BACKGROUND: In response to the World Health Organization call for research on alternative routes for tranexamic acid (TXA) administration in women with postpartum haemorrhage, we examined the pharmacokinetics of TXA after i.v., i.m., or oral administration. METHODS: We conducted a randomised, open-label, crossover trial in 15 healthy volunteers who received i.v. TXA 1 g, i.m. TXA 1 g, or oral TXA solution 2 g. Blood samples were drawn up to 24 h after administration. Tranexamic acid concentration was measured with liquid chromatography-mass spectrometry, and the parameters of the pharmacokinetic models were estimated using population pharmacokinetics. RESULTS: The median time to reach a concentration of 10 mg L-1 was 3.5 min for the i.m. route and 66 min for the oral route, although with the oral route the target concentration was reached in only 11 patients. Median peak concentrations were 57.5, 34.4, and 12.8 mg L-1 for i.v., i.m., and oral routes, respectively. A two-compartment open model with body weight as the main covariate best fitted the data. For a 70 kg volunteer, the population estimates were 10.1 L h-1 for elimination clearance, 15.6 L h-1 for intercompartmental clearance, 7.7 L for the volume of central compartment, and 10.8 L for the volume of the peripheral compartment. Intramuscular and oral bioavailabilities were 1.0 and 0.47, respectively, showing that i.m. absorption is fast and complete. Adverse events were mild and transient, mainly local reactions and low-intensity pain. CONCLUSIONS: The i.m. (but not oral) route appears to be an efficient alternative to i.v. tranexamic acid. Studies in pregnant women are needed to examine the impact of pregnancy on the pharmacokinetics. CLINICAL TRIAL REGISTRATION: EudraCT 2019-000285-38; NCT03777488.