The investigator's aim with this study is to address, with the new paradigm of tranexamic acid, the role of wound drainage following total knee arthroplasty in blood loss, blood transfusion requirements and functional recovery

All participants will attend a preoperatory appointment with an anaesthesiologist where preoperative haemoglobin will be optimized (Hg > 13 g/L).

Randomization will be done by opening a closed envelop with the selected group from a bag at the time of wound closure.

Collected preoperative information will include age, gender, weight, height, body-mass index, underlying diseases, operated side, haematocrit, haemoglobin level, midpatellar knee circumference and range of movement. Blood work was performed within 1 hour of the beginning of the surgery. Clinical parameters will be collected before anaesthesia.

In all cases, the surgeon will perform a medial parapatellar approach with a standard incision with the participants in the supine position. The SIGMA PFC total knee system (DePuy, Warsaw, In) and its posterior-stabilized design with a rotatory platform will be used. All participants will be under a subarachnoid spinal block.

A tourniquet will be used (pressure of 350 mmHg) from the beginning of the surgery and deflated for haemostasis before wound closure. At this time, before complete water-tight fascial closure, the patients will be allocated to group With or Without drain as described, and a drain will be placed in a subfascial position. The drain will be left clamped for 1 hour and will be left in place for 24 hours. Compressive dressings will be applied at the end of the surgery. Continuous passive motion will be started within 24 hours of surgery (0-60°, progressing as tolerated).

In all participants, about 30min before tourniquet release, intravenous tranexamic acid will be used with an initial bolus of 10 mg/kg administered in 10min followed by a perfusion of 10mg/kg over 4h. For postoperative analgesia, a femoral triangle and a popliteal block (anterior approach) will be done in all participants.

A standard fluid therapy of a balanced crystalloid solution (perfusion 10ml/kg/h) will be administered to all patients. After tourniquet release, during haemostasis and in the recovery room, an eventual need of increasing fluid therapy perfusion or a bolus will be noted in the patient's file.

Hidden blood loss will be calculated according to the haemoglobin balance method as it suggested to be the most reliable method. Total blood volume will be calculated as described by Nadler et al.

The normality of the data will be tested with the Shapiro-Wilk test/Kolmogorov-Smirnov test. The data will be compared and analyzed using Student's t-test for independent samples, or the Mann-Whitney test, according to the normality of the data. The correlation of data will be analyzed with the Pearson or Spearman correlation coefficient according to its normality. Nominal variables will be compared using the chi-square test. P-values less than 0.05 will be considered significant. Dedicated statistical software (SPSS version 21, SPSS Inc., Chicago, USA) will be used.

Inclusion Criteria:

• Consecutive patients admitted to perform a primary unilateral total knee arthroplasty

Exclusion Criteria:

• Contraindication to take tranexamic acid
• hematologic disease with a higher risk of bleeding
• abnormal coagulation tests
• refusal of blood transfusions
• inadequate preoperatory optimization
• deformity >15⁰ of varus or valgus