Design and study setting
A double-blind randomized placebo-controlled trial with two parallel groups was conducted in women scheduled to undergo a vaginal delivery at the Maternity Teaching Hospital, Erbil City, Kurdistan region, Iraq, between February 1, 2020, and October 10, 2020. The women were randomly assigned to receive TA or placebo immediately, along with the administration of a uterotonic agent, after fetal delivery in the third stage of labor, and their data were collected. Blood loss was measured during two periods, from fetal to placental delivery and from placental delivery to 2 hours after.
Eligible participants were women aged ≥ 18 years who had a singleton pregnancy at ≥ 35 weeks of gestation, grand multiparity, a twin pregnancy, polyhydramnios, a previous history of PPH, a previous history of cesarean section, suspected macrocosmic fetus, prolonged labor, HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome, been receiving low-molecular-weight heparin and aspirin during pregnancy, an intention to deliver vaginally, and agreed to participate.
Patients with intrauterine fetal death, history of thromboembolic disease, current, or previous history of heart disease, renal and liver disorders, history of seizure or epilepsy, placenta previa, and placental abruption or refusal to participate were excluded from the trial. An obstetrician (one of the authors) provided the women with information about the trial during early labor when the obstetrician considered that vaginal delivery was likely (≥ 4 cm of cervical dilation). The women confirmed participation at the labor ward and provided written informed consent.
Randomization and procedures
Eligible women were randomly assigned in a 1:1 ratio to receive 1 g of TA or placebo (glucose water) administered intravenously. A computer-generated randomization code list was created using the program accessible at www.randomization.com. Two blocks of randomly varied sizes were used for the two arms. An independent statistician generated the randomization numbers. TA and placebo were prepared at a single site and by the same person (an independent pharmacist in the labor ward pharmacy). They were numbered and labeled in infusion bags containing a 30-cc syringes labeled as bag A (experimental group), containing 1 g/10 mL tranexamic acid diluted with 20 mL of 5% glucose water, and bag B (placebo group), containing 30 mL of 5% glucose water, each with a 30-mL vial of the trial regimens (1 g of TA or normal saline) depending on the randomization number. Neither the investigators nor the participants were aware of the trial-group assignments.
At the end of the delivery, the randomized number of the bag was applied to the questionnaires containing information about the patient and the details of the procedures. A statistician in the College of Medicine of Hawler Medical University independently analyzed the data until the trial was completed and the database was closed. All the participating women, researchers, and data handlers were blinded to the individual allocations throughout the study.
Interventional drug and grouping
The participants in the interventional group received two ampules of 5 mL TA added to 20 mL of 5% glucose water (TRENAXA 500 mg, Macleod Pharmaceuticals Ltd., India). The placebo group received 30 mL of glucose water 5% [Glucose (B Braun) 50 mg/mL] and oxytocin (5 IU/mL, 2 mL; Gland Pharma Limited).
The intravenous trial regimen was administered slowly (over a period of 60 seconds) immediately after fetal delivery, coinciding with the routine prophylactic intravenous injection of oxytocin and clamping of the umbilical cord. All the other aspects of the management of the third stage of labor were the same in the two groups.
The duration of the third stage was measured and recorded in minutes starting from injection of both medications (oxytocin + TA and oxytocin + glucose water) and placental delivery. During the fetal delivery, a sterile disposable pad of known weight was placed beneath the patient’s buttocks to collect blood loss and then weighed. Blood loss was measured during two periods, from fetal to placental delivery and from placental delivery to 2 hours after childbirth. Blood soaked gauzes, gowns, sheets, and tampons were all weighed before and after use (when blood soaked), and blood loss was estimated using the formula of Gai et al.  as follows: quantity of blood (mL) = (weight of used materials − weight of materials before use)/1.05.
Maternal observations were recorded every 15 minutes in the first hour and every 30 min in the second hour after delivery, and these data were recorded.
Sample size estimation
The sample size was estimated using the openepi.com computer program. The information entered in the program was based on the results of a pilot study involving 15 women in each study group (oxytocin vs. oxytocin + TA). The mean (± SD) blood loss in group 1 was 176.467 ± 39.169 mL, and that in group 2 was 162.4 ± 13.081 mL. The power was set at 90%; and the confidence interval, at 95%. Accordingly, the estimated sample size for the clinical trial was 100 in each group.
Data were analyzed using the Statistical Package for Social Sciences (version 25). The Student t test for two independent samples was used to compare two means. Multiple regressions were used where the dependent variables were the amount of blood loss and the duration of the third stage of labor. A p value of ≤ 0.05 was considered statistically significant.