A number of injuries can produce a GCS 3 and interventions are multiple with varying evidence of efficacy

Pragmatic, multicentre multinational RCT

Open-label with blinded follow-up

Could be randomised up to 10/7 post injury and must have had first-level interventions to treat it (head up, CSF drainage, etc)

If could not be controlled with hypothermia had barbiturates or hypertonic

Screen 2498 patients, 387 randomised

  • 188 randomised and analysed in hypothermia
  • 187 randomised and analysed in controls

No significant differences in baseline characteristics including age, severity of injury, APACHE, etc AND neurosurgical intervention prior to randomisation

Hypothermia did control ICP more than standard interventions in control group

Stopped early as greater incidence of unfavourable outcome (poor GOSE score) in hypothermia compared with controls

Unadjusted hazard ratio for mortality 1.45 (1.01,2.10) at 6/12

No difference in pneumonia rate between two groups


Conclusion: Titrated hypothermia to reduce raised ICP in addition to standard care did not improve outcomes but and increased mortality

Hypothermia for Intracranial Hypertension after Traumatic Brain Injury