• 
• 

Three Part Question

[In a child who has suffered a cardiac arrest] does [induced hypothermia] improve [mortality and neurodisability]?

Clinical Scenario

A six year old boy with an asystolic cardiac arrest is successfully resuscitated in the A&E department, but he remains comatose and on a ventilator. The paediatric retrieval team is on its way. The anaesthetic consultant asks you whether, as is the case in adult medicine, induced hypothermia should be initiated.

Search Strategy

Medline, Embase and the Cochrane library were searched.
Secondary sources:

The Cochrane library was searched on 10/11/2012 using the following terms:

1. Heart arrest (MeSH)
2. “Cardiac arrest”
3. Cardiopulmonary resuscitation (MeSH)
4. #1 OR #2 OR #3
5. Induced hypothermia (MeSH)
6. “Induced hypothermia”
7. “Therapeutic hypothermia”
8. “Cooling”
9. #5 OR #6 OR #7 OR #8
10. #3 AND #9

Primary sources:

Medline was searched on 10/11/2012 using the following terms:

1. Heart arrest (MeSH)
2. “Cardiac arrest”
3. Cardiopulmonary resuscitation (MeSH)
4. #1 OR #2 OR #3
5. Induced hypothermia (MeSH)
6. “Induced hypothermia”
7. “Therapeutic hypothermia”
8. “Cooling”
9. #5 OR #6 OR #7 OR #8
10. #3 AND #9

Embase was searched on 10/11/2012 using the following terms:
"Heart arrest\" OR\\"Cardiac arrest\" OR \"cardiopulmonary resuscitation\") AND "induced hypothermia\"therapeutic hypothermia\" OR "Cooling\\") AND (Child*)

Search Outcome

198 reviews and trials were reviewed for relevance from the Cochrane library. One Cochrane protocol was identified .

Using the filters “humans” and “child: birth-18 years” and “English” 154 titles reviewed for relevance from Medline. Two studies were identified.

Using the filters “humans” and “English” 167 titles were reviewed for relevance from Embase. No new studies were identified that had not already been found on Medline.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Doherty 2012 Canada, UK	79 children (>40 weeks postconception and <18 years) with CA >3 minutes duration, survival for >12 hours after return of spontaneous circulation (ROSC), no birth asphyxia 29/79 HT vs. 50/79 standard therapy (ST)	Retrospective cohort.	Mortality at six months	Mortality: OR 1.99 (95% CI 0.45, 8.85; p=0.502)	- Groups not treated equally. Heterogeneity in practice (two centres did not use HT at all) - Two groups dissimilar: HT had longer duration of arrest (median 30mins vs. 10mins; p=0.002), and more use of ECMO (79.3% vs. 20.4%, p=<0.001) - Most CAs of a cardiac aetiology (55/79), thus not representative of all paediatric CAs
			Paediatric Cerebral Performance Category (PCPC) at six months	Unfavourable PCPC: OR 2.00 (95% CI 0.45, 9.01; p=0.364)
Fink 2010 USA	181 children (1 week to 21 years) received chest compressions and went on to have ROSC 40/181 HT vs. 141/181 ST	Retrospective cohort	Adverse events in first four days	a) RR of requiring insulin in HT 2.3 (95%CI 1.6 – 3.4, p<0.01)*	- Groups not treated equally after resuscitation - Two groups dissimilar. Children with longer duration of CA more likely to undergo HT - Children with congenital heart disease excluded (a group that may benefit most from HT) - Mortality and disability were secondary outcomes only - Multivariate analysis not performed for disability, and no long-term follow up
			Hospital mortality	Multivariate analysis HT vs ST OR 0.47 (95% CI 0.15, 1.45; p=0.2)
			Disability - Glasgow Outcome Score (GOS) at discharge	No difference in GOS score between HT (2.4 +/-1.7) and ST (2.4 +/- 1.6) p=0.9

Comment(s)

Childhood cardiac arrest (CA) is rare but has poor outcomes. In children suffering an out-of-hospital CA only 0.3%-4.0% survive neurologically intact . Induced hypothermia (HT) is an established therapy for improving neurological outcome and mortality in neonates with hypoxic ischaemic encephalopathy , and in adults following CA . International resuscitation guidelines suggest HT may be considered for children who remain comatose after CA . In practice the therapy is only occasionally used in the UK. In a survey of 77 paediatric emergency department consultants, 63% said that they had never used HT. Most CAs in children are due to gradual respiratory or cardiovascular decompensation leading to bradycardia or asystole; whereas in adults the majority of CAs are cardiac in origin and due to ventricular arrhythmia. In neonates HT is used specifically in the context of birth asphyxia. It is therefore not possible to confidently infer efficacy of HT in children on the basis of neonatal or adult studies. The evidence in children is insufficient to recommend or refute the use of HT after CA. Two relevant studies succeed in demonstrating the feasibility of HT in children; however both studies have similar methodological weaknesses. They are observational studies and thus cannot demonstrate causality. Data was gathered retrospectively from clinical notes and is prone to information bias (particular with regard to adverse events). Study numbers are small with adjusted results for mortality and disability that do not show a statistically significant difference between groups. Data from Fink et al suggests an association between HT and hyperglycaemia and electrolyte imbalance, and that ‘over-cooling’ might be dangerous. The studies are beset by heterogeneity in the allocation of children to HT (selection bias) and in the way HT is carried out. Both studies are subject to confounding where the sickest children are more likely to undergo HT. The results of a randomised, controlled trial are awaited .

Clinical Bottom Line

• Based on extrapolation from adult data, international resuscitation guidelines suggest induced hypothermia may be considered following cardiac arrest in children. • There is currently insufficient evidence to be able to establish the efficacy or safety of induced hypothermia in children.

References

1. Doherty et al Hypothermia Therapy After Pediatric Cardiac Arrest Circulation 2009;119:1492-1500
2. Fink et al A tertiary centre\'s experience with therapeutic hypothermia after pediatric cardiac arrest Pediatr Crit Care Med 2010;11:66 –74