Introduction Exercise capacity in chronic obstructive pulmonary disease (COPD) is limited both by abnormal pulmonary mechanics, reported as breathlessness, and by leg muscle fatigue. To improve functional capacity it is important to understand the primary physiological constraint. Neural respiratory drive (NRD), measured using the diaphragm electromyogram expressed as a proportion of maximum (EMG_di%max), quantifies the load on the respiratory muscles imposed by abnormal pulmonary mechanics, and relates closely to breathlessness. We hypothesised that end-exercise EMG_di%max would be higher in patients stopping because of breathlessness than in those stopping because of leg fatigue.

Methods EMG_di, ventilation (V_E), oxygen consumption (VO₂) and ventilatory reserve (V_E/MVV%) were measured in 23 COPD patients (median (IQR) FEV₁ 39 (30.0 to 56.8)%predicted) during exhaustive cycle ergometry. Differences in physiological variables between groups of patients stopping because of breathlessness, leg fatigue or both were examined using 1-way ANOVA.

Results EMG_di%max was higher in patients stopping because of breathlessness (n = 12, EMG_di%max 75.7 (69.5 to 77.1)%) than in those stopping because of leg fatigue (n = 8, EMG_di%max 44.1 (39.4 to 63.3)%, p < 0.05). There were no significant differences in end-exercise V_E or VO₂. V_E/MVV% tended to higher levels in the breathless group.

Discussion These results suggest that patients limited by breathlessness due to ventilatory constraints can be identified as those reaching near-maximal levels of NRD during exercise. Measurement of EMG_di%max during exercise could prove useful in identifying patients whose functional performance would be best optimised by improving pulmonary mechanics rather than interventions to train peripheral muscle groups.