We compared the pharmacodynamic effects and hospital costs of three long-acting neuromuscular blocking drugs in a prospective, randomized, double-blind manner. Each neuromuscular blocking drug was administered with fentanyl (50 micrograms/kg) for intravenous induction of anesthesia for coronary artery bypass surgery. Each patient received twice the 95% effective dose (ED95) of either pancuronium (0.14 mg/kg, n = 10), pipecuronium (0.10 mg/kg, n = 10), or doxacurium (0.05 mg/kg, n = 10). Hemodynamic measurements were recorded at baseline, 5 min after completion of anesthetic induction, immediately after endotracheal intubation, and 5 min after intubation. Only small hemodynamic differences between neuromuscular blocking drugs were observed. Doxacurium (but not pancuronium or pipecuronium) significantly decreased mean arterial blood pressure (from 94 +/- 4 mm Hg before induction to 83 +/- 3 mm Hg 5 min after intubation); nevertheless, there were no significant between-group differences at any time. Pancuronium increased heart rate (from 68 +/- 4 beats/min before induction to 76 +/- 5 beats/min 5 min after intubation); however, pancuronium differed significantly from doxacurium and pipecuronium only 5 min after induction and 5 min after intubation. Central venous pressure, pulmonary artery occlusion pressure, cardiac index, and systemic and pulmonary vascular resistance indices did not change. Electrocardiographic abnormalities were observed in two pipecuronium patients: ST segment depression in one and premature ventricular contractions in another. No other electrocardiographic changes were observed. There were no significant between-group differences in the need for hemodynamic interventions.(ABSTRACT TRUNCATED AT 250 WORDS)