High Quality Targeted Temperature Management (TTM) After Cardiac Arrest

Post-anoxic brain damage is the most dramatic complication of cardiac arrest [1]. In international guidelines, targeted temperature management (TTM) is the only neuroprotective intervention currently recommended after out-of-hospital cardiac arrest (OHCA) [2]. Nevertheless, the scientific community has raised concerns about the level of evidence supporting this recommendation [3]. Two early randomized clinical trials (RCTs) showed that TTM at 33 °C for 12–24 h was associated with a greater proportion of survivors with intact neurological recovery compared to standard care in OHCA survivors with witnessed shockable rhythm [4, 5], but subsequent observational studies questioned the efficacy of this intervention in other settings, such as non-shockable rhythms and in-hospital cardiac arrest (IHCA) [6, 7]. After publication of the so-called “TTM trial” in 2013, which showed similar survival and neurological recovery rates in OHCA patients treated at 33 °C or at 36 °C for 24 h [8], the use of TTM decreased significantly [9, 10] as many physicians considered that keeping patient body temperature within normothermic ranges (i.e., at about 37 °C) would likely be as effective as using TTM at 36 °C, without the adverse events related to cooling procedures, including use of sedative drugs.

Many “supporters” of TTM criticized the “TTM trial” [8], emphasizing that a number of features, including the high patient heterogeneity, the very short time to resuscitation, the slow induction phase of TTM, and the rapid rewarming period, may have influenced the main results, and still consider TTM at 33 °C as the best therapeutic option in cardiac arrest survivors. This position is supported by the publication of the recent HYPERION study, which showed a significant improvement in neurological outcome at 3 months for patients with OHCA or IHCA associated with a non-shockable initial rhythm who were treated with TTM at 33 °C, compared to a control group kept at 37 °C [11]. Although some of the criticisms of the “TTM trial” may have been reasonable, it was the largest study in this field and was conducted using sound up-to-date statistical methodology [8]. Moreover, looking at the early evidence supporting the use of TTM, one can argue that populations were highly selected and results were not generalizable to all cardiac arrest victims. These early studies also had many methodological biases (e.g., no power calculation, relatively small cohorts, early stopping because of lack of funding, no blinded assessors of primary outcome, no prognostication guidelines) and the control group experienced fever (i.e., temperature > 38 °C), which might have been responsible for detrimental effects, thus overestimating the beneficial effects of TTM [4, 5]. Importantly, the significant improvement in the clinical management of cardiac arrest patients (including early coronary angioplasty, standardized hemodynamic and ventilatory targets, avoidance of early withdrawal of life-sustaining therapies) between the early trials [4, 5] and the “TTM trial” [8] may in part explain the blunted effects of TTM at 33 °C in this study. Finally, the enthusiastic results supporting the effectiveness of TTM in experimental cardiac arrest [12] may not be directly translated in humans because the animals used in these models do not have comorbidities and/or underlying cardiac disease; resuscitation is standardized and cooling is immediate; the brain size is smaller; and some measured outcomes only included histological lesions and/or biomarkers of brain injury, which cannot reflect “cognitive function” as assessed in human studies.

Today, when treating patients resuscitated after cardiac arrest, the medical community is separated into TTM “believers” and “neutral,” with a significant impact on patient management and a trend towards a less accurate TTM prescription or, in the worst scenario, a “nihilistic” approach, with the total abandon of any temperature control in a number of centers.

When prescribing a drug, physicians consider its mechanism of action and the appropriate route (oral or intravenous), dose, and duration, according to specific information collected from clinical trials. As an example, to compare two anti-inflammatory agents for pain relief, the patients randomized into the two study arms will receive the regimens that would result in the most potent anti-inflammatory effects for both molecules. Unfortunately, this “most effective” protocol for TTM is undefined. We could have, for example, five OHCA survivors admitted to five different intensive care units (ICUs), who could receive different TTM protocols, as indicated in Fig. 1; despite the differences in treatment modalities and targets, they would all be included and considered in the “TTM group” of a pragmatic multicenter RCT, thus adding significant heterogeneity to the delivery of TTM and its effects on outcome.

Returning to our analogy with drugs, we need to define the optimal way of delivering TTM, specifying the characteristics that could provide the best neuroprotective effects after anoxic brain injury with minimal adverse effects. This approach is also similar to the concept of “high-quality cardiopulmonary resuscitation (CPR)” [13], which considers the correct rate and depth of compression, with minimal interruptions, to increase the probability of success. As such, “high-quality TTM” should be considered in clinical protocols when TTM is initiated.

In 2009, a consensus of five scientific societies introduced the concept of “targeted temperature management” to replace the previous term of “therapeutic hypothermia” [14], to underline the clinical relevance not only of the cooling or maintenance period, but also of the other phases of therapy, including induction, rewarming, and normothermia. However, we still lack good clinical data and knowledge about the optimal method, including when best to initiate TTM, the target temperature, the duration, and the rewarming rate. The “TTM study” from Nielsen et al. [8] only investigated the most effective target temperature but did not explore the other questions related to optimal TTM. We will summarize the current evidence for each aspect in the next sections. The discussion will not include the selection of the patients who would benefit the most from TTM, which is a relevant and unresolved issue, but beyond the scope of the viewpoint.

The definition and the methods to achieve high quality of TTM are intrinsically linked. To achieve precise and accurate TTM, two issues are crucial: the use of sedatives/analgesics and the choice of device.

TTM is a complex therapy and requires a “bundle” of interventions and clinical decisions that require standardization in order to maximize their neuroprotective effects (Fig. 2). As such, protocols should be used in clinical practice to improve the homogeneity of TTM prescription; these protocols could result in an improvement in patients’ outcome [40]. If body temperature and shivering are not adequately monitored, induction is delayed, body temperature remains variable using non-automated methods, and spontaneous and fast rewarming are allowed, the patient will be exposed to “low-quality TTM” and the likelihood of a beneficial effect will be compromised (Fig. 3a). By contrast, early initiation after the anoxic injury, continuous core temperature monitoring, the use of drugs to facilitate the decrease in temperature and prevent shivering, the selection of a specific target temperature during the cooling phase, a maintenance phase with well-regulated constant temperature, a prolonged rewarming phase, and the avoidance of fever after TTM are the main components of “high-quality TTM” (Fig. 3b). It is interesting to note that this “high-quality TTM” has been used in several experimental studies [15] and may explain why marked neuroprotective effects could be obtained with only a few included animals, whereas “low-quality TTM” has often been observed in large pragmatic RCTs [8].

Before clinical studies can finally validate the clinical significance of this concept on measured relevant outcomes, future studies should further explore TTM characteristics, such as target temperature or duration) while ongoing studies will provide relevant information in the next years; one ongoing RCT will evaluate whether TTM at 33 °C for 24 h is more effective than early treatment of fever (i.e., when body temperature exceeds 37.8 °C) after OHCA (NCT02908308) while a second RCT will compare two different rewarming strategies (0.25 °C/h vs. 0.50 °C/h) in OHCA patients treated with TTM at 33 °C for 24 h (NCT02555254). Using large databases, reports on the specified aspects comprising the quality of the delivered therapy will help to better understand how each of the components may influence patient trajectories and how the clinical practice of TTM can be improved.