Well before Han Solo was frozen in carbonite, before Sigourney Weaver crossed galaxies in cryostasis, or Walt Disney was cryopreserved, we have been fascinated by the stasis-like powers of hypothermia. Given this enthusiasm, we most likely would have reacted with just as much vigor even if the initial trials of hypothermia for out of hospital cardiac arrest (OOHCA) were half as positive as they appeared.
The two trials that ushered in the age of therapeutic hypothermia were published in the February 2002 issue of the NEJM. The first by Bernard et al was a study of 77 subjects, who presented with ventricular fibrillation. Patients were randomized to either a reduction in core body temperature to 33 degrees Celsius for 12 hours, or a control group where patients were left to their own thermoregulatory devices (1). The authors found an absolute mortality benefit of approximately 16% favoring the hypothermia group and a 23% absolute difference in patients discharged from the hospital with what was considered to be a “good” neurological outcome. The second trial by Holzer et al appeared in the same issue of NEJM and examined a total of 275 patients (2). This trial included patients who presented with either pulseless ventricular tachycardia or ventricular fibrillation. Patients randomized to the hypothermia group were reduced to a core temperature of 33 degrees Celsius for 24 hours. Conversely in the normothermic group “patients were placed on a conventional hospital bed and normothermia was maintained”. In this trial not only was there a 16% absolute difference in favorable neurological outcomes, a 14% absolute decrease in mortality was reported in the hypothermia group.
These trials were stunningly successful and not soon after their publication there was a cry for global implementation of hypothermia protocols. There were those who were more hesitant along the way. They reminded us that trials of this size performed without proper blinding or controls were hypothesis generating only and more definitive research was needed before moving forward. Seemingly these voices went unnoticed. Ignored in part because of the excitement over the arrival of futuristic medicine. It was not wrong for us to look towards therapeutic hypothermia as a new hope, even as a possible cure for what once was a terminal illness. Our mistake was jumping in wholeheartedly without requiring further confirmatory studies. Based on a total of 352 patients and a science fiction dream we mobilized a healthcare system and created an industry. Instead of taking a step back, we expanded. No longer were V-fib or V-tach arrests enough, we soon began cooling any soul who at some point may have lost a pulse. There was even talk of cooling STEMI patients with full neurological function(9), so it was only a matter of time before the bubble burst. This reality check came in the form of the Targeted Temperature Management (TTM) Trial (3).
The TTM trial was published by Nielsen et al online in the NEJM on November 17th 2013. To date it is the largest RCT examining therapeutic hypothermia in out-of- hospital cardiac arrests (OOHCA). The trial included 939 people, close to three times the amount found in the original two trials combined. What was unique about this trial was that the authors chose not to randomize patients to a hypothermic and normothermic group. Rather they randomized patients to either a goal temperature of 33 or 36 degrees Celsius. Patients were maintained at their target temperature for 36 hours before being rewarmed. The authors found no statistical difference in mortality or neurological outcome between the groups. Even the meager trends toward benefit were actually found in the near-normothermic group, who had a lower mortality rate (48% vs 50%), better functional outcomes (CPC of 3-5 52% vs 54%), and shorter ICU and hospital stays(3).
Already many have speculated on the meaning of these results. Dr. John Rittenburger, a leader in the field of EM research, in the editorial published alongside this paper proposes that it is not the hypothermia that is beneficial in post-cardiac arrest patients, but rather preventing the hyperthermia that is commonly associated with the disease process(4). He elegantly explains that in the early trials of therapeutic hypothermia the normothermic groups were allowed to become hyperthermic during the study protocol and it was this thermal insult which resulted in the dramatic difference in outcomes.
