathing frequency fell significantly during CPAP application (at isotime) by 17% (p less than 0.02); other steady-state ventilatory variables and end-expiratory lung volumes were not significantly different during CPAP and control.(ABSTRACT TRUNCATED AT 250 WORDS)