Perelman School of Medicine at the University of Pennsylvania
Center for Resuscitation Science


Welcome to the CRS Frequently Asked Questions page. The following are questions that have been received through our website and answered by one of our expert faculty. We are happy to answer any questions you may have about CPR, therapeutic hypothermia, and cardiac arrest. Please note that the responses posted on this page are only opinions from our medical professionals. We do our best to answer these important questions, but we are dealing with a complex field and nothing below can be considered an absolute certainty. To submit questions please use the Contact page.

In addition, we have created the Penn Hypothermia Discussion Forum providing an opportunity for clinicians and medical professionals to discuss hypothermia implementation success stories and challenges. The responses by the medical professionals on this forum do not necessarily represent the opinion of the Center for Resuscitation Science and should be considered carefully. To request information or become a member of this forum, please email Audrey Blewer at

Thank you for your questions!

Top FAQs:

October 29, 2013

Q: How do you ensure that the IV fluid being administered is at 4 degrees C? Our medication refrigerators are maintained between 2-8 degrees C according to USP and CDC guidelines. Is this an acceptable range for the administration of cooled fluid? Do you have a separate storage refrigerator for the IV fluid used in mild therapeutic temperature management?

A: Storing the IV fluid in the medication refrigerator is perfectly fine and what most institutions do to keep the fluid cold . The cold fluid is used to help induce TH and drop the patient’s temperature more quickly; the devices are used to regulate and maintain the temperature at goal temperature of 32-34 degrees C.

-Marion Leary MSN, RN


October 3, 2013

Q: When is it ok to terminate hypothermia early?

A: Although we have had a few folks for which hypothermia has been terminated per family request, we are hesitant to stop cooling and will recommend to the family to continue cooling through rewarming to give the patient the best shot at neurological recovery.  If it is clear that the family understands the intent of hypothermia and chooses to terminate cooling efforts anyway, depending upon the DNR discussion, we recommend gradual, controlled rewarming so that hemodynamic instability does not complicate an already difficult situation. If the family is reluctant to endure any more uncertainty and wants to terminate the entire process immediately, the physician will discuss the possible side effects of too-rapid passive rewarming; if the family is clear about those side effects and chooses to terminate immediately, the physician will certainly honor those wishes and the staff supports the family through the process.

Patients who appear to ‘wake up’ during hypothermia are extremely unusual under our protocol.  We do monitor and titrate sedation/paralytic effect  to do our best to be sure that the patient is well-sedated.  After adequate sedation, our protocol calls for maintenance of paralysis during the course of cooling/and rewarming.  In the circumstance you described of the patient who followed commands while cold, there was probably not a choice but to initially rewarm at the usual rate to prevent hemodynamic/electrolyte instability then perhaps to titrate the rewarming rate as the patient tolerates.  Consider giving a touch of sedation so that the patient is comfortable during this period.

There are very few reasons to terminate TH and these decisions are made on a patient-by-patient basis.  We recently heard from a colleague that hypothermia is akin to boarding an airplane – after the airplane takes off, you really want to stay on until you arrive at your destination!  This is not to say that there aren’t rare situations, but they are very patient-specific.  

Thanks for the thought-provoking questions!

-Gail Delfin MSN RN CCRN


July 19, 2013

Q: I am in the process of making an evaluation tool for therapeutic hypothermia (TH) for our facility.  I would like to ask your expertise regarding some variables/scenarios we have come across with: 1) pt. who had an MI from ER went straight to cath. We were not able to start it right away because of the time spent in cath 2)another MI pt. who they had a hard time intubating and cryc due to the pt's anatomy had to go to emergency sx for tracheostomy. Was the latter considered a disqualification because of the most recent sx? And TH was also delayed because of this.  3) the pt. undergoing HT --BP dropped , no significant change in labs and EKG, MD ordered to start vasopressor.

A: I will describe how we might handle these scenarios at our hospitals.

The first situation is a common one in most hospitals that see large numbers of out-of-hospital (OOH) cardiac arrests.  It can shake out a few different ways.  Because we never want to interfere with door-to-balloon times, cardiac cath for the STEMI takes precedent.  Cardiac cath/intervention requires so little time in most cases – usually an hour to hour and a half – that, while we want to begin cooling as quickly as possible, if the cath lab is immediately ready for the patient, he will go right to the lab and we begin to cool right after.  If the cath lab gives us a little time (‘a patient is on the table, we will bring him up in 10 minutes’), we do as much as we can fit in.  That might involve getting intubation, sedation and paralysis initiated and maybe beginning cold fluids. (For hypothermia patients in our institutions, we paralyze and use surface cooling devices.) Maybe we will have a chance to place the patient into the wraps.  We have had several patients that have been transported to the cath lab with the wraps on and cold fluids completed or infusing and, after a little coaching for the cath lab staff, that staff is willing to monitor the temperature during induction.  As a former cardiac ICU clinical specialist, I was able on occasion to hang around for some of the interventions just to be present in case the staff had questions. Usually we feel that if hypothermia is started asap after cath for STEMI (1-2 hours after arrest), that usually is an acceptable time frame. We also encourage consideration of immediate cath for most OOH, for disease is present in a majority of them.

