November 08, 2017
Immediate treatment with tranexamic acid reduces the risk for death by 70% for people with severe bleeding, researchers say. However, rapid treatment initiation is important, as even short delays are associated with reductions in benefit.
"We have to make sure tranexamic acid is available before patients reach hospital and whenever a woman gives birth," Ian Roberts, MD, from the London School of Hygiene & Tropical Medicine in the United Kingdom, said in a news release. Dr Roberts and colleagues published a meta-analysis of research on the drug online November 7 in the Lancet.
Every year, more than 2 million people worldwide die from traumatic extracranial bleeding, often as a result of road traffic injuries and violence. In addition, postpartum hemorrhage is the leading cause of maternal death worldwide, killing around 100,000 women a year.
Antifibrinolytic drugs such as tranexamic acid, aminocaproic acid, and aprotinin work by stabilizing blood clots and reducing bleeding. They have been used for many years to reduce heavy menstrual bleeding and are often given during surgery to reduce the need for blood transfusions.
In the WOMAN and CRASH-2 trials, tranexamic acid cut deaths resulting from postpartum hemorrhage and bleeding after serious injury by about a third if given within 3 hours of bleeding onset.
To understand better the importance of time, Dr Roberts and colleagues identified a total of 13 trials on tranexamic acid conducted between 1946 and 2017. The WOMAN and CRASH-2 trials were the only two that assessed the effect of time to treatment on treatment effectiveness. So the authors merged individual patient data on 40,138 patients enrolled in those two trials for the current meta-analysis. There were 3558 deaths, of which 1408 (39.6%) were from bleeding. Overall, 62.8% of deaths from bleeding (n = 884) occurred within 12 hours of bleeding onset. However, deaths resulting from postpartum hemorrhage peaked 2 to 3 hours after childbirth.
Among women with postpartum hemorrhage, 1.5% of those who died after receiving tranexamic acid died of bleeding (155/10,034) compared with 1.9% of those who died after receiving a placebo plus standard care (190/9977).
Among trauma patients, 4.9% died after receiving tranexamic acid (489/10,060) compared with 5.7% who died after receiving a placebo and standard care (574/10,067). Overall, compared with the patients who received placebo, those who received tranexamic acid were 20% more likely to survive (odds ratio, 1.20; 95% confidence interval, 1.08 - 1.33; P = .001). However, if given immediately, tranexamic acid improved survival by more than 70% (odds ratio, 1.72; 95% confidence interval, 1.42 - 2.10; P < .0001).
For every 15-minute delay in treatment, survival benefit was cut by about 10%, even after taking into account age and systolic blood pressure, which are strong risk factors for death resulting from bleeding. No benefit was seen if treatment was delayed beyond 3 hours.
The researchers also found no evidence of increased complications or risk for clotting such as heart attack, stroke, pulmonary embolism, or deep vein thrombosis in the patients receiving tranexamic acid. And they noted fewer cases of heart attacks with tranexamic acid than with the placebo.
Treatment delay may be underestimated in trauma because many injuries are not witnessed. However, the delay may be overestimated in postpartum hemorrhage because birth is taken as the time of bleeding onset. So the researchers used a sensitivity analyses to test a range of plausible errors. Their results support the conclusion that prompt treatment is essential.
"Tranexamic acid is safe, cheap, easily administered, and does not need to be refrigerated," Dr Roberts said. He called for national or regional benchmarks of time from bleeding onset to early treatment.
He also called for more research into the drug's mechanism of action. In an accompanying commentary, David J. Dries, MD, from the University of Minnesota in Minneapolis, explored what is known about that mechanism.
He noted that tranexamic acid binds to plasmin, which has a role in inflammation, angiogenesis, and wound healing, and that the drug has protective effects on the endothelium.
Patients who bleed heavily after injury tend to have hyperfibrinolysis, he said, and the patients in the CRASH-2 and WOMAN trials mostly fall into that category. But patients with severe brain injury or multiorgan trauma are more likely to experience shutdown of fibrinolysis than massive bleeding. "In these patients, administration of tranexamic acid might not be effective," Dr Dries cautions.
However, he notes that there are numerous ongoing trials testing tranexamic acid in a variety of settings. "As data from additional trials with tranexamic acid become available, the spectrum of applications for this agent should become apparent," he concludes.