Intra-arrest hypothermia during cardiac arrest: a systematic review

Use of mild therapeutic hypothermia, or "targeted temperature management" as recently suggested [1], has been recommended in cardiac arrest (CA) patients since the publication of two randomized clinical trials in 2002, the results of which demonstrated a significant improvement in neurologically intact survival for comatose CA patients presenting with ventricular fibrillation (VF) or ventricular tachycardia (VT) [2, 3]. Current guidelines suggest that mild therapeutic hypothermia should also be considered in patients presenting with other rhythms although this has been less well studied [4].

Although therapeutic hypothermia has been widely implemented [5], its benefits are still questioned and several issues remain unanswered, including the optimal time to initiate cooling. Animal data have indicated that early cooling after return of spontaneous circulation (ROSC) produces better brain function and neurological recovery than does normothermia, whereas delaying therapeutic hypothermia significantly limited these beneficial effects [6, 7]. There are also experimental data suggesting that hypothermia initiated during cardiopulmonary resuscitation (CPR), that is, intra-arrest, is superior to cooling initiated after ROSC, both in terms of increased rates of successful CPR and improved survival [8‐11]. Experimental investigations have also shown that intra-arrest therapeutic hypothermia (IATH) increases the success rate of defibrillation attempts in VF [12] and has beneficial effects on heart function, including improved left ventricular function and reduced myocardial infarct size [13]. Clinical investigations have shown that pre-hospital induction of therapeutic hypothermia is feasible [14], without major adverse events even when used intra-arrest [15], and may provide some additional benefits over delayed in-hospital cooling [16].

The objective of this systematic review of the literature was, therefore, to evaluate whether IATH affects survival and neurological and cardiac function in experimental and human CA.

This systematic review revealed that: a) in experimental models of CA, IATH improves survival and neurological outcome when compared to normothermia and/or PATH; b) IATH improved ROSC rates and was associated with improved cardiac function, including better left ventricular function and reduced myocardial infarct size, when compared to normothermia; c) data on the efficacy of IATH in humans remain limited; d) IATH is feasible and is associated with few adverse events.

Therapeutic hypothermia has been shown to provide protective effects on brain and heart cells through different pathways. Hypothermic mechanisms providing myocardial protection have not been completely elucidated but include improved energy production during ischemia [38], inhibition of apoptosis [27], increased calcium sensitivity of myocytes [39], regulation of mitochondrial oxidative phosphorylation [38], attenuation of reactive oxygen species generation after ischemia-reperfusion [40], and preserved myocardial vascular autoregulation [41]. All these mechanisms translate into increased myocardial contractility [39, 42], which should potentially be enhanced by IATH. Animal studies demonstrated a significant improvement in contractility when IATH was used in comparison with normothermia and PATH [13, 27, 30, 31] and IATH also had the potential to reduce the infarction size after coronary vessel occlusion and CA [13]. Nevertheless, conflicting results have been reported between IATH and normothermia and/or PATH in terms of ROSC rate, duration of CPR, CoPP, total dose of epinephrine administered during CPR and defibrillation success rate [6, 11, 12, 21, 31]. Variations in the experimental models could explain, at least in part, these differences. A very short ischemic period is unlikely to produce sufficient cardiovascular alteration to make it possible to demonstrate any superiority of IATH over other therapies. In one study, time from CPR to ROSC was as short as two to three minutes [8] and in another, all the animals achieved ROSC within one minute after CPR was initiated [23]. The complexity and severity of the animal model, the different durations of VF, and the methods used to induce IATH (for example, systemic vs. selective brain) may also confound comparison of these studies. A further concern is related to the limited numbers of experimental animals used, as sometimes this restricted the power of statistical analyses [35]. Unfortunately, human studies have reported only data on the ROSC rate; in the RCT, TNEC resulted in similar ROSC rates to those seen in the control group [16], whereas in a retrospective study, IATH using cold fluids was associated with a significantly increased rate of ROSC compared to historical controls [18]. Further studies are needed to investigate myocardial function during IATH after human CA.

Therapeutic hypothermia may also preserve cerebral function after CA through inhibition of release of neurotransmitters, such as glutamate and dopamine, involved in brain damage [43]; preservation of the blood brain barrier [44]; protection of cerebral energy stores [45] and microcirculation [46]; and decrease in intracranial pressure and increase in CBF [47]. Importantly, cerebral reperfusion occurring after ROSC can also trigger the production of free radicals and other neurotoxic mediators, which may enhance anoxic damage to the brain [48]. Thus, the beneficial effects of therapeutic hypothermia after CA should be even more pronounced when cooling is initiated during the no-flow state and/or CPR before reperfusion. The positive effects of IATH on neurological outcome after experimental CA have been well-demonstrated. IATH affects brain perfusion and metabolism, as demonstrated by increased supra- and infra-tentorial CBF [23, 33, 34] and decreased cerebral oxygen uptake [25, 33, 34], thus preventing anaerobic metabolism and excito-toxicity [25]. All the experimental studies showed better neurological function in animals treated with IATH compared to normothermia or PATH, as well as a reduction in the number of injured neurons on histological brain examination. In the human RCT, IATH increased the intact neurological outcome rate, particularly in patients with a short no-flow time [16].

