Article Citation: CRASH-2 trial collaborators, Shakur H, Roberts I, Bautista R, et al. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet. 2010 Jul 3;376(9734):23-32. PMID: 20554319

What we already know about the topic: Traumatic injuries are a major cause of death worldwide. Approximately one-third of in-hospital traumatic deaths are a result of hemorrhage. The body’s hematologic response to trauma and surgery involves fibrinolysis, or clot breakdown, which can become exaggerated and pathologic. Antifibrinolytic agents reduce blood loss in patients with normal and exaggerated fibrinolytic response to surgery. In many studies tranexamic acid (TXA) has been found to reduce bleeding in patients undergoing elective surgery. It is thought that the hematologic response in trauma and surgery are similar and TXA may reduce mortality secondary to bleeding in trauma patients.

Why this study is important: To assess the effects of early administration of TXA on death, vaso-occlusive event and the receipt of blood in trauma patients.

Brief overview of the study: This study was a randomized placebo controlled trial in 274 hospitals in 40 countries. Adult trauma patients with significant hemorrhage (SBP <90 or HR >110) or who were considered to be at high risk were included. Patients were randomly assigned to either receive TXA or placebo within 8 hours of injury. Patients that met a clear indication for TXA were not randomly assigned and received TXA. Primary outcome was in-hospital death within 4 weeks of injury. Intention to treat analysis found TXA significantly reduced all-cause mortality, with a relative risk reduction of 0.91. Furthermore, the risk of death secondary to bleeding was also reduced with a relative risk reduction of 0.85. There was no significant difference in vaso-occlusive events or amount of blood products between the groups.

Limitations: The randomization process for the trial was not a true randomization. Patients who clearly met use for TXA or clearly did not met use were excluded and only those patients where it was uncertain were randomized. Only about 50% of patients in both placebo and TXA received blood products. In the US we would likely be giving a much higher percentage of patients blood products and massive transfusion. Furthermore, the patients who were included may not have been that sick: hypotensive (~32%) and tachycardic (~48%)in both groups. The most common cause of death was head injury, with only ~5% of the study population dying from hemorrhage. Finally, they were unable to assess the mechanism of TXA.

Take home points: The study was flawed and has its fair share of critics, but followup systemic reviews and studies, including MATTERs demonstrated clear benefit to early use of tranexamic acid to hemorrhaging trauma patients.