Midazolam vs. Pentobarbital for Pediatric Sedation

Abstract & Commentary

Source: Moro-Sutherland DM, et al. Comparison of intravenous midazolam with pentobarbital for sedation for head computed tomography imaging. Acad Emerg Med 2000;7: 1370-1375.

These authors sought to determine whether pentobarbital is a more effective sedative than midazolam for young patients undergoing computed tomography (CT) of the head. Children ages 6 months to 6 years were eligible if they required a head CT for any reason, needed intravenous sedation, and had no underlying cardiopulmonary disease. After obtaining parental consent, the children were randomized to receive pentobarbital or midazolam. Pentobarbital was given as a 2.5 mg/kg IV dose followed by two 1.25 mg/kg doses at 1-minute intervals (total dose of 5 mg/kg over 3.5 minutes). Midazolam was given as a 0.1 mg/kg IV dose followed by two 0.05 mg/kg doses. Each midazolam dose was given over 2 minutes and was followed by a 2-minute wait (total dose of 0.2 mg/kg over 10 minutes). The authors found it impossible to blind the staff to the medication being administered.

The primary end point of the study was the quality of sedation, which was judged to be "good" if the patient was sedated and the CT completed; "adequate" if the patient was not sedated but the study completed; and "poor" or "unsuccessful" if the study was non-diagnostic or was not performed because of poor sedation.

Fifty-five subjects were enrolled over 2.5 years. In the pentobarbital group, 28 of 29 patients had good sedation and were successfully scanned. In the midazolam group, three of 26 had good sedation and another two had adequate sedation. Most of the subjects in the midazolam group subsequently were given pentobarbital to allow completion of the CT. There were no serious adverse events in either group; however, four patients receiving pentobarbital had mild oxygen desaturation. Because so few patients receiving midazolam had successful scans, the authors presented no formal analyses of the differences between the groups, and concluded that IV pentobarbital is more effective than IV midazolam in sedating young children for head CT.

Comment by David J. Karras, MD, FAAEM, FACEP

The authors attempted to perform a carefully controlled study of two sedative agents for young children. Unfortunately, major limitations of this study’s design leave the question unsettled. The study was unblinded, introducing the possibility of observer bias, and data collection was performed by staff who were aware of which drug was being administered. Secondly, midazolam was administered over a much longer period of time than was pentobarbital. The most important concern, however, is that the total dose of midazolam (0.2 mg/kg) was on the low end of the therapeutic range for achieving sedation in young children. The manufacturer’s current recommendation is that a total midazolam dose of 0.6 mg/kg (given in 0.05-0.1 mg/kg increments) may be necessary to achieve procedural sedation in this age group. These limitations detract significantly from what appears to be a fairly straightforward study, making it difficult to accept the authors’ conclusions regarding any advantages of pentobarbital over midazolam for pediatric sedation.