To compare the two administration methods of tranexamic acid for patients undergoing the first unilateral THA, to explore the effects of tranexamic acid on perioperative blood loss from the aspects of dominant blood loss, hidden blood loss, blood transfusion rate, mean blood transfusion volume and safety so as to further clarify what kind of medication is more effective.

Background Total hip arthroplasty (THA) is extensively used in various hip diseases. Abundant blood supply at surgical area of bone tissue, intraoperative extensive cleaning, acetabular formation, femoral expansion, and medullary cavity and bone wound bleeding, will lead to extensive intraoperative and postoperative blood loss. The blood transfusion rate is high, which affects patients' rehabilitation. Previous studies have shown that 16%-69% patients required blood transfusion after first THA, which greatly affected patient's rehabilitation, and increased the risk of transfusion-related infection, hemolysis, immunosuppression, acute lung injury, and death. How to reduce perioperative blood loss in total hip arthroplasty has become a hot spot for joint surgeons.

Tranexamic acid as a synthetic antifibrinolytic drug is a synthetic analog of the amino acid lysine. Its mechanisms of action are the competitive inhibition of fibrinolytic zymogens, and noncompetitive inhibition of fibrinolytic enzymes. Tranexamic acid strongly inhibits plasmin-induced fibrin decomposition, reduces fibrinolytic system activity, to achieve local hemostasis and reduce bleeding (Appendix 1). However, the route of administration of tranexamic acid is diverse, and the best way to administer is not clear. A large number of studies have confirmed that during total knee arthroplasty and THA, intravenous application of tranexamic acid can significantly reduce postoperative blood loss, blood transfusion volume and blood transfusion rate. Intra-articular application of tranexamic acid can reduce the drainage volume of drainage tube, total blood loss, blood transfusion volume and blood transfusion rate after joint arthroplasty. Moreover, hemoglobin levels remarkably increase after surgery. More and more studies and meta-analysis have verified that tranexamic acid has a "target effect". Reasonable use of tranexamic acid does not increase the risk of deep venous thrombosis or pulmonary embolism after hip and knee arthroplasties. Simultaneously, tranexamic acid has a good potency ratio.

Adverse events

1. To record adverse events during follow up, including incision pain, infection, hip pain, peripheral nerve injury, pulmonary embolism, lower extremity hematoma, deep vein thrombosis, and fixator falling off.
2. If severe adverse events occur, investigators would report details, including the date of occurrence and measures taken to treat the adverse events, to the principle investigator and the institutional review board within 24 hours.

Data collection, management, analysis, open access

1. Data collection: Case report forms were collected and processed using Epidata software (Epidata Association, Odense, Denmark). These data would be recorded electronically.
2. Data management: Tengzhou Central People's Hospital, China would preserve all of the data regarding this trial. Only the project manager has the right to query the database file. This arrangement will not be altered.
3. Data analysis: A professional statistician would statistically analyze the electronic database and create an outcome analysis report. An independent data monitoring committee would supervise and manage the trial data.
4. Data open access: Anonymized trial data will be published at www.figshare.com.

Statistical analysis

1. Statistical analysis would be performed using SPSS 19.0 software (IBM, Armonk, NY, USA) and would follow the intention-to-treat principle.
2. Normally distributed measurement data would be expressed as means ± standard deviation and minimums and maximums. Non-normally distributed measurement data would be expressed as the lower quartile (q1) and median and upper quartiles (q3). Count data would be expressed as a percentage.
3. The differences in hidden blood loss, dominant blood loss and mean blood loss in each group at intraoperative, postoperative 1 and 3 days would be compared using two-factor analysis of variance with repeated measures. Paired comparison of intergroup data would be conducted using least significant difference. Blood transfusion rate and incidence of adverse reactions in each group were compared using Pearson X2 test.
4. The significance level would be α = 0.05.

Inclusion Criteria:

• Femoral head necrosis or femoral neck fracture patients undergoing the first unilateral THA
• Bilateral hips with indications for THA in patients with femoral head necrosis, but after arthroplasty on one side, the arthroplasty on the other side will be conducted when choosing a good time and physical condition allows.
• Average age: 62.52 years
• Sex ratio of males to females: 11:19
• Signed informed consent

Exclusion Criteria:

• Coagulation disorders and anemia
• History of infection on the affected extremity
• History of vascular embolization and long-term oral anticoagulant drugs
• Contraindications for tranexamic acid or anticoagulant drugs

To recruit 90 patients undergoing the first unilateral THA at the Tengzhou Central People's Hospital from July 2015 to November 2016. The patients were equally and randomly assigned to three groups.