The Targeted Temperature Management (TTM) trial published in the New England Journal of Medicine in November 2013 was perhaps the most important article of the year in resuscitative and critical care medicine.1
In the setting of advanced post-arrest care and a protocolized neuroprognostication system, this large, multinational randomized controlled trial compared two targeted temperature management strategies (33°C versus 36°C) in out-of-hospital post-arrest patients and found no significant difference in mortality between the two groups.
If you haven’t heard about this study yet, this is not an introductory piece. But that’s okay. The website www.EMCrit.org offers an enormous set of resources about this trial and others dealing with therapeutic hypothermia (TH) that came before it. The purpose of this piece, rather, is to discuss some important details of the study and attempt to answer the question, “So, what do we do now?”
Basis for Mild Therapeutic Hypothermia and TTM Trial
TH in post-arrest patients has been used for decades; it’s a practice largely based on theory, animal models, and retrospective analyses.2 In 2002, two groundbreaking randomized control trials, Bernard and Hypothermia After Cardiac Arrest (HACA), demonstrated a benefit in the number of patients discharged neurologically intact when cooling patients with initial shockable ventricular dysrhythmias. These patients were cooled to mild hypothermic temperatures of 32–35°C for 12 and 24 hours after arrest versus then-standard-of-care ad-lib temperatures. The control groups did not receive active temperature control.3,4
Further nonrandomized studies have shown similar survival benefit in terms of relative risk in other arrest rhythms and groups, making moderate hypothermia the standard of care for out-of-hospital post-arrest patients around the world.5
36°C may be as good as 33°C with potentially fewer complications…it’s still mild hypothermia, it’s still preventing fever.
A 2011 Cochrane review estimates that for every six patients treated with mild hypothermia after cardiac arrest, one receives survival benefit, making TH one of the most effective interventions in our modern post-arrest care arsenal.6
However, moderate TH does not come without potential adverse events such as sepsis, coagulopathy, electrolyte imbalance, and life-threatening dysrhythmia, to name a few.7
That’s why this trial is such a big deal. The TTM trial suggests that by actively preventing fever with mild hypothermia at 36°C, we can achieve the same mortality benefit without added adverse consequences.
Online Reaction & Reception
Almost immediately after its publication, the TTM trial gained widespread attention in the FOAMed and social media communities, igniting discussion and debate with both praise and caution for its future implication on clinical practice.
None of this would have been possible a decade ago. Truly, it was interesting to watch and another collective win for free open access medical education (FOAM).
The EMCrit podcast featured interviews with both Jon Rittenberger, MD, MS, associate professor of emergency medicine at the University of Pittsburgh, who coauthored the accompanying New England Journal of Medicine editorial of the trial, and Stephen Bernard, MB BS, MD, senior intensivist at The Alfred Hospital and director of intensive care at Knox Private Hospital in Victoria, Australia, and lead author of the 2002 Bernard trial.
Dr. Rittenberger interprets the TTM trial as affirmation that targeted temperature management works, only more research may be necessary to find the appropriate temperature for every subpopulation. For instance, in patient populations stratified by their organ dysfunction, the most severely brain-injured patients may still benefit more from deeper degrees of hypothermia.
In what may be the most compelling affirmation of the TTM trial, Dr. Bernard calls the study “outstanding science” that “ticks all of the boxes for an influential [randomized controlled trial].” In a display of intellectual integrity, Dr. Bernard has adopted 36°C in his ICU but adds that patients who undergo intra-arrest cooling may still benefit from cooler temperatures. Dr. Bernard also plans to adopt the TTM trial’s standardized system for prognostication.
Other experts, such as Kees Polderman, MD, of the University of Pittsburgh, Simon Carley, MB ChB, of the St. Emlyns blog (www.stemlynsblog.org), and Mike Cadogan, MB ChB, of the LifeInTheFastLane blog (www.lifeinthefastlane.com), offer a more skeptical interpretation of the data. Some of their more compelling arguments are summarized below. But first…
Please Don’t Stop Cooling Your Post-Arrest Patients!
Before getting into any more details, it is imperative to address this far too common misconception of the TTM trial. It is important to keep in mind that the Bernard and HACA trials of 2002, which helped establish 33°C as the standard of care in aggressive treatment of post-arrest patients, compared moderate hypothermia (32°–34°C) to no active temperature management at all. The average temperature of the control group in the HACA was 37.8°C, and it is plausible that even this mild elevation of temperature could absolutely be detrimental to the injured brain.
We know that TH works and fever in the post-arrest state is bad. The TTM trial does not contradict earlier studies. It was a dose-finding trial, not a trial of hypothermia versus no hypothermia. It’s a trial of tight temperature management at one temperature versus another. This is similar to trying to find the best dose of a drug.
So please do not misinterpret this trial! Subjects in the 36°C arm were still actively cooled using ice packs, cooling blankets, and intravenous cooling catheters when necessary. It is unfortunate that the paper does not specifically describe interventions made in the 36°C arm at each site, but these patients were not simply given Tylenol and a hope for the best.
What we find now is that 36°C may be as good a dose as 33°C with potentially fewer complications, and that makes a lot of sense—it’s still mild hypothermia, it’s still preventing fever, and it’s still accompanied by a profoundly aggressive and meticulous post-arrest care package, which was not seen in the control groups of previous trials.
