Breathing is controlled by a negative-feedback system in which an increase in the partial pressure of arterial carbon dioxide stimulates breathing and a decrease inhibits it. Although enhanced sensitivity to carbon dioxide helps maintain the partial pressure of arterial carbon dioxide within a narrow range during waking hours, in some persons a large hyperventilatory response during sleep may lower the value below the apneic threshold, thereby resulting in central apnea. I tested the hypothesis that enhanced sensitivity to carbon dioxide contributes to the development of central sleep apnea in some patients with heart failure. This prospective study included 20 men who had treated, stable heart failure with left ventricular systolic dysfunction. Ten had central sleep apnea, and 10 did not. The patients underwent polysomnography and studies of their ventilatory response to carbon dioxide. Patients who met the criteria for central sleep apnea had significantly more episodes of central apnea per hour than those without central sleep apnea (mean [+/-SD], 35+/-24 vs. 0.5+/-1.0 episodes per hour). Those with sleep apnea also had a significantly larger ventilatory response to carbon dioxide than those without central sleep apnea (5.1+/-3.1 vs. 2.1+/-1.0 liters per minute per millimeter of mercury, P=0.007), and there was a significant positive correlation between ventilatory response and the number of episodes of apnea and hypopnea per hour during sleep (r=0.6, P=0.01). Enhanced sensitivity to carbon dioxide may predispose some patients with heart failure to the development of central sleep apnea.