Chapter 16. Therapeutic Hypothermia

• Induced hypothermia has been shown to reduce mortality when applied after resuscitation from cardiac arrest.
• While the ideal temperature for hypothermia is not known, most experts believe that cooling to 32 to 34°C is optimal.
• Induced hypothermia may have benefit for other disease processes such as myocardial infarction and stroke.
• The mechanisms by which hypothermia acts are multifaceted and a focus of much current investigation.

The notion of cooling patients for medical benefit is quite old. In 1814, Baron Larrey, a French surgeon in the service of Napoleon's army, reflected on soldiers who suffered major injuries on the frozen battlefields in Russia by commenting that “cold acts on the living parts … the parts may remain … in a state of asphyxia without losing their life.”1 A belated resurgence of interest in hypothermia has taken place in the past decade, expanding the possible medical indications for its use. Induced hypothermia, the intentional lowering of body temperature, has been explored in a number of acute critical care settings, including myocardial infarction, stroke, head trauma, and after cardiac arrest. While the optimal depth and timing of hypothermia are not yet established for these uses, most experts advocate a temperature goal of 32 to 34°C because it seems to provide significant benefit while avoiding most of the adverse effects associated with the intervention. Timing of hypothermia, with respect to both time of induction and duration of therapy, is even more uncertain, although general consensus holds that cooling should be initiated as soon as possible after the morbid event and should be maintained for at least 12 to 24 hours. Regarding specific uses, there is particularly good evidence that hypothermia is protective for the resuscitated cardiac arrest patient after return of spontaneous circulation.2,3 The use of hypothermia in other clinical scenarios remains promising but less clear at present.

This chapter addresses elements of the history of hypothermia, the laboratory and clinical data that have developed our understanding of its use, some of the various techniques used to cool patients, and the clinical syndromes for which hypothermia appears to offer the greatest advantage.

The protective effects of hypothermia induction have been suggested since the time of Hippocrates, who advocated packing bleeding patients in snow.4 Hypothermic protection also was noted by Napoleon's battlefield surgeon, Baron Larrey, during the French invasion of Russia. He observed improved survival of injured soldiers left in the snow compared with those treated with warm blankets and heated drinks.1 Induced hypothermia has been studied in a wide variety of illnesses, both ischemic and nonischemic in nature (reviewed in refs. 5 through 7). These include traumatic brain injury,8–10 status epilepticus,11 arrhythmia, sepsis, and the ischemic illnesses of myocardial infarction, stroke, and cardiac arrest.7,12 Interestingly, the first reported use of induced hypothermia was in the setting of malignancy. In 1939, Fay and colleagues treated patients with metastatic carcinoma, with the goal of both pain reduction and retardation of tumor growth.13 While hypothermia to 32°C for 24 hours did not prove effective for the stated goals, it was considered well tolerated.13

A decade later, Bigelow studied the induction of hypothermia in the setting of cardiac surgery, with the goal of cerebral protection.14 Two other studies using hypothermia as therapy for cardiac arrest also were published. Both these early cardiac arrest studies used moderate hypothermia of 30 to 34°C in patients after resuscitation from cardiac arrest. One of these pioneering papers presented a series of four patients, all of whom were cooled and survived arrest.15 In the other study, 12 patients were cooled with a survival rate of 50% compared with 14% survival in 7 normothermic control patients.16

During the 1960s and 1970s, the field of induced hypothermia lay relatively dormant for reasons that remain unclear. Some have suggested that more dramatic therapies were developed that overshadowed cooling as a possible therapy, such as controlled ventilation, monitored ICU management, and cardiopulmonary resuscitation (CPR).5 Additionally, several adverse effects of hypothermia were described, which may have dampened enthusiasm.17,18

Interest in “resuscitative hypothermia” was rekindled by Safar and others at the University of Pittsburgh, who demonstrated in a ventricular fibrillation dog model of cardiac arrest that mild to moderate hypothermia could be induced to improve outcomes.19,20 Trauma research also provided a motivation for the development of induced hypothermia. It was understood from military combat experience that definitive therapy for penetrating trauma was often delayed for practical reasons (e.g., transportation and access to surgical care) and that measures were needed to preserve exsanguinating soldiers until appropriate care could be delivered.21 Given the animal data on exsanguination and cooling, it appeared that hypothermia might be a suitable approach.22

Safar went on further to describe “suspended animation,” a process that allows “rapid preservation of viability of the organisms in temporarily unresuscitable cardiac arrest, which allows time for transport and repair during clinical death and is followed by delayed resuscitation, hopefully to survival without brain damage.”12 Hypothermia has been a primary component of this concept of stasis. In this paradigm, victims of cardiac arrest may be cooled to some target temperature and maintained at that temperature for a specific period of time. With advanced medical interventions, which may include cardiopulmonary bypass, metabolic correction, and controlled reperfusion, the patient is stabilized and rewarmed, and “reanimation” is initiated. While many methodologies have been studied under the rubric of suspended animation, including cardiopulmonary bypass and pharmacologic interventions,23 these are used most often as adjuncts to the use of hypothermia.

