This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Engel, J. M. Articles by Hempelmann, G. Search for Related Content PubMed PubMed Citation Articles by Engel, J. M. Articles by Hempelmann, G. Related Collections Blood Surgery

---

GENERAL ARTICLES

Regional Hemostatic Status and Blood Requirements After Total Knee Arthroplasty With and Without Tranexamic Acid or Aprotinin

Department of Anesthesiology and Intensive Care Medicine, and *Department of Orthopedic Surgery, Justus-Liebig-University, Giessen, Germany

Address correspondence and reprint requests to Dr. med. Jörg M. Engel, Department of Anaesthesiology and Intensive Care Medicine, Justus-Liebig-University, Rudolf-Buchheim-Str. 7, D-35385 Giessen, Germany. Address e-mail to Joerg.Engel{at}chiru.med .uni-giessen.de.

	Abstract

Top Abstract Introduction Patients and Methods Results Discussion References

 
Antifibrinolytics seem to reduce postoperative blood loss after total knee arthroplasty. Few studies have shown the impact of these drugs on the mechanisms of coagulation. The purpose of this study was to examine coagulation/fibrinolysis variables as well as blood loss after total knee arthroplasty with and without antifibrinolytics in the operated limb on a regional level. Thirty-six patients were randomized into one of three groups to receive aprotinin, tranexamic acid, or no medication. We took blood samples of the femoral vein before deflating the tourniquet and after 5, 10, 30, 60, 120 min and on the first postoperative day. The implantation of a knee prosthesis in artificial ischemia caused a significant activation of coagulation and fibrinolysis in the regional circulation. Tranexamic acid and aprotinin did not cause a significant modulation of fibrinolysis variables or a significant reduction of postoperative bleeding and transfusion requirements. One of the differences in comparison to other studies was the decreased total blood loss. The use of bone cement as well as surgical hemostasis before wound closure may be regarded as reasons for this. Therefore, primarily these methods should be used because there is no increased risk of adverse drug effects.

Implications: After total knee arthroplasty total blood loss may be kept in a low range if methods such as cemented knee prosthesis and surgical hemostasis are used. In this case aprotinin and tranexamic acid did not cause a significant modulation of fibrinolysis variables or a significant reduction of postoperative bleeding.

	Introduction

Top Abstract Introduction Patients and Methods Results Discussion References

 
Nowadays, the implantation of a knee prosthesis is performed in most cases after applying a pneumatic tourniquet. The decisive advantage of this procedure is the improvement of the surgeon’s intraoperative working conditions because of better visibility of the site of surgery, which makes the insertion of the implant easier. This stage is accompanied by negligible blood loss. The postoperative blood loss may, however, reach such dimensions that it becomes necessary to administer homologous blood transfusions with all associated risks (1–5). Results of various studies have shown that postoperative bleeding increased after artificial ischemia of the operated limb. This was attributed to an additional activation of the fibrinolytic system (6–8).

On the basis of this hypothesis, different working groups were administered antifibrinolytics with the result of reducing postoperative blood loss and the number of required blood transfusion units (2–4,9–13). Few studies have shown the impact on the mechanisms of coagulation. On one hand, a reduction of fibrinolytic activity in the systemic circulation was observed, when measuring the D-dimers. On the other hand, other fibrinolytic variables indicated an increase (11,12). There was no correlation between variables of fibrinolysis and blood loss.

Because a dilution effect in the systemic circulation always complicates the interpretation of coagulation and fibrinolysis variables, we decided to examine the effects of knee prosthesis implantation in the operated limb on a regional level. In addition, the effects of doses of tranexamic acid and aprotinin were tested. Blood loss and frequency of transfusions were recorded. Because the investigators recommended the administration of antifibrinolytics we expected to show a significant decrease in blood loss in combination with a modulation of the fibrinolytic activity.

	Patients and Methods

Top Abstract Introduction Patients and Methods Results Discussion References

 
The Ethics Committee approved the study. All patients gave their written, informed consent. Preoperative measures included an examination of global coagulation variables (activated partial thromboplastin time, prothrombin time, fibrinogen, bleeding time) and platelet count for deviations from the standard. Patients were instructed to stop taking medications containing acetylsalicylic acid 1 wk before surgery. All patients received low molecular weight heparin (Certoparin 3000 IU anti-Xa s.c.; Novartis Pharma, Nürnberg, Germany) to prevent thromboembolic complications. The treatment started on the evening before surgery and lasted until the patient was fully mobilized or discharged.

A combined spinal epidural anesthesia was chosen as the anesthetic method. Intraoperative analgesia was obtained by spinal and epidural administration of isobaric 0.5% bupivacaine. Postoperative analgesia started in the recovery room using the epidural catheter and included a mixture of morphine and dehydrobenzperidol. This medication was used until the patient was mobilized on the third postoperative day.

