1 Department of Orthopaedics and 2 Department of Anaesthesiology and Intensive Care, Faculty of Health Sciences, University of Linköping, S-581 85, Linköping, Sweden
Corresponding author. E-mail: bjorn.lisander@lio.se
Accepted for publication: January 14, 2003
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. 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 kg1 (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.
Results. The groups had similar characteristics. The median volume of drainage fluid after placebo was 845 (interquartile range 523990) ml and after tranexamic acid was 385 (331586) ml (P<0.001). Placebo patients received 2 (02) units and tranexamic acid patients 0 (00) units of packed red cells (P<0.001). The estimated blood loss was 1426 (11351977) ml and 1045 (7921292) ml, respectively (P<0.001). The hidden loss of blood (calculated as loss minus drainage volume) was 618 (3301347) ml and 524 (3309620) ml, respectively (P=0.41). Two patients in each group developed deep vein thrombosis.
Conclusions. 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.
Br J Anaesth 2003; 90: 5969
Keywords: blood, antifibrinolytics; blood, loss; surgery, arthroplasty
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In several studies, tranexamic acid decreased the blood loss associated with TKA.25 In those studies, tranexamic acid 1015 mg kg1 was given before the release of the tourniquet and in three,35 further doses of tranexamic acid were given. In one, the treatment began before application of the tourniquet.5 In these studies the total blood loss was reported as the loss during surgery plus the drainage volume. In general, tranexamic acid reduced the blood loss by about 50%. Since there may also have been hidden blood loss, the true effect of tranexamic acid on blood loss is not clear.
We set out to assess the total blood loss during TKA and how tranexamic acid influences it. The loss was determined from haemoglobin (Hb) balance.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Coded ampoules containing either tranexamic acid 100 mg ml1 (Cyklokapron®, Pharmacia) or saline were prepared by Apoteksbolaget, Umeå, Sweden. The contents of the ampoules were randomized in blocks of 10 (five saline, five tranexamic acid) by computer-generated numbers. At the end of the surgical procedure, just before release of the tourniquet, tranexamic acid 10 mg kg1 or placebo was infused i.v. (maximum dose 1000 mg). The dose was repeated after 3 h. Four randomized patients were excluded from the study before the code was broken. One in the tranexamic acid group suffered a large gastrointestinal bleed and one patient in the control group was given desmopressin because of extensive blood loss from the wound. Two randomized patients in the control group were found not to fulfil the criteria for inclusion: in one the serum creatinine was too great and the other had rheumathoid arthritis. Features of the remaining 51 patients are given in Table 1.
|
Anaesthetic and surgical procedures
Subarachnoid spinal anaesthesia was with isobaric bupivacaine (Marcain spinal®, Astra) 17.520 mg. Midazolam or propofol were given i.v. for sedation if needed. Non-invasive arterial pressure and heart rate were noted every 5 min and patients were given cloxacillin i.v. The patients underwent a standardized procedure performed by one of five surgeons, two of whom did 40 of the 51 operations. The surgeons were well balanced between groups (P=0.64, 2-test). After partial exsanguination of the limb by elevation for 1 min, a pneumatic tourniquet was inflated to 300 mm Hg. All patients received cemented prostheses with gentamycin (Nex-gen®, Zimmer, Scandinavia) and the lumen of the femur was plugged with autologous bone. The wound was closed and a compressive bandage was applied before release of the tourniquet. The joint was drained with a single closed suction drain until less than 50 ml was collected during a 6-h period [27 (SD 6) h and 27 (8) h in the placebo and tranexamic acid groups, respectively].
Volume substitution
Ringers acetate 500 ml was given i.v. before the subarachnoid injection. Volume replacement was with Ringers acetate and hydroxyethyl starch 200/0.5 (HAES-steril®, Fresenius) 10 mg ml1. After surgery, the venous Hb was measured when necessary in the ward using HemoCue® (HemoCue, Helsingborg, Sweden). At other times, venous Hb was determined with a modified cyano-methaemoglobin method (Celldyn 3500®, Abbott). If Hb was less than 90 g litre1, allogeneic leucodepleted red blood cell concentrate was given in 250-ml units containing about 150 ml cells and 1020 ml plasma.
