Traumatic brain injury (TBI) is a leading cause of death and long term disability, particularly in young adults. Studies from Australia have shown that approximately half of those with severe traumatic brain injury will be severely disabled or dead 6 months post injury. Given the young age of many patients with severe TBI and the long term prevalence of major disability, the economic and more importantly the social cost to the community is very high.

Pre-hospital and hospital management of patients with severe brain injury focuses on prevention of additional injury due primarily to lack of oxygen and insufficient blood pressure. This includes optimising sedation and ventilation, maintaining the fluid balance and draining Cerebrospinal Fluid (CSF) and performing surgery where appropriate. In recent years there has been a research focus on specific pharmacologic interventions, however, to date, there has been no treatment that has been associated with improvement of neurological outcomes.

One treatment that shows promise is the application of hypothermia (cooling). This treatment is commonly used in Australia to decrease brain injury in patients with brain injury following out-of-hospital cardiac arrest. Cooling is thought to protect the brain using a number of mechanisms. There have been a number of animal studies that have looked at how cooling is protective and also some clinical research that suggests some benefit. However at the current time there is insufficient evidence to provide enough proof that cooling should be used routinely for patients with brain injury and like all treatments there can be some risks and side effects.

The POLAR trial has been developed to investigate whether early cooling of patients with severe traumatic brain injury is associated with better outcomes. It is a randomised controlled trial, which is a type of trial that provides the highest quality of evidence.

The null hypothesis is that there is no difference in the proportion of favourable neurological outcomes six months after severe traumatic brain injury in patients treated with early and sustained hypothermia, compared to standard normothermic management.

Eligible patients will be randomised in the pre-hospital setting or on admission to the Emergency Department. POLAR study trained paramedics and physicians will screen patients in the pre-hospital setting. Eligible patients will be randomised if they fulfil the inclusion criteria with no pre-hospital exclusion criteria. Those randomised to the normothermia group will follow standard care. For those randomised to the "cooling arm", pre-hospital prophylactic hypothermia will be induced by exposure and by infusing up to 2 litres intravenous cold (4°C) 0.9% sodium chloride aiming for a core temperature of 35°C during transport. In the emergency department the "cooling arm" patients will be assessed to exclude significant bleeding and, once significant bleeding has been excluded, surface cooling vests/wraps will be applied to reach the target core temperature of 33°C. The patient will be then maintained at this temperature for a further 72 hours. Patients with significant bleeding will have cooling withheld until it is safe to decrease the temperature to the target core temperature of 33°C. Patients who have not been randomised pre-hospital will be re-screened in the ED. Eligible patients will be randomised if they fulfil the inclusion criteria with no ED exclusion criteria. Hypothermia will be induced by administration of up to 2L intravenous ice-cold (4°C) 0.9% sodium chloride followed by application of the surface cooling vests/wraps to achieve the target core temperature of 33°C. Patients allocated to standard 'normothermic' care will be maintained at a core temperature of 37°C ± 0.5°C.

Inclusion Criteria:

• Blunt trauma with clinical diagnosis of severe TBI and GCS <9
• Estimated age ≥ 18 and < 60 years of age
• The patient is intubated or intubation is imminent

Exclusion Criteria:

• Pre-hospital:

  • Clinical diagnosis of drug or alcohol intoxication as predominant cause of coma
  • Randomisation unable to be performed within 3 hrs of estimated time of injury
  • Estimated transport time to study hospital >2.5hrs
  • Able to be intubated without drugs
  • Systolic BP <90mmHg
  • Heart rate > 120bpm
  • GCS=3 + un-reactive pupils
  • Penetrating neck/torso injury
  • Known or obvious pregnancy
  • Receiving hospital is not a study site
  • Evidence of current anti-coagulant treatment

• Emergency Dept:

  • Clinical diagnosis of drug or alcohol intoxication as predominant cause of coma

  • Randomisation unable to be performed within 3 hrs of estimated time of injury

  • Able to be intubated without drugs

  • GCS=3 + un-reactive pupils

  • Persistent Systolic BP <90mmHg

  • Clinically significant bleeding likely to require haemostatic intervention, for example:

    • Bleeding into the chest, abdomen or retro-peritoneum likely to require surgery +/- embolisation
    • Pelvic fracture likely to require surgery +/- embolisation
    • More than two long bone fractures requiring operative fixation

  • Penetrating neck/torso injury

  • Positive urine or blood pregnancy test

  • Evidence of current anti-coagulant treatment

  • In the treating clinician's opinion, "cooling" is not in the patient's best interest