A comparison of propofol and ketamine/midazolam for intravenous sedation of children

OBJECTIVE: To compare ketamine and propofol sedation in children undergoing diagnostic and therapeutic procedures. DESIGN: Retrospective study. SETTING: A six-bed pediatric intensive care unit and a pediatric hematology oncology clinic. PATIENTS: From 1996 to 1998, 405 procedures were performed on patients between 1 month and 22 yrs of age. INTERVENTIONS: All patients but one were sedated intravenously with either propofol or ketamine; those who received ketamine also received midazolam and either atropine or glycopyrrolate. Vital signs were monitored continuously. Procedures included bone marrow biopsies, lumbar punctures, esophagoduodenoscopies, colonoscopies, and other miscellaneous procedures. A pediatric intensivist performed all sedations. MEASUREMENTS AND MAIN RESULTS: Two hundred sixty-one procedures were conducted with propofol and 144 with ketamine. The mean time (+/-sd) from administration of the first dose of medication until the patient was awake was 36.6 (15.0) mins for the propofol group and 69.2 (43.2) mins for the ketamine group. The mean time to awakening was significantly longer for the ketamine group, even after adjusting for the length of the procedure, American Society of Anesthesiologists score, and setting of procedure (inpatient or outpatient; p =.0001). Only one unplanned endotracheal intubation in the propofol group and two in the ketamine group occurred. Patients were significantly more likely to have airway (p =.01) or hemodynamic (p =.002) effects with propofol than with ketamine, although these effects were essentially minor in nature. CONCLUSIONS: Both propofol and ketamine provided safe and effective sedation for the short, painful procedures performed. Because the patients who received propofol awakened almost twice as quickly as the patients who received ketamine, the sedation service operated more efficiently when propofol was used. The major complication rates for propofol and ketamine were small, and the differences between the two groups were not statistically significant. We conclude that with proper monitoring, intravenous propofol can be used safely and effectively for short procedures in the pediatric setting.