Since these initial observations in the 1980s and early 1990s, much of the work pertaining to hypothermia and ischemic disease has focused on focal ischemia and reperfusion, e.g., animal stroke and myocardial infarction models. A number of ischemia-reperfusion (IR) model systems have been developed over the last two decades, including cellular,24 isolated organ,25 and whole-animal models in which arterial supply to the organ under study is temporarily occluded.26,27 In this latter category are included experiences with human IR, for example, during coronary vascular procedures.28 Recently, two seminal papers were published describing the use of hypothermia to successfully treat resuscitated cardiac arrest patients.2,3 With these studies, hypothermia has moved from the laboratory to active clinical use.

The mechanisms by which induced hypothermia protects against cellular and tissue injury are poorly understood. Given the importance of temperature in a wide range of physiologic processes, it is reasonable to conclude that multiple mechanisms may be involved in any given tissue (reviewed in refs. 5 and 6). Some mechanisms implicated in hypothermic protection include modulation of transcription and/or translation, suppression of reactive oxygen species (ROS) production, and inhibition of programmed cell death, or apoptosis. The mechanisms by which cooling protects tissues may overlap as well. For example, the effect of hypothermia on cellular metabolism may lead indirectly to modulation of programmed cell death, and cooling may have a direct effect on cell death machinery itself. Most of the data pertaining to mechanism of hypothermic protection comes from studies of IR injury in models of stroke and myocardial infarction.

A number of gross physiologic changes have been observed in the setting of hypothermia that may contribute to decreased injury. As early as 1954 it was noted that hypothermia induced by ice water immersion could lower cerebral oxygen consumption in dogs by approximately 7% per 1°C drop in temperature.29 Other studies have demonstrated that mild hypothermia in rats improves postischemic cerebral blood flow disturbances.30 Another marker of general physiologic injury after reperfusion, brain edema, also was found to be reduced by hypothermia in a rat model of global ischemia.30,31 Finally, hypothermia has been shown to minimize damage to the blood-brain barrier, which in turn may protect against blood-borne toxic metabolites reaching brain tissues through the compromised barrier.30,32

Intracellular signaling can alter the array of gene transcription activity of a cell quickly and dramatically, and this, in turn, can trigger a variety of injury processes. In a cardiac arrest mouse model, a group of signaling pathway genes known as the immediate early genes was activated after resuscitation.33 A study of liver IR demonstrated a drop in c-jun terminal kinase activity at 25°C when compared with normothermic controls.34 An extracellular signaling molecule thought to protect against injury, BDNF, was increased in a rat model of cardiac arrest when animals were cooled to 33°C.35

A number of biochemical changes during IR can be modified by the induction of hypothermia. In a gerbil stroke model, animals subjected to mild hypothermia were found to have decreased arachidonic acid metabolism compared with normothermic controls.36 In a rat brain ischemia model, hypothermia to 32°C reduced nitric oxide production, as measured in jugular blood.37 Whether these attenuations are simply markers of hypothermic effects or actually relevant factors in reperfusion injury remains to be clearly established. Other biochemical phenomena seem more likely to be linked directly to damage processes, such as the observation that hypothermia slows ATP depletion during IR..38 ROS production also appears to be attenuated by hypothermic conditions in a rat cerebral ischemia model.39

Programmed cell death is a complex yet ubiquitous process by which cells actively chose or are chosen to die. This cellular program can be activated as part of normal physiology, such as during embryonic development, or as an abnormal response in a wide variety of disease states.40,41 Much evidence implicates the induction of apoptosis as a component of reperfusion injury.42,43 A recent report showed that the apoptotic pathway enzyme caspase 3 was upregulated in brain tissue after resuscitation from cardiac arrest, as measured at autopsy in patients who died within days of undergoing resuscitative measures.44 Widespread evidence from animals also supports the notion that apoptosis is activated after reperfusion.25,45 Hypothermia may inhibit this process. Proteolysis of the cytoskeletal protein fodrin, a characteristic step in the apoptotic pathway, is inhibited by hypothermia to 32°C in a rat brain IR model.46 The process of apoptosis is an active one, requiring protein synthesis and enzymatic activity, both of which may be inhibited by lower temperatures. While some data suggest that the degree of apoptosis can be reduced by hypothermia, the topic certainly deserves more investigation in animal models.

