Propofol (Diprivan), a modern intravenous hypnotic, produces a reduction in both cardiac index (CI) and mean arterial pressure (MAP). Ketamine (Ketanest), a potent analgesic, in contrast, causes an increase in MAP and CI. The aim of the present study was to investigate whether the combination of propofol and ketamine can give better hemodynamic stability during the induction and maintenance of general anesthesia than propofol used with fentanyl, whose cardiodepressant actions may cumulate. METHODS. For induction of general anesthesia 10 patients (ASA I and II) each received 3-5 boluses of propofol (0.5 mg.kg-1 during 35 s until predetermined level of anesthesia was reached (stage D2/E0 according to ) followed by a continuous propofol infusion (0.120 mg.kg-1.min). Fentanyl 0.1 mg was administered to each patient in group A for induction of anesthesia and again if evident pain was present. In group B ketamine was given following a pharmacokinetic model based on computer-simulated calculation. After an initial bolus of 38 mg injected within 2 min further doses of 42 mg, 35 mg, 32 mg and 28 mg ketamine were administered over 30 min at a time. Signs of evident pain were treated by means of supplementary doses of 0.5 mg.kg-1. RESULTS. In both groups a moderate drop of MAP was observed after the induction of general anesthesia. Two patients in each group showed a distinct decrease in MAP (-32%). The heart rate dropped slightly (-9%) in group A, but did not change in group B. Following intubation the MAP rose by less in group A (+8%) than in group B (+21%). After the beginning of the operation the group treated with propofol/fentanyl showed major hemodynamic changes; in particular, bradycardia with less than 40 bpm was observed in more patients than in the propofol/ketamine group. Postoperatively, fewer patients in group B required rescue doses of analgesics (1 of 10) than these in group A (7 of 10), though vigilance was better in group B. DISCUSSION. The dose of ketamine administered during the induction of general anesthesia may have been not high enough to neutralize the cardiodepressant effect of propofol. But during the maintenance of anesthesia there was in fact better hemodynamic stability in group B than in group A as a result of the neutralization of opposing actions. Fentanyl even intensified the fall in MAP after propofol. Patients in group B showed better vigilance as well as better pain relief postoperatively. The population of the fentanyl group was obviously more deeply sedated and analgesia was still inadequate. In our study general intravenous anesthesia with propofol and ketamine offered the advantages of better analgesia, a higher state of vigilance and the absence of respiratory depression during the postoperative phase compared with the combination of propofol and fentanyl.