Total knee replacement (TKR) is performed with an occlusive tourniquet. As a result, TKR is associated with minimal intraoperative but extensive postoperative blood loss. Consequently, this surgical procedure is ideally suited to pharmacological manipulation of the fibrinolytic and procoagulant systems.
In previous studies designed to decrease blood loss after TKR, the administration of tranexamic acid (TA) has been associated with significant postoperative allogeneic blood-sparing when compared with control and desmopressin-treated patients (1–4). However, in these studies, TA was empirically administered via a continuous IV infusion. This complicates the treatment protocol and requires special organization in the postanesthesia care unit (PACU) and surgical department. Furthermore, the ideal dosage, route, and duration for TA administration are unknown. Therefore, the efficacy of alternative, simpler treatment protocols must be evaluated. Because TA is available in the oral form and undergoes rapid and complete absorption 5,6), we postulated that this preparation may be associated with blood-sparing similar to that achieved with IV TA. Furthermore, after TKR, the imbalance between the fibrinolytic and procoagulant systems is transient. Therefore, it is possible that TA administration for a shorter period may result in blood-sparing similar to that demonstrated with a 12-h administration. On the basis of this hypothesis, we performed a prospective, controlled, randomized, single-blinded study designed to assess the blood-sparing efficacy of TA when administered orally or via a variable IV infusion.
Eighty patients (ASA physical status I–III) undergoing elective TKR were enrolled in this IRB-approved, randomized, prospective, controlled, single-blinded study. Written, informed consent was obtained in all cases. Patients with a history of severe ischemic heart disease (New York Heart Association Class III and IV), chronic renal failure, cirrhosis, bleeding disorders, or current anticoagulant therapy were excluded from the study.
Upon arrival in the operating room holding area, monitoring equipment was applied. Throughout the perioperative period, the following variables were recorded at 1- to 5-min intervals: noninvasive blood pressure, electrocardiogram, and peripheral hemoglobin oxygen saturation.
According to a computer-generated randomization table, patients were allocated to one of four treatment groups. In group TA-long, 30 min before the limb tourniquet was deflated, an IV bolus dose of TA (TEVA™; Biogal Pharmaceutical Works Ltd., Debrecen, Hungary) 15 mg/kg was administered over 30 min. Thereafter, a constant IV infusion of 10 mg · kg−1 · h−1 was administered until 12 h after final deflation of the limb tourniquet. In group TA-short, 30 min before deflation of the limb tourniquet, an IV bolus dose of TA 15 mg/kg was administered over 30 min, followed by a constant IV infusion of 10 mg · kg−1 · h−1 until 2 h after final deflation of the limb tourniquet (time of discharge from the PACU). Thereafter, oral TA 1 g was administered after 6 and 12 h. In group TA-oral, 60 min before surgery, an oral dose of TA 1 g was administered. After surgery, the same dose of TA was administered every 6 h for the next 18 h. In the control group, TA was not administered.
A standardized general anesthetic was administered. This included IV thiopental 4–6 mg/kg and fentanyl 5 μg/kg for the induction of anesthesia. Thereafter, isoflurane (0.5%–1.5%; end-tidal) and 70% nitrous oxide in oxygen was administered for maintenance of anesthesia. Neuromuscular blockade was induced with IV vecuronium, and the trachea was intubated. After the induction of anesthesia, the end-tidal carbon dioxide concentration was monitored, and the urinary bladder was catheterized.
Before surgical incision, the operative limb was isolated by inflation of an occlusive limb tourniquet. After placement of the prosthesis, the tourniquet was released and hemostasis was performed.
Intraoperative fluids were administered to maintain the heart rate and mean arterial blood pressure within 20% of baseline values and the urine output of at least 2 mL · kg−1 · h−1. Allogeneic blood was transfused to maintain a hematocrit >28%.
One day before surgery and for the duration of the hospital admission, subcutaneous enoxaparin 40 mg/d was administered for deep vein thrombosis (DVT) prophylaxis. Blood loss in the surgical drain was measured 12 and 24 h after surgery. Thereafter, to decrease possible infection, the drain was removed.
Allogeneic blood transfusion was verified by an independent observer (ME), who was blinded to the treatment modality. In all cases, a hematocrit <28% constituted the postoperative transfusion trigger.
Before surgery, immediately after surgery, at 2 and 12 h after surgery, and once daily for eight postoperative days, hematocrit was measured. In addition, hemoglobin, prothrombin time, partial thromboplastin time, platelet count, and fibrinogen were measured before surgery, 12 h after surgery, and once daily for eight postoperative days. In the event that patients were discharged before completing the required number of blood tests, these were performed on an outpatient basis.