The second patient, if I understand it correctly, had emergent surgery for tracheostomy?  This might be a muddy situation for some practitioners … as you know, surgeons are very concerned about the risk of bleeding in a fresh postoperative patient – cooling drops platelets and may affect the clotting cascade.  Bleeding, however, does not seem to be significantly associated with cooling.  Each patient situation has to be carefully considered, so that if there was any evidence of uncontrolled hemorrhage that didn’t respond to blood products or if there is bleeding with really aberrant coags, hemoglobin and/or platelets, one might be hesitant to cool.  On the other hand, recent publications of case series/reports looking at recently operated cardiothoracic and neurosurgery patients did not demonstrate significant bleeding, given that signs of bleeding and labs were carefully monitored. In summary, it is likely that our group would not consider the tracheostomy a contraindication to cooling.

With regard to your third question, hypotension can be attributable to a number of things – all else being equal (no new EKG ischemic changes, no bleeding, no volume depletion, no sepsis, etc), we consider reperfusion syndrome or ‘post-cardiac arrest syndrome’ - a syndrome causing myocardial stunning and drop in the systemic vascular resistance. For that reason, in our protocol, we generally consider the following: What is the BP? The CVP? The SvO2 and the lactate? Our approach is similar to the sepsis early goal-directed approach.  (See Gaieski DF, et al. Early goal-directed hemodynamic optimization combined with therapeutic hypothermia in comatose survivors of out-of-hospital cardiac arrest. Resuscitation (2009), doi:10.1016/j.resuscitation.2008.12.015) Subsequent authors have taken different approaches to managing this algorithm (and we have fallen out of love with dopamine for the time being, using norepinephrine and epinephrine more frequently); it usually boils down to what pressor the physician is comfortable using.  However, it helps to think about hypotension in a pressure, volume, tissue/cellular oxygenation framework. Generally speaking, when presented with a falling BP, volume challenge in bolus form is always our first response (unless the CVP > 15 and/or the patient has poor sats).  If you are dealing with true shock and there is little response to fluid, pressors/inotropes follow.

Good luck with your program and thanks for your questions!

-Gail Delfin MSN RN CCRN


April 24, 2013

Q: We are trying to implement the use of continuous EEG monitoring at our site but we are running into some roadblocks and I thought you might be able to help me out since your program is so well established. Also, our neurologists would like us to put together a protocol specific to cEEG monitoring. It appears in your protocol that you are doing cEEG monitoring for the duration of cooling and rewarming and this seems to be the standard most places, correct? Also, how often do you suggest the EEGs be read?

A: Readings of the EEGs are a separate situation.  Since the EEG is sent to a remote site for reading, we usually expect reports every 8 to 12 hours or so.  That sometimes is initiated by the housestaff calling the epilepsy fellow to get a reading.  However, if there is anything unusual (patient difficult to cool/keep cool, emg interference, evidence of seizure/myoclonus on BIS or high BIS reading, weird breakthrough patient motor activity/twitching despite paralysis, new episodes of hypertension or changes in HR, etc), the nurse or housestaff can call the remote monitoring number or epilepsy fellow to ask them to read the EEG at that point. I am not sure that we mentioned it at HART – we do somatosensory evoked potentials on day #3 to evaluate midbrain function.  It also lives in the neuro world and might be of interest.  We are looking at the information right now and it is just another piece of the neurodiagnostic puzzle. 

-Gail Delfin MSN RN CCRN


April 11, 2013

Q: Our last 2 therapeutic hypothermia patients had significant bradycardia during maintenance. One had an adequate blood pressure without the use of Dopamine (MAP> 80). The other required Dopamine to get her pressure up because the MAP was < 70. What I found concerning is that the RNs and MDs were only really concerned about the heart rate. They didn’t look at what the pressure was or the total picture and stopped cooling due to bradycardia.  Is there an absolute number you become concerned with and what other measures can you recommend to mediate the bradycardia

A: Like you  mentioned, we really look at all of the other parameters. If the MAP is ok, if the CVP is ok, if the urine output is ok, then we are ok with a HR in the 30s. Sometimes what we tell providers that are becoming worried that they could bump the temperature up to 34 degrees C, and usually that will bump the HR up to the 40s and then you don’t have to worry as much.

