WOMAN

Population

• Women aged 16 years and older with a clinical diagnosis of postpartum hemorrhage after a vaginal birth or caesarean section.

Intervention

Intravenous tranexamic acid 1 g plus usual care. A second 1 g could be given if bleeding continued after 30 minutes or stopped and restarted within 24 hours of the first dose.

Comparator

Matching placebo plus usual care.

Key results

• 20,060 women were randomly assigned (10,051 tranexamic acid, 10,009 placebo); 10,036 and 9,985 were included in the analysis.
• The composite primary end point of death from all causes or hysterectomy was not reduced: 534 of 10,036 (5.3%) versus 546 of 9,985 (5.5%); RR 0.97 (95% CI 0.87-1.09; p=0.65).
• Death due to bleeding: 155 of 10,036 (1.5%) versus 191 of 9,985 (1.9%); RR 0.81 (95% CI 0.65-1.00; p=0.045). The authors called this a significant reduction; the interval reached 1.00.
• Subgroup, death due to bleeding among women treated within 3 hours of birth: 89 (1.2%) versus 127 (1.7%); RR 0.69 (95% CI 0.52-0.91; p=0.008).
• Hysterectomy (overall) was not reduced: 358 of 10,036 (3.6%) versus 351 of 9,985 (3.5%); RR 1.02 (95% CI 0.88-1.07; p=0.84). Other causes of death did not differ significantly.

Harms

• Adverse events, including thromboembolic events, did not differ significantly between tranexamic acid and placebo. Event rates are not given.

Practice impact

• Give 1 g intravenous tranexamic acid as soon as postpartum hemorrhage is diagnosed, with a second gram if bleeding continues or restarts.
• Do not use TXA expecting fewer hysterectomies. The composite of death or hysterectomy was not reduced.

Applicability

• Women 16 years or older with clinically diagnosed postpartum hemorrhage after vaginal birth or caesarean section, treated in addition to usual care.

Limitations

• The named composite primary end point was not reduced.
• The overall bleeding-death confidence interval reached 1.00 even though p=0.045.
• Hysterectomy was often decided at randomization, which is why the sample size was increased from 15,000 to 20,000.
• Analyzed Ns (10,036 and 9,985) are slightly smaller than the randomized Ns (10,051 and 10,009).

Common misinterpretations

• Do not cite the composite of death or hysterectomy as a positive primary result.
• Hysterectomy was not reduced.
• The clearer bleeding-death difference was in the subgroup treated within 3 hours of birth; that is not the overall primary comparison.
• Late treatment is not supported by the 3-hour analysis.