Ensure that patients who incur serious trauma receive tranexamic acid (TXA) within 3 hours of the injury.1
STRENGTH OF RECOMMENDATION
B: An analysis of a large randomized controlled trial (RCT).
CRASH-2 collaborators; Roberts I, Shakur H, Afolabi A, et al. The importance of early treatment with tranexamic acid in bleeding trauma patients: an exploratory analysis of the CRASH-2 randomised controlled trial. Lancet. 2011;377:1096-1101.
You are working in the emergency department (ED) of a rural hospital when a 26-year-old man is brought in with multiple injuries sustained in a high-speed collision and motor vehicle rollover. The nearest trauma center is more than an hour away. Should you administer TXA to the patient before transferring him?
Trauma is a leading cause of death among those younger than 40 years, and in 30% of such fatalities, hemorrhage is the cause.2 Tranexamic acid (TXA) minimizes blood loss by inhibiting lysine binding sites on plasminogen, thereby preventing the conversion of plasminogen to plasmin. This inhibits fibrinolysis and reduces clot breakdown, resulting in a reduction in bleeding.3
TXA has a proven track record
TXA is not new. It has been used to minimize blood loss associated with surgery for decades.4 A retrospective cohort study involving recent military engagements in Afghanistan showed a reduction in both coagulopathy and mortality in trauma patients who were given TXA.5 The CRASH-2 randomized controlled trial (RCT), initially published in 20106 and further analyzed in the study detailed below,1 was the first extensive multicenter trial to evaluate the use of TXA in civilian trauma care.
STUDY SUMMARY: TXA saves lives— within a 3-hour window
CRASH-2 studied the early administration of TXA in adult trauma patients in 274 hospitals in 40 countries.1,6 Patients (N = 20,211) were enrolled if the treating physician judged them to have or be at risk for significant hemorrhage and were randomized to either TXA or placebo, administered in identical-looking packs. Within 8 hours of injury, participants received a 1-g intravenous (IV) loading dose of either TXA or placebo over 10 minutes; a 1-g infusion over 8 hours followed. Patients and study staff were blinded to the treatment groups.
The primary outcome was overall mortality in the 4 weeks after injury. Secondary outcomes included vascular occlusive events (myocardial infarction, stroke, pulmonary embolism, and deep venous thrombosis), major surgical intervention, quantity of blood transfusion (if any), and cause of death (bleeding, vascular occlusion, multi-organ failure, head injury, or other cause). In this analysis, the researchers considered the effect of TXA on mortality based on time to administration of treatment after injury, severity of blood loss as assessed by systolic blood pressure, Glasgow coma scale score, and type of injury. All analyses were intention to treat, and follow-up was 99.6%.
TXA reduced all-cause mortality in the first month after trauma (relative risk [RR] = 0.91; 95% confidence interval [CI], 0.85-0.97; P = .0035; number needed to treat [NNT] = 68). There were 3076 deaths from all causes in both groups, 35% of which were the result of bleeding. Among patients who received TXA, the overall risk of death due to bleeding was 4.9%, vs 5.4% in the placebo group (RR = 0.85; 95% CI, 0.76-0.96; P = .0077; NNT = 119).
After 3 hours, TXA may do more harm than good
For those treated with TXA within the first hour of injury, the risk of death due to bleeding was 5.3%, vs 7.7% for the placebo group (RR = 0.68; 95% CI, 0.57-0.82; P<.0001; NNT = 41). Giving TXA between one and 3 hours of injury also reduced the risk of death due to bleeding, to 4.8% vs 6.1% for the placebo group (RR = 0.79; 95% CI, 0.64-0.97; P = .03; NNT = 77).
TXA administered more than 3 hours after injury, however, appeared to increase the risk of death due to bleeding, to 4.4% compared with 3.1% for the placebo group (RR = 1.44; 95% CI, 1.12-1.84; P = .004; number needed to harm = 77). The researchers found no evidence that TXA’s effect on death due to bleeding varied on the basis of systolic pressure, Glasgow coma score, or type of injury.
The rate of occlusive vascular events over the 4-week study period was similar in both groups (RR = 0.84; 95% CI, 0.68-1.02; P = .08). Of note, the rate of myocardial infarction was reduced by TXA (RR = 0.64; 95% CI, 0.42-0.97; P = .035; NNT = 504).