We hypothesized that (1) patients with congestive heart failure (CHF) and Cheyne-Stokes respiration with central sleep apnea (CSR-CSA) would have greater nocturnal urinary and daytime plasma norepinephrine concentrations (UNE and PNE, respectively) than those without CSR-CSA because of apneas, hypoxia and arousals from sleep and (2) attenuation of CSR-CSA by nasal continuous positive airway pressure (NCPAP) would reduce UNE and PNE concentrations. Eighteen patients with and 17 without CSR-CSA (Non-CSR-CSA group) were studied. Left ventricular ejection fraction was similar in the two groups, but overnight UNE and awake PNE concentrations were greater in the CSR-CSA group (30.2 +/- 2.5 nmol/mmol creatinine and 3.32 +/- 0.29 nmol/L) than in the Non-CSR-CSA group (15.8 +/- 2.1 nmol/mmol creatinine, p < 0.005, and 2.06 +/- 0.56 nmol/L, p < 0.05, respectively). Patients with CSR-CSA were randomized to a control group or to nightly NCPAP for 1 mo. CSR-CSA was attenuated in the NCPAP but not in the control group. The NCPAP group experienced greater reductions in UNE and PNE concentrations (-12.5 +/- 3.3 nmol/mmol creatinine and -0.74 +/- 0.40 nmol/L) than did the control group (-1.3 +/- 2.8 nmol/mmol creatinine, p < 0.025 and 1.16 +/- 0.66 nmol/L, p < 0.025, respectively). In conclusion, in patients with CHF, CSR-CSA is associated with elevated sympathoneural activity, which can be reduced by NCPAP.