Podcast 66 – …Until they are warm and dead: Severe Accidental Hypothermia

Severe Accidental Hypothermia

It is winter and that means cardiac arrests coming in with extremely low body temperatures after environmental exposure. How do you treat these patients? How do you rewarm if you don’t have bypass?

Predisposing factors

hypoglycemia, malnutrition, ETOH, Addison’s, infection, and Myxedema (especially if failure to rewarm)

In urban environments, in patients > 32° C, failure to passively rewarm at least 1 C per hour should make you suspect one of the above factors. (Acad Emerg Med 2006;13(9):913)

Do not need to worry much until temp hits ~32° C

Bradycardia (refractory to atropine), but should not be treated anyway as it is appropriate to body temperature as long as it is sinus brady; but if you needed to, you can pace hypothermia internally (Ann Emerg Med 2007;49(5):)


FS, CBC, Lytes, TFTs, Cortisol, and blood cultures if you can’t figure out why a patient got hypothermic or is not warming appropriately

Get Temperature Probe in early for sick patients

Place rectal probe in 15 cm or much better IMNHO is an esophageal probe

See this post for how to place the esophageal temperature probe

Active Rewarming

Active rewarming if pt temp <32° C, CNS sx, or age extremes

Rewarming Methods

Shivering 1.5° C/hr

Warming Blanket 2° C/hr

Warm O2 1 C/hr with mask; 1.5° C/hr ET tube

IV Fluids do not add, but do not take away either

Peritoneal Lavage 3° C/hr

Thoracic Lavage with Chest Tubes 3-6° C/hr

Cardiac Bypass 9-18° C/hr

When to Stop Rewarming

If K>10, pt is not coming back, even if cold and dead

Must be greater than 30-32° C degrees to be considered dead

Rewarming with Chest Tubes

32-36 F Chest tubes one anterior and one posterior lateral

Use Level 1 Device or similar to pump warm fluids into the anterior chest tube

attach auto-transfuser or pleur-evac to posterior-lat chest tube to allow cont. emptying

Review article with two case reports (Resuscitation 2005;66:99-104)


Instead of salem-sump adaptors, use the luer to XMAS tree adapters made for this purpose. See the Blakemore Post for more on this item.

Extra-corporeal rewarming

Easiest method is to place an HD catheter and then get a dialysis machine to do CVVH or standard HD

CAVR Level I Rewarming

Here is an actual protocol from a Trauma Nursing Journal

J Trauma 1991;31:1151 and 1992;32:316 both by Gentilello

Bypass Rewarming

Crit Care Med 2011;39:1064


Best Review Article (NEJM 2012 367(20):1930)
Give a bunch of fluids
Afterdrop is a myth
30-32 is cutoff

Now on to the Podcast…


You finished the 'cast,
Now get CME credit

Already an EMCrit CME Subscriber?
Click Here to Get CME Credit for the Episode

Not a subcriber yet? Why the heck not?
By subscribing, you can...

  • Get CME hours
  • Support the show
  • Write it off on your taxes or get reimbursed by your department

Sign Up Today!


Subscribe Now

If you enjoyed this post, you will almost certainly enjoy our others. Subscribe to our email list to keep informed on all of the ED Critical Care goodness.

This Post was by , MD, published 4 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.


  1. Chris Menard says


    You advocate using iced saline IV to induce hypothermia. Why do you say warmed saline is ineffective to rewarm patients? Seems that if one is effective the other should be too.

    Having just attempted–ultimately unsuccessfully–to resuscitate a hypothermic immersion victim for nearly five hours, I appreciate you covering this topic, although I wish you had done it just a couple of weeks sooner. We used a poor man’s version of two chest tubes. I infused saline warmed in a microwave using a 60 cc syringe about 300 cc at a time through the anterior tube while the posterior tube was clamped, then drained by gravity from the posterior tube once the saline cooled, which took about 3 – 5 minutes per cycle. I like your way better, both because it’s continuous and because you don’t need to throw out your scrubs after the case.

