Objective: to compare the efficacy of therapeutic hypothermia versus therapeutic normothermia in children and adolescents who were resuscitated after in-hospital cardiac arrest.
Design: multicentre randomised clinical trial with blind evaluation.
Setting: paediatric intensive care units at 37 children’s hospitals in the United States, Canada, and the United Kingdom.
Study sample: 329 children aged 48 hours to 18 years who experienced in-hospital cardiac arrest, received chest compressions for at least 2 minutes and remained dependent on mechanical ventilation after return of spontaneous circulation. The exclusion criteria were a score of 5 or 6 in the Glasgow Coma Scale motor response subscale, severe bleeding, pre-existing illness with a life expectancy of less than 1 year or decision by the clinical team to withhold aggressive treatment. The 329 children were randomly assigned to therapeutic hypothermia or therapeutic normothermia with the use of permuted blocks stratified according to clinical centre and age category. In the final sample, 161 received hypothermia and 160 normothermia.
Intervention: treatment was initiated within 6 hours of the return of spontaneous circulation, and consisted in maintaining the target temperature for 120 hours in both groups. The target central temperature in the hypothermia group was 33°C (range, 32°C to 34°C) for 48 hours, followed by rewarming for a minimum of 16 hours until reaching 36.8°C (range, 36°C to 37.5°C), which was maintained for the remainder of the 120 hours. The normothermia group was maintained at a target central temperature of 36.8°C (range, 36°C to 37.5°C) for the entire 120 hours.
Outcome measurement: the primary outcome was survival with a favourable outcome at 12 months of followup, defined as score of 70 or higher on the Vineland Adaptive Behavior Scales, second edition (which has a mean of 100 and a standard deviation of 15). The authors excluded 31 patients in the hypothermia group and 29 in the normothermia group from the analysis because they had a baseline score in the Vineland-II scale of less than 70 before cardiac arrest, or a score of 3 to 6 in the Pediatric Overall Performance Category or Pediatric Cerebral Performance Category scales (moderate disability , severe disability , coma  or brain death ). Secondary outcomes were survival at 12 months and change in neurobehavioral function, compared to baseline (before cardiac arrest).
Main results: the median age of the patients was 1 year, and the cause of arrest was cardiac in 65%. Fifty-one percent of patients in the hypothermia group and fifty-eight percent in the normothermia group were receiving extracorporeal membrane oxygenation at the time of randomization.
The trial was stopped prematurely due to futility (lack of efficacy) after randomization of 329 patients (out of the 558 initially expected). Among the 257 children with a baseline Vineland-II score of 70 or higher before arrest, there was no difference in the proportion alive at 1 year with a score of 70 or higher between the hypothermia and normothermia groups: 36% versus 39%, with a relative risk (RR) of 0.92 and a 95% confidence interval (95 CI) of 0.67 to 1.27. There were also no differences between these groups in mortality at 1 year (RR, 1.07; 95 CI, 0.85 to 1.34) or changes in neurobehavioural functioning at 1 year.
Conclusion: in comatose children that survived in-hospital cardiac arrest, therapeutic hypothermia did not confer a significant benefit with respect to survival with a good functional outcome at 1 year compared with therapeutic normothermia.
Conflicts of interest: several authors disclosed having receiving funds during the study.
Funding sources: grants from the National Heart, Lung and Blood Institute and other institutions