Cardiac arrest is a highly mortal condition that leads to at least 300,000 deaths each year in the United States alone.47 Survival from cardiac arrest remains dismal some 50 years after introduction of chest compressions and electrical defibrillation, with only 1% to 11% of patients surviving until hospital discharge after out-of-hospital cardiac arrest.48–50 While initial survival from in-hospital cardiac arrest ranges from 25% to over 50%, subsequent survival until hospital discharge is much lower, from 5% to 22%, suggesting that a high mortality rate is seen shortly after initial return of spontaneous circulation. Some of this mortality after return of normal circulation may be due to events related to reperfusion injury. For further discussion of cardiac arrest and resuscitation therapies, see Chap. 15.

Cardiac arrest remains a major medical challenge despite research efforts over the past few decades. There is little time after arrest to defibrillate the heart and thereby stop ongoing ischemic injury to key organs such as the heart and brain, and few therapies are proven to be useful during the postresuscitation phase of cardiac arrest—when up to 90% of patients go on to die despite successful defibrillation. New approaches are desperately needed to improve cardiac arrest survival, and induced hypothermia may be one of the most promising new approaches.51

Hypothermia may be helpful during three periods of time in a cardiac arrest (Fig. 16-1): (1) prearrest, (2) intraarrest, and (3) postarrest. Prearrest cooling can only be used practically as a preoperative intervention when the heart is stopped in a controlled fashion during cardiac surgery. Postarrest cooling to 32 to 34°C was found to be protective in recent human cardiac arrest trials, induced by applying cooling blankets or ice packing in the subset of cardiac arrest patients having return of spontaneous circulation (ROSC) who remained unresponsive.2,3 The goal of this level of hypothermia was to prevent neurologic injury and decrease mortality while maintaining a temperature warm enough to prevent the other adverse effects of more profound cooling (e.g., cardiac arrhythmia, coagulopathy, and infection—see discussion of adverse effects later in this chapter). This cooling was protective despite taking 4 to 8 hours to reach target temperature after ROSC (Fig. 16-2). Intraarrest cooling, when cooling is induced after failed initial CPR, has the potential to induce a protective state long enough for more definitive circulation (e.g., cardiac bypass) to be established and life restored. Little is known about the optimal depth or clinical potential of such hypothermia owing in large part to the technical difficulties involved in inducing hypothermia during the low-flow states of sudden cardiac arrest.

Given the current data on hypothermia in cardiac arrest, guidelines for resuscitation likely will include hypothermia as a recommended intervention in the not too distant future. Recent large clinical trials support the use of hypothermia to 32 to 34°C in the successfully resuscitated patient with postarrest global neurologic dysfunction.

Cooling in these patients should be achieved as rapidly as possible using cooling blankets or ice. When ice is used, cloth or other material should be placed between the skin and the ice to avoid frostbite. Temperature should be measured via bladder probe, pulmonary artery catheter, or tympanic probe if invasive temperature monitoring is not available. Temperature should be taken every 15 to 30 minutes during the cooling protocol and until a stable cooled temperature state is achieved. Avoidance of temperatures below 32°C is important, and further temperature drops need to be addressed.

Cooling almost always will induce shivering and could result in discomfort. In the published studies of hypothermia after cardiac arrest, patients uniformly received neuromuscular blockade and sedation, and it is likely most patients will require such pharmacologic treatment. The disadvantages of this management strategy is that neurologic examination is not interpretable, and paralytic agents carry the risk of long-term neuromyopathy.

Rewarming often requires active intervention, which can include warmed intravenous fluids and warming blankets.

Although induced hypothermia has not been formally included in Advanced Cardiac Life Support (ACLS) guidelines at the time of this writing, it is recommended that ICU physicians and staff develop protocols for cooling after cardiac arrest because this recommendation likely will be forthcoming soon. Development of cooling protocols will require some time because a number of hospital-specific technical issues (e.g., how to cool, who will do the cooling, how to monitor temperature) will need to be established.

