**4. Clinical studies**

Even though the idea of neuroprotection by decreasing body core temperature is not new, its implementation became sustainable in 2002, when the results of two crucial randomized clinical trials were published. Both were of similar design and reached analogical results, strongly supporting post-cardiac arrest cooling (table 2).

The European Hypothermia After Cardiac Arrest (HACA) multicenter trial has been the largest one (Hypothermia after Cardiac Arrest Study Group, 2002). A total of 275 patients who had been resuscitated from cardiac arrest due to ventricular fibrillation or pulseless ventricular tachycardia was randomised to conventional therapy or to induced mild hypothermia of 32 – 34 °C for 24 hours. In addition to a significant neuroprotective clinical effect, mortality at 6 months was 41 % in the hypothermia group, as compared with 55 % in the normothermia group (RR 0.74; CI 95% 0.58-0.95, p<0.05).

In the Australian study by Bernard et al, 77 patients succesfully resuscitated from cardiac arrest of presumed cardiac arrest origin, having ventricular fibrillation or pulseless ventricular tachycardia as an initial rhythm,were assigned to mild hypothermia of 32 – 34 °C for 12 hours or to standard treatment (Bernard et al., 2002). In the hypothermia group, more patients reached the favourable neurological outcome (cerebral performance category score 1 or 2) at hospital discharge than in the control group. In both studies, there was no significant diference in the frequency of adverse events.


Table 2. Outcomes of therapeutic hypothermia in randomized clinical studies.

Following this, a series of clinical studies have confirmed a positive role of TH in postcardiac arrest care. It was also shown that a close relation of TH with other interventions such as control of blood glucose, haemodynamics, ventilation and handling of seizures can improve long term survival of patients (Sunde et al., 2007; Knafelj et al., 2007; Nolan et al, 2009).

Hypothermia reduces cerebral metabolism by 7 % for each degree Celsius reduction in Body Temperature (BT) and reduces whole-body energy demand (Erecinska et al., 2003). Other mechanisms which play a role in hypothermia-induced neuroprotection are listed in table 1 (Ostadal, 2009; Polderman, 2009; Liu & Yenari, 2007; Busto et al., 2007; Kataoka & Yanase, 1998; Globus et al., 1995; Lei et al., 1994; Xu et al., 2002; Huang et al., 1999; Fischer et

Even though the idea of neuroprotection by decreasing body core temperature is not new, its implementation became sustainable in 2002, when the results of two crucial randomized clinical trials were published. Both were of similar design and reached analogical results,

The European Hypothermia After Cardiac Arrest (HACA) multicenter trial has been the largest one (Hypothermia after Cardiac Arrest Study Group, 2002). A total of 275 patients who had been resuscitated from cardiac arrest due to ventricular fibrillation or pulseless ventricular tachycardia was randomised to conventional therapy or to induced mild hypothermia of 32 – 34 °C for 24 hours. In addition to a significant neuroprotective clinical effect, mortality at 6 months was 41 % in the hypothermia group, as compared with 55 % in

In the Australian study by Bernard et al, 77 patients succesfully resuscitated from cardiac arrest of presumed cardiac arrest origin, having ventricular fibrillation or pulseless ventricular tachycardia as an initial rhythm,were assigned to mild hypothermia of 32 – 34 °C for 12 hours or to standard treatment (Bernard et al., 2002). In the hypothermia group, more patients reached the favourable neurological outcome (cerebral performance category score 1 or 2) at hospital discharge than in the control group. In both studies, there was no

 **favourable neurological outcome at hospital discharge** 

**Trial** 53 36 1.50 0.006 **Bernard et al.** 49 26 1.75 0.052

**Trial** 52 36 1.44 0.009

Following this, a series of clinical studies have confirmed a positive role of TH in postcardiac arrest care. It was also shown that a close relation of TH with other interventions such as control of blood glucose, haemodynamics, ventilation and handling of seizures can improve long term survival of patients (Sunde et al., 2007; Knafelj et al., 2007; Nolan et al,

 **favourable neurological outcome at 6 months** 

Table 2. Outcomes of therapeutic hypothermia in randomized clinical studies.

**STANDARD TREATMENT GROUP (%)** 

**RELATIVE RISK <sup>P</sup>**

strongly supporting post-cardiac arrest cooling (table 2).

the normothermia group (RR 0.74; CI 95% 0.58-0.95, p<0.05).

significant diference in the frequency of adverse events.

