Anna Ings. Therapeutic hypothermia, Don't Forget the Bubbles, 2013. Available at:
A 4 year old girl presents to ED in cardiac arrest. After a dramatic resus, you and the team were able to bring her back but she remains unconscious.
People around you start to mention therapeutic hypothermia (TH).
You start to panic and sweat as you have heard of it in adults, but aren’t sure if it works in children.
- TH may be beneficial post paediatric cardiac arrest, but most of our evidence is extrapolated from adult data and the best evidence for this is for VF/VT out of hospital arrests
- There is no evidence for TH in severe traumatic brain injury (TBI)
- Being too cool (<320C) or too hot (>380C) leads to adverse outcomes
- TH at 32-340C may benefit adolescents following a witnessed VF or children who remain comatose after resuscitation
Why do we use Therapeutic Hypothermia (TH)?
We know that major morbidity and mortality post-arrest is due to neurological injury (hypoxic-ischaemic encephalopathy). The aim of TH is to protect the brain and allow recovery.
In adults, mild TH (32-340C for 12-24 hours) is used in survivors of adult cardiac arrest who are initially comatose. Its goal is to reduce secondary brain injury after reperfusion.
When should we do it?
TH should be initiated early and quickly, ideally within 6 hours of the arrest. It requires sedation, ventilation and paralysis to prevent shivering.
How do we do it?
There are several methods for TH: surface cooling (e.g. a mattress or blanket); ice packs; selective head cooling (with a helmet); cold IV fluids (30ml/kg saline infusion); body cavity/peritoneal or hollow organ lavage; or extracorporeal cooling (e.g. ECMO).
What do we know about adults and why are kids different?
TH improves survival and neurological outcomes in adults.
Initial data showing TH is neuroprotective after VF arrest came from two 2002 NEJM landmark studies. Subsequently there have been similar results in several non-randomised studies
'Mild Hypothermia to Improve the Neurologic Outcome After Cardiac Arrest.' Holzer et al. NEJM (2002)
In this European study, 55% in the hypothermia group vs 39% normothermia group had favourable neurological outcomes.
6m mortality was 55% in normothermia group, Vs 41% in the hypothermia group
Conclusion: In patients successfully resuscitated after VF-cardiac arrest, therapeutic mild hypothermia to 32-34°C increased survival & favourable neurologic outcome compared to standard therapy
Treatment of Comatose Survivors of Out-of-Hospital Cardiac Arrest with Induced Hypothermia. Bernard et al. NEJM (2002)
Smaller study – but an Aussie-based crew in Melbourne. Of 77 patients – 49% treated with hypothermia survived with good outcomes, compared to 26% who had standard post-arrest care
Important to note that these original studies were in arrests where VF/VT is the first rhythm…..this is where cooling kids may differ! Aetiology of paediatric arrest often differs to adults & only ~10% are primary cardiac.
And even in adults – the evidence for benefit in non-VF/VT arrests is inconclusive.
But – International Liaison Committee for Resuscitation (ILCOR), recommend ‘considering use of TH for 12-24h in infants & children who remain comatose after resuscitation’.
This recommendation is extrapolated from adult data, but is mentioned in APLS guidelines.
TH decreases mortality and disability from HIE in neonates.
Tagin et al published an update of a previous Cochrane review in 2012 – this included 1214 newborns and 7 trials. It did show a reduction in the risk of death from neurological abnormalities and concluded that TH improves survival and neurodevelopment at 18 months in newborns with moderate to severe HIE.
Unfortunately evidence for TH in paediatrics is pending, and inferences are being made from adult data. There are ongoing RCTs, to answer the question, but none published yet.
There have been 2 retrospective large trials published –
Doherty et al (2009) did a retrospective study in Canada & the UK from 2001-2003. There were 222 cardiac arrests identified, but only 79 eligible for study. Of these, 29/79 were cooled for 10-30 hours. Effectiveness of TH was neither supported nor refuted – there was no statistical difference in mortality between the two groups.
Fink et al (2010) did a retrospective cohort of 181 patients admitted to PICU in the US from 2001-2006 in the largest study on this topic. TH was used at the doctor’s discretion. In this study 91% of the arrests were asphyxial in origin. Those whose temperature was below 320C had a higher mortality. Overall, the groups had a similar mortality (55%) and there was not enough statistical difference to be able to support or refute the use of TH.
THAPACA is an ongoing trial looking at the effect of TH on survival and neurobehavioural outcome in children post-arrest. This is one to watch.
A study by Hutchinson (2008) looked at 225 children with TBI. There was no improvement in outcome in children who had TH. The TH group actually showed an increase in mortality but this was not statistically significant.
Also remember…there is a cost to cooling!
Most studies show no significant difference in complications in TH groups.
However, potential adverse effects (which tend to occur more at very low temperatures <320C) include:
Haematological – infection, impaired coagulation & bleeding
Cardiovascular effects – cardiac arrhythmias
Metabolic effects – hyperglycemia, electrolyte disturbance
Doherty, DR, Parshuram CS, Gaboury I, Hoskote A, Lacroix J, Tucci M, Joffe A, Choong K, Farrell RF, Bohn DJ, Hutchison JS. Hypothermia Therapy After Pediatric Cardiac Arrest. Circulation. 2009;119:1492-1500.
Hutchinson JS, Ward RE, Lacroix J, Hebert PC, Barnes MA, Bohn DJ, Dirks PB, Doucette S, Fergusson D, Gottesman R, Joffe AR, Kirplani HM, Meyer PG, Morris KP, Moher D, Singh RN, Skippen PW. Hypothermia therapy after traumatic brain injury in children. N. Engl. J. Med. 2008;358(23):2447-56.
Tagin MA, Woolcott CG, Vincer MJ, Whyte RK, Stinson DA. Hypothermia for Neonatal Hypoxic Ischemic EncephalopathyAn Updated Systematic Review and Meta-analysis. Arch Pediatr Adolesc Med. 2012;166(6):558-566.