In the PACU, patients were connected to a patient-controlled analgesia device programmed to deliver 1-mg bolus doses of morphine with a 10-min lockout interval and no basal infusion.
All patients were examined daily (by an independent observer) for signs of lower limb DVT (swelling or increase in calf diameter) and underwent lower limb Doppler ultrasound imaging on the fifth postoperative day. Three months after hospital discharge, all patients were interviewed by a blinded researcher. During this interview, the incidence of DVT, pulmonary embolus, myocardial infarction, transient ischemic attack, and stroke was recorded.
Prestudy power analysis determined a sample size of 20 patients per group to have a 90% chance (β = 0.01) for detecting a difference in postoperative blood administration of 1 U at 95% confidence interval limitations (∝ = 0.05).
In all cases, normality was assessed with the Kolmogorov-Smirnov test. Depending on the results of the Kolmogorov-Smirnov analysis, parametric or nonparametric analysis was performed. Analysis of variance with Bonferroni’s test for selected pairs of columns was used to analyze demographic and anesthetic data, perioperative blood accumulation, and blood administration, as well as laboratory investigations. The number of patients receiving allogeneic blood was analyzed with the χ2 test. Data are expressed as numbers or mean values ± sd or sem. In all comparisons, P < 0.05 was considered statistically significant.
The four study groups were comparable in age, weight, height, sex, and ASA physical status ((Table 1). The anesthetic time, surgical time, tourniquet time, and hospital admission time were similar among groups (Table 2). In addition, the perioperative arterial blood pressures and heart rates were similar among groups.
After the first 12 postoperative hours, blood accumulation in the surgical drain was significantly (P < 0.02) more in group Control and group TA-oral when compared with group TA-long and group TA-short, respectively. However, at the end of the subsequent 12 h, blood accumulation in the surgical drain was significantly (P < 0.05) larger in group Control when compared with group TA-long, group TA-short, and group TA-oral (Table 3). Significantly more blood transfusions were administered in group Control when compared with each of the three TA treatment groups (Table 3). The timing of the allogeneic blood transfusion is presented in Table 4.
Baseline preoperative hematocrit values were comparable among groups. In group Control, despite the significantly larger allogeneic blood transfusion, the postoperative hematocrit recordings were smaller than those among the three TA treatment groups. A significant difference in hematocrit was demonstrated on Day 1 (group Control versus group TA-long, group TA-short, and group TA-oral) and on Days 3 and 4 (group Control versus group TA-long and group TA-short) (Fig. 1). At the 3-mo follow-up, no postoperative thromboembolic events were documented.
The results of this prospective, randomized study demonstrate that, for TKR, TA is associated with allogeneic blood sparing at different dosages and routes of administration. This is supported by the fact that postoperative allogeneic blood administration was significantly more frequent in group Control when compared with each of the three TA treatment groups. Furthermore, even though TA-Long was associated with the slowest transfusion rates, both TA-short and TA-oral significantly reduced allogeneic blood requirements when compared with the TA-control group.
After the first 12 postoperative hours, the amount of blood in the surgical drain was similar in the TA-oral and TA-control groups. However, at the end of the subsequent 12 hours, blood accumulation in the surgical drain was significantly larger in group Control when compared with group TA-oral, group TA-long, and group TA-short. We suggest that the fact that group TA-oral and group Control were associated with a similar volume of blood in the surgical drain during the first 12 hours is likely due to an imbalance between fibrinolysis and TA-induced fibrinolysis inhibition. Because tourniquet release activates the fibrinolytic pathway, we hypothesize that maximal fibrinolysis occurs at this time. However, the maximal antifibrinolytic effect of oral TA occurs approximately two hours after ingestion (5,6). Therefore, perhaps at the time of maximal fibrinolysis, the blood concentration of TA was insufficient to adequately counter the fibrinolytic load. Consequently, TA-oral did not reduce postoperative bleeding at the conclusion of the first 12 postoperative hours, and at this time the amount of blood in the surgical drain was similar between group TA-oral and group TA-control. By contrast, during the second 12 hours, the extent of fibrinolysis decreased in a time-dependent manner. However, because of repeated oral drug administration, the blood concentration of TA increased, and a balance was achieved between the antifibrinolytic effect of TA and tourniquet-related fibrinolysis. As a result, blood accumulation in the surgical drain at the end of the second 12-hour collection period was similar among the three TA treatment groups and was significantly larger in group Control.
