EMCrit Wee – The Targeted Temperature Trial Changes Everything


The TTM Trial was just published today in the NEJM (Nielsen et al. Targeted Temperature Management at 33°C versus 36°C after Cardiac Arrest NEJM 2013;epub Nov 17, 2013)

For my take and the take of Jon Rittenberger, come to the 2nd TTM Post

but even more important is to hear from the primary author himself: Niklas Nielsen on the TTM Trial

Study Design

International multicenter RCT

Inclusion criteria: Age >= 18 years, out-of-hospital cardiac arrest of presumed cardiac cause, unconsciousness (Glasgow Coma Score <8) after sustained return of spontaneous circulation (ROSC) (20 minutes of circulation).

Exclusion criteria: Conscious patients, pregnancy , out-of-hospital cardiac arrest of presumed non-cardiac cause, cardiac arrest after arrival in hospital, known bleeding diathesis, suspected or confirmed acute intracranial bleeding, suspected or confirmed acute stroke, temperature on admission <30°C, unwitnessed asystole, persistent cardiogenic shock, known limitations in therapy, known disease making 180 day survival unlikely, known pre-arrest cerebral performance category 3 or 4, >240 minutes from ROSC to randomisation.

Primary outcome: Survival to end of trial (at least 180 days).

Secondary outcomes: Composite outcomes of all-cause mortality and poor neurological function (Cerebral Performance Category (CPC) 3 and 4 and modified Rankin Scale (mRS) 4 and 5) at 180 days. All – cause mortality and CPC and mR S at 180 – days. Adverse events: Bleeding, pneumonia, sepsis, electrolyte disorders, hyperglycaemia, hypoglycaemia, cardiac arrhythmia, renal replacement therapy.

Tertiary outcomes: Complete neurological recovery. Best neurological outcome during trial perio d. Quality of life according to SF – 36. Biomarkers at 24, 48 and 72 hours

Intervention: The core body temperature will be set as quickly as possible at the predefined target temperature, according to intervention allocation, with 4°C intravenous solutions, 43 ice – packs 8, 44 and commercially available cooling devices 45 at the discretion of the treating physician . The target core temperature is then maintained for 24 h. After the maintenance period core temperature is gradually raised to normothermia of 37°C during 8 hours with a rewarming rate of 0.5°C/hour in both groups. Body temperature is then maintained at normothermia 37 ±0.5°C until 72 hours from sustained ROSC in both treatment groups, as long as the patient is in the ICU, using pharmacological treatment and temperature management systems when applicable

See all of the nitty-gritty in the appendix
rewarmed at 0.5 C/hr and then induced normothermia (37.5 C) for 36 hrs post arrest

Nonblinded to temp allocation, but neuro assessment was blinded

fluids icepacks surface and intravasc cooling– 1/4 intravasc and rest surface
900 pts give 90% power to detect 20% difference
80% shockable, 12% asystole
Table S2 – SOFA-C Scores in the first 72 hours


*SOFA denotes Sequential Organ Failure Assessment, SOFA-C denotes the cardiovascular subcomponent of the SOFA score. SOFA-C=0 No need for inotrope or vasopressor, mean arterial pressure (MAP) > 70mmHg, SOFA-C =1 MAP < 70mmHg, SOFA-C=2 any dose of dobutamine or dopamine <5 ìg/kg/minute, SOFA-C=3 dopamine 5-15 ìg/kg/minute or epinephrine or nor-epinephrine <0.1 ìg/kg/minute, SOFA-C=4 dopamine >15 ìg/kg/minute or epinephrine or nor-epinephrine >0.1 ìg/kg/minute.

Neuroprognostication from the TTM Trial

Excerpted from Life in the Fast Lane CCC

The TTM trial (Nielsen et al, 2013) used a standardised protocol for neurological prognostication to guide decisions regarding treatment withdrawal following targeted temperature management post-cardiac arrest:

  • All patients in the trial were actively treated until a minimum 72 hours after the intervention period, i.e. after rewarming, when neurological evaluation was done on patients not regaining consciousness.

Exceptions from this rule were

  • (1) patients with myoclonus status in the first 24 hours after admission and a bilateral absence of N20-peak on median nerve somatosensory evoked potentials (SSEP)
  • (2) patients who became brain dead due to cerebral herniation, and
  • (3) because of ethical reasons described below.