**HYPOTHERMIA GROUP (%)** 

al., 1999).

**4. Clinical studies** 

**HACA Study Group** 

**HACA Study Group** 

2009).

Analyses of the registries confirmed the results of clinical studies (Oksanen et al., 2007; Arrich, 2007). Several meta-analyses have however produced inconsistent results (table 3). While Holzer et al calculated that six patients have to be treated by TH to save one more life (95% CI 4–13, *p* < 0,05), Nielsens meta-analysis pointed out to some gaps in the evidence (Nielsen et al., 2010). Anyway, we have still quite firm evidence compared to other commonly used procedures in intensive care medicine.


TH…therapeutic hypothermia, RR…relative risk, CI…confidence interval, \*…p<0.05

Table 3. The results of the analyses of the registries and metaanalyses.

### **5. Indications and contraindications**

Therapeutic hypothermia is indicated in successfully resuscitated cardiac arrest patients, persistence of coma and requirement of mechanical ventilation. While there is good evidence supporting cooling of out-of-hospital adult cardiac arrest patients presenting with ventricular fibrillation, induction in other groups of in-hospital cardiac arrest patients, such as those with non-defibrillating initial rhythm and pediatric patients is supported by lower degrees of evidence (Holzer et al., 2006; Storm et al., 2008; Don et al., 2009; Polderman et al., 2003; Biarent et al., 2010; Testori et al., 2011). Nevertheless, clinical, experimental and patophysiological data suggest that hypothermia may also have protective effects in patients other than those with ventricular fibrillation OHCA. Therefore, it may be considered reasonable to also use TH in these groups.

Experimental data suggest that TH should be started as soon as possible after ROSC. Thus, in OHCA patients, pre-hospital initiation of cooling appears to be a method of choice and this is discussed further. On the contrary, there is weak evidence of what is an acceptable delay from ROSC to start chilling. It is generally considered that it should not be delayed more than 6 hours.

There is not any general agreement on contraindications to TH and there are differences among the guidelines. However, Table 4 lists all conditions we should keep in mind.

2007).

Michelson et al.,

Frelinger AL et al.,

antithrombotics.

circulatory support.

2003

Bjelland et al., 2010 clopidogrel

Therapeutic Hypothermia in Cardiac Arrest Survivors 77

PCI performed during TH is not associated with an increased rate of major bleeding events (Schefold et al., 2009; Knafelj et al., 2007). It is an important issue because major bleeding has been shown to be a strong predictor of mortality in patients after PCI (Manoukian et al.,

Third, it has been shown that antithrombotic and / or fibrinolytic therapy may be administered as the part of complex post-cardiac arrest care including therapeutic hypothermia. It is not a contraindication to TH and it should be used in accordance with the recent guidelines regardless of cooling (Van de Werf et al., 2008; Torbicki et al., 2008). On the other hand, most enzymatic processes in the body exhibit temperature dependency. Pilot studies presented alterations in the pharmacokinetics of many drugs used in critical care during TH. Antithrombotics are not an exclusion (table 5). Further studies are neccessary to

**AUTHOR DRUG EFECT OF HYPOTHERMIA ON METABOLISM** 

platelet aggregation

Fourth, it has been previously described that post-cardiac arrest myocardial dysfunction is a common phenomenon. In AMI patients, apart from myocardial stunning, myocardial ischemia due to coronary artery occlusion can play a role. There are several experimental and clinical data that TH does not ameliorate post-cardiac arrest myocardial dysfunction (Murphy et al, 2010; Hsu et al., 2009). Nevertheless, there is a room for further investigation. Complex hemodynamic clinical presentation of patients post-cardiac arrest can be the shock syndrome. It is a condition reflecting not only myocardial dysfunction but also systemic ischemia-reperfusion injury. A few studies have reported on the feasibility of TH in the setting of cardiogenic shock (Hovdenes et al., 2007; Skulec et al., 2008). Further investigation is needed but TH should be always considered in circulatory unstable patients who are responsive to volume expansion and/or vasopressoric support and/or mechanical

Accurate and safe practicing of TH following a written protocol is a prerequisite for achieving optimal neuroprotective effect. Till now, only few papers summarizing practical procedural issues of the TH procedure in detail have been published (Bianchin et al., 2009;

induced platelet dysfunction in vivo

in patients with hypothermia, absence of effect on day 1 after administration, with some improvement

mild hypothermia augments eptifibatide- and tirofiban- but not abciximab-induced inhibition of

1999 aspirin aspirin did not significantly augment hypothermia-

on day 3

Table 5. Effects of hypothermia on pharmacokinetics and pharmacodynamics of

tailor its dosage regimen for the specific conditions during TH.

glycoprotein IIb/IIIa antagonist

**7. Protocol for practicing therapeutic mild hypothermia** 

Škulec et al., 2010). Local therapeutic protocols differ from one to the other.