-Marion Leary MSN RN


February 12, 2013

Q: Can anyone provide what special considerations should be taken into account when performing therapeutic hypothermia on hemodialysis and / or peritoneal dialysis dependent patients ? E.g. Turn off heaters on hemodialysis PTs. , consider continuous veno dialysis , etc...

A: Your question is a really good one and comes up often - you have already proposed most of the steps that can be taken.  In patients receiving CRRT, blood warmers are simply turned off, and the patient's blood is circulating against room air temperatures, albeit relatively slowly compared to the speed of excursion in hemodialysis or ultrafiltration.  As I understand it, ultrafiltration can be done at a lower temperature than hemodialysis due to the possibility that clotting can occur at temperatures < 36 degrees in dialysis. This may be a machine-specific or institution-specific policy, however, so I would recommend checking with the folks in your specific hospital.  If the patient is cool already, the short amount of time the blood courses through the dialysis circuit shouldn't make keeping the patient cool too difficult; the cooling device should be able to compensate. 

I am not familiar with the PD literature about warmed dialysate vs not warmed; would you have a renal fellow to be able to comment re: whether warming is only a comfort measure - could it be eliminated or is there another physiologic reason to warm the fluid? 

-Gail Delfin MSN RN CCRN


February 8, 2013

Q: How do you manage hyperglycemia while the patient is undergoing hypothermia.

A:  In our hospital, we manage our patients to around the 110 – 150 range generally.  (There may be some small variances in protocols in some ICUs).  The majority of  patients become hyperglycemia during hypothermia induction and maintenance, many requiring insulin infusions.  During rewarming, hypoglycemia is always lurking in the background and can appear pretty rapidly as these patients start making more of and become sensitive once again to their own insulin.

- Gail Delfin MSN RN CCRN


October 24, 2012

Q: Why are there EKGs done only every 8 hours x 2. Also, do you have any data on cardiac ischemia during the therapy. We are not a cath center, but if there is not a lot of ischemia or infracting during the process, either cooling or rewarming, may we breathe a little easier?

A: The Q 8 hour x 3 EKG is our basic rule-out MI protocol here at HUP.  However, we have a very low threshold to do additional 12 lead EKGs if there is any indication of rate/rhythm change or hemodynamic instability.

With regard to ischemia identification either prior to or during cooling, decision-making about a trip to the cath lab can still be quite a challenge.  It is a super straight-forward decision to go to the lab when there is obvious ST segment elevation or a new L bundle.  But because probably about 45 -50% of the patients with shockable out-of-hospital arrests have coronary artery disease, we try to be aggressive about sending those folks to the lab even without signs of STEMI.  We have had some experience with patients who did not have clear cut worrisome EKG changes after arrest but experienced ischemic changes on the EKG during rewarming.  Rewarming is similar to a stress test – there is vasodilatation during rewarming, the protective effect of hypothermia is being withdrawn, the patients often have BP swings.  Those patients showing ischemia during rewarming did go immediately to the lab and significant disease was seen in the majority of patients. 

In contrast, cooling induction and maintenance appears to have a stabilizing effect on the myocardium – the metabolism slows, the heart rate drops, the blood pressure often stabilizes.  It is unusual to see new ischemia while the patient is cooling.

So, to sum up, go for the cooling but if there is any suspicion at all about coronary artery disease in the story, history or on the EKG, it’s best to begin to cool but to not delay in transporting the patient to a hospital with a cath lab.  Think too about having a plan if the patient begins to show signs of ischemia during rewarming (new hemodynamic instability, tachypnea when paralytics wear off, restlessness).  You might want to do more frequent EKGs during that time so that you could monitor ST-T wave changes. 

-Gail Delfin MSN RN CCRN


July 23, 2012

Q: I am searching for guidelines regarding termination of resuscitation efforts during hypothermia.  Does the saying "not dead until they are warm and dead?"apply?  Our dayshift worked a multi-hour v-fib arrest today while pt was at temp (31.8c)as the MD was uncomfortable calling the pt while hypothermic.  

A: This is a frequent question we get at the website and during our HART courses.

The confusion comes from equating an old axiom related to environmental hypothermia with possible side effects of therapeutic hypothermia.