At that time-point, limitations in and withdrawal of therapy could be instituted by the treating physicians. The neurological evaluation was based on:

  • clinical neurological examination (including Glasgow Coma Scale (GCS), pupillary and corneal reflexes)
  • SSEP and electroencephalogram (EEG)
  • Biomarkers for brain damage were not used for operational prognostication

Findings allowing for discontinuation of active intensive care:

  • Brain death due to cerebral herniation
  • Severe myoclonus status in the first 24 hours after admission and a bilateral absence of N20-peak on median nerve SSEP
  • Minimum 72 hours after the intervention period: persisting coma with a Glasgow Motor Score 1-2 and bilateral absence of N20-peak on median nerve SSEP.
  • Minimum 72 hours after the end of the intervention period: persisting coma with a Glasgow Motor Score 1-2 and a treatment refractory status epilepticus (see TTM trial supplement for definition)

Patients with Glasgow Motor Score 1-2 at 72 hours or later who had retained N20-peak on the SSEP, or patients in hospitals where SSEP was not  available were:

  • re-examined daily
  •  the limitations/withdrawal of intensive care considered if GCS-Motor did not improve and metabolic and pharmacological affection was ruled out

Active treatment could be withdrawn prior to 72 hours after the intervention period for ethical  reasons

  • for instance: previously unknown information about disseminated end-stage cancer  or refractory shock with end-stage multiorgan failure
  • However assumptions of a poor  neurological function were not allowed be the sole reason for withdrawal of active treatment  prior to 72 h after the intervention period (exception: brain death and early myoclonus status including a negative SSEP)

Should we change right now?

In the coming months, you may see induced hypothermia protocols changing to induced normothermia protocols. Read the study and tell me what you think in the comments. This article just published in Resuscitation adds additional credence to the concept of: fever=bad.(23153649)


Here is Ilcor’s interim statement on the TTM Trial

And here is an excellent rebuttal from Kees Polderman on the ditching of 33 (Polderman Kees H. and Varon Joseph. Therapeutic Hypothermia and Temperature Management. June 2015, 5(2): 73-76. doi:10.1089/ther.2014.0031. )

Now on to the Wee…


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 2 years ago. We never spam; we hate spammers! Spammers probably work for the Joint Commission.


  1. tac57 says

    NOTE: These are more questions for my own learning (I’m new to this whole EBM thing). Take any of these with a pound of salt.

    Why are the p-values in the NEJM study so high for most of their findings?

    The studies were both well done with regards to methodology.

    The NEJM study had a very good bystander CPR rate (median 1 min for both groups). What was the rate of bystander CPR in the original studies?

    Another question – what were the vent settings used/how was oxygenation managed in these patients?
    What were the PaO2 values at each time-point for the TH vs 36 deg C groups? We know that hyperoxia alone +/- TH/TTM will worsen outcomes. Just wondering.

    Would “lumping” the VF/VT arrests in with the asystole/PEA arrests change the numbers at all?

    • says

      This is holding trials such as this up to a much different standard than any of the other clinical trials out there that compare a new therapy to standard care. To say a 1000 patient hypothermia study doesn’t do much is probably to set one’s expectations way too high for future research.

      • Sean Marshall says

        I’m not disagreeing with you that an adequately powered study would be difficult to do here but I don’t think that takes away from Simon’s statistical analysis. It’s pretty easy to let studies like this lead us in directions that the data doesn’t support. We have prior data that did find significant difference between TH and nothing. Now we have a study that didn’t find significant difference between 33 and 36C, but was only powered to find a pretty huge difference. So we can trust this study to say there is <11% mortality difference between these 2 groups, but not that they are the same. Perhaps subgroup analysis, as suggested in your podcast, is where this should go next. Fascinating discussion… gotta love FOAMed

      • tac57 says

        Also, it is interesting to note how St. Emlyn’s mentions CRASH 2. This trial did manage to enroll a cohort of ~20 000 pts.

        Running a RCT of that scale for TTM at 33 vs 36 would be possible and has been done for RCT’s of other interventions – one of the the only problems being that the logistics would be (VERY) nasty.

        • Roslan says

          Any future RCTs will have this problems as advancement in medical care kicks in. As mortality drops, we need bigger sample size. The solution is either change the primary outcome (ie MOPETT trial) or bundling it in al together (ie EGDT). However even if we were unable to find 20 000 pt to validate the 2% differences, it is too small and the benefit and conveniences of 36 C is too great to pass. For us who has not even started TTM extensively due to practicality/cost/logistic issues, this study is a gift from heaven.

          Emergency Physician
          Sarawak, Malaysia

  2. John Hinds says

    Will this change our practice much?
    Probably not.

    Here’s what we will most likely be taking out of it, based on all that’s come through from the AHA meeting and other publications:

    We will continue to not warm in the ED, and will still be inserting a CVC cooling catheter in these patients once we get them into ICU.
    We may well remove the “2 litres of 4 degree saline IV” section from our protocol in the ED also; as there is now prehospital evidence of harm/no benefit which might be transferable. Jury is out.

