Tranexamic acid decreases external blood loss but not hidden blood loss in total knee replacement

BackgroundTotal knee arthroplasty (TKA) is often carried out using a tourniquet and shed blood is collected in drains. Tranexamic acid decreases the external blood loss. Some blood loss may be concealed, and the overall effect of tranexamic acid on the haemoglobin (Hb) balance is not known.MethodsPatients with osteoarthrosis had unilateral cemented TKA using spinal anaesthesia. In a double-blind fashion, they received either placebo (n=24) or tranexamic acid 10 mg kg−1 (n=27) i.v. just before tourniquet release and 3 h later. The decrease in circulating Hb on the fifth day after surgery, after correction for Hb transfused, was used to calculate the loss of Hb in grams. This value was then expressed as ml of blood loss.ResultsThe groups had similar characteristics. The median volume of drainage fluid after placebo was 845 (interquartile range 523–990) ml and after tranexamic acid was 385 (331–586) ml (P<0.001). Placebo patients received 2 (0–2) units and tranexamic acid patients 0 (0–0) units of packed red cells (P<0.001). The estimated blood loss was 1426 (1135–1977) ml and 1045 (792–1292) ml, respectively (P<0.001). The hidden loss of blood (calculated as loss minus drainage volume) was 618 (330–1347) ml and 524 (330–9620) ml, respectively (P=0.41). Two patients in each group developed deep vein thrombosis.ConclusionsTranexamic acid decreased total blood loss by nearly 30%, drainage volume by ∼50% and drastically reduced transfusion. However, concealed loss was only marginally influenced by tranexamic acid and was at least as large as the drainage volume.Br J Anaesth 2003; 90: 596–9 Total knee arthroplasty (TKA) is often carried out using a tourniquet and shed blood is collected in drains. Tranexamic acid decreases the external blood loss. Some blood loss may be concealed, and the overall effect of tranexamic acid on the haemoglobin (Hb) balance is not known. Patients with osteoarthrosis had unilateral cemented TKA using spinal anaesthesia. In a double-blind fashion, they received either placebo (n=24) or tranexamic acid 10 mg kg−1 (n=27) i.v. just before tourniquet release and 3 h later. The decrease in circulating Hb on the fifth day after surgery, after correction for Hb transfused, was used to calculate the loss of Hb in grams. This value was then expressed as ml of blood loss. The groups had similar characteristics. The median volume of drainage fluid after placebo was 845 (interquartile range 523–990) ml and after tranexamic acid was 385 (331–586) ml (P<0.001). Placebo patients received 2 (0–2) units and tranexamic acid patients 0 (0–0) units of packed red cells (P<0.001). The estimated blood loss was 1426 (1135–1977) ml and 1045 (792–1292) ml, respectively (P<0.001). The hidden loss of blood (calculated as loss minus drainage volume) was 618 (330–1347) ml and 524 (330–9620) ml, respectively (P=0.41). Two patients in each group developed deep vein thrombosis. Tranexamic acid decreased total blood loss by nearly 30%, drainage volume by ∼50% and drastically reduced transfusion. However, concealed loss was only marginally influenced by tranexamic acid and was at least as large as the drainage volume.