Navigating the Nitty-Gritty
There are plenty of well-respected experts in post-arrest care who are not so quick to adopt 36°C based upon this single trial. Here is a list of their concerns, with brief summaries and analysis.
- Power of the Study: Dr. Carley was probably the first to point out that the TTM trial was a negative trial of difference, not a positive trial of equivalence. Dr. Cadogan echoed these concerns. The TTM trial was powered to find a 20 percent relative risk reduction, or 11 percent absolute risk reduction, between the two groups, and it failed to show a difference. This is not the same as proving the two therapies are equivalent in terms of mortality benefit. Statistically speaking, 33°C could confer a 10 percent absolute risk reduction versus 36°C that would have been detected had more subjects been enrolled.
Retort: That being said, there was no trend toward benefit in the data, and 939 subjects nearly triples the total used in the 2011 Cochrane review. To say that a randomized controlled trial with nearly 1,000 patients doesn’t offer much is raising the bar extremely high for future studies.
- “Small but Cumulative” Differences: In a 2014 editorial, Dr. Polderman and Joseph Varon, MD, of the University of Texas Health Science Center in Houston, suggest that while no significant differences were found in favorable versus unfavorable clinical factors between the two groups, the trend was for less-favorable factors, such as absence of pupillary reflexes for the 33°C group and more-favorable factors such as bystander-witnessed arrest and shockable rhythm in the 36°C group.8 As a whole, these small, statistically insignificant differences would accumulate to create unequal study populations favoring the 36°C group.
Retort: In terms of Glasgow Coma Scale and Sequential Organ Failure Assessment–Cardiovascular (SOFA-C) scores, these two groups were nearly identical. Another interesting piece of data was that patients in the 36°C arm dramatically improved their SOFA-C over days two to three, as opposed to the 33°C arm.
- Patient-Selection Bias? TH was already the standard of care, so patients not screened for the trial would by default receive TH. Thus, physicians may have had a subconscious incentive to not screen the patients they felt would most likely benefit from TH. On average, the 36 ICUs involved in the TTM trial screened 18 patients per year and only enrolled about 12 patients per year, or one per month, a number Dr. Polderman points out is suspiciously low and evidence of the potential effect of the subconscious incentive described above.
Retort: Lead TTM trial study author Niklas Nielsen, MD, PhD, writes, “The baseline characteristics, active care (60 percent early angiography and 40 percent coronary intervention), and survival rates strongly contradict a selection of patients with a presumed poor outcome.”
- Rapid Rate of Rewarming? Dr. Polderman also expresses concern over a rapid rate of active rewarming in the TTM trial, from 33°C to 36°C in six hours, or 0.5°C per hour, which is greater than in previous trials, and potentially enough to negate the potential benefits of TH.
Retort: Dr. Nielsen states on EMCrit that the guideline for each institution was to actively re-warm at a rate no greater than 0.5°C per hour, and the actual average of rate was closer to previous trials at 0.36°C per hour, which closely correlates with the HACA trial rewarming rates. Also, there is no definitive evidence that the rate of rewarming leads to a change in outcome.
At the virtual hospital Janus General, Dr. Weingart is targeting a temperature range between 35°C and 36°C for ventricular fibrillation, ventricular tachycardia, and pulseless electrical activity patients who qualify for an aggressive treatment path. Unwitnessed asystolic arrest patients were left out of the HACA, Bernard, and TTM trials. In this group, there is little guidance, and it may be reasonable to continue cooling to 33°C because these patients are most likely to have the most severe post-arrest neurologic injury. Certainly, more evidence may surface that identifies subpopulations that benefit at different temperatures or duration of temperature managements, but for now, the TTM appears to be the best evidence out there.
It also important to mention that for low-volume centers that do not see large numbers of post-arrest patients, 36°C may be an easier and equally effective option, without the difficulty of managing all the potential complications of hypothermia.
Dr. Weingart is an ED intensivist. This column is a distillation of the best material from the EMCrit Blog and Podcast (http://emcrit.org).
Dr. Cox is a resident in emergency medicine at the Icahn School of Medicine at Mount Sinai in New York City.
- Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C versus 36°C after cardiac arrest. N Engl J Med. 2013;369:2197-2206.
- Benson DW, Williams GR Jr, Spencer FC, Yates AJ. The use of hypothermia after cardiac arrest. Anesth Analg. 1959;38:423-428.
- Bernard SA, Gray TW, Buist MD, et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346:557-563.
- The Hypothermia After Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve neurologic outcome after cardiac arrest. N Engl J Med. 2002;346:549-556.
- Kim YM, Yim HW, Jeong SH, Klem ML, Callaway CW. Does therapeutic hypothermia benefit adult cardiac arrest patients presenting with non-shockable rhythms? A systematic review and meta-analysis of randomized and non-randomized studies. Resusitation. 2012;83:188-196.
- Arrich J, Holzer M, Herkner H, Müllner M. Hypothermia for neuroprotection in adults after cardiopulmonary resuscitation. Cochrane Database Syst Rev. 2009;(4):CD004128.
- Bernard S. Post-cardiac arrest care in 2013 with Stephen Bernard. EMCRIT Podcast. 2013. Available at http://emcrit.org/podcasts/post-arrest-care-2013-i/.
- Polderman KH, Varon J. We should not abandon therapeutic cooling after cardiac arrest. Crit Care. 2014;18:130.