#### **CONTRAINDICATIONS TO THERAPEUTIC HYPOTHERMIA**


Table 4. Contraindications of therapeutic hypothermia.

The use of TH in children has been classified as a class IIb recommendation by pediatric guidelines 2005 and the recent guidelines have accepted this treatment in children too (Polderman et al., 2001; Losert et al., 2008). Therefore, pediatric cardiac arrest patients should not be excluded from the candidates for PTH.

#### **6. Therapeutic hypothermia in the setting of acute myocardial infarction and PCI**

Acute myocardial infarction (AMI) is the most common cause of cardiac arrest.. In a patient with cardiac arrest caused by AMI, there may arise several questions:


First, it should be noticed that successfully resuscitated cardiac arrest patient with AMI must be considered as a critically ill patient in general and not onlyanAMI patient. Therefore, PCI must not cause a delay of initiation of cooling and it is even more paramount vice versa. Performance of both procedures in parallel has been shown to be feasible and safe and PCI centers should establish TH in the catheterization laboratory as routine procedure (Koester et al., 2011). The use of common cooling methods do not interfere with the procedure of cardiac catheterisation.

Second, there is evidence, that the concomitant use of TH does not compromise angiographic results of PCI (Noc, 2010). Moreover, several studies have demonstrated, that

**CONTRAINDICATIONS TO THERAPEUTIC HYPOTHERMIA** 

• circulatory failure defined as hypotension irresponsive to volume expansion and/or

The use of TH in children has been classified as a class IIb recommendation by pediatric guidelines 2005 and the recent guidelines have accepted this treatment in children too (Polderman et al., 2001; Losert et al., 2008). Therefore, pediatric cardiac arrest patients

**6. Therapeutic hypothermia in the setting of acute myocardial infarction and** 

Acute myocardial infarction (AMI) is the most common cause of cardiac arrest.. In a patient

• What is more important and what should be prefered – TH or percutaneous coronary

• Should the patient planned for TH be treated by antithrombotics and / or

First, it should be noticed that successfully resuscitated cardiac arrest patient with AMI must be considered as a critically ill patient in general and not onlyanAMI patient. Therefore, PCI must not cause a delay of initiation of cooling and it is even more paramount vice versa. Performance of both procedures in parallel has been shown to be feasible and safe and PCI centers should establish TH in the catheterization laboratory as routine procedure (Koester et al., 2011). The use of common cooling methods do not interfere with

Second, there is evidence, that the concomitant use of TH does not compromise angiographic results of PCI (Noc, 2010). Moreover, several studies have demonstrated, that

• What is the impact of TH on myocardial function altered by AMI and cardiac arrest?

• established multiple organ failure or underlying terminal illness

• patient conscious after short cardiopulmonary resuscitation

vasopressoric support and/or mechanical circulatory support

• do-not-resuscitate or do-not-intubate status • coma of origin other than cardiac arrest

Table 4. Contraindications of therapeutic hypothermia.

should not be excluded from the candidates for PTH.

with cardiac arrest caused by AMI, there may arise several questions:

• Does TH induction interfere with the angiographic result of PCI?

• cardiac arrest of traumatic origin • preexisting hypothermia <34 °C

• pregnancy

intervention (PCI)?

the procedure of cardiac catheterisation.

thrombolytics?

**PCI** 

• pre-existing coagulopathy • severe systemic infection • severe active bleeding

PCI performed during TH is not associated with an increased rate of major bleeding events (Schefold et al., 2009; Knafelj et al., 2007). It is an important issue because major bleeding has been shown to be a strong predictor of mortality in patients after PCI (Manoukian et al., 2007).