The combination of these beneficial effects of IATH on cardiac and cerebral functions should be expected to improve overall survival. However, IATH reduced mortality rates in only 7 of the 16 animal studies in which it was compared to normothermia [8, 27‐32] and 2 of the 6 studies in which it was compared to PATH [8, 13]. Nevertheless, some of the negative studies had methodological limitations reducing the possibility of demonstrating any beneficial effects of IATH over standard treatment. In two of the studies, the observation period after CA was less than three hours [10, 22], so that no effects on long-term outcome could be reported. In two studies, all animals survived after arrest because of a short no-flow and CPR period [23, 24], this situation being far different from the clinical scenario of CA in humans. Two studies showed an improvement, although not statistically significant, in survival rate when IATH was used [26, 35] and the authors acknowledged a limited sample size as the main reason for their findings.

Human data on the effects of IATH on mortality after CA are also scarce. External cranial cooling did not improve survival rates in OHCA compared to a control group without ice [19]; however, the reduction in tympanic temperature, a surrogate of brain temperature, was similar between groups. In a large retrospective database, intra-arrest cold intravenous fluids were not associated with increased overall survival; importantly, only 13% of the whole cohort received in-hospital therapeutic hypothermia. In the RCT using intra-arrest cooling [16], OHCA patients admitted to the hospital and receiving IATH had improved overall survival when compared to the control group although the difference was not statistically significant; these effects appeared to be even more beneficial in the sub-groups of patients with short no-flow time and VF.

If IATH is to be applied, it remains unclear which technique/device provides optimal cooling and protection. Speed of achieving target temperature in experimental studies was dependent on animal size and the duration of exposure to hypothermia during CA. These factors may limit the reproducibility of these data in the human setting. Moreover, although the neuroprotective effect of IATH may be related to how quickly the brain is cooled, systemic cooling is important for myocardial protection. Also, IATH using cold fluids resulted in a lower CoPP and decreased LVEF after CA [13]; the need for epinephrine and duration of CPR were also significantly longer than control animals, suggesting that post-ROSC myocardial dysfunction could be exacerbated by cold fluid loading. Cooling with an intravascular catheter, PFC-TLV or TNEC has been shown to improve ROSC rate, survival and intact neurological outcome when compared to cold fluid infusion [10, 13, 37]. Hence, although intravenous cold fluids have been shown to be effective and safe when applied after ROSC [2, 14], the safety of large amounts of fluid during CPR may be questioned.

Other potential methods for IATH induction also have limitations when applied to the human setting. Internal cooling with catheters is invasive and not applicable in the field. Intrapulmonary PFC with TLV would not be easy to apply and may potentially exacerbate lung injury because of increased intrapulmonary pressure during ventilation [10, 49]. Trans-nasal cooling devices spray a PFC fluid through a nasal catheter system into the nasal cavity and induce brain hypothermia via the cooled blood and by direct convection; body cooling occurs later. Further studies are needed to investigate the beneficial effects of this technique in the human setting as well as the occurrence of adverse events.

The use of hypothermia has been associated with improved outcomes for survivors from CA; however, it has been suggested that the timing of the induced cooling may influence its beneficial effects. Experimental studies have shown that IATH can protect the heart against the ischemic processes occurring after CA and reduce the neuronal injury secondary to global ischemia. These beneficial effects seem to be significant when IATH is compared not only to normothermia, but also to PATH. Nevertheless, not all of these studies have reported similar conclusions, probably because of the different experimental conditions that have been used. Also, the evidence that IATH is superior to PATH in animal studies is more limited than the evidence comparing IATH to normothermia. Human data on IATH remain limited; however, several large cohort studies have suggested some beneficial effects of IATH on ROSC rates and neurological outcomes, especially if initiated within a short no-flow time. Selective brain cooling may have potential advantages in protecting the brain before reperfusion and has shown promising results in experimental and clinical studies; however, this technique may limit potentially beneficial effects of hypothermia on cardiac function and needs to be further evaluated in the human setting.

FST received honoraria for lectures from Benechill, Inc. FST, MC, JLV and PN have conducted a study supported by Benechill, Inc. SC and KD have no conflicts of interest to declare.