The mechanism responsible for the attenuated heart rate (HR) response to exercise in patients with congestive heart failure (CHF) was investigated in 46 normal subjects and 59 patients with CHF stratified by peak exercise oxygen consumption (VO2). The peak exercise HR and the increment in HR from rest to peak exercise were decreased in CHF patients, and both correlated strongly with peak VO2 (r = 0.810, p less than 0.0001; r = 0.863, p less than 0.0001, respectively). Peak exercise norepinephrine level (NE) and the increment in NE from rest to peak exercise were not attenuated in CHF patients. Resting NE was elevated in CHF patients and correlated inversely with peak VO2 (r = -0.595, p less than 0.001). However, no significant correlation occurred between peak VO2 and either peak exercise NE or the exercise increment in NE. The ratio of the exercise increments in HR and NE, and indirect index of sinoatrial node sympathetic responsiveness, was markedly reduced in CHF patients and was inversely related to the severity of exercise impairment. Likewise, the HR response to a graded isoproterenol infusion was markedly reduced in CHF patients. Age-matching of normal subjects and CHF patients did not affect the foregoing observations. Infusion of CHF patients with the phosphodiesterase inhibitor milrinone caused a significant increase in the ratio of the exercise increments in HR and NE. These data strongly suggest that the attenuated HR response to exercise in CHF patients is due, at least in part, to postsynaptic desensitization of the beta-adrenergic receptor pathway.