    Also, do you have any advice about coagulopathy in these patients? My patient oozed from all sites slowly by inexorably.

    As always, thank you for your excellent work,

    • says

      Chris, All I can tell you is that iced saline has been proven to cool patients. I have seen no such data on warming accidental hypothermia with warmed fluids. In beagles, fluid heated to 140 F given centrally will rewarm dogs, but good luck on giving that temp fluid to humans without a study. I have seen little effect on my patients and even in lethal triad hypothermia, warmed fluids don’t seem to warm patients. I am sure some of the physics geeks who read the blog might give an explanation.

      Best way to fix the coagulopathy is to get the patient to temp, less than useful but there you go.

      • Søren Rudolph says

        Ok – here is a “geeky” explanation:
        – In order to understander why the rewarming effect of fluids are minimal it is helpful to review the concept of specific heat. Specific heat is defined as the number of kilocalories (kcal) required to warm 1 kg of a substance by 1°C. The specific heat of water is 1 kcal/kg/ °C, and thus 10 kcals are required to raise the temperature of 1 kg (1 L) of water by 10°C. The specific heat of the human body is 0.83 kcal/kg/°C. One can calculate the heat needed to raise the temperature of a 70-kg patient from 25°C to 35°C as follows:

        70 kg x 0.83 kcal/kg/°C x 10°C = 581 kcals (or 58 kcal/°C in temp gain).

        If 1 L of 42°C saline is infused into a patient with a core temperature of 25°C, the heat transfer will be:

        1 kcal/kg/°C x 1 kg x (42°C – 25°C) = 17 kcals.

        This is sufficient heat to raise the temperature of the aforementioned 70 kg patient by 17 / 58 = 0.29°C. The actual heat transferred with infusion of 42°C fluids can be substantially less than that. The temperature of warmed saline decreases rapidly in room-temperature air, and further cooling is proportional to the length of tubing traversed prior to reaching the patient.

        ref : Kempainen, R. R. and Brunette, D. D. The Evaluation and Management of Accidental Hypothermia. Respir.Care 2004;49(2):192-205.

        • says

          ahhh, now that you say that it makes complete sense. When we are using iced saline, the patient is 36 C and the fluid is 4 C–big difference. When we want to use warmed fluids, the patient is 28 C and the fluid is 40 C–much smaller transfer of energy. Brilliant. That is why the rat studies with the fluids much hotter, given centrally work. Thanks, Soren.

  2. R says

    What do you tell your prehospital providers to do (or not to do) with respect to rewarming the hypothermic patient?

  3. Scott Gallagher says


    Excellent review as always!

    I recently reviewed our protocols for prehospital EMS/SAR treatment of accidental hypothermia and was surprised by the lack of data for anything we do with regard to hypothermia cardiac arrest management. The standard mantra of limiting ACLS drugs and shocks until target temp of 30C is achieved is based on almost nothing, but gets repeated in perpetuity.

    Similarly, there are significant changes in the 2010 AHA guidelines regarding the management of accidental hypothermia arrest – also based on very little data. AHA now suggests that standard ACLS drugs and electricity protocols may be followed, but the references cited are a case reports and pig/dog models. (Part 12.9: Cardiac Arrest in Accidental Hypothermia [http://circ.ahajournals.org/content/122/18_suppl_3/S829.full])

    The biggest protocol hurdle I have not been able to resolve, is when to provide chest compressions. Some references state that when there is no palpable central pulse (PEA) after a good attempt to find one, that compressions should be initiated (Auerbach: Wilderness Medicine, 6th ed.; Chapter 5 – Accidental Hypothermia). Other references, suggest that if an ‘organized rhythm’ is visible on the monitor, that compressions are not necessary.

