ObjectiveTo compare the safety and effectiveness of propofol (2,6-diisopropylphenol) to midazolam for sedation of mechanically ventilated patients after coronary artery bypass grafting.DesignOpen, randomized, prospective trial.SettingCardiothoracic intensive care unit (ICU), Cleveland Clinic Foundation.PatientsEighty-four patients with normal or moderately impaired left ventricular function who underwent elective coronary artery bypass graft surgery under high-dose opioid anesthesia.InterventionsPatients were randomly selected to receive either propofol (mean loading dose 0.24 mg/kg; mean maintenance dose 0.76 mg/kg/hr) or midazolam (mean loading dose 0.012 mg/kg; mean maintenance dose 0.018 mg/kg/hr). Infusion rates were titrated to keep patients comfortable, drowsy, and responsive to verbal stimulation. Study duration, 8 to 12 hrs; infusions were started in the ICU when patients were awake and hemodynamically stable.Measurements and Main ResultsDuring therapy, both groups had lower mean arterial pressures and heart rates compared with baseline measurements; however, the propofol group had significantly lower heart rates than the midazolam group during the first 2 hrs of infusion. The propofol group also had significantly lower blood pressure measurements 5 and 10 mins after the initial dose, although there was no difference during infusion. Baseline cardiac output was measured before starting the infusion, and measurements were repeated during continuous infusion at 4, 8, and 12 hrs. Cardiac output values were similar. Propofol maintenance infusions ranged from 3 to 30 μg/kg/min and midazolam infusions ranged from 0.1 to 0.7 μg/kg/min. At these infusion rates. both groups had adequate sedation, based on nurse and patient evaluations; however, the propofol group used significantly lower total doses of sodium nitroprusside and supplemental opioids.ConclusionsBoth propofol and midazolam provided safe and effective sedation of coronary artery bypass graft patients recovering from high-dose opioid anesthesia. The reduced need for both antihypertensive medication and opioids seen in the propofol group may be advantageous. However, the hypotension seen after the initial bolus dose of propofol may be a concern. No difference between the two drugs could be demonstrated in time to extubation or ICU discharge, although it is probable that time to extubation was governed more by residual operative opioids than the study agents. (Crit Care Med 1994; 22:1415–1423)