No Benefit of Hypothermia After In-Hospital Cardiac Arrest in Children

Pauline Anderson

January 31, 2017

Therapeutic hypothermia does not provide additional benefits to comatose children who survive an in-hospital cardiac arrest, results of a randomized trial show.

Researchers found children undergoing therapeutic hypothermia had similar rates of 12-month survival, with a favorable neurobehavioral outcome, compared with those who received care that involved actively keeping the body temperature at normal levels, referred to as therapeutic normothermia.

However, the study was not large enough to detect whether any subset of patients may have benefited.

"Our study found no evidence of improved survival or better functional outcome with therapeutic hypothermia compared to therapeutic normothermia in infants and children experiencing a cardiac arrest in-hospital, and that's important to know," said lead author Frank W. Moler, MD, professor of pediatrics, University of Michigan Health System, Ann Arbor.

The study, published January 25 in the New England Journal of Medicine (NEJM), was halted early because of futility, as assessed by a review of interim efficacy analyses by the data and safety monitoring board, before attaining the target trial enrollment.

Earlier research in adults had shown that cardiac patients who were cooled to 33.0°C rather than receiving usual care had better neurologic outcomes.

"Everyone thought it was due to the cooling, but they didn't account for the fact that a lot of people in the usual treatment group were getting fever," said Dr Moler.

He and his colleagues carried out two Therapeutic Hypothermia after Pediatric Cardiac Arrest (THAPCA) studies, with the same design except that one was in the out-of-hospital (OH) setting and the other was in the in-hospital (IN) setting.

The studies compared the efficacy of therapeutic hypothermia (target temperature 33.0°C) with that of therapeutic normothermia (target temperature 36.8°C) in comatose children and adolescents who were resuscitated after cardiac arrest.

Therapeutic hypothermia was achieved by using a commercial unit (Blanketroll III, Cincinnati Sub-Zero), consisting of a blanket applied front and back. Targeted temperature management was actively maintained for 120 hours in each group. Other than temperature targets, the two groups received identical care.

The OH study, published in the NEJM in 2015, found no significant between-group difference in survival, with a good neurobehavioral outcome at 12 months. It was not halted early.

The current study included children in pediatric intensive care units at 37 children's hospitals in the United States, Canada, and the United Kingdom. Eligible participants were older than 48 hours and younger than 18 years, had experienced a cardiac arrest within the hospital, received chest compression for at least 2 minutes, and remained dependent on mechanical ventilation after return of circulation.

Primary Outcome

The primary outcome was 12-month survival with a favorable neurobehavioral outcome, defined as an age-corrected standard score of 70 or higher (on a scale of 20 to 160) on the Vineland Adaptive Behavior Scales second edition (VABS-II).

This measure has an age-corrected mean score of 100 and a standard deviation of 15. Higher scores indicate better function.

Children with a VABS-II score of less than 70 before cardiac arrest (as gleaned from caregiver questionnaires) were excluded from the primary efficacy analysis.

Of those screened, 329 patients underwent randomization; 166 were assigned to hypothermia and 163 to normothermia. Due to not receiving the intervention, ineligibility, or missing data, 257 patients could be evaluated for the primary outcome, 317 for the secondary outcome of change in neurobehavioral function, and 327 for the secondary outcome of 1-year survival.

The median age of the study patients was 1 year; 60% of patients were male, and 91% had a pre-existing medical condition. The initial rhythm was bradycardia in 57% and ventricular fibrillation or ventricular tachycardia in 10%.

The median time from cardiac arrest to cardiopulmonary resuscitation (CPR) was 0 minutes, and the median duration of CPR was 22.0 minutes.

"The kids in these hospitals have the best resuscitation teams in the world; almost all of them were being monitored, so resuscitation started immediately," commented Dr Moler.

The median time from return of circulation to the initiation of treatment was 4.9 hours in the hypothermia group and 4.7 hours in the normothermia group.

