CRASH-3

Population

• Adults with TBI within 3 hours of injury, GCS 12 or lower or any intracranial bleeding on CT, and no major extracranial bleeding.

Intervention

Tranexamic acid 1 g over 10 minutes, then 1 g over 8 hours.

Comparator

Matching placebo.

Key results

• Among patients treated within 3 hours, head injury-related death was 18.5% with TXA vs 19.8% with placebo; RR 0.94 (95% CI 0.86-1.02).
• Excluding GCS 3 or bilateral unreactive pupils, death was 12.5% vs 14.0%; RR 0.89 (95% CI 0.80-1.00).
• TXA reduced head injury-related death in mild-to-moderate injury (RR 0.78, 95% CI 0.64-0.95) but not in severe injury (RR 0.99, 95% CI 0.91-1.07).
• Earlier treatment was more effective than later treatment in mild-to-moderate injury (P=0.005), with no obvious time-to-treatment effect in severe injury (P=0.73).
• Vascular occlusive events and seizures were similar between groups.

Harms

• The risk of vascular occlusive events was similar (RR 0.98, 95% CI 0.74-1.28).
• The risk of seizures was similar (RR 1.09, 95% CI 0.90-1.33).

Practice impact

• Give TXA within 3 hours of TBI without major extracranial bleeding, especially in mild-to-moderate injury; the overall primary result was not statistically significant.

Applicability

• Most applicable to adults with TBI who can be treated within 3 hours, have GCS 12 or lower or intracranial bleeding on CT, and do not have major extracranial bleeding.

Limitations

• The eligibility window was shortened from 8 hours to 3 hours during the trial.
• Patients with major extracranial bleeding were excluded.
• The primary comparison in patients treated within 3 hours had a confidence interval that included no effect.

Common misinterpretations

• This is not a license for late TXA; recruitment was limited to treatment within 3 hours after the protocol change.
• Benefit was not seen in severe head injury; do not expect TXA to reverse a GCS of 3 with unreactive pupils.