Most patients hospitalized at tertiary care pediatric institutions receive at least 1 medication outside the terms of the Food and Drug Administration product license. Substantial variation in the frequency of off-label use was observed across diagnostic categories and drug classes. Despite the frequent off-label use of drugs, using an administrative database, we cannot determine which of these treatments are unsafe or ineffective and which treatments result in substantial benefit to the patient.

Objectives To describe a large multicenter cohort of pediatric cardiac arrest (CA) with return of circulation (ROC) from either the in-hospital (IH) or out-of-hospital (OH) setting in order to determine if significant differences related to pre-event, arrest event, early post-arrest event characteristics and outcomes exist that would be critical in planning a clinical trial of therapeutic hypothermia (TH). Design Retrospective cohort study Setting Fifteen Pediatric Emergency Care Applied Research Network (PECARN) sites. Patients Patients from 24 hours (h) to 18 years with either IH or OH CA who had a history of at least 1 minute of chest compressions and ROC for at least 20 minutes were eligible. Interventions None Measurements and Main Results A total of 491 patients met study entry criteria with 353 IH cases and 138 OH cases. Major differences between the IH and OH cohorts were observed for patient pre-arrest characteristics; arrest event initial rhythm described; and arrest medication use. Several post-arrest interventions were utilized differently, however, the use of TH was similar (<5%) in both cohorts. During the 0–12 h interval following ROC, OH cases had lower minimum temperature and pH, and higher maximum serum glucose recorded. Mortality was greater in the OH cohort (62% vs. 51%, p=0.04) with the cause attributed to a neurological indication much more frequent in the OH than IH cohort (69% vs. 20%; p < 0.01). Conclusions For pediatric CA with ROC, several major differences exist between IH and OH cohorts. The finding that the etiology of death was attributed to neurological indications much more frequently in OH arrests has important implications for future research. Investigators planning to evaluate the efficacy of new interventions such as TH should be aware that the IH and OH populations differ greatly and require independent clinical trials.