Thirteen patients with advanced congestive heart failure (CHF) were treated with prazosin. Following the first dose, cardiac output (CO) (mean +/- SD) rose from 3.2 +/- 1.2 to 4.3 +/- 1.1 L/min, pulmonary artery diastolic pressure (PAD) decreased from 23 +/- 12 to 18 +/- 11 mm Hg, mean arterial pressure (MAP) decreased from 85 +/- 10 to 76 +/- 10 mm Hg, and heart rate did not change (92 +/- 15 vs 92 +/- 14 bpm). At the end of a 48 to 72 hour titration to an optimal regimen, significant effects on CO (3.2 +/- 1.1 vs 4.5 +/- 1.3 L/min), PAD (24 +/- 12 vs 18 +/- 8 mm Hg), and MAP (84 +/- 10 vs 76 +/- 10 mm Hg) were still seen. The patients were restudied after 3 months of treatment. In contrast to reports of rapid development of tolerance to prazosin, we found continued beneficial effects on CO (3.0 +/- 1.3 vs 3.8 +/- 1.0 L/min) and PAD (23 +/- 12 vs 18 +/- 10 mm Hg), without significant change in MAP (81 +/- 11 vs 78 +/- 8 mm Hg). We found wide variability in the CO rise in response to prazosin, which was not accounted for by differences in plasma prazosin concentration. Systemic vascular resistance in the untreated state did correlate with the percentage change in CO. In addition, excessive lowering of the PAD appeared to blunt the CO response in some cases.