Patients with heart failure exhibit an increased ventilatory response to exercise; their slope of the overall ventilation/carbon dioxide production ratio (VE/VCO2) is elevated. This elevation is related to impaired exercise performance and is commonly explained by an increased ventilation-perfusion mismatch. However, the concept of afferents to the respiratory centre modulating ventilation during exercise has been raised. In healthy subjects, ventilation during exercise is related to the hypercapnic ventilatory response during rebreathing. This is explained by a similar response of the respiratory centre to different stimuli. The aim of this study was to analyse the relationship between the ventilatory response to exercise and to carbon dioxide in patients with chronic heart failure. The hypercapnic ventilatory response was measured at rest using the rebreathing method in 31 patients with chronic heart failure and a left ventricular ejection fraction < 40% and 25 controls. Thereafter a maximal bicycle exercise test with evaluation of VE/VCO2 was performed. The maximal oxygen uptake during exercise was 13.1 +/- 5.2 ml. min-1.kg-1. The slope of the hypercapnic ventilatory response was normal (1.1 +/- 0.81. min-1. mmHg-1) but minute ventilation breathing room air was increased in the patients as compared to the controls. VE/VCO2 during exercise was positively correlated to the hypercapnic ventilatory response (r = 0.70; P < 0.00001). This relationship supports the concept that in patients with heart failure, ventilation during exercise is not only influenced by an increased ventilation-perfusion mismatch but is in part mediated by the responsiveness of the respiratory centre. Different afferents to the respiratory centre, such as central command or muscle ergoreflex may play a role in modulating ventilation during exercise.