In the last few years several Authors reported positive results using beta-blocking therapy in chronic heart failure. They also observed a group of patients who did not improve after this treatment. The aim of our study was comparing hemodynamic, neurohumoral and heart rate responses to beta-blockers in 2 groups of patients: Group A, patients who improved and Group B, patients who did not improve their clinical conditions after beta-blockade. We studied 26 patients with chronic heart failure of different origin: coronary artery disease and idiopathic dilative cardiomyopathy. They were treated with atenolol 50 mg/die for at least 1 year. Patients were divided into 2 groups: Group A including patients who increased the exercise time and the peak VO2 (> or = 2 ml/min/kg) and Group B, patients who did not increase exercise time and peak VO2 during beta-blockers. In Group A patients we observed a correlation between heart rate and end-diastolic volume by scintigraphy. We did not observe this correlation in Group B patients. The ejection fraction, evaluated by scintigraphy, significantly increased in Group A, while in Group B did not change. In both groups of patients we observed that plasmatic norepinephrine decreased significantly. In our opinion, this reduction is not due to an amelioration of organ clearance, because it was observed even in patients who did not show an increase of cardiac index and of ejection fraction. We suggest that the decrease of plasmatic norepinephrine might be the effect of beta-blockade and that it is not related to clinical response after treatment.