BACKGROUND: Postpartum haemorrhage (PPH) is the leading cause of maternal death worldwide. Tranexamic acid (TA), an antifibrinolytic drug, reduces bleeding and transfusion need in major surgery and trauma. In ongoing PPH following vaginal delivery, a high dose of TA decreases PPH volume and duration, as well as maternal morbidity, while early fibrinolysis is inhibited. In a large international trial, a TA single dose reduced mortality due to bleeding but not the hysterectomy rate. TA therapeutic dosages vary from 2.5 to 100 mg/kg and seizures, visual disturbances and nausea are observed with the highest dosages. TA efficiency and optimal dosage in haemorrhagic caesarean section (CS) has not been yet determined. We hypothesise large variations in fibrinolytic activity during haemorrhagic caesarean section needing targeted TA doses for clinical and biological efficacy. METHODS/DESIGN: The current study proposal is a blinded, randomised controlled trial with the primary objective of determining superiority of either 1 g of TXA or 0.5 g of TXA, in comparison to placebo, in terms of 30% blood-loss reduction at 6 h after non-emergency haemorrhagic caesarean delivery (active PPH > 800 mL) and to correlate this clinical effect in a pharmacokinetics model with fibrinolysis inhibition measured by an innovative direct plasmin measurement regarding plasmatic TA concentration. A sample size of 342 subjects (114 per group) was calculated, based on the expected difference of 30% reduction of blood loss between the placebo group and the low-dose group, out of which 144 patients will be included blindly in the pharmaco-biological substudy. A non-haemorrhagic reference group will include 48 patients in order to give a reference for peak plasmin level. DISCUSSION: TRACES trial is expected to give the first pharmacokinetics data to determinate the optimal dose of tranexamic acid to reduce blood loss and inhibit fibrinolysis in hemorrhagic cesarean section. TRIAL REGISTRATION: ClinicalTrials.gov, ID: NCT02797119 . Registered on 13 June 2016.