The objective of the study is to assess the effects of emphysema on peak oxygen uptake ([Formula: see text]) during a cardiopulmonary exercise test in patients with chronic obstructive pulmonary disease (COPD). We measured [Formula: see text] and oxygen pulse in 80 patients with stable COPD exercising maximally. Oxygen saturation was measured by pulse oximetry (SpO(2)), and the ventilatory response assessed by the ratio of tidal volume (V (T)) at peak to slow vital capacity (SVC) at baseline, and by the percent increase of peak V (T) over baseline. Computed tomography imaging (CT scan) served as the reference diagnostic standard for emphysema. Based on the panel-grading (PG) method, emphysema was rated absent or mild (PG ≤ 30, n = 54), or moderate to severe (PG > 30, n = 26). Multiple quantile regression was applied to estimate the effects of PG > 30 on [Formula: see text]. At peak exercise, the patients with PG > 30 had significantly lower [Formula: see text], oxygen pulse and SpO(2), and featured a blunted ventilatory response with respect to those with PG ≤ 30 (p < 0.001). With multiple quantile regression, the effects of PG > 30 on [Formula: see text] were only partially explained by the degree of lung hyperinflation, a substantial component being imputable to impairment of lung diffusing capacity. In conclusion, chronic obstructive pulmonary disease patients with moderate to severe emphysema feature significantly lower exercise tolerance than those with no or mild emphysema. Our findings underscore the need of tailoring therapeutic interventions for COPD to the predominant clinical phenotype to improve exercise capacity.