Anti-fibrinolytic therapy in traumatic brain injury

Introduction
The CRASH-2 trial showed that in trauma patients (n=20,211) with major extracranial injury, administration of tranexamic acid (TXA) within 3 hours of injury significantly reduced all cause mortality and death from bleeding. This prompted TXA to be included in the guidelines for prehospital care with trauma, but traumatic brain injury (TBI) patients were not included as they were not included in the CRASH-2 trial. However, significant additional data has been obtained since that time that suggests a potential benefit for TXA in patients with TBI. TXA (1g IV over 10 minutes, then 1g IV over 8 hours) is likely of benefit in reducing head-injury related mortality in patients with TBI when administered within 3 hours in patients with GCS 9-15. In patients with GCS of ≤8 the benefit is less certain, but there is little evidence of harm. Overall NNT based on the post-CRASH-3 meta-analysis is ~63.

CRASH-2 IBS (intracranial bleeding) trial (2012)
Study was a subset of 270 CRASH-2 participants randomized to TXA (1g IV over 30 minutes, then 1g IV over 8 hours) or placebo, and showed a trend towards less hemorrhage growth (-3.8 mL, 95% CI -11.5 to 3.9 mL) and fewer deaths in the TXA group (OR 0.54, 95% CI 0.20 to 1.46).

Yutthakasemsunt et al. trial (2013)
Single center, double-blind, randomized, placebo-controlled trial of 238 TBI patients randomized to TXA (1g IV over 30 minutes, then 1g IV over 8 hours) or placebo. It showed a trend towards less hemorrhage expansion (RR 0.65, 95% CI 0.40 to 1.05), and no differences in death or unfavorable outcome. TXA appeared to be safe.

Jokar et al. trial (2017)
Single-center, single-blind, randomized, placebo-controlled trial of 80 TBI patients randomized to TXA (1g IV over 30 minutes, then 1g IV over 8 hours) or placebo. It showed less hemorrhage expansion in the TXA group (1.7 mL) vs placebo (4.3 mL) (p=<0.001). Functional outcomes and detailed adverse events were not reported.

Fakharian et al. trial (2018)
Single-center, double-blind, randomized, placebo-controlled trial of 149 patients with TBI randomized to TXA (1g IV over 30 minutes, then 1g IV over 8 hours) or placebo. Is showed a trend of fewer deaths and adverse outcomes in the TXA group but they were not statistically significant.

Chakroun-Walha et al. trial (2019)
Single-center, open-label, randomized, non-placebo-controlled trial of 180 TBI patients randomized to TXA (1g IV over 30 minutes, then 1g IV over 8 hours) or "no TXA." It showed no significant differences in outcomes, but the rate of pulmonary embolism was higher in the TXA group (11.5% vs 2.4%, p=0.02). However, the TXA group had much higher rates of abdominal and pelvic trauma, which likely increased the risk of DVT.

NCT01990768 trial (2019, as yet not formally published)
Multicenter, double-blind, randomized, placebo-controlled trial of 967 moderate to severe TBI patients randomized to TXA 1g IV bolus followed by 1g IV over 8 hours, TXA 2g IV bolus followed by placebo over 8 hours, or placebo. No significant difference in favorable function outcome or adverse events.

Weng et al. Meta-analysis (2019)
This meta-analysis included data from all five of the prior small trials listed above except for the NCT01990768 trial. It showed no differences in outcomes or mortality in the combined data set. However when including only CRASH-2 IBS, Yutthakasemsun et al. 2013, and Fakharian et al. 2018 in the analysis (excluding the other two which were low quality), it showed a nearly significant benefit on mortality with RR 0.64 (95% CI 0.41 to 1.00), and a significant benefit in achieving independence with RR 1.10 (1.01 to 1.20). There was no difference in thrombotic events.

CRASH-3 (2019)
Multicenter, randomized, double-blind, placebo-controlled trial, enrolling adults with TBI within 3 hours of injury, with GCS of 12 or lower OR any intracranial bleeding on CT, and no major extracranial bleeding. Patients were randomized to TXA (1g IV over 10 minutes, then 1g IV over 8 hours) or matching placebo. After excluding patients given treatment after more than 3 hours, as well as those lost to follow-up or who withdrew consent, they had 9,127 patients. Baseline characteristics were well-matched. Including all patients there was no significant difference in head injury-related death at 28 days post-injury with RR 0.94 (95% CI 0.86 to 1.02). However, in a prespecified analysis excluding patients with GCS 3 or bilateral unreactive pupils, there was a reduction in head injury-related death that just missed statistical significance with RR 0.89 (95% CI 0.80 to 1.00). Additionally, the RR for head injury-related death within 24 hours in the TXA group was 0.81 (95% CI 0.69 to 0.95) and this was even stronger when excluding patients with GCS 3 or bilateral unreactive pupils with RR 0.72 (95% CI 0.56 to 0.92). In the subgroup of patients with mild-moderate head injury (GCS 9-15) there was a clear reduction of head injury-related death (5.8% vs 7.5%) with RR 0.78 (95% CI 0.64 to 0.95) but no evidence of benefit in those with severe had injury (GCS 3-8). Earlier treatment was more effective than later treatment. The risk of vascular occlusive events and other complications was no different in the two groups and disability was similar between the two groups.

Post-CRASH-3 meta-analysis (2019)
In the publication of the CRASH-3 trial itself, a meta-analysis was performed (including the new CRASH-3 data with the CRASH-2 IBS data, Yutthakasemsunt et al. 2013 data, and the NCT01990768 trial) and showed a reduction in head-injury related death with TXA with RR 0.89 (95% CI 0.80-0.99). Of note, the NNT for this analysis would be ~63.