Frequently, patients with COPD with similar spirometric impairment have marked differences in dyspnea and exercise limitation. As the classic "blue bloater" with attenuated respiratory drive is described as being less dyspneic than his "pink puffer" counterpart, we wondered whether the variability in dyspnea and exercise tolerance in a group of patients with COPD with relatively similar degrees of air-flow obstruction might be partly explained by the variability in resting respiratory drives (unstimulated P0.1 and hypoxic and hypercapnic P0.1 responses). Therefore, we measured unstimulated mouth occlusion pressure (P0.1), hypoxic response (-delta P0.1/delta SaO2), hypercapnic response (delta P0.1/delta PCO2), 6-min walk distance, VO2max, steady-state exercise VE/VO2, exercise SaO2, and dyspnea using an oxygen cost diagram in 15 subjects with severe COPD (mean FEV1% 35.2 +/- 1.9 SEM). No correlations between spirometric impairment and either dyspnea or exercise performance were seen. Unstimulated P0.1 correlated inversely with spirometric impairment but did not correlate with dyspnea, VO2max or 6-min walk distance. Both hypoxic and hypercapnic responses were significantly correlated with greater exercise ventilation (VE/VO2), less exercise O2 desaturation, and a greater VO2max, but not with dyspnea or 6-min walk distance. The results of this study do not support the concept that depressed respiratory drives are associated with less dyspnea or greater exercise capability in COPD.