The CRASH3 study: prehospital TXA for every injured patient?

The CRASH3 results are out, but do we know what to do? The study enrolled 9202 head injured patients within 3 hours of injury with a Glasgow Coma Scale (GCS) of 12 or less, or any intracranial bleeding on CT scan and randomised them to Tranexamic Acid (TXA) or placebo. The relative risk (RR) of all-cause mortality (RR 0.96, 95% CI 0.89-1.04) and head injury death (RR 0.94, 95% CI 086-1.02) among those receiving TXA were not significant. However there were significant differences in subgroups who were less severely injured (RR 0.89 95%CI 0.80-1.00 if those with GCS=3 or bilateral fixed pupils were excluded, and RR 0.78, 95% CI 0.64 to 0.95, in the GCS 9 to 15 subgroup) or treated earlier (p=0.005 for time effect)[1].The results need to be considered in the context of earlier CRASH2[2] results, which showed a reduction in all-cause mortality (RR 0.91, 0.85 to 0.97) and death due to bleeding (RR 0.85, 0.76-0.96) if trauma patients who were bleeding or at risk of bleeding were given TXA. In both CRASH2 and CRASH3 the TXA was given in the emergency department.