A standard surgery team performed the surgical interventions. After the extremity had been drained of blood by applying a rubber bandage, a pneumatic tourniquet was put around the upper thigh and inflated to a pressure of 420 mm Hg. It is the routine tourniquet pressure chosen by the orthopedic staff because in 99% of all cases this pressure enabled a bloodless field. A midline skin and parapatellar capsular incision was made to expose the knee joint. A knee prosthesis (Duracon; Howmedica Osteonics, Rutherford, NJ) of an appropriate size was inserted and fixed with bone cement. Before deflating the pneumatic tourniquet, the joint was stuffed with surgical gauzes, and an elastic bandage was applied around the knee for 5 min. After removal of the surgical gauzes, electrical cauterization and/or ligature were applied to obtain hemostasis. One intraarticular and one extrafascial wound drain were inserted and connected to a vacuum system after wound closure.

Lactated Ringer’s solution was infused for basic volume replacement. When there were signs of inadequate intravascular volume such as a decrease of central venous pressure, 10% hydroxyethyl starch solution 200 up to a dose of 1.0 mL · kg^-1 · day^-1 were used as replacement fluid. If the hemoglobin concentration decreased to <10 g/dL, blood transfusions were administered. The intraoperative and postoperative blood loss was determined by measuring the quantity in the suction and/or drainage reservoir, and a qualitative description of the bloodstained surgical swabs and/or the bandage. The number of units of transfused red blood cell concentrates was recorded. Patients with clinical symptoms of deep venous thrombosis underwent diagnostic venography. Thromboembolic complications were assessed using the definition of Baker and Bick (14).

Thirty-six patients were randomized into one of the three groups. The first group received 1 million KIE aprotinin immediately before deflating the tourniquet, followed by an infusion of 500,000 KIE per hour for 4 h. The second group received 15 mg/kg tranexamic acid, followed by a repeated dose of 10 mg/kg after 3 h. The third group did not receive any medication and served as a control group. For reasons of comparability, we followed the recommendations of other studies regarding the doses of antifibrinolytics and the number of patients (2,4,8). A power analysis after an evaluation of the first nine patients had resulted in an extremely large number of patients (n = 939) for each group.

We took blood samples of the downflowing vein of the operated leg to obtain a homogeneous impression of the coagulation and fibrinolysis activities in the operated leg. For this purpose, a catheter was inserted preoperatively in the femoral vein of each patient. The catheter was inserted into the vessel to a maximum of 3–5 cm, so that the blood sample was taken from the distal circulation and there was no admixture with blood from visceral iliac vessels. A simultaneous pH measurement demonstrated the right position of the catheter. For the first five patients, this was confirmed by radiographic evaluation. Because the decisive changes occur in the first 24 h (12), blood samples were taken immediately before deflating the tourniquet and after 5, 10, 30, 60, and 120 min and on the first postoperative day. Five minutes after tourniquet release is an adequate time point to register maximal changes, as the washout process needs some time.

Thrombin-antithrombin III (TAT) was measured by using an enzyme-linked immunosorbent assay (ELISA) (Enzygnost-TAT; Behring Werke AG, Marburg, Germany). D-dimers and plasminogen activator inhibitor were also quantified by using ELISA (Asserachrom D-dimer, Boehringer Mannheim AG, Mannheim, Germany; Imubined^® PAI-I-ELISA, American Diagnostica Inc., Greenwich, CT). The tissue plasminogen activator concentration was measured by using Imubined^® total tPA-ELISA (American Diagnostica). Factor XII and -2-antiplasmin were determined by using chromogen substrates (Berichrom -2-antiplasmin, Behringwerke AG, Marburg, Germany). The plasmin-antiplasmin complex (PAP) was determined by using an ELISA (Enzygnost-PAP micro; Behringwerke AG, Marburg, Germany). All measurements were performed twice and the mean values of the results taken.

Hemoglobin concentration, platelet count, prothrombin time, thromboplastin time, and thrombin time were measured on the day before the operation, in the recovery room, and on the postoperative days 1, 2, 4, and 6 applying routine methods of the clinical laboratory.

All data are presented as means ± SD or standard error of the mean (SEM). Normality was assessed by applying the Kolmogorov-Smirnov test. The statistical evaluation was performed by analysis of variance or the Kruskal-Wallis test followed by post hoc Scheffe-test or Mann-Whitney U-test for interpretation of hemostatic markers. Descriptive variables were analyzed by means of ² test. A value of P smaller than 0.05 was considered statistically significant.

	Results

Top Abstract Introduction Patients and Methods Results Discussion References

 
Patient characteristics, the duration of tourniquet application, and the surgery were comparable for all groups ( Table 1).

View larger version (15K): [in this window] [in a new window]   Figure 1. Plasmin-antiplasmin complex (PAP): Blood samples were taken before deflation of the tourniquet (1), and 5 (2), 10 (3), 30 (4), 60 (5), 120 min (6) after deflation, and on the morning of the first postoperative day (7) (mean ± SEM, * P < 0.05 control vs other groups).

View larger version (16K): [in this window] [in a new window]   Figure 2. -2-antiplasmin: Blood samples were taken before deflation of the tourniquet (1), and 5 (2), 10 (3), 30 (4), 60 (5), 120 min (6) after deflation, and on the morning of the first postoperative day (7) (mean ± SEM, * P < 0.05 aprotinin vs other groups).

