Hyperthermia after resuscitation from cardiac arrest has been shown to be associated with poor outcomes and an increase in mortality.
Use of induced hypothermia to goal 32-34˚C, after return of spontaneous circulation (ROSC) in cardiac arrest patients, has been widely adopted and supported based on small studies from 2002 showing benefit in survival and neurologic outcome.
More recent and robust data seems to refute these findings, and has opened the debate even further, leaving the future of post-cardiac arrest care up for discussion.
Landscape of post-cardiac arrest care changed after two studies were published in NEJM in 2002.
Bernard et al performed a randomized study of 77 patients with ROSC after ventricular fibrillation (VF) arrest.
Patients were assigned to either a reduction in core body temperature to 33˚C, or a control group where temperatures were not controlled.
Results showed a mortality benefit of 16% in the hypothermic group and a 23% increase in patients discharged with a good neurologic outcome in the same intervention group.
The second trial, the Hypothermia after Cardiac Arrest (HACA) study by Holzer et al, included 275 patients with either ventricular tachycardia (VT) or VF arrest after ROSC.
Patients were randomized to a reduction of core body temperature to 33˚C or to a group where “normothermia was maintained.”
Findings included a 16% increase in favorable neurologic outcome, and a 14% decrease in mortality, both favoring the hypothermia group.
These statistically significant findings seemed to indicate that further studies were needed, as the trials were small in size and proper blinding was not used.
The studies also only analyzed VF and VT arrests.
Despite the limited evidence, a large amount of support was generated for the use of hypothermia in these patients, and EMS and hospital policies were completely redesigned to include hypothermia protocols post-cardiac arrest.
Some hospital policies even began to include hypothermic recommendations for patients without VF arrest.
A Cochrane Database systematic review in 2009 recommended the use of therapeutic hypothermia, based on moderate level of evidence.
International Resuscitation Guidelines also supported mild therapeutic hypothermia as the best medical practice.
For the decade following the release of the 2002 studies, the healthcare industry took the presented evidence despite its limitations, and has made it the standard of care.
It was not until an RCT published in 2013 that this widely accepted intervention is now back to the forefront of controversy.
In late 2013, one large multi-center RCT about Targeted Temperature Management (TTM) was published in NEJM by Nielsen et al.
TTM trial is the most extensive RCT examining hypothermia in out-of-hospital cardiac arrest.
Unlike the 2002 studies, patients were included regardless of initial rhythm.
Sample size of 939 patients was included, with randomization to two groups.
In contrast to the earlier studies, patients were not randomized to hypothermic and normothermic groups, but rather to groups with goal temperature of 33˚C and 36˚C.
Target temperatures were maintained for 36 hours before rewarming.
Mortality was slightly improved in the 36˚C group compared to the 33˚C group (48% to 50%), and better functional outcomes were seen as well.
Shorter ICU and hospital stays were seen in the 36˚C group as well.
Overall, results showed small benefits in the near-normothermic group, though not statistically significant findings. No differences were found in subgroup analyses.
The findings of this large trial seem to oppose the results of the previous studies that prompted great reform.
Potential causes of these contradictory findings include the small sample size of the previous studies, and the fact that patients in the control groups of the earlier studies did not have temperature closely monitored, and were even allowed to become hyperthermic.
The allowance of fever in these patients may have increased mortality, as opposed to the intervention of hypothermia decreasing mortality.
In the TTM trial, both groups were essentially intervention groups – though to different target temperatures – and were maintained at those temperatures with great attention and care.
In this sense, both groups were treated with the same level of care. Staff caring for patients could not be blinded, but independent neurologic prognostication and data interpretation were blinded.
Prior trials have suggested better outcome in VF arrests.
The 2002 studies analyzed only VT/VF arrests, while the TTM trial included patients irrespective of initial rhythm.
Of note, subgroup analysis of the TTM trial showed no benefit in cooling to 33˚C in the shockable rhythm group, compared to 36˚C.
Another possible effect on the data is the improvement in ICU care in the decade between the landmarks studies. The mortality rate in both TTM groups is lower than the control group of the HACA trial.
2013 study in JAMA investigated pre-hospital cooling initiation.
1,359 patients with out-of-hospital cardiac arrest with ROSC were included, with patients randomized to either rapid cooling in the field or standard pre-hospital care.
No improvement in survival or neurologic outcome was seen in either those with initial VF or those without VF.
Of note, rapidly cooled patients had a statistically significant rate of re-arrests in the field, with more complications as well – with a number needed to harm of 20.
The results of this large study seem to indicate that pre-hospital cooling is a thing of the past, as it harms patients.
There is great debate as to the evolution of post-cardiac arrest care after the TTM trial. The general consensus is that it was overzealous to take the findings of the smaller, earlier studies and to accept them as standard of care prior to further investigation and validation.
Bottom Line/Pearls & Pitfalls
The future of post-cardiac arrest temperature management is up in the air with the current debate.
Many agree that with the robust Targeted Temperature Management study’s findings, a goal temperature of 36˚C for out-of-hospital cardiac arrest with return of spontaneous circulation should be the target. Temperature control lower than 36˚C is unlikely to improve outcomes.
An emphasis on preventing hyperthermia is very important in these patients, and it appears to be responsible for poorer outcomes.
As a result, hospital protocols are expected to change the new target temperature post-arrest to 36˚C, with strict control of the goal temperature.
Target temperature appears to work mainly by avoidance of fever leading to a decrease in metabolic demand and temperature-related tissue injury.
The key point is that cooling to 36˚C (low-normal temperatures) is not the same as not regulating temperature. Paying close attention to temperature will make survival a more likely outcome.
Controversy still exists about post-cardiac arrest temperature management for best neurologic outcome and mortality, and the debate will continue.
Further studies are already underway to further investigate the topic.
Uncertainty persists regarding optimal temperature, different target temperatures for different patients (possible lower goal temperature for asystole), when to start cooling, stratification of patients, whether treatment should differ based on initial rhythm, and duration of cooling.
This is still a contentious topic, and there will undoubtedly be more information and opinions on the subject to come.