    Lastly, a commonly overlooked portion of the managment of accidental hypothermia is that the target temperature is 32-34C, not normothermia. This mild hypothermia is to be attained and then maintained for 24 hours or so, then slowly rewarmed to normothermia in accordance with standard post-arrest therapeutic hypothermia protocols.

    • says

      Well it’s great to have someone share my frustration at the COMPLETE absence of evidence-backing for any of there recs.

      For me, if I can’t feel a pulse, I’m doing compressions–not just for perfusion but b/c my rewarming depends on blood circulation.

      I would modify your last paragraph to say, if patient hits 33 C and is not waking up enough to follow commands, then leave them there.

      thanks for the comments, Scott

      • Scott Gallagher says

        Just FYI…not to belabor the point…the latest version of UpToDate written by well-known expert on the topic Dr. Daniel Danzl (among others) agrees with your decision to start compressions if no pulse appreciated: “We believe that chest compressions should be performed in patients who manifest an organized rhythm on a cardiac monitor but have no pulses and no other signs of life. ”

        This is probably a best expert guess/consensus, as I have seen recommendations waiver over the years. Others have considered an organized rhythm to be a sign of life and recommended holding chest compressions in the presence of PEA with an organized rhythm on the monitor…

        AR Mulcahy, MR Watts. Accidental Hypothermia. Emergency Medicine Practice. Jan 2009; 11(1) (Review article)

        Jurkovich GJ. Environmental cold-induced injury. Surg Clin North Am. Feb 2007;87(1):247-267, viii. (Review article)

  4. Ram says


    I didn’t hear you mention bladder or whole bowel irrigation as options.

    It is my understanding that bladder irrigation won’t get you rapid rewarming rates due to limited surface area, but it still seems to me to be a viable option in the non-crashing severe hypothermic since it is so easy to do.

    I imagine that electrolyte fluctuations make whole bowel irrigation less appealing and, perhaps, not worth the effort.

    My go-to for the peri-crash or the coding hypothermic has also been thoracic lavage but the literature describes relatively rapid rates of re-warming (2-4 degrees C/hr) with peritoneal lavage (at least twice as fast as rates of airway re-warming), so I was a little surprised that you dismissed it.

    Lastly, at least one source does not advise irrigating the left chest to avoid thermal irritation of the electrical tissue of the heart. Though it isn’t clear what evidence the author was citing to base that recommendation on. (Danzl 1994 Dec. p 1756-1760; NEJM).


    • says

      As you mention bladder doesn’t get you much, so not worth it to me. Whole bowel is fraught with all sorts of peril in the best of cases, I’d avoid it in my sick patients.

      All I care about is heating the heart enough to get a perfusing rhythm, so I go to the chest. Peritoneal is effective, but I don’t want to violate the belly. Way to easy to screw up. You are essentially doing 2 perc. dpls; when you look at the complication rate for that, it becomes not worth it for me.

      As I mentioned on the cast, go left sided when the patient is in arrest–irritation is irrelevant at that point. Go right if they are profoundly hypothermic with a pulse–even though that is theoretical. These recs are from the review in the shownotes.

      You Canadians probably have far more experience on this than us Southerners–it’s cold up there!

      • Sarah says

        I am a new EMT and just started listening to your pod casts. I have found them to be very informative. I have worked in the colder northern states, Alaska being one of them. I know pulmonary edema can be present when patients are in the colder climates for an extended amount of time. I know when you must take extreme measures to warm a patient your main concern is getting the patient warm and then getting their heart to start if it hasn’t already with the rewarming. So after rewarming, and restarting the heart, will pulmonary edema subside or is their another measure you would have to put into action.

        • says

          Sarah, I am only familiar with cold-immersion pulmonary edema, not cold induced probably b/c I live in NYC and am not a wilderness guy. Can any of you other folks help Sarah out with an answer?

