Background and Significance A growing amount of evidence linking transfusion of allogeneic blood products with negative patient outcomes and increased cost continues to drive interest into strategies and technologies that limit patient exposure to this risk. The single largest consumer of this resource continues to be cardiac surgery, with 20% of the world wide use of allogeneic blood products accounted for by this cohort. The lysine analogs tranexamic acid (TXA) has gained wide spread use in cardiac surgery as a blood-sparing agent. Mounted evidence has proved its efficacy and safety in cardiac surgery. However, the optimal dose regimen of TXA and the impact on patients' outcomes remains debated.

Study Objectives The primary objective of the study is to analyze the primary efficacy (superiority) and primary safety (non-inferiority) of the two dose regimen of tranexamic acid.. The primary efficacy endpoint includes perioperative allogeneic transfusion rate, and the primary safety endpoint includes the 30-day rate of the composite of perioperative renal dysfunction, myocardial infarction, ischaemic stroke, seizure, deep venous thrombosis, pulmonary embolism and all-cause mortality. The secondary objectives are to demonstrate the efficacy of the two dose regimens in reducing perioperative allogeneic transfusion volume, postoperative bleeding (chest tube drainage), reoperation rate, mechanic ventilation duration, ICU stay, hospital length of stay (LOS), and total hospitalization cost.

Study Endpoints The primary endpoints include efficacy and safety. The primary efficacy endpoint includes perioperative allogeneic transfusion rate, and the primary safety endpoint includes the 30-day rate of the composite of perioperative renal dysfunction, myocardial infarction, ischaemic stroke, seizure, deep venous thrombosis, pulmonary embolism, and all-cause mortality. The key secondary endpoints of the study are defined as perioperative allogeneic transfusion volume, postoperative bleeding (chest tube drainage), reoperation rate, mechanic ventilation duration, ICU stay, hospital length of stay (LOS), and total hospitalization cost.

Study Population Adult patients aged 18-70 years undergoing elective cardiac surgery with cardiopulmonary bypass are included. Totally 3008 patients will be required for this study (1504 in each of the 2 groups).

Study Design The study is a multicenter, randomised, double-blind trial. Cardiac surgery patients with cardiopulmonary bypass will be randomised to Dosage 1 regimen group or Dosage 2 regimen group of tranexamic acid.

Study Treatment The dosage regimen is implemented with dose of loading (intravenous infusion in 20 mins), maintenance (throughout the surgery), and pump prime (added into the bypass machine). The Dosage 2 regimen contains an intravenous bolus of 10 mg/kg after anesthetic induction followed by an intravenous maintenance of 2 mg/kg/h throughout the surgery, and a pump prime dose 1 mg/kg. As for the Dosage 1 regimen, the intravenous bolus and the maintenance are 30 mg/kg and 16 mg/kg/h respectively, and a pump prime dose 2 mg/kg. Patients, surgeons and research staff interviewing patients postoperatively will be blind to treatment allocation.

Statistical Considerations The study hypothesis is that the Dosage 1 regimen of tranexamic acid is superiority to the Dosage 2 regimen in the primary efficacy endpoint, while at the same time, the Dosage 1 regimen is non-inferiority to the Dosage 2 regimen in the primary safety endpoint in cardiac surgery with cardiopulmonary bypass. The sample size calculation is mainly based on the blood transfusion rate, and 30-day rate of the composite of perioperative renal dysfunction, myocardial infarction, ischaemic stroke, seizure, deep venous thrombosis, pulmonary embolism and all-cause mortality. For the primary efficacy endpoint, a sample size estimate of 1,214 randomized subjects (607 for each group) has 90% power to detect a 12.5% reduction (61.7% vs 70.5% between Dosage 1 regimen and Dosage 2 regimen ), by means of a single-sided α = 0.025 Chi-square test. For the primary safety endpoint, a sample size estimate of 2,698 randomized subjects (1349 for each group) has 90% power to detect a noninferiority margin for the difference of 5%, by means of a single-sided α = 0.025 log rank test. In order to conduct an interim analysis, the sample size in each group is 1504(10% drop-out rate) for the adjusted significance level (from 0.025 to 0.0245 in accordance with α spending function by Lan-DeMets Method). Finally, the investigators decided to enroll 3008 study patients (1:1 ratio) for the OPTIMAL trial.

A face to face visit (review in hospital, or remote video interview via smart phone and social media) is required to screen the occurrence of 30-day rate of the composite endpoints of renal dysfunction, myocardial infarction,stroke, seizure, deep venous thrombosis, pulmonary embolism and all-cause mortality, specific examinations are needed to confirm the diagnosis.

Inclusion Criteria:

1. Male or female adult patients aged 18~70 years.
2. Patients receiving cardiac surgery with cardiopulmonary bypass
3. Written Informed consent obtained

Exclusion Criteria:

1. Acquired chromatic disorder
2. Active intravascular coagulation
3. Previous convulsion or seizure
4. Allergy or contraindication to tranexamic acid injection or its components
5. Feeding or pregnancy women
6. Terminal illness with a life expectancy of less than 3 months
7. Patients with mental or legal disability
8. Currently enrolled in another perioperative interventional study

Participants, care provider, investigator and outcomes assessor will be blinded to treatment allocation.