Traumatic brain injury: tranexamic acid reduces head injury-related mortality

  • Lancet

  • International Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.


  • Tranexamic acid reduced head injury-related mortality by 22% among patients with mild to moderate traumatic brain injury (TBI) treated within 3 hours.

Why this matters

  • Intracranial bleeding after TBI increases risks for poor outcomes.

 Key results

  • Among the 72.2% of patients treated within 3 hours of injury, risk for head injury-related death with tranexamic acid vs placebo: 18.5% vs 19.8%.
    • Risk ratio, 0.94 (95% CI, 0.86-1.02).
  • Greater reduction after excluding patients with Glasgow Coma Scale score of 3 or bilateral unreactive pupils: 12.5% vs 14.0%.
    • Risk ratio, 0.89 (95% CI, 0.80-1.00).
  • By severity:
    • Mild to moderate injury: risk ratio, 0.78 (95% CI, 0.64-0.95).
    • Severe injury: risk ratio, 0.99 (95% CI, 0.91-1.07).
  • Difference in benefit going from time to treatment of 0-240 minutes:
    • Significant in mild and moderate head injury (P=.005).
    • Not significant in severe head injury (P=.73).
  • Tranexamic acid, placebo similar on risks for vascular occlusive events, seizures.

Expert comment

  • In a Comment, Andrew P. Cap, MD, PhD, writes, "Despite its limitations, CRASH-3 is a remarkable study that will change practice, and tranexamic acid will benefit future patients with TBI who might reasonably have a chance of recovery from their injuries."

Study design

  • Randomized controlled trial, 12,737 adults with TBI (CRASH-3 trial):
    • ≤3 hours after injury (originally ≤8 hours).
    • Glasgow Coma Scale score ≤12 or intracranial bleeding on CT scan.
    • No major extracranial bleeding.
  • Randomization: tranexamic acid vs placebo.
  • Main outcome: in-hospital head injury-related death ≤28 days after injury in patients treated ≤3 hours after injury.
  • Funding: National Institute for Health Research Health Technology Assessment; others.


  • Wide confidence intervals.
  • Possible missed thrombotic/embolic events.
  • Inclusion of patients with unilateral unreactive pupils.