Patients, as you know, are prone to arrest when their core temperature drops below 28-30 degrees C and one of the central components of treating that arrest is rewarming; on the other hand, it is very unusual to arrest at 32-34 degrees C (therapeutic hypothermia) FROM the temperature; rather, the patients receiving TH have recently had an arrest, are critically ill, and, until proven otherwise, arrest because of the pathology leading to the original arrest (primary arrhythmia; cardiac ischemia; post-cath VF, etc). If a patient arrests at 32 or 34 degrees C, no efforts to rewarm the patient should be undertaken; the patient should be treated like someone who arrested @ 37 degrees C and normal CPR/ACLS protocols followed; after a time frame that someone would be comfortable calling an arrest at 37 degrees C has passed, the clinician should be comfortable making the same (albeit difficult) decision to terminate resuscitation efforts; that said, if there is significant overshoot and it is felt that the excessive hypothermia is contributing to the arrest, that should be taken into consideration and warming efforts undertaken.

So, to directly answer your question, the saying "not dead until they are warm and dead" doesn't apply.

-David Gaieski, MD


March 25, 2011

Q: Do you initiate/complete for all causes of arrest, including respiratory that leads to cardiac?

A: Yes; if comatose after arrest, of any type, we treat (with caveat of above).

-David Gaieski, MD

Q: Will you initiate/complete TH on patients who have a baseline CPC of 3 or 4, if known?

A: In general, no. This is a complicated question because it is pretty clear that TH confers a survival as well as a neurological benefit. However, none of the studies or implementation studies/databases included patients w/ the baseline CPC you are describing and so it is impossible to know if the survival benefit is experienced by that patient population as well.

-David Gaieski, MD

March 22, 2011

Q: Have you read any research that specifically addresses outcomes in regard to time it took to reach target temperature? I'd like to find some data to support cooling SOONER rather than LATER after ROSC. Why is 6-8 hours after ROSC the cutoff? Worse outcomes?

A: There are very little data on this... 6-8 hours is mentioned only because that was the average time to target temp in the randomized trials from 2002 (the HACA and Bernard trials). Based on a number of animal studies, we think cooling earlier is better, but no clinical studies have conclusively shown this.

-Benjamin Abella, MD

February 2, 2011

Q: I am teaching staff and one of my colleagues questioned why a hemorrhagic bleed would be contraindicated. I explained that hypothermia increases the incidence of coagulopathies, however he brought up a good point, if we are treating post arrest, the chance of an MI is there, therefore they will go to the cath lab.... were anticoagulation is started. So what do you do and how can I explain that?

A: Good questions ? while true that TH may increase the risk and/or severity of bleeding (i.e., cooling causes coagulopathy) ? the extent of this increased risk is small. Therefore it's a question of balancing risks and benefits. If someone has clinically dangerous bleeding (big GI bleed or brain bleed, for example), something that could worsen this would be high risk. If someone is on heparin, Coumadin, or going to the cath lab, without actual clinical bleeding, the risk is small, so it's much more comfortable for us to proceed with TH.

It's kind of analogous to the treatment of elevated INR. If someone's INR is found to be very high, you might just give vitamin K if there is no actual bleeding (low risk for bad outcomes). But if an INR is very high AND they are coughing up blood (high risk for bad outcomes), you wouldn't rest with just vitamin K, you would also give FFP and/or other clotting factors.

-Benjamin Abella, MD MPhil

December 14, 2010

Q: I understand that hypothermia causes decreased cardiac output, but how much?

A: Cardiac output decreases 25-40% depending upon exact target temperature and is mostly related to bradycardia; this is usually more than balanced (in regard to oxygen delivery) by decreased metabolism (and decreased oxygen consumption) so that the oxygen delivery/consumption profile is improved.

Q: What effect does a temperature of 33 degrees ALONE have of ejection fraction?

A: Mostly related to bradycardia. See previous comment above.

Q: Combined with a stunned myocardium, 1 day post MI, is there any value to this ECHO?

A: Yes, there is a value to the ECHO as it is telling you what the patient's EF and wall motion are @ that time; most of the stunning from arrest is reversible; the injury from MI leading to arrest is dependent upon how quickly reperfused, how much was penumbra vs. necrosis, and effects of TH on myocardial recovery; Data suggest good outcomes can be attained in this exact patient population if IABB, inotropes, and aggressive post-cath care is used in conjunction with TH.

Q: How long after the patient has been rewarmed would we be seeing "real" LV function data on ECHO?

A: The change in EF related to TH itself will go away as the patient is rewarmed and heart rate increases; the myocardial stunning of post-arrest period usually resolves by hour 40 or so; the injury from MI is dependent upon how much necrosis occurred.

