The investigators aimed to compare two doses of intravenous (IV) tranexamic acid (TXA) with a combined single dose of topical and IV TXA on haemoglobin decline 24 hours after total joint arthroplasties.

The investigators conducted randomized, double-blind trial. The participants were randomized to either intrvenous group receiving 2 doses of 1 gram TXA in intravenous route 3 hours apart, or a combined application group receiving the first 1 gram IV and the topical dose was 1.5 gram after reduction of the fascia.

The investigators provided the pre-anaesthetic consultation. They carried out a rigorous clinical examination, a balancing of the defects and an adjustment of the therapies. The patients' written and informed consent was obtained. All patients had a preoperative blood count, blood grouping, renal assessment with dosage of uremia and creatinine. The investigators calculated the tolerable losses (mL). Randomization and allocation were carried out at this stage by an anesthesiologist other than the one managing the patient perioperatively.

On arrival in the operating room, the investigators set up the following systematic monitoring: electrocardioscope, pulse oximetry, gas analyzer, monitoring of capnia and curarization in the event of general anesthesia, non-invasive blood pressure. Two peripheral venous approaches were put in place. All patients received antibiotic prophylaxis with 2 g of Cefazolin and in case of allergy 900 mg of Dalacin. The anesthetic protocol was standardized. The choice between general or locoregional anesthesia was left to the discretion of the anesthesiologist treating the patient. For each patient, the investigators specified: The surgical approach, the type of prosthesis (cemented or not), the intraoperative posture, the duration of the procedure and of the anesthesia. During intraoperative monitoring, any hypotension with a decrease in mean arterial pressure of 20% of the baseline value for a period of more than 3 minutes or other complications were mentioned. The estimated bleeding (contents of the suction jar, number of drapes and compresses soaked in blood) with determination of the percentage of total losses intraoperatively as well as the results of the intraoperative blood count (if deemed necessary) were also specified. Transfusion of labile blood products was managed according to the patient's terrain, the degree of anemia, the speed of onset of the anemia and his hemodynamic tolerance. The analgesic protocol was also standardized.

Inclusion Criteria:

• Patients aged of 18 years or more;
• Primary total hip or knee arthroplasty;
• A traumatic, degenerative or malformative setting.

Exclusion Criteria:

• Revisions;
• Bilateral procedures;
• Polytrauma;
• Pregnant patients;
• Contraindications for the use of TXA, Coagulation disorders or anaemia with Haemoglobin (Hb) less than or equal to 9 g/dl preoperatively;
• Patients treated with anticoagulants, Heparin, Warfarin, Oestrogen.
• Serious anaesthesia-related complications (impossible orotracheal intubation , anaphylactic shock);
• Severe transfusion reaction such as haemolysis.