Abstract
Agitation in children in acute care settings poses significant patient and staff safety concerns. While behavioral approaches are central to reducing agitation and oral medications are preferred, parenteral medications are used when necessary to promote safety. The goal of this systematic review was to evaluate the effectiveness and safety of an ultra-short-acting parenteral medication, droperidol, for the management of acute, severe agitation in children in acute care settings. A systematic review of randomized controlled trials, observational studies, and case series/reports examined the effectiveness and safety of parenteral droperidol for management of acute agitation in patients ≤21 years old in acute care settings. Effectiveness outcomes included time to sedation and need for a subsequent dose of medication. Safety outcomes were adverse effects such as QTc prolongation, hypotension, respiratory depression, and dystonic reactions. A total of 431 unique articles were identified. Six articles met inclusion criteria: two in the prehospital setting, one in the emergency department, and three in the inpatient hospital setting. The articles included a prospective observational study, three retrospective observational studies, and two case reports. The largest study reported a median time to sedation of 14 min (interquartile range 10-20 min); other studies reported a time to sedation of 15 min or less. Across studies, 8%-22% of patients required a second dose of medication for ongoing agitation. The most frequent adverse effects were dystonic reactions and transient hypotension. One patient had QTc prolongation and another developed respiratory depression, but both had significant comorbidities that may have contributed. The risk of bias in included studies ranged from moderate to critical. Existing data on droperidol for management of acute agitation in children suggest that droperidol is both effective and safe for acute, severe agitation in children. Data are limited by study designs that may introduce bias.