Whereas both oxygen therapy and nasal continuous positive airway pressure (CPAP) decrease the apnea-hypopnea index (AHI) in patients with Cheyne-Stokes respiration (CSR) and congestive heart failure (CHF), only nasal CPAP is known to affect the left ventricular ejection fraction (LVEF). We therefore evaluated the effects of 1 month of nocturnal oxygen therapy on LVEF. Ten patients (52 +/- 12 years) with CHF (LVEF of 12% +/- 5%) and CSR (AHI 57 +/- 61 events/hour) were studied. Polysomnograms identified CSR and were repeated on oxygen initially (oxygen night 1 [2 L/min]) and after 30 nights (oxygen night 2). LVEF was measured by radionuclide ventriculography. Oxygen therapy decreased the AHI from a baseline of 57 +/- 61 to 9 +/- 11 and 12 +/- 17 events per hour during oxygen nights 1 and 2, respectively (p < .05), with no difference between treatment nights. The lowest oxygen saturation increased during oxygen nights 1 and 2, from a baseline of 87% +/- 7% to 94% +/- 4% and 91% +/- 7%, respectively (p < .05), with no difference between treatment nights. The LVEF did not significantly change from a baseline of 22% +/- 11% to 19% +/- 9% after 1 month of nocturnal oxygen (p = .05). Compared to baseline, there was no change in circulation time during oxygen nights 1 and 2, from 24 +/- 8 seconds to 30 +/- 15 seconds and 23 +/- 6 seconds, respectively (p = .2). Total sleep time, sleep efficiency, and sleep architecture, when compared with baseline, remained unchanged during both oxygen therapy nights. Although 1 month of nocturnal oxygen therapy decreases the AHI in patients with CSR and CHF, there is no improvement in left ventricular function.