Objective At Boston Children’s Hospital, over 3000 children are sedated annually using established protocols.

Methods Since 1997, a computerized database has tracked quality assurance (QA) data for each sedation. Prior to 2005, routine sedation consisted of intravenous pentobarbital (Nembutal) with fentanyl added if needed. Although the safety and efficacy of this regimen was acceptable, the risk of a prolonged sedation, lengthy recovery time and pentobarbital rage motivated the Sedation Committee to seek alternative agents. Dexmedetomidine, an alpha-2 agonist, was trialled as the primary intravenous sedative to replace pentobarbital.

Results Our original Dexmedetomidine protocol reflected the Food and Drug Administration recommendations for adult dosing. The outcome data for 16,000 pentobarbital sedations showed an 0.4% incidence of significant adverse events: 0.35% incidence of oxygen desaturation, 0.06% incidence of brief positive pressure ventilation and no incidences of cardiac arrest or endotracheal intubation. The outcome data for Dexmedetomidine would have to be comparable to pentobarbital in order to justify a practice change. Over 3 years, following careful review of all QA data, the dosing increased. Currently we have replaced pentobarbital with Dexmedetomidine, a medication with a shorter half life, less risk of respiratory depression and a shorter recovery time.

Conclusion The QA tool must be carefully designed to reliably query critical information. Adverse events should only be entered after agreement by independent blinded reviewers. With a reliable QA database, information may be queried, protocols modified and new protocols may be trialled and implemented.