During the postoperative period, control patients were associated with consistently lower hematocrit recordings when compared with the three TA treatment groups. Furthermore, during the late recovery period, patients in group Control required significantly more blood to maintain the predefined target hematocrit. Because the decision to administer blood was made by an investigator blinded to the treatment modality and because hematocrit was the only transfusion trigger, we believe that our results are a true reflection of the blood-sparing properties of both oral and IV TA administration.
The transition from an experimental technique to routine clinical practice is dependent on numerous factors. First, the experimental entity must be associated with desired positive results. Second, the technique must be devoid of unacceptable adverse effects. Third, the new intervention must be simple, easy to perform, not labor intensive, and cost-effective. Since our results support the findings of previous investigations that demonstrated the allogeneic blood-sparing efficacy of TA after TKR (1–4), we suggest that pharmacological manipulation of the fibrinolytic and procoagulant systems is appropriate after this surgical procedure. In addition, despite the theoretical possibility that antifibrinolytic drug administration could increase the incidence of unwanted thromboembolic events, neither our results nor those of other investigators support such concerns (2,7,8). Finally, our study has demonstrated that oral TA, in the dose administered, is associated with allogeneic blood sparing similar to that induced by IV drug administration. However, the oral route has the added advantage of not requiring dedicated IV access, electronic drug-delivery systems, or expensive nursing care. Therefore, we suggest that oral TA may be routinely administered to patients undergoing TKR.
This study may be criticized in that equipotent doses of TA were not administered. As a result, it is possible that serum TA blood levels were not similar among groups. However, the aim of the study was to assess the allogeneic blood-sparing efficacy of TA when administered in a user-friendly manner. Therefore, in group TA-short, we limited the IV drug administration to the standard PACU admission time (two hours) for patients recovering from TKR. Furthermore, considering the favorable absorption profile of TA, we included oral TA in our comparison. Because oral TA has not been administered after TKR, our dosage regimen was based on the standard oral dosage of TA when administered for the management of bleeding due to thrombocytopenia, menorrhagia, and hematuria of prostate or bladder origin (9).
We conclude that TA is associated with significant allogeneic blood sparing when compared with control. Furthermore, no significant difference in blood sparing was demonstrated among TA treatment groups. Because oral drug administration is simple and does not require specific infusion equipment, we suggest that oral TA is more advantageous than IV drug administration.
1. Hiippala S, Strid LJ, Wennerstrand MI, et al. Tranexamic acid radically decreases blood loss and transfusions associated with total knee arthroplasty. Anesth Analg 1997;84:839–44.
2. Zohar E, Fredman B, Ellis MH, et al. A comparative study of the postoperative allogeneic blood-sparing effects of tranexamic acid and of desmopressin after total knee replacement. Transfusion 2001;41:1285–9.
3. Ellis MH, Fredman B, Zohar E, et al. The effect of tourniquet application, tranexamic acid and desmopressin on the procoagulant and fibrinolytic systems during total knee replacement. J Clin Anesth 2001;13:509–13.
4. Good L, Peterson E, Lisander B. Tranexamic acid decreases external blood loss but not hidden blood loss in total knee replacement. Br J Anaesth 2003;90:596–9.
5. McNicol GP, Fletcher AP, Alkjaersig N. The absorption, distribution and excretion of epsilon-aminocaproic acid following oral and iv administration to man. J Lab Clin Med 1962;59:15–22.
6. Hedner U, Hirsh J, Mader VJ. Therapy with antifibrinolytic agents. In: Colman RW, Hirsh J, Mader VJ, eds. Hemostasis and thrombosis: basic principles and clinical practice. 4th ed. Philadelphia: Lippincott, Williams & Wilkins, 2001:798.
7. Zohar E, Fredman B, Ellis M, et al. A comparative study of the postoperative allogeneic blood-sparing effect of tranexamic acid versus acute normovolemic hemodilution following total knee replacement. Anesth Analg 1999;89:1382–7.
8. Ho KM, Ismail H. Use of intravenous tranexamic acid to reduce allogeneic blood transfusion in total hip and knee arthroplasty: a meta-analysis. Anaesth Intensive Care 2003;31:529–37.
© 2004 International Anesthesia Research Society
9. Bonnar J, Sheppard BL. Treatment of menorrhagia during menstruation: randomized controlled trial of ethamsylate, mefenamic acid, and tranexamic acid. BMJ 1996;313:579–82.