-Dave Gaieski, MD

November 29, 2010

Q: How does TH affect lab values?

A: The main effect of TH on lab values is on blood gases. Need to know whether your lab corrects values or not to the body temperature. In other words, most blood gas analyzers warm blood to 37 degrees C before running; some labs report the values @ 37 degrees C, others @ the patient's temp @ time labs were drawn. That is what you need to know to manage the patient clinically.

Can approximate as follows:
for every 1 degree C below 37 degrees C, for PO2, substract 5 mmHg; for PCO2, substract 2 mmHg; for pH, add 0.012.

-Dave Gaieski, MD

November 19, 2010

Q: What is the danger of infusing large quantities of cold IV fluids via a SC, IJ or PICC? My understanding is potential arrythmias.

A: This is a theoretical concern, that is why we recommend chilled saline via peripheral IVs, preferably in the antecubetal location. Adequate peripheral access (2 minimum) should be obtained in the peri-arrest period and these can be used for induction of TH while central venous access is obtained.

-David Gaieski, MD

November 16, 2010

Q: Is any time delay recommended to see what the neuro assessment is prior to sedating and cooling?

A: The way we do this at our institution, if there is a question regarding neurological status of a patient right after the arrest and sedation was given, sometimes, we will wait an hour, let the medications wear off and then decide if they are waking up. And by waking up I mean following commands appropriately. The patient has to be able to give a thumbs up or squeeze your hands appropriately for us to decide not to cool. Just moving extremities or "reaching of the ET tube" is not, for our determination, following commands appropriately. In our protocol we go with a Glasgow Motor Score of <6, which is not following commands, since a GMS of 6 is "following commands". I would recommend using GMS over the full GCS.

We always error on the side of cooling, since there are very little risks and very big benefits. We reason that we only have one chance to save the brain and if we decide not to cool because they some what look like they might be following commands or opening their eyes, there is no going back and we have lost our chance.

As always, these things are a case by case basis. If they arrested in the field and they are already paralyzed and sedated, we may just cool regardless.

-Marion Leary BSN, RN

Q: I heard recently that a rectal temperature is contraindicated in a pt post MI. Can you explain to me why that is?

A: The general concern about rectal temperatures post-MI is that it might stimulate the vagus nerve and cause bradyarrhythmias. There aren't a lot of data to support this. Some institutions have successfully used rectal temp probes in post-arrest patients (without reported increases in bradyarrhythmias), though we don't feel this temp modality is as accurate as esophageal or bladder.

-David Gaieski, MD

October 06, 2010

Q: Can you tell me your experience with temperature monitoring - rectal probe, bladder probe etc.?

A: We use the temperature probe foley at Penn. If the patient is not producing enough urine (4cc/hr) we will put in an espophgeal probe. We try to stay away from rectal probes as there is a temperature lag that could be dangerous when initiating cooling. There was a Case Study published in 2009 that showed a significant delay between esophageal temperature (core temp) versus rectal probe. The patient was given saline boluses and the esophageal probe showed a temperature of 33.5 at 19 minutes while the rectal probe had the patient temperature at 35.5 at 19 min. The concern is that you could over-cool a patient and get into a dangerous range (30 degree C or lower) where ventricular rhythms could occur. There is also a lag with bladder temperature as well, but it is more evident with the rectal probe.

-Marion Leary BSN, RN

August 10, 2010

Q: What does the evidence say about neurological assessment after a cardiac arrest patient is brought back to a normothermic level? Are there studdies available on how long we need to wait and be patient to see if the patient will make further progress?

A: In terms of neurological prognostication after cardiac arrest, currently the standard guidelines are to not neuro prognosticate until at least 72 hours post ROSC, by that time the patient should be re-warmed. Although the guidelines say 72 hours we have seen in our patients a range anywhere from 2 day post ROSC up to 14 days or more, with the average being 3-5 days. There was a Consensus statement put out by the International committee on Resuscitation, for which the AHA is a part of, which explains why older neurologic prognostication tests are not reliable and it also goes into more detail about prognostication in this population. There are new International/AHA guidelines coming out at the end of this year, so things may change even more, we will just have to wait and see.

-Marion Leary BSN, RN

Q: We have only recently (in the last year) developed our order set and protocol for therapeutic hypothermia in cardiac arrest. In our first 14 patients, we have had trouble getting to our goal temperature very quickly. I think much of this is a learning curve. I've seen that some facilities have a "Code Ice" team that is paged with any cardiac arrest patient to help facilitate the process. I'm thinking of initiating that here. I see from your website that you have a "Resuscitation Team", and I was wondering about that ... is this different from your code blue team? What are their responsibilities?

