ObjectiveThe aim of the study was to compare the safety concerning cerebral hemodynamics of ketamine and sufentanil used for sedation of severe head injury patients, both drugs being used in combination with midazolam.DesignProspective, randomized, double-blind study.SettingIntensive care unit in a trauma center.PatientsTwenty-five patients with severe head injury.InterventionsTwelve patients received sedation with a continuous infusion of ketamine-midazolam and 13 with a continuous infusion of sufentanil-midazolam. All patients were mechanically ventilated with moderate hyperventilation.Measurements and Main ResultsPrognostic indicators (age, Glasgow Coma Scale scores, computed tomography diagnosis, and Injury Severity Scale score) were similar in the two groups at study entry. Measurements were carried out during the first 4 days of sedation. The average infusion rates during this time were 82 ± 25 &mgr;g·kg−1·min−1ketamine and 1.64 ± 0.5 &mgr;g·kg−1·min−1midazolam in the ketamine group and 0.008 ± 0.002 &mgr;g·kg−1·min−1sufentanil and 1.63 ± 0.37 &mgr;g·kg−1·min−1midazolam in the sufentanil group. No significant differences were observed between the two groups in the mean daily values of intracranial pressure and cerebral perfusion pressure. The numbers of intracranial pressure elevations were similar in both groups. The requirements of neuromuscular blocking agents, propofol, and thiopental were similar. Heart rate values were significantly higher in the ketamine group on therapy days 3 and 4 (p< .05). With regard to arterial pressure control, more fluids were given on the first therapy day and there was a trend toward greater use of vasopressors in the sufentanil group. Sedative costs were similar in the two groups.ConclusionThe results of this study suggest that ketamine in combination with midazolam is comparable with a combination of midazolam-sufentanil in maintaining intracranial pressure and cerebral perfusion pressure of severe head injury patients placed under controlled mechanical ventilation.