Hypothermia was reported as a method for resuscitation as far back as 1803, when Russians covered a patient in snow hoping for a return of spontaneous circulation (ROSC). More recently, in the late 1950s, there were several case reports of the use of therapeutic hypothermia (TH) for neurologic injury after cardiac arrest, but it was subsequently abandoned because of uncertain benefit and difficulties with its use. Interest was rekindled in the early 1980s by Safar and colleagues at the University of Pittsburgh using the dog model. In the February 21, 2002, issue of The New England Journal of Medicine, two sentinel studies were published that led to new recommendations for Therapeutic Hypothermia after Cardiac Arrest from the International Liaison Committee on Resuscitation:

  • Unconscious adult patients with spontaneous circulation after out-of-hospital cardiac arrest should be cooled to 32°C to 34°C for 12 to 24 hours when the initial rhythm was ventricular fibrillation (VF).
  • Such cooling may also be beneficial for other rhythms or in-hospital cardiac arrest.

(International Liaison Committee on Resuscitation. Therapeutic hypothermia after cardiac arrest. Circulation 2003;108:118-121 )

Out-of-hospital sudden cardiac arrest is common, occurring in approximately one in 2000 adults per year. Even though the victim may initially survive to reach the hospital, discharge rates are dismal at 4%-9%. Hypoxic ischemic brain injury is an important cause of morbidity and mortality following resuscitation. The goal of TH is to reduce brain injury in survivors of cardiac arrest so there is less death and an increased chance of good neurological survival. It was shown that by treating seven patients with TH, one life can be saved. Similarly, the number needed to treat to improve neurologic outcome in one patient is five. 

Scientific Evidence Leading to the New Recommendations

A multi-center trial was conducted by the Hypothermia after Cardiac Arrest (HACA) Study Group in nine centers in five European countries with patients who had been resuscitated following a witnessed VF (ventricular fibrillation)/VT (ventricular tachycardia LTC) arrest and randomly assigned to undergo TH with a target temperature of 32-34°C, measured in the bladder. TH was induced using a mattress to deliver cold air over the entire body, and ice packs were often added to maintain the lower temperature. Seventy-five of the 136 patients in the hypothermia group had a favorable neurologic outcome (55%) (i.e., able to live independently and work at least part time) measured at 6 months, as compared with 54 of 137 (39%) in the normothermia group (p=0.009). Mortality at 6 months was 41% in the TH group compared to 55% in the normothermia group (p=0.02).

(The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. NEJM 2002;346:549-556)


The second study was undertaken in four hospitals in Melbourne, Australia. Seventy-seven patients with VF arrest were randomly assigned to either normothermia or TH with a core body temperature goal (pulmonary artery thermometry) of 33°C initiated in the field and reached within 2 hours using cold packs and maintained for 12 hours. Twenty-one of the 43 patients treated with TH (49%) survived and had a good outcome (i.e., they were discharged home or to a rehabilitation facility) as compared with nine of the 34 treated with normothermia (26%), [with a p=0.046].

(Bernard, S.A. et al. Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. NEJM 2002;346:557-563)


Selection of Patients for Therapeutic Hypothermia

The inclusion and exclusion criteria from the original research studies mentioned above are found in the table.

Inclusion Criteria (include all)

Exclusion Criteria

Patients resuscitated from out-of-hospital witnessed arrest with VF/VT as the initial rhythm

Severe cardiogenic shock (systolic BP < 90 mm Hg despite use of inotropes and fluids)


An estimated interval of 5-15 minutes from the patient’s collapse to the first attempt at resuscitation by emergency medical service (EMS)

Cause of coma other than cardiac arrest (e.g., head injury, drug overdose, cerebrovascular accident)

An interval of no more than 60 minutes from collapse to ROSC



Persistent coma after ROSC


A known coagulopathy


Endotracheal intubation with mechanical ventilation

Life-threatening arrhythmias


Adult age range


Initial temperature < 30°C



Pre-existing DNR code status


Some institutions are using TH with all cardiac arrest patients who remain in a coma no matter the initial rhythm. Polderman in 2003 reported on the use of TH in 68 patients matched with 70 historical controls – all in either asystole or pulseless electrical activity upon arrival of the ambulance. Mortality at six months was 56% in the TH group, whereas mortality in the control group was 78% (p=<0.02). Forty-four percent of the TH patients had a Glasgow Coma Score (GCS) of 15 at ICU discharge vs 20% of the control patients (p=0.01).

(Polderman, D. H. et al. Induced hypothermia improves neurological outcome in asystolic patients with out-of-hospital cardiac arrest. Circulation 2003;108:IV-581 [Suppl])


Therapeutic Hypothermia Protocols 

TH should be induced as early as possible after ROSC to gain the greatest neurologic benefit. The patient should be cooled to 32° to 34°C for a period of 12 to 24 hours. Core temperature should be monitored continuously and accurately using either bladder or esophageal thermometry. Unintentional overcooling occurs more frequently than originally thought, along with rebound hyperthermia at the end of active cooling. TH protocols from various hospitals can be found at the following websites: 


Post Resuscitation Care

Therapeutic hypothermia is a critical component of supportive care post resuscitation. Efforts are underway to define elements of a standardized post resuscitation care plan that optimizes hemodynamics, oxygenation and metabolism. Since it was found that coronary artery disease was the cause of arrest in 78% of cardiac arrest victims in a post mortem study in Finland, revascularization should be considered.

(Silfvast, T. Cause of death in unsuccessful prehospital resuscitation. Journal of International Medicine 1991;229:331-335)

Neurologic outcome cannot be predicted from clinical evaluation the first 24-48 hours post arrest. Prognostication is made even more difficult with the heavily sedated comatose victim who has TH. Thus, active treatment of the cardiac arrest victim should be pursued early and prognostication delayed until the patient is rewarmed and the effects of sedation and paralytics are minimized.