Induced hypothermia to 32 to 34°C has been shown in a variety of model systems to limit the size of a myocardial infarction in ischemia without reperfusion and to limit myocardial injury when reperfusion is established. Given that current clinical practice is directed toward early reperfusion of myocardial tissue via thrombolytic therapy or percutaneous coronary intervention (PCI), it is the IR model that has received the most current attention. A recent trial in pigs used an endovascular cooling catheter to induce hypothermia during ischemia and after reperfusion of 60-minute coronary occlusions. Hypothermia to 34°C had a substantial protective effect, with smaller infarct size, better cardiac output, and improved microvascular flow.52

Endovascular cooling also has been studied to induce hypothermia during PCI in humans. In a pilot study, patients were randomized to cooled (34 to 35°C) or normothermic PCI. Cooling was maintained for 3 hours after intervention. No patients suffered adverse effects from cooling, and there was a trend toward smaller infarct size in the cooled patients.53 Larger studies will be required to both confirm this finding and test the induction of hypothermia in the setting of thrombolytic therapy as well. Given that no major randomized studies have shown benefit from induced hypothermia in myocardial infarction, there is currently no clinical recommendation for its use. However, it is likely that hypothermia will become a standard technique to reduce infarct damage in the next decade.

The rationale for the consideration of induced hypothermia in acute ischemic stroke is similar to that for cardiac arrest and myocardial infarction. Stroke is a difficult clinical problem with severe outcomes; not only is stroke the second most common cause of death in the world, but it is also the leading cause of disability in patients over 65 years of age.54 Moreover, few improvements in acute stroke care have improved the general outcome of the disease.55,56 Treatment with thrombolytic agents such as tissue plasminogen activator, while a powerful option when a patient presents within 3 hours of ischemic stroke onset, carries significant risks and has only demonstrated improvement in a subset of stroke patients.57,58 In part, this is so because irreversible damage to brain tissue may have occurred already before restoration of blood flow. However, reperfusion of ischemic neuronal tissues also may produce additional damage via free radical production and induction of programmed cell death and the other so-called IR injury mechanisms described earlier59 (reviewed in ref. 42).

The use of induced hypothermia in the setting of stroke remains experimental, although promising data exist to support its use (reviewed in refs. 60 and 61). In a variety of model systems, hypothermia reduces injury, as measured by biochemical and histologic markers. Cooling also appears to extend the time in which brain tissue can suffer ischemia without irreversible damage.62,63 In several animal models, cooling to 32 to 34°C for greater than 24 hours provided significant benefit, whereas shorter periods of hypothermia provided more inconsistent results.64,65 Several pilot studies have evaluated the feasibility and safety of cooling after ischemic stroke in humans. In one study, cooling to 33 to 34°C appeared to reduce intracranial pressure in patients with severe middle cerebral artery strokes.66 In a trial of induced hypothermia after thrombolysis, cooling stroke patients to 32°C for at least 12 hours demonstrated a nonsignificant trend toward improved functional outcomes compared with normothermic controls.56 These pilot studies were limited by small sample size. Larger randomized trials of induced hypothermia in stroke are currently underway in Europe.

While the induction of hypothermia after ischemic stroke requires additional refinement before becoming the established standard of care, aggressive prevention of hyperthermia is clearly indicated. A number of epidemiologic and clinical studies have shown that even modest elevations in temperature increase stroke severity and patient mortality.67,68 Antipyretic medications in combination with either forced-air cooling or cooling blankets are sufficient to maintain core temperatures below 37°C. A clinical investigation using such modest measures demonstrated significant mortality risk reduction.69

The induction of hypothermia has been used experimentally in a number of other clinical scenarios from surgical to critical care applications. One such indication is traumatic brain injury, in which an extensive body of laboratory work suggests that hypothermia may improve outcomes for severe trauma.70 In one clinical study of 87 patients, hypothermia to 33 to 35°C for at least 72 hours provided a significant functional benefit at 1 year after injury.71 However, a larger randomized trial of 392 patients failed to find a significant benefit at 6 months after cooling to 33°C for 48 hours.10 While this has been widely interpreted as a “negative study,” the authors have pointed out that the likely reason for finding no benefit was that owing to technical challenges (including taking 4 hours to get informed consent), the average time to cooling was nearly 8 hours after injury. There are no animal studies that suggest any improvement in neurologic results at this late time—the longest delay in cooling that shows any positive effect is about 1 hour following injury. Another clue to these conflicting results was provided by a clinical trial in which hypothermia was found to have no benefit in patients who did not exhibit elevated intracranial pressure.72 Thus hypothermia may be useful in only a subset of brain injury patients. Taken together with studies demonstrating worrisome electrolyte abnormalities and pneumonia in cooled head injury patients,73,74 no definitive recommendations exist currently for treatment of traumatic brain injury with induced hypothermia, although it is likely that future work will establish a role for cooling in this disease entity.