Third, it has been shown that antithrombotic and / or fibrinolytic therapy may be administered as the part of complex post-cardiac arrest care including therapeutic hypothermia. It is not a contraindication to TH and it should be used in accordance with the recent guidelines regardless of cooling (Van de Werf et al., 2008; Torbicki et al., 2008). On the other hand, most enzymatic processes in the body exhibit temperature dependency. Pilot studies presented alterations in the pharmacokinetics of many drugs used in critical care during TH. Antithrombotics are not an exclusion (table 5). Further studies are neccessary to tailor its dosage regimen for the specific conditions during TH.


Table 5. Effects of hypothermia on pharmacokinetics and pharmacodynamics of antithrombotics.

Fourth, it has been previously described that post-cardiac arrest myocardial dysfunction is a common phenomenon. In AMI patients, apart from myocardial stunning, myocardial ischemia due to coronary artery occlusion can play a role. There are several experimental and clinical data that TH does not ameliorate post-cardiac arrest myocardial dysfunction (Murphy et al, 2010; Hsu et al., 2009). Nevertheless, there is a room for further investigation. Complex hemodynamic clinical presentation of patients post-cardiac arrest can be the shock syndrome. It is a condition reflecting not only myocardial dysfunction but also systemic ischemia-reperfusion injury. A few studies have reported on the feasibility of TH in the setting of cardiogenic shock (Hovdenes et al., 2007; Skulec et al., 2008). Further investigation is needed but TH should be always considered in circulatory unstable patients who are responsive to volume expansion and/or vasopressoric support and/or mechanical circulatory support.

### **7. Protocol for practicing therapeutic mild hypothermia**

Accurate and safe practicing of TH following a written protocol is a prerequisite for achieving optimal neuroprotective effect. Till now, only few papers summarizing practical procedural issues of the TH procedure in detail have been published (Bianchin et al., 2009; Škulec et al., 2010). Local therapeutic protocols differ from one to the other.

Therapeutic Hypothermia in Cardiac Arrest Survivors 79

those induced by cardiac arrest and the precipitating pathology. Anyhow, early goal directed therapy approach should be applied (Table 6), using standard procedures of volume expansion, vasopressor and/or inotropic support and/or mechanical circulatory

> **mean arterial pressure (mm Hg)** 65 – 100 **systolic blood pressure (mm Hg)** > 100 **central venous pressure (mm Hg)** 8 – 12 **diuresis (ml/kg/hour)** 1 **central venous oxygen saturation (%)** > 70

Peptic ulcer and deep vein thrombosis prophylaxis, oral care, positioning, ventilator setting assessment govern the same indications as in critically ill patients in general. There have been no publications recommending the timing and selection of suitable formulation of clinical nutrition in cardiac arrest patients. In most cases, it is initiated immediately after

Monitoring the body core temperature (BT) should be continuous with a probe inserted in the pulmonary artery, rectum, urinary bladder or oesophagus. Pulmonary artery catheter is considered to be the gold standard for BT measurement but this should not be the only indication of pulmonary artery catheter insertion. In the absence of other conditions such as the need for invasive haemodynamic monitoring, an alternative approach should be used.

Key principles for proper practicing of TH are to keep patients deeply sedated to suppress

Currently, there is a wide range of cooling techniques available. They can be classified into two major groups – the methods inducing whole body hypothermia and the methods primarily targeted to local brain cooling (Table 7) (Škulec et al., 2009). The latter should be considered as a supplementary approach because evidence for outcome benefit has been

In spite of the unproven superiority of device-based cooling techniques (endovascular catheter cooling, surface cooling via mattress or cooling pads), it is convenient for several reasons when compared with conventional approaches (surface cooling via ice-packs, cold intravenous infusion, cold gastric or urinary bladder lavage). It provides more reliable maintenance of target temperature including less indidence of overcooling, it is more friendly for nursing and allows control of BT even in the phase of controlled normothermia (Hoedemaekers et al., 2007; Škulec et al., 2009). However, it must be stressed that this does not disqualify conventional methods from routine practice. Pivotal clinical studies were based on simple surface cooling and its low cost predetermines it to serve as a starting method for new TH users (Hypothermia after Cardiac Arrest Study Group, 2002; Bernard et al., 2002).

Table 6. Goals for early directed therapy in cardiac arrest survivors.

Tympanic temperature is an acceptable for pre-hospital monitoring.

shivering and watch for potential side effects of TH carefully.

completion of the TH process (Škulec et al., 2010).

