Animal and human models of the effects of therapeutic hypothermia, now called targeted temperature management (TTM), began to surface in the late 1980s. The first randomized clinical trial employing TTM as a neuroprotective strategy following cardiac arrest did not appear until the early 2000s. When compared with normothermia, the HACA trial (Holzer M, et al. N Engl J Med. 2002;346[8]:549-56) demonstrated a 14% reduction in mortality and improved neurologic outcomes following out of hospital cardiac arrest (OHCA) due to ventricular fibrillation (VF) or ventricular tachycardia (VT) when maintaining body temperature between 32˚C and 34˚C post-arrest. Following the results of this trial, TTM in comatose patients following cardiac arrest was recommended by international guidelines and became the standard of care. It was not until the publication of the TTM1 trial (Nielsen N, et al. N Engl J Med. 2013;369[23]:2197-206) about a decade later, that serious questions regarding the efficacy of TTM were raised. The TTM1 trial showed no difference in mortality or neurologic outcomes when comparing TTM at 33˚C vs 36˚C for OHCA. The results of this trial heralded widespread practice change, with many abandoning deep cooling, and often active cooling measures, in favor of fever avoidance. The HYPERION trial (Lascarrou J, et al. N Engl J Med. 2019;381:2327-37) came next, comparing TTM at 33˚C to normothermia (<37.5˚C) for cardiac arrest with non-hockable rhythm. This study did not identify any improvement in mortality with utilization of TTM but suggested it may be associated with more favorable neurologic outcomes, albeit in a small number of patients.
The TTM2 trial (Dankiewicz J, et al. N Engl J Med. 2021;384:2283-94) is the most recent trial to address the question of TTM post-cardiac arrest. The TTM2 trial was an international, randomized controlled superiority trial of TTM at 33˚C vs normothermia (≤37.8˚C) for patients with coma following OHCA with any initial rhythm. It was conducted by the same group as the TTM1 trial and, to date, represents the largest (N= 1,850) and most robust trial conducted in this area. The trial spanned 61 institutions across 14 countries and had nearly complete follow-up at 6 months. Once again, there was no significant difference in all-cause mortality at 6 months in the TTM group when compared with the normothermia group. Equally important, there were no differences observed in secondary outcomes, including functional neurologic status and health-related quality of life at 6 months. With the results of the TTM1 and TTM2 trials failing to show any neurologic or mortality benefit to TTM, we are left wondering, is there anything therapeutic about “therapeutic hypothermia”?
Both the 2020 American Heart Association (AHA) and 2021 European Resuscitation Council (ERC) guidelines predate this trial; they recommend cooling any OHCA or in-hospital cardiac arrest (IHCA) patient who remains unresponsive after return of spontaneous circulation (ROSC) regardless of initial rhythm. They further suggest maintaining a target temperature between 32˚C and 36˚C for at least 24 hours, followed by avoidance of fever (>37.7˚C) for at least 72 hours after ROSC in patients who remain comatose. While it will be interesting to see what future iterations of the guidelines recommend, the results from the TTM1 and TTM2 trials support a shift in clinical practice away from TTM and toward more active fever avoidance. Additionally, careful review of adverse events in the TTM2 trial suggests that induced hypothermia is not without risk of harm. When compared with the normothermia group in the TTM2 trial, the hypothermia group experienced higher rates of arrhythmias with hemodynamic instability (16% vs 24%), increased exposure to sedation, increased use of neuromuscular blockade, and increased duration of mechanical ventilation.
While the results of the TTM2 trial move the needle away from therapeutic hypothermia for OHCA patients, there is some nuance that warrants further discussion. First, the initial HACA trial, upon which the standard of TTM was based, included only patients with an initial shockable rhythm (VT/VF). Inherently, the etiology of these arrests is likely to be cardiac and more reversible in nature. Most subsequent landmark trials on TTM, including the TTM2 trial, have included OHCA patients with both shockable and nonshockable initial rhythms. Still, the majority of patients in the TTM2 trial had an initial shockable rhythm on presentation (72% hypothermia vs 75% normothermia). This may limit broad generalizability of study findings as an increasing number of OHCA patients are presenting with nonshockable initial rhythms. Next, it is well known that bystander CPR improves outcomes following OHCA. Impressively, over 75% of patients in both groups in the TTM2 trial received bystander CPR compared with an average of 46% of arrest patients in the US according to AHA data. Finally, like most of its predecessors, the TTM2 trial only included OHCA patients meaning no real conclusions can be drawn regarding application of TTM to IHCA patients. Of the major trials to date, only the HYPERION trial included IHCA patients – representing about 25% of the study population. Thus, the utility of TTM in the setting of IHCA remains largely unknown.
Taken in summation, recent trials, including TTM2, suggest that fever-avoidance post-cardiac arrest is likely the best option for improving mortality and neurologic outcomes while mitigating risk to the patient. We must remain vigilant in our enforcement of normothermia though as worse neurologic outcomes have been observed with hyperthermia in the early post-arrest period (Zeiner A, et al. Arch Intern Med. 2001;161[16]:2007-12). A key takeaway from recent trials is that maintaining normothermia without active temperature control measures is likely to be difficult to achieve. A criticism of the HYPERION trial was that a “substantial proportion” of patients in the normothermia group had temperatures above 38˚C. Similarly, 10% to15% of patients in the TTM2 trial had body temperatures above 37.7˚C, 40 to 72 hours after randomization and, ultimately, 46% of patients in the normothermia group required cooling with a temperature management device. Thus, we can conclude that maintenance of strict normothermia will likely continue to require active control with a temperature management device.
Despite an increasing number of well conducted studies in this area, there are several questions that remain unanswered. The first is whether cooling patients even earlier post-arrest is felt to increase the likelihood of survival with improved neurologic outcomes. Like HACA and HYPERION, the rate of cooling in the TTM2 trial was relatively quick with a time to randomization after onset of cardiac arrest of about 2 hours in both groups and a median time from intervention until reaching target temperature of 3 hours. While some retrospective data suggest ultra-early cooling may be beneficial, neither induction of therapeutic hypothermia during OHCA using a rapid infusion of cold saline (Bernard SA, et al. Circulation. 2016;134[11]:797-805) nor transnasal evaporative cooling in the pre-hospital setting (Nordeberg P, et al. JAMA. 2019;321(17):1677-85) has shown improvement in survival with good neurologic outcomes. Next, if we are going to continue TTM, the TTM2 trial does not provide guidance on optimal duration of cooling. Although the current guidelines are to cool for at least 24 hours after ROSC, it is unclear for how long strict temperature control should be continued. The currently enrolling ICECAP study aims to further elucidate the optimal duration of TTM for OHCA patients with both shockable and non-shockable initial rhythms.
Post-cardiac arrest management continues to be a significant area of interest in clinical research and for good reason. Although steady improvement has occurred with regards to survival and neurologic function for IHCA, of the approximately 350,000 nontraumatic OHCA that occur in a year in the United States, only about 10.2% of those patients will survive their initial hospitalization, and only about 8.2% of those who survive will have good functional status (American Heart Association. Circulation. 2020;142(suppl 2):S366-S468). There remains much room for continued study and improvement.
Dr. Capp is a Pulmonary and Critical Care Fellow; and Dr. Pendleton is Assistant Professor of Medicine; Division of Pulmonary, Allergy, Critical Care, and Sleep Medicine; University of Minnesota, Minneapolis, Minnesota.