Tranexamic acid, an anti-fibrinolytic, helps reduce bleeding. When bleeding can be attenuated, patient-oriented outcomes related to bleeding can be improved. Its effect on outcomes, however, mirrors its clinical effect: quite small. In CRASH-2, general adult trauma with significant extracranial bleeding, the 28-day all-cause mortality reduction was 1.5%. In WOMAN, death due to bleeding from post-partum hemorrhage was reduced 0.4%.
Now, we address the use of TXA for intracranial bleeding. This has been evaluated before, however, in the TICH-2 trial. The primary outcome for TICH-2 was functional status at day 90, which was not reliably different between groups, nor was mortality at 90 days. There were, however, only 2,325 participants in this trial. CRASH-3, on the other hand, has enrolled 12,737.
With nearly six times as many patients, there are likewise about six times as many events – leading to detection of a small difference in head injury-related death, 18.5% in those receiving TXA versus 19.8% receiving placebo. Broken down further, the authors refine source of this difference primarily to those receiving TXA within three hours, and in those with mild and moderate head injury. Beyond three hours and in those with severe head injury, TXA administration was not associated with any reliable benefit (or harm).
However, any enthusiasm for TXA must evaporate when the results are no longer parsed based solely on head injury-related death. All-cause mortality is ultimately no different, with a reported RR of 0·96 (0·89–1·04). It would certainly be preferable if TXA were clearly beneficial, as opposed to observing an excess of non-head injury-related deaths to counterbalance its seeming benefit. Furthermore, looking at their various measures of disability, no clear advantage is seen favoring TXA. The ultimate flavor of their observations are somewhat embittered by these loamy morsels.
What should we do with these results? Well, probably, what we’ll likely do is expand the use of TXA. The good news is this is unlikely to be harmful and TXA is inexpensive. The bad news is, the minimal effect TXA does have on attenuating bleeding simply doesn’t result in a easily measured excess of functional adults post-injury. Practice variation is certainly acceptable surrounding the use of TXA, but we certainly ought not be dogmatic about the necessity of its application in isolated head injury.
“Effects of tranexamic acid on death, disability, vascular occlusive events and other morbidities in patients with acute traumatic brain injury (CRASH-3): a randomised, placebo-controlled trial”
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(19)32233-0/fulltext