Background Postpartum haemorrhage (PPH) ‐ heaving bleeding within the first 24 hours after giving birth ‐ is one of the main causes of death of women after childbirth. Antifibrinolytics, primarily tranexamic acid (TXA), have been shown to reduce bleeding in surgery and safely reduces mortality in trauma patients with bleeding without increasing the risk of adverse events. An earlier Cochrane review on treatments for primary PPH covered all the various available treatments ‐ that review has now been split by types of treatment. This new review concentrates only on the use of antifibrinolytic drugs for treating primary PPH. Objectives To determine the effectiveness and safety of antifibrinolytic drugs for treating primary PPH. Search methods We searched Cochrane Pregnancy and Childbirth’s Trials Register, ClinicalTrials.gov, the WHO International Clinical Trials Registry Platform (ICTRP) (28 May 2017) and reference lists of retrieved studies. Selection criteria Randomised controlled trials (RCTs), including cluster‐randomised trials of antifibrinolytic drugs (aprotinin, TXA, epsilon‐aminocaproic acid (EACA) and aminomethylbenzoic acid, administered by whatever route) for primary PPH in women. Participants in the trials were women after birth following a pregnancy of at least 24 weeks' gestation with a diagnosis of PPH, regardless of mode of birth (vaginal or caesarean section) or other aspects of third stage management. We have not included quasi‐randomised trials, or cross‐over studies. Studies reported as abstracts have not been included if there was insufficient information to allow assessment of risk of bias. In this review we only identified studies looking at TXA. Data collection and analysis Two review authors independently extracted data from each study using an agreed form. We entered data into Review Manager software and checked for accuracy. For key review outcomes, we rated the quality of the evidence as 'high', 'moderate', 'low' or 'very low' according to the GRADE approach. Main results Three trials (20,412 women) met our inclusion criteria. Two trials (20,212 women) compared intravenous (IV) TXA with placebo or standard care and were conducted in acute hospital settings (labour ward, emergency department) (in high‐, middle‐ and low‐income countries). One other trial (involving 200 women) was conducted in Iran and compared IV TXA with rectal misoprostol, but did not report on any of this review's primary or GRADE outcomes. There were no trials that assessed EACA, aprotinin or aminomethylbenzoic acid. Standard care plus IV TXA for the treatment of primary PPH compared with placebo or standard care alone Two trials (20,212 women) assessed the effect of TXA for the treatment of primary PPH compared with placebo or standard care alone. The larger of these (The WOMAN trial) contributed over 99% of the data and was assessed as being at low risk of bias. The quality of the evidence varied for different outcomes, Overall, evidence was mainly graded as moderate to high quality. The data show that IV TXA reduces the risk of maternal death due to bleeding (risk ratio (RR) 0.81, 95% confidence interval (CI) 0.65 to 1.00; two trials, 20,172 women; quality of evidence: moderate). The quality of evidence was rated as moderate due to imprecision of effect estimate. The effect was more evident in women given treatment between one and three hours after giving birth with no apparent reduction when given after three hours (< one hour = RR 0.80, 95% CI 0.55 to 1.16; one to three hours = RR 0.60, 95% CI 0.41 to 0.88; > three hours = RR 1.07, 95% 0.76 to 1.51; test for subgroup differences: Chi² = 4.90, df = 2 (P = 0.09), I² = 59.2%). There was no heterogeneity in the effect by mode of birth (test for subgroup differences: Chi² = 0.01, df = 1 (P = 0.91), I² = 0%). There were fewer deaths from all causes in women receiving TXA, although the 95% CI for the effect estimate crosses the line of no effect (RR 0.88, 95% CI 0.74 to 1.05; two trials, 20,172 women, quality of evidence: moderate). Results from one trial with 151 women suggest that blood loss of ≥ 500 mL after randomisation may be reduced (RR 0.50, 95% CI 0.27 to 0.93; one trial, 151 women; quality of evidence: low). TXA did not reduce the risk of serious maternal morbidity (RR 0.99, 95% CI 0.83 to 1.19; one trial, 20,015 women; quality of evidence: high), hysterectomy to control bleeding (RR 0.95, 95% CI 0.81 to 1.12; one trial, 20,017 women; quality of evidence: high) receipt of blood transfusion (any) (RR 1.00, 95% CI 0.97 to 1.03; two trials, 20,167 women; quality of evidence: moderate) or maternal vascular occlusive events (any), although results were imprecise for this latter outcome (RR 0.88, 95% CI 0.54 to 1.43; one trial, 20,018 women; quality of evidence: moderate). There was an increase in the use of brace sutures in the TXA group (RR 1.19, 95% CI 1.01, 1.41) and a reduction in the need for laparotomy for bleeding (RR 0.64, 95% CI 0.49, 0.85). Authors' conclusions TXA when administered intravenously reduces mortality due to bleeding in women with primary PPH, irrespective of mode of birth, and without increasing the risk of thromboembolic events. Taken together with the reliable evidence of the effect of TXA in trauma patients, the evidence suggests that TXA is effective if given as early as possible. Facilities for IV administration may not be available in non‐hospital settings therefore, alternative routes to IV administration need to be investigated.