Effects of inhaled carbon dioxide and oxygen on Cheyne-Stokes respiration in patients with heart failure

We hypothesized that in patients with congestive heart failure (CHF), reductions in Pa-CO2 sensed at the peripheral chemoreceptors trigger central apneas during Cheyne-Stokes respiration (CSR-CSA), and that raising Pa-CO2 by inhalation of a CO2 would eliminate these events. The effects of CO2 inhalation on the frequency of apneas and hypopneas during stage 2 (S2) sleep were studied in 10 CHF patients with CSR-CSA. The time from the breath with the minimal end tidal fraction of CO2 (FETCO2) during hyperpnea until the onset of apnea correlated strongly with the lung to ear circulation time (LECT) (r(2) = 0.90, p < 0.0001), a measure of lung to carotid body circulatory delay. Among the six patients who also inhaled O-2, CO2 inhalation increased transcutaneous Pco(2) (Ptc(CO2)) (36.4 +/- 4.6 mm Hg versus 38 +/- 4.4 mm Hg, p < 0.002), abolished central apneas and hypopneas (43.0 +/- 8.4 per hour on air versus 1.6 +/- 2.6 per hour on CO2, p < 0.0001), and increased Sa(O2). In contrast, O-2 inhalation causing a similar rise in Sa(O2) had no significant impact on either Ptc(CO2) or the frequency of central events. We conclude that central apneas in patients with CHF are triggered by a low Pa-CO2 most likely sensed at the peripheral chemoreceptors, and that inhalation of CO2 reverses central apneas by raising Pa-CO2.