Background
Head trauma is a leading cause of trauma related morbidity and mortality worldwide. Tranexamic acid (TXA) inhibits fibrinolysis and can decrease ongoing hemorrhage. In the CRASH-2 Trial, TXA given within 3 hours of injury was shown to decrease mortality from major extracranial injury (CRASH 2 Investigators, Lancet. 2010. PMID: 20554319). A subgroup analysis of the CRASH 2 study that included 270 patients who met criteria for trauma with concern for hemorrhage and a GCS < 15 and a brain CT scan compatible with traumatic brain injury did not demonstrate a mortality benefit of Tranexamic acid but did demonstrate less hematoma expansion and fewer new intracranial hemorrhages.
Clinical Question
In adults with head trauma and a Glasgow Coma Scale or less than 12 or traumatic brain injury on CT, does the administration of Tranexamic Acid within 3 hours of injury when compared to Placebo reduce the incidence of head trauma related death in hospital within 28 days of injury without an increase in thromboembolic or other adverse events?
Population
Inclusion:
- Adults
- Traumatic brain injury within 3 hours of injury
- GCS 12 OR any traumatic brain injury on CT
Setting: 175 hospitals in 29 countries, 7/2012-1/2019
Intervention
Tranexamic Acid: 1 gram over 10 minutes then 1 gram over 8 hours
Control
Placebo (0.9% Saline): Appearance and packaging identical to TXA
Outcomes
Primary Outcomes: Head Injury Death in hospital within 28 days of injury
Sensitivity Analysis: Excluded GCS=3 or Bilateral unreactive pupils (unlikely to responded to TXA)
Subgroup Analyses:
- Head injury severity: Mild-Moderate (GCS 9-15), Severe (GCS 3-8)
- Age: 30 years, 31-60 years, > 60 years
- Timing of TXA administration: 1 hour, > 1hour to 3 hours, > 3 hours
Secondary Outcomes:
- Early Head injury related death in hospital within 24 hours of injury
- All disability and disability subcategories
- Vaso-occlusive events: MI, DVT, PE, Stroke
Design
Interventional: Randomized Clinical Trial
Excluded
Major extracranial bleeding
Primary Results
n = 9,202
Efficacy:
- There was no statistically significant difference in the primary outcome or in the sensitivity analysis excluding those with GCS3 or bilateral unreactive pupils
- All: RR = 94 (0.86, 1.02), RRR = 6.6%
- Sensitivity Analysis: RR = 89 (0.80, 1.00), RRR = 10.7%
- There was a statistically significant reduction in mortality within 28 days (TXA < Placebo) in the primary outcome in the subgroup analyses
- GCS 9-15: RR = 78 (0.64, 0.95), RRR = 22.7%. This difference met the author’s criteria for clinical significance of a relative risk reduction of > 15%.
- Bilaterally reactive pupils: RR = 87 (0.77, 0.98), RRR 13%
- NNT = 58 (1/ARD = 1/0.017). In patients with a GCS of 9-15 or bilaterally reactive pupils for every 58 patients treated TXA within 3 hours, 1 less patient would die from head trauma.
- There was a statistically significant reduction mortality within 24 hours (TXA < Placebo) in all patients and in the sensitivity analysis
- All: RR 0.81, 95% CI (0.69, 0.95), RRR 19%
- Sensitivity analysis: RR 0.72, 95% CI (0.56, 0.92), RRR 28%.
- Both of these outcomes meet the authors criteria for clinical significance
- The benefit of TXA was statistically significant only if administered within 3 hours of injury
Safety
- No statistically significant difference in combined or individual disability categories, combined or individual vaso-occlusive events or other adverse events.
Strengths
- Multinational, multicenter setting likely improving the generalizability of the results
- Allocation was concealed and participants and staff were masked to treatment group
- Treatment groups similar in assessed confounding variables
- Since the primary outcome was measured until hospital discharge, 99.1% were included in the primary intention to treat analysis
- Sub-analysis did not show a difference in the primary outcome and sensitivity analysis when comparing high income to low-middle income countries
- Included patient-oriented outcomes of mortality and disability
Limitations
- The randomization method not presented and it is unclear if randomization wa stratified within study site or country
- While 98.5% received the intended study bolus intervention, only 92.4% received the intended maintenance dose. The study’s results may underestimate the efficacy of TXA and underestimate its safety. A per protocol analysis would have been helpful.
- Study intended a sample size of 10,000 but only included 9,202. The study may be underpowered to detect rare adverse outcomes
- Included only 1 center in Canada and none in the US though no reason to believe results our would be different from other high-income countries
Author's Conclusions
“On the basis of the CRASH2 trial results, tranexamic acid was included in guidelines for the prehospital care of patients with trauma. However, patients with isolated traumatic brain injury were specifically excluded. The CRASH3 trial provides evidence that tranexamic acid is safe in patients with traumatic brain injury and that treatment within 3 hours of injury reduces head injury related deaths.”
Our Conclusions
Tranexamic acid, when administered within 3 hours of injury did not result in a decrease in head injury related mortality in hospital within 28 days but did result in decrease in head injury related mortality in hospital within 1 day which may better reflect death associated with head trauma. There was reduced head injury related mortality in the subgroups of patients with a Glasgow come score of 9-15 and in those with bilaterally reactive pupils compared to 0.9% saline placebo. There was no difference in disability measures, venous thromboembolic events or other adverse events when tranexamic acid is compared to placebo.
Potential Impact To Current Practice
The use of tranexamic acid in the patient with isolated head injury requires collaboration with our trauma surgery and neurosurgery colleagues.