Does adjunctive midazolam reduce recovery agitation after ketamine sedation for pediatric procedures? A randomized, double-blind, placebo-controlled trial

Study objective: Despite widespread use of adjunctive benzodiazepines durin g ketamine sedation, their efficacy in reducing recovery agitation in child ren has never been studied. We wished to characterize the nature and severi ty of recovery agitation after ketamine sedation in children treated in the emergency department and to determine whether the addition of adjunctive m idazolam reduces the magnitude of such recovery agitation. Methods: The study was a randomized, double-blind, clinical trial of adjunc tive midazolam versus placebo during ketamine sedation. We enrolled 104 chi ldren aged 12 months to 15 years (median age, 6 years) at a combined univer sity medical center and children's hospital. Subjects received either intra venous midazolam (0.05 mg/kg up to 2 mg) or placebo after intravenous admin istration of a ketamine loading dose (1.5 mg/kg). Treating physicians and n urses independently noted the presence of crying, hallucinations, and night mares during recovery and graded recovery agitation by using a 100-mm visua l analog scale. Preprocedure agitation and external stimulation during reco very were also graded. The time from ketamine injection until each subject met the recovery criteria was recorded. Results: Fifty-three subjects received midazolam, and 51 received placebo. Potentially confounding variables were similar between the groups. Sedation efficacy, adverse effects, and recovery time were also similar between gro ups. Interobserver agreement between physician and nurse assessments was su bstantial. Median physician assessment of recovery agitation was 4 mm (inte rquartile range, 2 to 19) in the midazolam group and 5 mm (interquartile ra nge, 3 to 14) in the placebo group (difference -1; 95% confidence interval -3 to 2; P=.705). Recovery agitation was moderately correlated with preproc edure agitation (p=0.486) but not with external stimulation during recovery (p=0.147). Conclusion: Recovery agitation is common but generally of very low magnitud e after ketamine sedation in children in the ED. We observed a median physi cian rating of 5 mm on a 100-mm visual analog scale, a score that we believ e to be clinically insignificant. The degree of recovery agitation after ke tamine sedation is significantly related to the degree of preprocedure agit ation. In this study, concurrent midazolam did not diminish such agitation and had no measurably beneficial effect. Use of adjunctive benzodiazepines in pediatric ketamine sedation appears unnecessary.