iPhoneIcon_BigIn accidental hypothermic cardiac arrest we are to continue CPR until the patient has been rewarmed to around 34. If available, and appropriate, these patients are to be transferred to a hospital with ECMO capability. ECMO is the most efficient means we have for reheating hypothermic patients. Rewarming rates up to 12 C/h are possible with the fastest devices. ECMO will also provide circulatory and respiratory support during resuscitation and after ROSC when these patients are notoriously unstable. A small retrospective study in AJEM looks at outcomes. It is a tiny study but nevertheless interesting as the results are good but also as it comments on the hypothermic cardiac arrest patients with asphyxia as part of their mechanism.

Case background
The study touches on a HEMS case I had some years ago. A couple of years ago a capsized boat was spotted of the coast. As the life boat service arrived they found the owner submerged next to it. The victim was dragged out of the water and CPR was ongoing as our HEMS crew rendezvoused with them in the harbour. ECG showed asystole. Pupils were dilated. The patient was cold to the touch. He had been submerged in ice-cold northern Atlantic waters. Estimated down-time was anything between 45 minutes (when he was spotted floating by rock fishermen) and 10 hours (when our patient was seen leaving the harbour.)

Now what? Should we pronounce life extinct or transport him with sub-par CPR ongoing to nearest ECMO-centre for rewarming? Is this a primary drowning followed by a corpse becoming hypothermic? Or did the poor guy cling on to his capsized boat until he had hypothermic cardiac arrest and then became submerged? No-one is dead until warm and dead? It was a difficult decision to make and afterwards I could tell my decision would have been controversial no matter what I did.

The study
The study is a retrospective study reviewing all hypothermia patients admitted to the ED  in Sapporo, Japan between 1994 and 2012. All patients who had hypothermic cardiac arrest and were resuscitated with ECMO-CPR were included. Mechanism of hypothermia was categorised into exposure to cold air, immersion to cold water, submersion and avalanche. The victims were also grouped into asphyxiation (submersion and avalanche) or non-asphyxiation  (exposure and immersion). For the study purposes this meant the victims´ airways were obstructed as part of the mechanism. The authors never conclude this group actually died from asphyxiation.

That last part is interesting as this is were ECMO in hypothermia becomes controversial. People who die of asphyxiation and then become hypothermic probably have as poor outcomes as other cardiac arrests. Then again, differentiating between the two is notoriously hard. Ie did the patient die from hypothermia and then became submerged or did the patient drown then became hypothermic?

The authors identified 68 patients who suffered primary hypothermic arrest.
6 of those were excluded because of futility.
30 patients were resuscitated with ECMO-CPR.
All of them achieved ROSC.
Of these 4 were excluded as there were causes other than hypothermia explaining the cardiac arrests.
10 out of the 26 primary hypothermia cardiac arrests recovered with good neurological outcomes. (CPC 1 or 2)

Tiny numbers of course, but still quite impressive.

It gets interesting when they look at the 14 patients who had asphyxia, the submersion and avalanche patients. 12 of those 14 patients had poor neurological outcomes.

These patient scored a CPC of 3 or 4 meaning inability to independently perform activities of daily living or a vegetative state. Asphyxia was also associated with worse pH, potassium, PaCO2 and lactate values compared to non-asphyxic patients.

In the non-asphysic group 8 out of 12 patients recovered with good neurological outcomes.

Screen Shot 2014-05-08 at 02.26.27

Take home message
ROSC was achieved in all hypothermia patients who had ECMO-CPR for rewarming and circulatory support. I don´t know how the ECMO patients were selected so I´m not sure what that means. In the paper there are 30 hypothermic arrests that aren´t put on ECMO. The authors never tell us why. Only 10 of those achieve ROSC.

Non-asphyxial hypothermic cardiac arrest was associated with good outcomes most of the time while the presence of an obstructed airway, by water or snow, was associated with a poor neurological outcome in 85% of patients.

Study lives here:

Outcome from severe accidental hypothermia with cardiac arrest resuscitated with extracorporeal cardiopulmonary resuscitation. Sawamoto K, Bird SB, Katayama Y, Maekawa K, Uemura S, Tanno K, Narimatsu E. Am J Emerg Med. 2014 Apr;32(4):320-4. doi: 10.1016/j.ajem.2013.12.023. Epub 2013 Dec 14.

This entry was posted in ECMO, Prehospital Medicine, SAR. Bookmark the permalink.


  1. Jeff Bishop says:

    Here is a news report from a case last month from British Columbia. Have to be careful not to apply the technology to all cases but this was an example of what happens when things line up (i.e. hypothermia not asphyxia, early continuous bystander CPR, a great prehospital service and a well primed ECMO program).

  2. Pingback: The LITFL Review 136 - LITFL

  3. Pingback: Episode 19 – Environment: Mushrooms and Hypothermia | FOAMcast

  4. David Nash says:

    Hello stumbled randomly upon your article. I was the Perfusionist on the Jeff Bishop followup comment, it was I might add, the most significant save of my 30 years of providing extracorporeal support. One thing stands out for me personally is how quickly we were able to rewarm this physically fit, slender young woman; about an hour. While on pump I maintained a 10C gradient between arterial and venous blood temps but obviously her core opened up and rewarmed quickly in this case. Initially I drew criticism from my peers for rewarming her so rapidly but as it turned out a remarkable save that went viral globally. Extremely heart warming to meet this beautiful vibrant young person again months later. Just a interesting adjunct I thought Id share to this eCPR review. David, Vancouver Canada

Leave a Reply to David Nash Cancel reply

Your email address will not be published. Required fields are marked *