**8. Cooling methods** 

registered only for whole body cooling.

support (Cvachovec et al., 2009).

There are four phases of the therapeutic hypothermia protocol: cooling phase, maintenance phase, rewarming phase and period of control of normothermia (Figure 2).

Fig. 2. The protocol for practicing therapeutic hypothermia. ROSC…return of spontaneous circulation.

During the cooling phase, the aim is to reach the target therapeutic temperature of 33 °C as soon as possible after the return of spontaneous circulation. This is followed by BT regulation in the therapeutic range of 32 – 34 °C for 12 – 24 hours. In randomized clinical studies, both TH durations were associated with an improved outcome. It is not known, whether one protocol is better than the other or if cooling for a longer time is better. However, durations shorter than 12 hours should be avoided. It is advisable to beware of overcooling below 32 °C for its potential association with worse outcome (Merchant et al, 2006; Škulec et al., 2008). The rewarming phase is as important as the maintenance of TH. The rewarming should not be a passive but rather an active process resulting in slow and controlled reaching of normothermia with a rewarming rate of 0.1–0.33°C/hour (Seder & Van der Kloot, 2009). It can be performed by the same technique that was used for chilling. Once a BT of 36 °C is reached, normothermia should be controlled carefully for up to 48 hours since further hypothermia may worsen the outcome.

In the course of cooling, adequate sedation should be tailored for suppression of involuntary shivering. In some patients, neuromuscular blockade can be required during the cooling and maintenance phases (Hékimian et al., 2004). In the next phases, muscle paralysis should be strictly individual to avoid adverse effects of lengthy neuromuscular blockade. After rewarming, sedatives, analgesics, and muscle relaxants should be discontinued and standard intensive care should be provided. Other drugs decreasing the shivering threshold are buspirone and meperidine.

Induction of hypothermia can evoke metabolic disturbances, especially hypokalemia, hypomagnesemia, hypophosphatemia, hypocalcemia and hyperglycemia (Michelson et al., 1994; Polderman et al., 2001; Losert et al., 2008). Therefore, regular measurements are advisable. While potassium and magnesium supplementation is favourable during the maintenance phase, calcium may worsen ischemia-reperfusion injury and permissive hypocalcemia may be tolerated. However, mild hypokalemia may be tolerated to avoid excessive increase of Potassium during rewarming. Control of Glucose levels should follow conventional rather than strict targets.

Decrease of body temperature is physiologically associated with haemodynamic changes. Relative and absolute hypovolemia, bradycardia, decrease of arterial blood pressure and cardiac output are common. In general, it is impossible to distinguish these changes from

There are four phases of the therapeutic hypothermia protocol: cooling phase, maintenance

During the cooling phase, the aim is to reach the target therapeutic temperature of 33 °C as soon as possible after the return of spontaneous circulation. This is followed by BT regulation in the therapeutic range of 32 – 34 °C for 12 – 24 hours. In randomized clinical studies, both TH durations were associated with an improved outcome. It is not known, whether one protocol is better than the other or if cooling for a longer time is better. However, durations shorter than 12 hours should be avoided. It is advisable to beware of overcooling below 32 °C for its potential association with worse outcome (Merchant et al, 2006; Škulec et al., 2008). The rewarming phase is as important as the maintenance of TH. The rewarming should not be a passive but rather an active process resulting in slow and controlled reaching of normothermia with a rewarming rate of 0.1–0.33°C/hour (Seder & Van der Kloot, 2009). It can be performed by the same technique that was used for chilling. Once a BT of 36 °C is reached, normothermia should be controlled carefully for up to 48

In the course of cooling, adequate sedation should be tailored for suppression of involuntary shivering. In some patients, neuromuscular blockade can be required during the cooling and maintenance phases (Hékimian et al., 2004). In the next phases, muscle paralysis should be strictly individual to avoid adverse effects of lengthy neuromuscular blockade. After rewarming, sedatives, analgesics, and muscle relaxants should be discontinued and standard intensive care should be provided. Other drugs decreasing the shivering threshold

Induction of hypothermia can evoke metabolic disturbances, especially hypokalemia, hypomagnesemia, hypophosphatemia, hypocalcemia and hyperglycemia (Michelson et al., 1994; Polderman et al., 2001; Losert et al., 2008). Therefore, regular measurements are advisable. While potassium and magnesium supplementation is favourable during the maintenance phase, calcium may worsen ischemia-reperfusion injury and permissive hypocalcemia may be tolerated. However, mild hypokalemia may be tolerated to avoid excessive increase of Potassium during rewarming. Control of Glucose levels should follow

Decrease of body temperature is physiologically associated with haemodynamic changes. Relative and absolute hypovolemia, bradycardia, decrease of arterial blood pressure and cardiac output are common. In general, it is impossible to distinguish these changes from

phase, rewarming phase and period of control of normothermia (Figure 2).