Induced hypothermia also has been studied in the setting of acute liver failure, such as from acetaminophen toxicity.75 As many as 50% of patients with acute liver failure succumb from complications of increased intracranial pressure,76 and induced hypothermia to 32 to 33°C has been shown to reduce intracranial pressure as much as 20 to 30 mm Hg in patients awaiting orthotopic liver transplantation.77 Cooling also lowers arterial ammonia concentration and ammonia delivery to the brain, which may be one of the mechanisms by which hypothermia acts, because ammonia delivery to the brain contributes to increased brain edema and intracranial pressure.78 It remains an open question whether hypothermia could be used to treat the hepatic encephalopathy of cirrhosis, also felt to be mediated by ammonia or at least compounds for which ammonia acts as a surrogate marker.

Neonatologists have studied induced hypothermia in the setting of hypoxic-ischemic encephalopathy after perinatal asphyxia, an important cause of acute neurologic injury at the time of birth.79 In a piglet model of brain ischemia, early cooling to 33 to 35°C improved neurologic outcomes by both functional and histologic criteria, whereas a delay in cooling by even 30 minutes negated this benefit.80,81 Initial clinical trials have demonstrated the feasibility of selective head cooling in infants, with minimal differences in complications between cooled patients and normothermic controls.82 Larger randomized clinical trials are in progress at the time of this writing.

A number of surgical applications of induced hypothermia have been evaluated, including aortic arch repair83 and cardiac bypass surgery.84 In these settings, hypothermia induction and rewarming can be controlled precisely and used for presumed important portions of the procedure in question. Unlike the clinical scenarios described earlier, surgical hypothermia does not attempt to treat a pathophysiologic problem but rather to prevent one from occurring. On a cautionary note, however, an extensive meta-analysis showed only a modest trend toward morbidity benefit in the setting of cardiac bypass surgery, with a trend toward decreased stroke rate offset by increased perioperative complications and myocardial injury.84

Hypothermia is not without risk of adverse effects. As the body cools, a series of physiologic changes take place, including shivering, alterations in the clotting cascade, immunologic suppression, and cardiac membrane changes that increase the risk of arrhythmia (see Chap. 110). These effects are poorly characterized and depend on the extent and duration of cooling. Therefore, the optimal therapeutic window of hypothermia needs to be established.

Shivering, a common occurrence during the induction of hypothermia, can serve to counteract the mechanisms thought to be beneficial from the cooling itself, raising metabolic rates and myocardial oxygen demands. Pharmacologic options to control shivering include opioid and neuroleptic medications. Simply covering the arms and legs during central cooling has demonstrated some reduction in shivering as well, supporting the notion that an increase in peripheral to central temperature gradient may be responsible for triggering a shivering response.85 A recent study demonstrated that meperidine and buspirone acted synergistically to reduce shivering,86 and this technique has been used successfully with endovascular cooling in myocardial infarction.53

Hypothermia induces a poorly characterized diuresis in which patients may develop significant hypokalemia, hypomagnesemia, and/or hypophosphatemia.74 Careful attention to electrolytes during the period of cooling and rewarming is required, and aggressive supplementation should be given when indicated.

A study of stroke patients demonstrated a risk of pneumonia with hypothermia, consistent with the cardiac arrest study results.66 It is unclear whether the immune system is globally suppressed by hypothermia or whether some pulmonary-specific mechanism is the culprit for this increased pneumonia risk. Induced hypothermia does not seem to lead to a significantly increased risk of bacteremia, urinary tract infection, or other infectious complications. Whether patients undergoing induced hypothermia should be treated prophylactically with antibiotics remains to be determined.

While cardiac arrythmias are a theoretical risk of induced hypothermia, human studies in which target temperatures were above 32°C did not exhibit significant arrhythmia. It is possible that deeper cooling may provoke such electrical dysfunction. Therefore, no additional clinical precautions need be taken when hypothermia to 32 to 34°C is performed.