A: Thanks for the email. Our resuscitation team is different than our hospital code team. We (the resuscitation team) do not respond to codes in the hospital for the purpose of "coding" the patient, but will follow up after every code to assist with hypothermia eligibility if the patients get their pulse back and then if the patient is eligible for cooling we will assist the team in getting the protocol started and will follow the patient throughout their course of hypothermia care. We are available to the treating team 24/7 with any and all hypothermia and post-resuscitation questions. We may be a little different as one of our physicians is actually credentialed to leave notes in the chart as an actual resuscitation consult, although that may not be necessary for what your needs are, which sound like assisting the teams in initiating hypothermia more quickly.

I think that starting that type of service is definitely a good way to get hypothermia started more quickly, and it is always helpful to have a hypothermia champion within any given hospital, but it is also very time intensive depending on how you set it up.

-Marion Leary, RN, BSN

August 6, 2010

Q: At our small community hospitals our ED docs are concerned about inserting alines. Without residents and a "cool team" on call, who inserts the aline is still undetermined. We will have to call the on call surgeon to come in and insert the aline.

My concern is that the patients will be at goal when they arrive in the ICU, or even in the ED after EMS starts the process. I know getting a radial aline will be very difficult, how about a femoral aline, have you had trouble with that when the patients are cooled?

A: We would not delay implementation of therapeutic hypothermia because of questions of who is going to place the arterial line. If the patient is at goal on arrival to the ICU, the benefits of that probably far outweigh the increased difficulty of placing a radial a-line in a patient vasoconstricted from therapeutic hypothermia. You have to work within the resources of your system and if the emergency physicians are not going to place a-lines but will start hypothermia then those constraints can be worked within to set up a successful program.

You are correct in saying that a radial arterial line may be more difficult to obtain in these patients but experienced providers such as the on call surgeon you are consulting should be able to achieve placement of radial a-line in a number of patients (especially if they are using ultrasound to assist them). Femoral a-lines are a completely reasonable alternative and remain easier to place properly in hypothermic patients because of the size of the femoral artery. Our standard approach is not to delay induction of therapeutic hypothermia to attempt a-line access; to make a few attempts at a radial a-line, then proceed with femoral access if necessary, with or without ultrasound guidance.

Hope this is helpful.

-David F. Gaieski, MD

February 2, 2010

Q: Has your facility done any research on the survival of morbidly obese patients who suffer cardiac arrest? Some specific concerns:

Are we able to deliver enough joules?

How effective is manual CPR for obese patients vs average wt. patients?

Are there mechanical devices available to provide compressions for obese patients?

Is special airway equipment available for the obese patient?

A: We have not specifically examined outcome differences in morbidly obese patients who suffer cardiac arrest versus patients with normal BMI. This is an important question, however, and one I am interested in examining. I am just finishing up a paper on BMI in severe sepsis and septic shock, which shows no differences in outcomes (but differences in sources of infection) in the morbidly obese versus non-obese.

We have noted challenges in reaching target temperature in therapeutic hypothermia in morbidly obese patients but have not yet systematically studied the topic. We have successfully treated a >150 kg post-arrest patient using therapeutic hypothermia and our post-arrest algorithm. The patient was discharged to home in pre-arrest condition with normal mental status. The issue of quality of CPR (QCPR) in the morbidly obese is an important issue.

We are currently trialing the Lucas mechanical compression device in our ED and it will fit many obese patients but there is a limit in the AP diameter it can accept and we were unable to use it on one patient who was approximately 300 pounds.

Regarding your other questions, I am attaching an abstract from Dana Edelson, at University of Chicago, and some of my colleagues at CRS, which addresses many of the QCPR issues.

Regarding airway management--the most important issue with intubating the morbidly obese is positioning, not specialized equipment; placing the patient on a ramp of folded towels or a commercially available ramp, which helps achieve proper head positioning, is the most important single intervention that can be performed; if a morbidly obese person is positioned properly and remains unintubatable, an algorithm for rescue airway management should be in place at the institution – we typically use a CMAC as a video-assisted rescue device if the patient is able to be ventilated and a King airway or LMA if the situation is more tenuous.

-David Gaieski, MD

Abstract link: Click here for the abstract

December 14, 2009

Q: Thrombocytopenia is considered a contraindication for TH - is there a platelet level you would consider as an absolute contraindication in the absence of bleeding? AND if hypothermia was initiated post arrest , at what platelet count would you consider stopping the protocol and rewarming?