          • Søren Rudolph says

            I think you are referring to what is know as subacute or chronic accidental hypothermia. The patient suffers prolonged, milder cold exposure and will correct the hyperthermia induced fluid losses by oral intake. Aggressive fluidresuscitation i these patients can precipitate pulmonary edema. Like everything else in the accidental hypothermia business – which is essentially non-evidence based – no specific guidelines exist (to my knowledge). I believe the treatment follows standard treatment for pulmonary edema. At least thats what I would do.

  5. Raj Kanji says

    Does pH help to prognosticate (like hyperkalemia >10) and if so at what level.

    I was at a talk where, a cardiac surgeon indicated that one should warm up a body titrating pH level. If the pH is low ( I cannot remember the level) one should cool them again. I believe the rationale is that the “frozen” metabolic waste is released upon warming and hence to warm slowly.

    Any comments?

    • says

      haven’t seen any lit on pH. The card surgeon’s approach seems unsupported and is predicated on the patient they would see…the patient already on bypass–good luck having precise control on the non-bypass patient.

  6. says

    Hey Scott

    Love the podcast as always. Accidental (or at least non-therapeutic) Hypothermia is a subject close to my heart (!), and our shop down in darkest Tasmania has a bit of experience with it. The pleural lavage via ICC’s has been successfully used here. Direct rapid warming of pericardium and subsequent early ROSC with simple equipment (despite lag of core temperature measurement) is the obvious advantage in theory and in practice! I have access to a couple of great images and bits of data from a case managed by a colleague which I will share with you if I can get his permission…will keep you posted


  7. says

    And for what it is worth, 45 litres of normal saline were microwaved to approximately 40 degrees celsius by a chain of orderlies and poured down the ICC’s into the left chest – low tech, but simpler than a level 1 warmer, plus faster!

      • says

        They were pretty slick and they did get 45 litres down that tube fairly fast, but I concede probably not any faster than a Level 1! Jo Kippax has a great graph of core temp measured by urethral probe mapped against time and blood pressure with ECG and timing of clinical events superimposed, plus a great CXR and clinical plus logistical images. He will hopefully get in touch with you himself. Well worth a look!

  8. Jeff says

    Hi Scott,
    We’ve found that the 8.5 F biliary drains we use as pigtail chest tubes have the standard IV connector already built in, which we can then use for warm fluids off the level 1 infuser.

    Essentially the set up is one pigtail chest tube hooked to the level 1 and then one standard chest tube to suction per thoracic cavity.

  9. Andrew says

    Are you preferential to the esophageal temp detector for a particular reason, or would you be just as happy with a temp-sensing foley? Seems like that might kill two birds with one stone.

  10. Gregory Raines says


    I did a pretty extensive search on what constitutes a temperature sufficient to terminate resuscitation efforts in the severely hypothermic patient. There is “insufficient evidence” to dictate what is “warm and dead.” I went back to the Bible of ACLS, the Textbook of ACLS, 2nd edition (when they published all of the science in the manual along with recs – what an intimidating textbook) and from that point forward, the recs have been “the clinical maxim that patients who appear dead after prolonged exposure to cold temperature should not be considered dead until they they have restored to near-normal core temperature and remain unresponsive to resuscitation cannot be applied literally to all cases. Instead, the decision to terminate resuscitation must be individualized by the physician in charge based on unique circumstances of each incident.”

    Sorry I couldn’t find you a number – but like all things from the AHA, its a little murky…

    Thanks for another excellent podcast,

    Greg Raines, PA-C

  11. Marie says

    Are there any general guidelines in hospital for what is too cold to resuscitate?

    I work prehospital in a very cold place, and we are allowed to terminate resuscitation if certain criteria are met, such as ice crystals in the airway, body too frozen for chest compressions or a rectal temp below 60 degrees F after 5 minutes of assessing the temp. Is that 60 degree number also accepted in hospitals or is a person that cold so rare there aren’t any specific guidelines?