There are a number of unresolved questions regarding the use of induced hypothermia. First, the optimal degree of cooling has yet to be established. As described earlier, hypothermia can have adverse effects such as shivering, reduced immune system function, and other possible complications.87–89 Therefore, the therapeutic window of hypothermia must be defined to provide maximum benefit. Most clinical research has concentrated on the use of mild (34 to 36°C) or moderate (30 to 34°C) hypothermia. Evaluation of deep hypothermia has been confined largely to specific applications such as brain cooling during surgery and reduced cerebral perfusion.90 Some animal investigations have compared different depths of cooling directly.91 It is possible that different disease states may require different depths of cooling to provide optimal protection.

The timing of hypothermia is another crucial question. There are three general time periods in which hypothermia may be considered, although only two are realistic for clinicians. Preinsult hypothermia, or cooling some time before the onset of an ischemic process, is only possible in the setting of surgical intervention, in which ischemia is iatrogenic and controlled.92,93 Induced hypothermia initiated immediately during an ischemic process, such as during stroke or myocardial infarction but before reperfusion has been initiated, is feasible but poorly studied in the clinical setting. A more dramatic example of this, the induction of hypothermia during CPR, is theoretically attractive and a promising area of current research but has not been attempted clinically. Postreperfusion hypothermia, or cooling after definitive treatment such as thrombolytic therapy or resuscitation, has been the most commonly attempted timing strategy and is certainly the most clinically convenient. Recent large-scale clinical trials have demonstrated significant benefit from induced hypothermia after resuscitation from cardiac arrest.2,3 These studies raise a secondary timing question related to postreperfusion hypothermia: Is the delay in hypothermia initiation clinically important? Cooling to target core body temperatures may take several hours in clinical settings.94,95 A variety of laboratory data suggests that earlier induction of hypothermia confers a greater benefit.27,96

The duration of the hypothermic state required for protection also remains to be clearly established. Clinical studies to date have all maintained mild hypothermic conditions for at least 12 hours.2,3,87 However, animal studies have shown benefit from hypothermia lasting only 2 to 6 hours.26,27 Certainly, a shorter duration of cooling would have a number of clinical advantages, including minimizing the logistical difficulties of maintaining a constant cooled state and also minimizing the potential adverse effects from hypothermia. This question has not been addressed directly in clinical studies at this time.

Finally, a number of other aspects of induced hypothermia remain to be examined. How quickly should patients be rewarmed at the end of a period of hypothermia? Would certain drugs serve as useful adjuncts to hypothermia, either to provide additional benefit or to protect against certain adverse effects of a cooled state? A number of pharmacologic strategies have been employed to protect against IR injury, including barbiturates, benzodiazepines, and gas anesthetic agents. These and other issues surrounding induced hypothermia await randomized clinical trials.

Before induced hypothermia becomes a standard therapy for such critical illnesses as myocardial infarction, stroke, and cardiac arrest, many of the questions just posed will require resolution. It is also important to note that the optimal combination of these factors may be different for each disease state, and therefore, such principles may not be generalized easily. It is possible, for example, that adverse effects such as electrolyte abnormalities may limit the depth of cooling in the more critically ill state of resuscitated cardiac arrest compared with myocardial infarction or that cardiac arrest patients require a slower rewarming period.

One such parameter, the time between injury and the initiation of cooling, may be the most consistent between different disease states. That is, the sooner a patient is cooled after a given insult, the more likely it is to prevent tissue injury from that disease process. This has been shown in a large variety of IR models at the cellular, organ, and whole-animal levels.20,52 Therefore, techniques to allow for early and rapid cooling will need to be developed. Current external cooling methods such as ice packing and cooling blankets take several hours to lower core body temperatures adequately in humans. Endovascular cooling catheters appear promising, especially for localized cooling in stroke or myocardial infarction. Additional techniques involving multiphase coolants are currently under development. One goal for induced hypothermia research in cardiac arrest would be to develop a cooling method that could be initiated in the prehospital setting by paramedics.

Hypothermia induction in cardiac arrest fits squarely within the metabolic phase of cardiac arrest, as part of the three-phase time-sensitive model of CPR (see further discussion of the three-phase model in Chap. 15).97 With the failure of defibrillation and medications to restore viable hemodynamics, patients could be cooled rapidly to protect organs from ongoing ischemia and as a bridge to additional therapies such as establishment of cardiopulmonary bypass with controlled reperfusion.98–103 Under such intensive control, further attention could be given to reversal of metabolic injury caused by reactive oxygen species, mediators of apoptosis, compromise of cell membrane integrity, and so forth. The limits for reversal of ischemic injury when treated under such a “suspended animation” paradigm remain to be determined.