The percentage of children with a VABS-II score of 70 or higher at 12 months did not differ significantly (36% for the hypothermia and 39% for the normothermia groups; relative risk 0.92; 95% confidence interval [CI], 0.67 - 1.27; P = .63).

The secondary outcome of change in the VABS-II score from baseline to 12 months didn't differ significantly between groups (P = .70). The overall percentage of patients with 12-month VABS-II scores that did not decrease by more than 15 points also did not differ significantly (30% for the hypothermia and 29% for the normothermia groups).

The rate of survival at 12 months among the 327 patients with known vital status did not differ significantly (49% for the hypothermia and 46% for the normothermia groups; relative risk, 1.07; 95% CI, 0.85 - 1.34; P =.56).

As for safety, the incidence of infection, blood product use, and serious arrhythmias within 7 days of randomization were similar between groups. Mortality at 28 days also did not differ significantly.

The leading cause of cardiac arrest was a cardiac cause (50%). In contrast, in the OH study, 72% of patients had a respiratory cause of cardiac arrest.

Brain death or withdrawal of life support owing to a poor neurologic prognosis was the cause of death in about 79% of patients in the OH trial but in only 36% of those in the IH trial.

Unknown Factors

While this new research shows that the two therapeutic approaches are essentially equal, Dr Moler stressed that "some unknowns" are involved in therapeutic hypothermia. For example, doctors may not know how lower body temperature affects clearance of chemotherapy or other potentially toxic drugs.

Although it's a relatively simple approach, therapeutic hypothermia is not taught in medical school and requires some level of training, said Dr Moler.

"For the vast majority of patients, it looks like you can use either temperature target; but it's probably a bit safer to be closer to normal if you don't do this intervention all the time."

Dr Moler acknowledged that there may be some patients in whom therapeutic hypothermia might be preferable, but the study wasn't large enough to capture them. "We don't know whether there was a subset within the whole population that benefited."

Halting the trial early was an important limitation to the study, according to the authors. Termination of enrollment was based primarily on the low conditional power of the trial to show a significant treatment effect if continued since no trend was observed with response to the primary or secondary outcomes.

Another possible limitation, they said, was the relatively long time from the return of circulation to the achievement of a temperature within the target temperature range (a median of about 6 hours).

Many questions remain unanswered. Among them is whether a shorter therapeutic window for attaining the target temperature would have affected the outcomes.

It's also not clear whether a longer or shorter duration of temperature control, or a higher or lower target temperature, would have made a difference.

"Nobody knows what the outcome would be if you make patients colder, or colder for a longer period of time," said Dr. Moler.

But he stressed that there are potential dangers — for example, sepsis — to lowering the body temperature too much.

Reached for a comment, Allan de Caen, MD, pediatric intensivist, Stollery Children's Hospital, Edmonton, Alberta, Canada, said that although terminated early "due to futility of statistically discerning a direction for the primary endpoint," the study had an "excellent design."

"The message that will need to be maintained and reinforced is that this is not a negative study as such; it actually reinforces that outcomes can be reasonable as long as fever post-ROSC [return of spontaneous circulation] is avoided, using either of these therapeutic approaches."

Some centers continue to use moderate hypothermia after ROSC for pediatric IH cardiac arrest, said Dr de Caen.

"This does not negate that as an option; it just reinforces that using normal core temperature — and avoidance of fever — post-ROSC is equally acceptable."

Would a different target temperature have changed things? "Potentially," he said. "But only further study can determine whether a different target, or a different way of reaching the target, or a different duration of cooling, would alter the outcome."

The trial was supported by grants from the National Heart, Lung, and Blood Institute, cooperative agreements from the Pediatric Emergency Care Applied Research Network and the Collaborative Pediatric Critical Care Research Network, and a National Emergency Medical Services for Children Data Analysis Resource Center Demonstration grant. Disclosure forms for coauthors are available at

N Engl J Med. Published online January 24, 2017. Abstract

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.