We studied 99 patients who were undergoing total knee arthroplasty (TKA) to determine the optimum protocol for the administration of tranexamic acid (TNA) in order to reduce blood loss. It decreased by more than 40% after the administration of TNA. The haemostatic effect was greatest when TNA was given preoperatively and on deflation of the tourniquet. There was no increase in the incidence of adverse affects in the patients receiving TNA, compared with a control group. We conclude that two injections of TNA, one given preoperatively and one on deflation of the tourniquet, significantly reduce blood loss without increasing the risk of thromboembolic complications.