Fig. 2. The protocol for practicing therapeutic hypothermia.

hours since further hypothermia may worsen the outcome.

ROSC…return of spontaneous circulation.

are buspirone and meperidine.

conventional rather than strict targets.

those induced by cardiac arrest and the precipitating pathology. Anyhow, early goal directed therapy approach should be applied (Table 6), using standard procedures of volume expansion, vasopressor and/or inotropic support and/or mechanical circulatory support (Cvachovec et al., 2009).


Table 6. Goals for early directed therapy in cardiac arrest survivors.

Peptic ulcer and deep vein thrombosis prophylaxis, oral care, positioning, ventilator setting assessment govern the same indications as in critically ill patients in general. There have been no publications recommending the timing and selection of suitable formulation of clinical nutrition in cardiac arrest patients. In most cases, it is initiated immediately after completion of the TH process (Škulec et al., 2010).

Monitoring the body core temperature (BT) should be continuous with a probe inserted in the pulmonary artery, rectum, urinary bladder or oesophagus. Pulmonary artery catheter is considered to be the gold standard for BT measurement but this should not be the only indication of pulmonary artery catheter insertion. In the absence of other conditions such as the need for invasive haemodynamic monitoring, an alternative approach should be used. Tympanic temperature is an acceptable for pre-hospital monitoring.

Key principles for proper practicing of TH are to keep patients deeply sedated to suppress shivering and watch for potential side effects of TH carefully.

### **8. Cooling methods**

Currently, there is a wide range of cooling techniques available. They can be classified into two major groups – the methods inducing whole body hypothermia and the methods primarily targeted to local brain cooling (Table 7) (Škulec et al., 2009). The latter should be considered as a supplementary approach because evidence for outcome benefit has been registered only for whole body cooling.

In spite of the unproven superiority of device-based cooling techniques (endovascular catheter cooling, surface cooling via mattress or cooling pads), it is convenient for several reasons when compared with conventional approaches (surface cooling via ice-packs, cold intravenous infusion, cold gastric or urinary bladder lavage). It provides more reliable maintenance of target temperature including less indidence of overcooling, it is more friendly for nursing and allows control of BT even in the phase of controlled normothermia (Hoedemaekers et al., 2007; Škulec et al., 2009). However, it must be stressed that this does not disqualify conventional methods from routine practice. Pivotal clinical studies were based on simple surface cooling and its low cost predetermines it to serve as a starting method for new TH users (Hypothermia after Cardiac Arrest Study Group, 2002; Bernard et al., 2002).

Therapeutic Hypothermia in Cardiac Arrest Survivors 81

Fig. 3. Watter cooling mattress system Blanketrol III, Cincinnati Sub-Zero Products, Inc., Cincinnati, USA. Published with the permission of P. Telekes, M.D., Cardiac Center, Liberec

Fig. 4. Highly effective cooling pads EMCOOLS pad, EMCOOLS, Emergency Medical

Cooling Systems AG, Vienna, Austria, with the permission.

Hospital, Czech republic.


RIVA…rapid intravenous administration of cold crystalloids

+…low cooling rate, incidence of overcooling or clinical evidence, +++…high cooling rate, incidence of overcooling or clinical evidence, €…low financial costs, €€€…high financial costs

Table 7. Cooling methods.

Surface cooling techniques are used most commonly. There is a variety of different modifications: simple ice-packs, more effective manufactured precooled special pads and sophisticated devices with automated feedback control of BT, chilling via gel pads or a cooling mattress (figures 3 and 4).