Estimation of blood loss
Intraoperative blood loss was negligible in all patients. Blood loss after surgery was estimated by two different methods. The first was the standard clinical method where blood loss was taken as the volume recovered in drains. The second method was based on Hb balance. We assumed that blood volume (BV in ml) on the fifth day after surgery was the same as that before surgery. BV was estimated according to the method of Nadler and colleagues6 taking sex, body mass and height into account. The loss of Hb (in grams) was then estimated according to the formula:7
Hbloss=BVx(HbiHbe)x0.001+Hbt
where Hbloss (g) is the amount of Hb lost, Hbi (g litre1) the Hb concentration before surgery, Hbe (g litre1) is the Hb concentration on the fifth day after surgery and Hbt (g) is the total amount of allogeneic Hb transfused. A unit of banked blood was considered to contain 52 (SD 5.4) g Hb, according to measurements at our hospitals blood centre using the modified cyano-methaemoglobin method (Dr J Strindberg, personal communication). The blood loss (ml) was related to the patients preoperative Hb value (g litre1):
Blood loss =1000xHbloss/Hbi
The blood loss minus the drainage volume gave the hidden loss.
Statistical methods
The size of the study was calculated as follows. Total blood loss after TKA was 1410 (480) ml in control patients in a previous study.3 Blood loss of 400 ml was considered important. In patients with a Hb concentration of 135 g litre1, 400 ml blood would contain 54 g Hb, similar to the Hb content of one red cell unit (52 g). To obtain a power of 0.80 and an alpha value of 0.05, 23 patients would be required in each group. To allow for potential exclusions, and since we knew little about the magnitude of the hidden blood loss beforehand, 55 patients were included. Values are reported as mean (SD). However, Hb concentration, blood loss, the number of red cell units transfused and volume of fluids infused were not normally distributed and are therefore given as median (2575% range). Groups were compared using the MannWhitney U-test, except for categorical data, for which Fishers exact test was used for 2x2 tables and the 2-test for larger tables. Comparisons were made using StatView 4.51 on a Power Macintosh 7500 computer.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Eight patients in the control group and 12 in the tranexamic acid group were given ibuprofen. The data in these patients did not differ in any respect from those of the others.
Two patients in the control group and two in the tranexamic acid group had clinical symptoms of deep vein thrombosis, which was verified by ultrasound. One patient in the tranexamic acid group developed a wound infection.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In our control patients, the hidden loss of blood was of similar magnitude to that in the drains, supporting findings by Sehat and colleagues.1 In that study, as in ours, virtually all external blood loss was postoperative and minor bleeding from the ischaemic limb during the surgery was disregarded. In some studies,24 but not in others,1 5 the tourniquet was released before the wound was closed to allow for haemostasis, a strategy that would increase the blood loss during surgery. We expressed the drainage loss in ml, whereas the total blood loss was estimated as ml of the preoperative venous Hb concentration. The Hb concentration in the drainage fluid must have been less than the preoperative venous value, however, as the blood loss took place after some haemodilution was caused by the fluid given before and during the operation and later to compensate for the loss of blood volume. Also, Hb in the drainage fluid can be less than in simultaneous samples of venous blood.12 This strongly suggests that the hidden loss of Hb during TKA exceeds that in the drains. Therefore, it must be highly questionable to use the volume of drainage fluid as a measure of blood loss in TKA.
Plasmin binds to fibrinogen or fibrin structures and promotes fibrinolysis. Tranexamic acid competitively blocks a lysine-binding site of plasminogen and thereby inhibits its conversion to the active enzyme plasmin. Tranexamic acid is also a weak non-competitive inhibitor of the active enzyme. Tranexamic acid reduced drainage volume by about 50%, supporting previous findings.25 In TKA under regional anaesthesia, mental stress before the surgery activates both coagulation and fibrinolysis and this is enhanced during surgery.8 This activation is more pronounced in blood from the wound just after tourniquet release than in simultaneous venous samples. There is increased release of the tissue plasmin activator from the endothelium in the limb, by tissue trauma, thrombin13 and ischaemia,14 accelerating fibrinolysis. In TKA, tranexamic acid exerts its effects mainly in the wound.8 Tranexamic acid does not influence fibrinolytic activity in vein walls,15 which may explain why previous studies25 and the present study have not shown a higher incidence of deep vein thrombosis in patients treated with tranexamic acid.