A: The exact risk of bleeding during therapeutic hypothermia (TH) is not known. TH has been used safely with thrombolytics and anticoagulants although this has not been carefully investigated. None of the large studies involving TH for cardiac arrest, stroke, or intracranial hemorrhage have reported large increases in bleeding. At 35 deg C, a mild decrease in platelet counts and qualitative dysfunction of platelets is seen, while at temp < 33 deg C, clotting factors will begin to be affected. That being said we do not have an absolute platelet count as a contraindication in our protocol. The only contraindications we have in terms of bleeding are if there is an intracranial bleed, bleeding due to significant trauma, or clinically relevant bleeding from other sources (GI bleed, dialysis graft bleeding etc). You can also consider prophylactically transfusing platelets (if it is not contraindicated for the patient) for counts < 20K with no bleeding, and < 50K with active bleeding. These are just our protocol recommendations, they have not been validated by scientific studies. I can not give you an absolute platelet count that I would not cool or stop cooling when present as It really depends on whether the patient is actively bleeding as well as many other patient specific factors. The risk of spontaneous bleeding from thrombocytopenia is classically noted at 20K so you may want to use this number, but again we may choose to cool someone with a platelet count below 20K if the specific situation calls for it. Bottom line is that in patients who are coagulopathic, the risk of bleeding should always be weighed against the benefits of cooling on a case by case basis.

November 9, 2009

Q: We have had several TH patients who were also Code Mi patients. Cardiologists have had questions regarding TH and these patients, specifically whether spasm is more easily induced or if medications such as bilvalirudin and thrombolytics interact with TH. Are there any resources or studies on this specific patient population? I have had little luck with a literature search – what do you see in clinical practice?

A: You ask an excellent question - alas, there are very little data published on the specifics of PCI/cath and cooling. The only data that DO exist suggest that morbidity and mortality of combining therapeutic hypothermia (TH) and cath is not worse than cath itself (so that, in aggregate, TH doesn't make things worse) – but no data are published on specific issues with drug metabolism, coronary flow, etc. I can say that we have performed cath on a number of patients while cooled and things seem about the same with regard to medication use, etc. We definitely need more research in this area.

-Benjamin S Abella, MD

July 27, 2009

Q: I can find references that say that overshooting the target temperature (too cold) is harmful and that rapid rewarming can create instability. Is there literature that addresses potential harm of mild fluctuation in temperature during the treatment period of 24 hours?

One of the selling points used by many of the equipment vendors is the claim that their equipment is superior at maintaining the temperature exactly at 33 degrees a high percentage of the time. Is there evidence to indicate that this is most beneficial or that failure to provide this is potentially harmful? Is time frame from initiation until reaching target temperature a higher priority in achieving the best outcomes or is steady maintenance a higher priority? Have protocols that use only iced IV fluids and conventional cooling blankets with less precise temperature maintenance been associated with worse outcomes?

A: As you have pointed out there are references for overshooting the target temperature (including my colleague Raina Merchant's paper from 2006) and for the potential risks of rapid rewarming. There is less available on the clinical relevance of variations from the target temperature during maintenance phase related to different devices. Hoedemaekers and colleagues published a paper in Critical Care looking at the speed of cooling and degree of variation with different devices but this didn't address clinical outcomes/relevance (Hoedemaekers CW, et al. Crit Care. 2007; 11(4): R91). It is highly unlikely that 0.3 degrees C variation versus 0.6 degrees C variation is clinically relevant. There currently is no evidence that such tight control is clinically relevant--that doesn't mean data won't emerge at any point.

A large French study comparing surface cooling to endovascular cooling has been completed and is being prepared for press. I don't know the results but one of the questions they will likely address is the precision of maintenance of target temperature. Joe Ornato in Richmond has discussed improvement in their outcomes after they switched from surface to endovascular cooling and attributes some of the improvement to the speed of cooling but this is not a randomized controlled trial.

As you have pointed out many of these details are "selling points" of different devices and focusing too much on those points is, to some extent, missing the forest for the trees. The important point is cooling and high quality ICU care. The recent publication by Friberg and colleagues (Neilson et al, "Outcome, Timing, and Adverse Events in Therapeutic Hypothermia After Out-of-Hospital Cardiac Arrrest," Acta Anesthesiol Scand 2009; 53: 926-936) doesn't discuss variation from goal temperature but does suggest NO relationship between time to target temperature and outcome (median time from arrest to goal temperature [< 34 degrees C] was 260 minutes, IQR 178-400 minutes). I always try to keep in mind that the two landmark studies (NEJM, 2002; 346) used fairly primitive cooling equipment (Bernard-->ice packs; HACA-->forced cold air blankets that were ineffective enough that 70% of the patients required supplemental ice packs) to achieve the statistically significant improvements in neurologic outcomes (Bernard and HACA) and mortality (HACA) that are the foundation for current therapy.