The method of rapid intravenous administration of cold crystalloids (RIVA) has been found efficient, simple and safe. Infusion of 15 – 30 ml/kg of 4 °C cold crystalloid (normal saline, Ringer solution, Hartmann solution) at an infusion rate >60 ml/h may induce the decrease of BT >1.4 °C (Kim et al., 2005; Kim et al., 2007; Kliegel et al., 2005; Kliegel et al., 2007; Virkkunen et al., 2004; Kämäräinen et al., 2008; Bernard & Rosalion, 2008; Bruel et al., 2008; Polderman et al., 2005; Vanden Hoek et al., 2004). However, additional cooling techniques are necessary for TH maintainance (Kliegel et al., 2007).

Endovascular catheter cooling is a highly sophisticated method of invasive cooling with a cooling rate 1.0 – 1.5 °C/h (Škulec et al., 2009). It is performed by inserting the cooling catheter into the inferior vena cava using the femoral approach. Circulation of cold normal saline through the catheter is directed by an extracorporeal unit to reach a satisfactory cooling rate and an excellent temperature stability of the patient (figure 5) (Hoedemaekers et al., 2007).

• **ice-packs** + +++ +++ € • **water cooling mattress** ++ ++ +++ €€ • **air cooling mattress** + +++ + €€ • **special cooling pads** ++ ++ + €€ • **cold air flow** ? ? - € • **RIVA** +++ ++ +++ €

**bladder lavage** ? ? - € • **cold peritoneal lavage** ? ? - ?

**cooling** ++ + +++ €€€

**device** +++ ? + €€€

+…low cooling rate, incidence of overcooling or clinical evidence, +++…high cooling rate, incidence of

Surface cooling techniques are used most commonly. There is a variety of different modifications: simple ice-packs, more effective manufactured precooled special pads and sophisticated devices with automated feedback control of BT, chilling via gel pads or a

The method of rapid intravenous administration of cold crystalloids (RIVA) has been found efficient, simple and safe. Infusion of 15 – 30 ml/kg of 4 °C cold crystalloid (normal saline, Ringer solution, Hartmann solution) at an infusion rate >60 ml/h may induce the decrease of BT >1.4 °C (Kim et al., 2005; Kim et al., 2007; Kliegel et al., 2005; Kliegel et al., 2007; Virkkunen et al., 2004; Kämäräinen et al., 2008; Bernard & Rosalion, 2008; Bruel et al., 2008; Polderman et al., 2005; Vanden Hoek et al., 2004). However, additional cooling techniques

Endovascular catheter cooling is a highly sophisticated method of invasive cooling with a cooling rate 1.0 – 1.5 °C/h (Škulec et al., 2009). It is performed by inserting the cooling catheter into the inferior vena cava using the femoral approach. Circulation of cold normal saline through the catheter is directed by an extracorporeal unit to reach a satisfactory cooling rate and an excellent temperature stability of the patient (figure 5) (Hoedemaekers et al., 2007).

• **intranasal cooling** + (++ brain) ? +++ ? • **cooling cap / helmet** + (++ brain) ? + ?

 **LOCAL BRAIN COOLING** 

overcooling or clinical evidence, €…low financial costs, €€€…high financial costs

RIVA…rapid intravenous administration of cold crystalloids

are necessary for TH maintainance (Kliegel et al., 2007).

**INCIDENCE OF OVERCOOLING**

**CLINICAL EVIDENCE**

**FINANCIAL COSTS** 

**RATE** 

 **WHOLE-BODY HYPOTHERMIA** 

**METHOD COOLING** 

• **surface cooling** 

• **cold gastric / urinary** 

• **endovascular cathether** 

• **extracorporal circuit** 

Table 7. Cooling methods.

cooling mattress (figures 3 and 4).

Fig. 3. Watter cooling mattress system Blanketrol III, Cincinnati Sub-Zero Products, Inc., Cincinnati, USA. Published with the permission of P. Telekes, M.D., Cardiac Center, Liberec Hospital, Czech republic.

Fig. 4. Highly effective cooling pads EMCOOLS pad, EMCOOLS, Emergency Medical Cooling Systems AG, Vienna, Austria, with the permission.