There is an alternative explanation for the difference in outcomes the TTM trial discovered when compared to the earlier trials. The near-normothermic group in the TTM trial may have served as an active placebo of sorts. In the earlier trial the patients in the normothermic groups were left to their own devices. Though they received all the traditional care given to patients in post-cardiac arrest they did not have the degree of attention paid to them compared to those in the hypothermia groups. We know that open trials often lead to unbalanced care even when precautions are taken to prevent this and the placebo response in such cases is powerful(5,6,7,10). The increased attention the hypothermia groups received because of the careful observation needed to maintain a cooler core temperature may have biased the results. It is possible for an intervention to provide a large placebo effect and yet provide very little specific benefits. In trials examining treatments for migraines both accupucture and sham-accupucture provided equal reduction in symptom burden, but were both equally superior to a placebo pill. They each demonstrated a large placebo effect with very little specific benefits (8). This is what is called the Efficacy Paradox and very well may account for positive results found in the early trials on therapeutic hypothermia and why TTM found no benefit (8,10).
The near-normothermic group (or for our purposes the “sham hypothermic group”) provides as close to a placebo group as can be obtained. In the sham hypothermia group the medical team was mandated to maintain a core temperature of 36 degrees C for a 36-hour period. Given these constraints the near-normothermic group now required a similar amount of attention to the hypothermia group. When the hypothermic treatment arm no longer received added care over and above its control, the benefits of hypothermia were no longer evident.
It is obvious that the definitive answer on temperature management in post-arrest resuscitation has not been written. It may be that there is still benefit to thermoregulatory stasis in post-arrest patients. The hopes of futuristic medicine made by science fiction visionaries are not completely dashed. More studies are needed to answer these questions. We may ultimately find that the benefit seen from the early hypothermia trials was due to the increased time and care invested in those patients. It is this time and care we invest in our patients that makes the ultimate difference, a far more traditional and “antiquated” thought but promising nonetheless.
Sources Cited
1. 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;346:557-63.
2. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002;346:549-56.
3.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
4. Rittenburg eta l Temperature Management and Modern Post–Cardiac Arrest Care. DOI: 10.1056/NEJMe1312700
5. Noseworthy et al. The impact of blinding on the results of a randomized, placebo-controlled multiple sclerosis clinical trial. Neurology 1994; 44:16–20.
6. Schulz KF, Chalmers I, Altman DG. The landscape and lexicon of blinding in randomized trials. Ann Intern Med 2002; 136: 254–59.
7. Schulz et al. Blinding in randomised trials: hiding who got what. THE LANCET • Vol 359 • February 23, 2002
8. Meissner et al. Differential Effectiveness of Placebo Treatments: A Systematic Review of Migraine Prophylaxis. JAMA Intern Med. Published online October 14, 2013. doi:10.1001/jamainternmed.2013.10391
9. Hung et al. A pilot study: The Noninvasive Surface Cooling Thermoregulatory System for Mild Hypothermia Induction in Acute Myocardial Infarction (The NICAMI Study). American Heart Journal Volume 150, Issue 5, November 2005, Pages 933.e9–933.e13
10. Kunz et al. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ Volume 317 31 October 1998
- EM Nerd-The Case of the Partial Cohort - May 24, 2020
- EM Nerd: The Case of the Sour Remedy Continues - January 20, 2020
- EM Nerd-The Case of the Adjacent Contradictions - December 23, 2019
Great read and excellent style. You might want to examine the problem behind a lot of these studies. Here, for instance, hypothermia is the ONE item that is mooted as being beneficial. Equally it could be some other treatment e.g. ARDSNET and low ventilation pressures for ARDS and then extended to all ventilated patients or the tight glucose control of the critically ill. All promote a one idea. Rivers, in contrast, did not focus on one item, but said we need to get going and treat the sepsis whilst getting antibiotics onboard, oxygen, fluids, and tests and all in a… Read more »
Thanks Andrew, I think there were multiple opportunities for bias in the early hypothermia trials which were open label with no placebo. You would hope the randomization and similar treatment protocols other than the hypothermia would control for such inequality but with such a small n and with no placebo these factors could have easily played a role. As for the Rivers trial, I don’t think we have the answer yet. Was it the EGDT that improved outcomes or just the fluids, antibiotics and the increased attention the EGDT group experienced. There are a numbers of large trials currently underway… Read more »