The present study was designed to investigate, in patients with severe heart failure, the dose-dependent acute hemodynamic effects of celiprolol versus those of esmolol. Celiprolol is a beta 1 -receptor blocker with vasodilating properties, whereas esmolol is an ultra-short-acting beta 1 -blocker. Included were 14 patients with decompensated chronic heart failure (NYHA class IV) due to coronary heart disease (n = 8) or to dilated cardiomyopathy (n = 6). Each patient received both celiprolol and esmolol in random fashion. The beta-blockers were administered in four dose tiers, with an increase in dosage every 15 min. Hemodynamic measurements were obtained with a Swan-Ganz thermodilution catheter. Administration of celiprolol (5, 10, 20, and 50 microg/kg) took place intravenously. After intravenous administration of a loading dose of 500 microg/kg, we continuously infused esmolol at increasing doses, which were individually titrated for each patient. Mean infusion rates of esmolol were as follows: 40, 75, 140, and 230 micromol/kg per minute. Celiprolol and esmolol induced a comparable dose-dependent decrease in heart rate to a minimum of -10% below baseline. Esmolol caused a significant dose-dependent decrease (-25% below baseline at the highest dose level) in cardiac index (CI). After administration of celiprolol, CI decreased only transiently (-10% below baseline at the second and third dose level) and did not differ from the baseline at the highest dose level. For treatment of severe heart failure, initiation of intravenous beta-blocker therapy with low doses of a beta 1 -blocker with vasodilating effects may have hemodynamic advantages over conventional beta-blockade.