Tranexamic acid reduced the estimated loss of blood by a third, which is less than the effect on drainage volume. The drug did not significantly reduce the hidden blood loss. The reasons for this are not clear. One explanation may be that the hidden loss, to a large extent, represents extravasation of red cells just after tourniquet release. At this time, the haemostasis may be by vascular and primary haemostatic mechanisms, before fibrinolysis has any effects. Haematoma in the joint and thigh could be limited by tissue pressure so that extravasation would not differ much between treatment groups.
Tranexamic acid effectively reduced the need for transfusion. Patients in the tranexamic acid group were given 7 units of blood in total, compared with 35 units in the control group. In our hospital the dose of tranexamic acid given would cost less than £7, compared with £46 for a unit of banked blood. Thus, the immediate saving in the patients given tranexamic acid would have been about £1100. To our knowledge, giving tranexamic acid is the only blood saving method that is cheaper, per saved unit, than banked blood in this type of surgery. This estimate does not include potential adverse effects from banked blood such as immediate transfusion reactions, transmission of infectious agents and disturbances of the immune system.16 17
We conclude that after TKA the hidden loss of blood is as large or larger than that in the drains. Tranexamic acid reduced the total loss of blood, but not as much as it decreased drainage volume.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Hiippala S, Strid L, Wennerstrand M, et al. Tranexamic acid (Cyklocapron) reduces perioperative blood loss associated with total knee arthroplasty. Br J Anaesth 1995; 74: 5347
3 Benoni G, Fredin H. Fibrinolytic inhibition with tranexamic acid reduces blood loss and blood transfusion after knee arthroplasty. J Bone Joint Surg Br 1996; 78: 43440[Medline]
4 Hiippala ST, Strid LJ, Wennerstrand MI, et al. Tranexamic acid radically reduces blood loss and transfusions associated with total knee arthroplasty. Anesth Analg 1997; 84: 83944[Abstract]
5 Jansen AJ, Andreica S, Claeys M, DHaese J, Camu F, Jochmans K. Use of tranexamic acid for an effective blood conservation strategy after total knee arthroplasty. Br J Anaesth 1999; 83: 596601
6 Nadler SB, Hidalgo JU, Bloch T. Prediction of blood volume in normal human adults. Surgery 1962; 51: 22432[ISI]
7 Lisander B, Ivarsson I, Jacobsson SA. Intraoperative autotransfusion is associated with modest reduction of allogeneic transfusion in prosthetic hip surgery. Acta Anaesthesiol Scand 1998; 42: 70712[ISI][Medline]
8 Benoni G, Lethagen S, Fredin H. The effect of tranexamic acid on local and plasma fibrinolysis during total knee arthroplasty. Thromb Res 1997; 85: 195206[CrossRef][ISI][Medline]
9 Johansson T, Lisander B, Ivarsson I. Mild hypothermia does not increase blood loss during hip arthroplasty. Acta Anaesthesiol Scand 2000; 43: 100510[CrossRef][ISI]
10 Benoni G, Fredin H, Knebel R, Nilsson P. Blood conservation with tranexamic acid in total hip arthroplasty: a randomized, double-blind study in 40 primary operations. Acta Orthop Scand 2001; 72: 4428[CrossRef][ISI][Medline]
11 Gaulrapp H, Eckstein S. Der Wert der Sonographie nach endoprothetischer Versorgung des Kniegelenks. Z Orthop Ihre Grenzgeb 2001; 139: 12733[CrossRef][ISI][Medline]
12 Harrap RS, Whyte GS, Farrugia A, Jones C. Some characteristics of blood shed into the Solcotrans postoperative orthopaedic drainage/reinfusion system. Med J Aust 1992; 157: 956[ISI][Medline]
13 Murphy WG, Davies MJ, Eduardo A. The haemostatic response to surgery and trauma. Br J Anaesth 1993; 70: 20513[ISI][Medline]
14 Fahmy NR, Patel DG. Hemostatic changes and postoperative deep-vein thrombosis associated with use of a pneumatic tourniquet. J Bone Joint Surg Am 1981; 63: 4615[Abstract]
15 Åstedt B, Liedholm P, Wingerup L. The effect of tranexamic acid on the fibrinolytic activity of vein walls. Ann Chir Gynaecol 1978; 67: 2035[ISI][Medline]
16 Nielsen HJ. Detrimental effects of perioperative blood transfusion. Br J Surg 1995; 82: 58287[ISI][Medline]
17 Landers DF, Hill GE, Wong KC, Fox IJ. Blood transfusion-induced immunomodulation. Anesth Analg 1996; 82: 187204[ISI][Medline]