    • says

      yep, the latter. Most prehospital hypothermia termination protocols are predicated on a balance between chance of recovery vs. difficulty of transport. In NYC transport doesn’t present too much trouble, so they all come in and then we decide.

      • Scott Gallagher says

        Might be rare, but the record-low (13.7 °C [56.7 °F]) fascinating case of Dr. Anna Bågenholm who not only survived but recovered sufficiently to finish medical training and return to skiing is a stark reminder that under the right conditions good outcomes are possible despite prolonged cardiac arrest at ridiculously low temperatures.

      • says

        great addition. i will add that palpating a pulse can be very difficult in these folks. we use a femoral a-line to make the rosc determination.

  12. Pedrinha says

    You seem to have ha a slip of the keyboard. Tha Annals paper is on external not internal pacing.
    Thanks fir the good work

  13. Pedrinha says

    Oops I forgot to ask:
    When would you want to pace a bradycardia ? Presence of shock and less than what ? Rule of thumb ?

  14. Pedrinha says

    Drugs supposedly do not work en severe moderate hypothermia:
    When you intubate uncouscious hypotherlic patients, do you use RSi drugs ?
    I did not find this anywhere so aI suppose people don’t know.
    What do you do ?

  15. says

    Thanks for the great education.
    One of our mountain rescue gurus used to recommend a K > 9mmol/L as an indication for stopping resuscitation. His thinking was that if the patient died then got cold the K would be high. If the patient was just hypothermic the K would be less than 9. Is there any truth to this?

  16. Taku says

    Hi Scott just catching up on my podcasts and heard this one the other day. I just wanted to add an emphasis and a reference.

    In this paper:

    Rewarming rates in urban patients with hypothermia: prediction of underlying infection
    Delaney, Kathleen A; Vassallo, Susi U; Larkin, Gregory L; Goldfrank, Lewis R
    2006 Sep;13(9):913-921, Academic emergency medicine

    They found that the rate of rewarming (or the inability to rewarm) a patient with hypothermia was correlated with underlying infectious etiology of the hypothermia. Although it was not the emphasis of your talk, with milder degrees of hypothermia I would think the safer approach would be to start with an assumption of infectious etiology…

    With regards to the severe hypothermia… a little factoid… there are human and animal studies out there that the bradycardia may be mediated through the release of an endogenous digoxin like substance. As far as I know there are no studies that have looked at the effectiveness of DigFAB in binding this endogenous digoxin like substance or the utility / safety of using digibind in the patient in cardiac arrest and hypothermia.

  17. Patrick Ryan says

    Hi Scott, Thanks for all your hard work, I love your podcasts! I am an EM pharmacist in Boston and we encountered this situation a few weeks ago, but actually were unable to rewarm our patient above 72F. Patient was found down outside, intubated, pulseless upon arrival and the entire time he was in our trauma bay (3.5 hours of CPR total). The patients temperature started in the low 60s, we did L sided thoracic lavage, warming blankets/lights, bladder irrigation, warm O2, gastric irrigation with warm fluids, rectal probe in place (changed twice and readjusted to make sure it was placed correctly) and we were still unable to warm the patient above 72-73F, we were trying to find information on failure to re-warm and I didn’t know if you’ve either encountered this or had any literature on this. Thanks so much and again appreciate all you do!!

      • Patrick Ryan says

        Thats the conclusion we drew after 3.5 hours of CPR, but we had multiple discussions of when we should stop, we could really find a definitive conclusion or resource.

        Their K was 4.2

  18. Matt Hendrickson says

    Matt Hendrickson, staff physician at St Agnes (Fresno, CA) and California Med Ctr (Los Angeles). Huge fan of yours for many years!! The 2012 NEJM review article by Brown referenced by you and in the 2015 LLSA reading list recommends using thoracic lavage only for patients with no vital signs which conflicts with your recommendation for lavage on patients <32. What are your thoughts?


Speak Your Mind (Along with your name, job, and affiliation)