    The target temp will alter to achieve low normothermia / mild hypothermia (34 – 36 degree as opposed to our current practice of 32 – 34 degree)

    I doubt that we will start to cuarise* them for the first 24 hours. Because we use a central catheter to cool; our shivering instance is pretty much zero, and always controlled by gentle surface warming. Need to dissect the mechanism for this a bit more and have a think about it.

    I suspect that the biggest impact and change since Bernard/HACA and the most recent NEJM paper is actually the overall quality of care package that ROSC patients are receiving. The impact of cooling/forced normothermia may well be buried now in an overall good package of care (ventilation, normocarbia, early revacularisation, accepted low oxygenation etc.)

    Is it analogous to the death of APC? Has improvements in ROSC care followed sepsis care back in the day; and getting good at simple measures outstripped the impact of the magic bullet….?

    Interesting times!


    *spell checker changed this to “I doubt we will start CURSING them for the first 24 hours”, which I found amusing. We already do that, after all….

      • John Hinds says

        Hi Scott,

        I suspect it will change our practice to targeting 34 – 36 degrees; though we are still thrashing it out looking at the overall balance of evidence

        The catheters we use are very, very good – when you dial in a temp, that’s exactly what you get. Our real term audited figures are bang-on what we set the machine for.

        As such, I’d be happy aiming for 34-36 degrees and not expect to have any unanticipated overshoots to hyperthermia.
        Indeed, we currently use them for the controlled rewarming phase in our patients at the moment, and never have any overshoots whilst the catheter is in.

        I don’t think the jump to from 32 -34 t0 34 – 36 is massive. The HACA trial, for example, targeted 32 – 34; but if you look at their measured temp figures, none of their 75th centile readings hit 32 degrees sustained at any stage.

        I’m still just speculating, but I suspect that yes, we as a hospital group will most likely be aiming for 34 -36 after the dust has settled and all the evidence scrutinised


          • John Hinds says

            Thanks Scott,

            SOFA C is just a surrogate marker of inotrope use though.
            We already know that hypothermia reduces inotrope requirements in cardiogenic shock, so would you not just chalk up the SOFA C variance as an observational vagrancy of the intervention itself?

            The inotrope requirement did not bear out a mortality difference, so is it not little more than a “number on the bedside pump?”


            • says


              I misread the table that same way the first time (b/c I expected hypothermia group to need less pressors). It was exactly the opposite–36 got off pressors more frequently than 33

  3. René Verbeek says

    Hi Scott,

    The original studies looked at shockable rhythms, this study at all rhythms. How does this study provide us evidence to change our hypothermia protocols for the shockable rhythms?

    Kind regards,

  4. says

    hey EMCrit family

    would like to comment on the target population of this study.

    The study excluded patients who are still in arrest at the time of arrival in the ED. Furthermore, a large percentage of the patients in the study (80%) had shockable rhythms, and 73% had bystander CPR, with the median time to initiation of BLS being 1minute. Though median time to ROSC was 25min, it strikes me that the patients included in the study may have substantially high degrees of brain perfusion during their arrest events as supported by moderate acidemia (pH 7.2 +/- 0.2) and moderate elevations in lactate (6.7 +/- 4.5). As such, the patients included here may not be the ones most likely to benefit from therapeutic hypothermia. The authors allude to this fact in the discussion — “The population of patients with cardiac arrest is heterogeneous, and the potential risks and benefits of temperature intervention may not be the same across subgroups.” — though stop short of acknowledging this explicitly in their limitations.

    I propose that the data reported in this study does not support the elimination of TH protocols for patients with prolonged arrest events who arrive in the ED in persistent arrest. Though these patients may never be shown to benefit from TH, no such conclusion may be derived from the current study.

  5. says

    Great podcast Scott, and key to point out as you did that 36 isn’t the old “standard of care.” I don’t think we’ve yet seen the last of hypothermia, and maybe really deep is better… Have you ever seen the vids of Peter Safar’s re-animated dogs?

    I ramble a little more about it here:




    • Greg Kelly says

      Hi all – I’m presenting this paper at a meeting today so I’m enjoying reading everyone’s insightful comments!

      Phil – I think the Safar paper that you mention is interesting but it shows only that hypothermia can prevent primary neurological injury as the dogs got cooled 2 minutes into their arrests. Hypothermia to prevent anticipated primary neurological injury is well established, e.g. in cardiac surgery.

      In an unexpected arrest that would require intra-arrest cooling. The cooling that Nielsen et. al performed was some time post arrest (median 25 minutes to ROSC, cooling started < 4hrs post ROSC) and can only prevent secondary neurological injury which is related to inflammation, oedema, etc, rather than acute hypoperfusion/ischemia.

      That's not to say that deep hypothermia wouldn't be better in the post arrest scenario though….


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