OBJECTIVE: To examine the safety, efficacy and pharmacology of intravenous (IV), intramuscular (IM) and oral tranexamic acid (TXA) use in pregnant women. DESIGN: Randomised, open-label trial. SETTING: Hospitals in Pakistan and Zambia. POPULATION: Women giving birth by caesarean section. METHODS: Women were randomised to receive 1 g IV, 1 g IM, 4 g oral TXA or no TXA. Adverse events in women and neonates were recorded. TXA concentration in whole blood was measured and the concentrations over time were examined with population pharmacokinetics. The relationship between drug exposure and D-dimer was explored. The trial registration is NCT04274335. MAIN OUTCOME MEASURES: Concentration of TXA in maternal blood. RESULTS: Of the 120 women included in the randomised safety study, there were no serious maternal or neonatal adverse events. TXA concentrations in 755 maternal blood and 87 cord blood samples were described by a two-compartment model with one effect compartment linked by rate transfer constants. Maximum maternal concentrations were 46.9, 21.6 and 18.1 mg/L for IV, IM and oral administration, respectively, and 9.5, 7.9 and 9.1 mg/L in the neonates. The TXA response was modelled as an inhibitory effect on the D-dimer production rate. The half-maximal inhibitory concentration (IC50 ) was 7.5 mg/L and was achieved after 2.6, 6.4 and 47 minutes with IV, IM and oral administration of TXA, respectively. CONCLUSIONS: Both IM and oral TXA are well tolerated. Oral TXA took about 1 hour to reach minimum therapeutic concentrations and would not be suitable for emergency treatment. Intramuscular TXA inhibits fibrinolysis within 10 minutes and may be a suitable alternative to IV.