Postpartum hemorrhage is the most common cause of maternal death internationally.1 Tranexamic acid, an agent that inhibits fibrinogen and fibrin breakdown by plasmin, has been shown to reduce bleeding following surgery2 and major trauma.3 Clinical investigators therefore turned to tranexamic acid as a potential candidate to reduce the severity and consequences of postpartum hemorrhage.
The WOMAN study was a double-blind, placebo-controlled trial conducted in 21 countries. Women (age ≥ 16 years) with a clinical diagnosis of postpartum hemorrhage (estimated blood loss ≥ 500 mL after vaginal delivery, ≥ 1,000 mL after caesarean section, or any blood loss causing hemodynamic instability) were randomly assigned to receive 1 gm of tranexamic acid intravenously or placebo, in addition to usual care.
At the start of the trial, the primary outcome was a composite of all-cause death or hysterectomy within 42 days of randomization. However, it was later changed to death due to bleeding (and the sample size was increased) based on two findings that emerged while the study was underway. First, it became clear that patients were being enrolled in the study and undergoing hysterectomy at the same time instead of waiting to see if the study drug stopped the bleeding in trial subjects. Second, the results of a large clinical trial evaluating tranexamic acid in trauma patients reported a reduction in mortality due to bleeding.3 Secondary outcome measures included thromboembolic events, surgical interventions, complications, other adverse events, and quality of life.
The majority of the 20,060 women in the trial underwent in-hospital vaginal deliveries, and approximately half were enrolled within one hour of delivery. There were 483 maternal deaths, of which 72 percent were attributed to bleeding. The risk ratio for death due to bleeding with tranexamic acid was 0.78 (95% CI, 0.62-0.98; p=0.03) after adjustment for baseline risk factors. If tranexamic acid was given within three hours of delivery, the risk ratio for death due to bleeding was 0.69 (95% CI, 0.52-0.91; p=0.008). No benefit was seen if tranexamic acid was given longer than three hours after delivery. The incidence of thrombosis and other complications did not differ between groups.
In Brief
The results of the WOMAN study show that tranexamic acid reduced maternal mortality due to postpartum hemorrhage by approximately 30 percent if given within three hours of delivery. Reduction in mortality is arguably always clinically significant; however, the effect size for this trial seems small, and the upper limit of the confidence interval lies close to no effect. Given that treatment of postpartum hemorrhage often consists of multiple interventions at once (e.g., manual compression, surgical intervention, use of uterotonics, transfusion of coagulation factors), it is difficult to tease out the role tranexamic acid played in controlling bleeding. Furthermore, only 20 percent of subjects had an estimated loss of greater than 1,500 mL of blood and/or a systolic blood pressure less than 90 mmHg. These clinical features suggest that the baseline risk of death due to bleeding in the study population was generally low. However, an important caveat to this conclusion is that small bleeds can rapidly, and unpredictably, become life-threatening bleeds if not treated.
Perhaps the most noteworthy finding of this large study is the low frequency of thrombosis. This patient population is known to have a high baseline thrombotic risk; therefore, the absence of an increased risk with tranexamic acid in a study of this size is reassuring. While the benefit of tranexamic acid for postpartum hemorrhage may be small (absolute risk reduction for hemorrhage death, 0.4%; NNT = 250), it does not appear to cause harm. These results therefore give health care professionals another tool for their toolbox when dealing with postpartum hemorrhage.
References
Competing Interests
Dr. Linkins indicated no relevant conflicts of interest.