Therapeutic Hypothermia in Cardiac Arrest Survivors 83

decrease the metabolic demand of organs and tissues has the potential to translate into a novel approach to prevention of ischemia-reperfusion injury. Neurotensin and its analogies

TH treatment may be associated with adverse events (Table 8) (Polderman et al., 2009; Skulec et al., 2008; Nielsen et al., 2009). These are related to the cooling device or caused by hypothermia itself. While the former are rare, the latter are quite common. However, we should keep in mind the fact that some of them are physiological responses to hypothermia

> **pneumonia** 40 – 50 **any arrhythmias** 33 **recurrence of cardiac arrest** 7 – 11

 **sustained hyperglycaemia >8 mmol/l** 37  **hypokalaemia <3 mmol/l** 18  **hypomagnesaemia** 18  **hypophosphataemia** 19  **hyperamylasaemia** 12 **seizures** 24 **major bleeding** 3 – 6 **sepsis** 4

Table 8. Adverse effects of therapeutic hypothermia (Nielsen et al., 2009).

Often, it is not possible to distinguish whether the event is caused by hypothermia or by post-cardiac arrest syndrome itself. This is especially the case of post-cardiac arrest shock syndrome. It may affect as much as 18 – 50% of cardiac arrest patients. Although TH induces cold diuresis, hypovolemia, bradycardia and decrease of cardiac output, post-cardiac arrest myocardial dysfunction and precipitating pathology arises independently of TH and all must be treated together. It was shown that TH does not worsen the course of the disease even in the presence of shock syndrome (Škulec et al., 2008). Pneumonia, cardiac arrhythmias, metabolite disturbances and seizures have been identified as the most common

Recently, Nielsen et al analyzed a set of 765 patients and identified that sustained hyperglycaemia and seizures are associated with increased mortality while other adverse events are not (Nielsen et al., 2011). Randomized clinical studies showed comparable incidence of adverse events except for pneumonia and sepsis, which exhibited a trend of increased appearance in cooled patients (Hypothermia after Cardiac Arrest Study Group,

**FREQUENCY (%)**

are studied as potential triggers of this process (Schneider et al., 2006).

**metabolic disturbances** 

**9. Side effects of cooling** 

rather than avderse events.

adverse events of cooling.

2002; Bernard et al., 2002).

Fig. 5. Endovascular catheter cooling via CoolGard 3000TM and Icy femoral cathether TM, Alsius Corp., Irvine, USA. Published with the permission of P. Ostadal, M.D., PhD., Department of Cardiology, Na Homolce Hospital, Czech republic.

It can be used also for controlling normothermia after the rewarming phase and guide the patients through the whole early post-resuscitation period. Cooling efficacy and safety was documented not only for cardiac arrest patients but also for patients with traumatic brain injury, stroke and acute myocardial infarction (Diringer et al., 2004; Schmutzhard et al., 2002; Lyden et al., 2005; Guluma et al., 2008; Georgiadis et al., 2001; Keller et al., 2003; De Georgia et al., 2004; Dixon et al., 2002; Kandzari et al., 2004; Stone et al., 2006). Dixon et al. randomized 42 patients with AMI to primary PCI with or without endovascular cooling (target core temperature 33 degrees C) for 3 hours after reperfusion. Reduction of infarct size ws not identifyed but the study confirmed the safety of endovascular cooling as an adjunct to primary PCI. Kandzari et al. reached similar results in a nonrandomized study of 18 patients.

Intranasal cooling is a new and safe method to easily cool down the brain. It is a very simple method with minimal burdening of the staff (Castrén et al., 2010). Like RIVA, it is convenient for the rapid induction of TH and suitable for pre-hospital care and emergency medicine departments.

There are many other experimental thermoregulatory methods. Some of them, like cardiopulmonary bypass, femorocarotic bypass with extracorporeal cooling of blood, peritoneal cooling or total liquid ventilation cooling are too invasive and/or complicated in comparison with the current effective techniques (Reed et al., 2002; Mori et al., 2001; Hong et al., 2002; Xiao et al., 1995). The others are just in the beginning of their development and perhaps represent the future. On the gate of science-fiction stands the possibility of pharmacological induction of hibernation. Hibernation is a behavioural, physiological, and molecular adaptation exhibited by diverse mammalian species to withstand protracted periods or seasons of insufficient or unpredictable food availability (Kelly, 2007). Distribution of this process amongst different species argues that hibernation genes required for hibernation are common to all mammals (Srere et al., 1992). Such capacity to profoundly decrease the metabolic demand of organs and tissues has the potential to translate into a novel approach to prevention of ischemia-reperfusion injury. Neurotensin and its analogies are studied as potential triggers of this process (Schneider et al., 2006).
