Patients with heart failure have increased vascular resistance and evidence for increased neurohumoral drive. High levels of circulating norepinephrine are found in patients with heart failure, but it is not known whether they reflect increased sympathetic neural activity or result from altered synthesis, release, or metabolism of norepinephrine. We used microneurography (peroneal nerve) to directly record sympathetic nerve activity to muscle (mSNA) and also measured plasma norepinephrine levels in patients with heart failure and in normal control subjects. Our goal was to determine whether sympathetic nerve activity is increased in patients with heart failure and whether plasma norepinephrine levels correlate with levels of mSNA in heart failure. Resting muscle sympathetic nerve activity in 16 patients with moderate to severe heart failure (54 ± 5 bursts/min, mean ± SE) was significantly higher (p < .01) than the levels of activity in either nine age-matched normal control subjects (25 ± 4 bursts/min) or 19 'young' normal control subjects (24 ± 2 bursts/min). We found a significant correlation between plasma norepinephrine levels and mSNA (r = .73, p < .05). Neither mSNA nor plasma norepinephrine levels correlated with total systemic vascular resistance, cardiac index, left ventricular ejection fraction, or heart rate. However, both mSNA and plasma norepinephrine levels showed significant positive correlations (p < .05) with left ventricular filling pressures (r = .80, mSNA vs filling pressures; r = .82, norepinephrine levels vs filling pressures) and mean right atrial pressure. The results of the study provide the first direct evidence of increased central sympathetic nerve outflow in patients with heart failure and the first direct evidence that plasma norepinephrine levels show a reasonable correlation with sympathetic nerve activity to muscle in these patients. Furthermore, the data suggest that preload is an important determinant of SNA in these patients.