Objective To investigate the effect and safety of intraarticular injection of tranexamic acid (TXA) at 2 concentrations on blood loss after total knee arthroplasty (TKA). Methods For this single blinded prospective randomised controlled trial, according to our inclusion and exclusion criteria we chose 24 patients diagnosed with degenerative joint disease who would receive bilateral TKA from October 2014 through July 2015. Comparisons were conducted between their bilateral knees. The knees were randomized into 2 groups to receive intraarticular injection of TXA at concentrations of either 15 g/L or 30 g/L before the tourniquet was released post-surgery. The TXA concentrations for each pair of knees were randomized by flipping a coin. The drainage, range of motion (ROM), visual analogue scale (VAS) and thigh circumference for each knee were recorded post-surgery. Blood routine examination and prothrombin time test were conducted for each patient at pre-surgery and 1, 3 and 5 days post-surgery. Vascular B-ultrasonography was performed for all the bilateral lower limbs at 5 days and one month post-surgery to assess incidence of thrombosis. Results The drainage for the 50 mL sides (65.4±33.4 mL) was significantly less than for the 100 mL sides (86.1±53.6 mL) (t=-3.873, P=0.001). There were no significant differences between the 50 mL and 100 mL sides regarding knee ROM (14.9°±3.1° versus 15.3°±3.3°), VAS (6.2±0.9 versus 6.0±1.0), or thigh circumference (46.8±3.0 cm versus 46.9±2.6 cm) (P>0.05). In this cohort, the post-surgery prothrombin time was significantly longer than the pre-surgery one (P<0.05). No deep infection, pulmonary embolism or deep vein thrombosis was observed in our trial. Conclusions Intraarticular injection of TXA can reduce blood loss after TKA and is safe. A higher concentration of TXA is more effective in decreasing post-surgery hemorrhage. Key words: Tranexamic acid; Arthroplasty, replacement, knee; Postoperative hemorrhage; Embolism and thrombosis