-David Gaieski, MD

June 18, 2009

Q: In a patient with cardiomyopathy EF 15%, and severe pulmonary hypertension RSVP in 60's, who arrests in hospital, what are your thoughts about post-arrest cooling potentially exacerbating pulmonary HTN and myocardial oxygen demand in a pt vulnerable to re-arresting? -Baltimore, MD

A: The use of hypothermia is definitely not without risks for complex patients such as this - alas, no clear data exist to answer your question, but I suspect one could still cool such a patient but with extreme caution.

What we often do, as a matter of practicality, is to cool more "gently" (only to 34 C) in such patients, under the assumption that this mild degree of cooling, while still supported by clinical data and AHA guidelines, might provoke less untoward hemodynamic effect.

-Benjamin S. Abella, MD

April 21, 2009

Q: Have you used Therapeutic Hypothermia in patients with non-cardiac etiologies for arrest? Specifically, we are looking at whether to cool overdose patients, and also we had a young man attempt suicide by hanging. We used TH in both, but ran into logistic issues when MICU critical care attendings balked at taking care on TH patients, as literature generally supports use only in cardiac arrest.

A: Yes, we use hypothermia on most etiologies of arrest — and this seems to be the practice of many other hospitals that use hypothermia. There are growing data from various case series reports that other arrest rhythms and causes of arrest benefit from hypothermia as well. However, we don't use hypothermia for non-arrest situations (if your near-hanging or OD patients didn't actually arrest from their problems, we wouldn't have cooled them).

-Benjamin S. Abella, MD

February 12, 2009

Q: If an athlete gets hot while playing outside with temp of 102 and he is immersed in a tub filled with ice, is there any danger of V.fib or other risk to the patient?

A: I am not aware of any risk of VF or other serious health risks to the rapid sudden cooling of a healthy person with hyperthermia, unless the patient was overcooled to a core body temperature of less than 30 C, when VF becomes a risk.

-Benjamin S. Abella, MD

February 9, 2009

Q: It seems to me that the primary damage mechanism as described by the article is oxidation by free radicals. Have you considered other methods of limiting the supply of oxygen? I am not a medical professional, but my image of a person being resuscitated involves chest compressions, electric shocks if necessary, and an oxygen mask. What would happen if instead of giving the patient extra oxygen, you gave them a limited supply of oxygen, intentionally keep the o2 level provided to the patent low through using a N2 rich blend of breathing air. You could monitor the oxygen levels in the blood and gradually bring them back up so that you don't cause cell death. Keeping the oxygen levels low might be less traumatic to the patient than intentionally causing hypothermia.

A: We have been considering a clinical trial of this, although there are some complex issues to figure out (how to easily manage the O2 "ramp up", what the comparison group should be in a trial, etc). Stay tuned!

-Benjamin S. Abella, MD

February 2, 2009

Q: I just read the article about you, Benjamin Abella and your research on hypothermia in Popular Science (Carnett, J.B. "Cold Relief" Popular Science 274(2) Pg. 54-59). Very interesting, but I was struck by how hard it was to get someone cold quickly. I just watched a TED talk by Peter Ward on mass extinctions and how breathing hydrogen sulfide can help lower body temperature. I don't know much about animal physiology, but perhaps breathing hydrogen sulfide while being on an "IV slushee" would make a person become colder faster.

A: It turns out that this is a gas of great current interest, it also appears to have cell-protective properties in small quantities. A number of laboratory experiments in simulated cardiac arrest are underway using this fascinating molecule...stay tuned! Not quite ready for human testing, I suspect. It's a tricky gas, and can be toxic at higher doses.

-Benjamin S. Abella, MD

January 13, 2009

Q: Could a Thermosuit spinoff, a 'Neck Vest' that both braces the neck and supercools it immediately, be used in cases where someone gets the same type of SCI, (Spinal Cord Injury) that crippled actor Chris Reeve, which happens a lot in sports arenas and car wrecks, to use cold temperatures in the area of their SCIs so that they DON'T get paralysed later on?

A: This is indeed an area of active investigation at a number of research centers around the country, especially the Miami Project to Cure Paralysis, a research group at the University of Miami. Many investigators believe that hypothermia, such as via a neck cuff as you describe, may hold great promise to prevent paraplegia or other injury from spinal cord trauma.

-Benjamin S. Abella, MD