The intraoperative blood loss was small in all groups with a mean of 150 (range 100–500) mL. Postoperative bleeding of patients in all groups was comparably strong; the total blood loss amounted to 810 (range 245–1370) mL ( Table 3). There were no significant differences between the control group, the Tranexamic Acid group and the Aprotinin group regarding the administered amounts of crystalloids, colloids, and transfused blood units (Table 3). Of all patients, 87.5% who needed blood transfusions had a baseline value of hemoglobin of <13 g/dL. With a hemoglobin value greater than 13.9 g/dL none of the patients required a blood transfusion.

	Discussion

Top Abstract Introduction Patients and Methods Results Discussion References

 
The implantation of a knee prosthesis in artificial ischemia caused a significant activation of coagulation and fibrinolysis, as was shown in the courses of TAT, PAP, and D-dimers on a regional level. Tranexamic acid and aprotinin did not bring about a decisive modulation of coagulation and fibrinolysis variables in the regional circulation. Furthermore, there was no significant reduction of postoperative blood loss and transfusion requirements. Tranexamic acid and aprotinin are used as antifibrinolytics because they interact (15) with the activation and function of plasmin, the central connecting link to fibrinolysis. Tranexamic acid forms reversible complexes with the lysine-binding sites of plasminogen, and thus inhibits the activation of plasminogen. In large doses, plasmin is also inhibited noncompetitively. Aprotinin is a serine protease inhibitor and affects fibrinolysis by various mechanisms (16). In addition to the direct effect on plasmin, inhibitory effects on the Factor XII kallikrein system and the release of tissue factor were also observed (17,18).

We discovered only discrete effects of antifibrinolytics on the examined fibrinolysis markers in the regional circulation. Plasmin activation, as demonstrated by the PAP formation, was only influenced insignificantly. An initial inhibitory effect does, however, seem possible, as the increased levels on the first postoperative day indicate a reactively increased formation at continuing fibrinolysis, as was observed in all groups. This increase was not accompanied by increased bleeding.

-2-antiplasmin increased after administering aprotinin. -2-antiplasmin is a major inhibitor of the fibrinolytic system. An increase signifies that this variable is released to an increased extent, but not consumed afterward. Because fibrinolysis did not increase clinically when administering aprotinin or on the level of factors, it must be assumed that the inhibitory function of -2-antiplasmin was fulfilled by aprotinin. This observation has also been made in other studies (18,19).

Effects of the postoperative analgesia modality on coagulation/fibrinolysis system, postoperative blood loss and blood transfusion requirements are possible, because the incidence of deep venous thrombosis is less with extradural anesthesia than with general anesthesia (20). However, this may have only a slight influence on our results because the postoperative analgesia regime consisted of epidural anesthesia in all groups. Sharrock et al. (5) found no significant differences in either thrombin generation or fibrinolytic activity during total knee arthroplasty with tourniquet after extradural and balanced general anesthesia, respectively. Thus a difference between general and epidural anesthesia may have a greater impact on coagulation/fibrinolysis system than the weaker effect of postoperative analgesia regime.

With regard to fibrinolysis variables the results of other studies are inconclusive. After administering tranexamic acid Jansen et al. (12) observed a smaller increase of D-dimers, indicating an inhibitory effect on fibrinolysis. However, a decrease of plasminogen was also observed, which is in accordance with an increased transformation to plasmin and thus with an increased fibrinolysis. Other variables such as fibrinogen remained uninfluenced. As a whole, there was no direct correlation between blood loss and variables of fibrinolysis.

We refrained from determining the coagulation and fibrinolysis variables in the wound blood, as described by Benoni et al. (11), as this did not guarantee a homogeneous taking of samples. The taking of samples from the central venous site of bloodflow reflects a much more representative picture.

Despite inconsistent findings regarding fibrinolysis variables and in contradiction to our results, a number of studies reported on a reduced postoperative blood loss after administering antifibrinolytics in the course of total knee arthroplasty. The majority of investigating scientists used tranexamic acid; a smaller group used aprotinin. Benoni et al. (9) recorded a reduction of blood loss by 340 mL and a smaller number of required blood transfusion units in patients who had received 10 mg/kg tranexamic acid before deflating the tourniquet. By administering another dose of tranexamic acid 3 h later, it was possible to reduce the blood loss by 680 mL and the transfusion requirements by 30% (10). Hiippala et al. (2) administered tranexamic acid 15 mg/kg and observed a reduction of blood loss from 1509 ± 643 mL to 689 ± 289 mL and a reduction of blood transfusion requirements by 50%, which was confirmed in another study (3). Jansen et al. (12) also recorded a significantly decreased blood loss and associated decreased blood transfusion requirements, when administering tranexamic acid. Thorpe et al. (4) observed a reduction of blood loss and transfusion requirements by using aprotinin, although the results had not acquired statistical significance.

A comparison of the total blood loss in our control group with that of the above-mentioned studies shows that the patients in our study bled significantly less. With an average of 847 ± 418 mL they were on a level that was only reached by administering antifibrinolytics in the above-mentioned studies. The transfusion rate of 0 (range 0–2) units in our Control group was also comparable to the Tranexamic Acid group in Hiippala’s study (1.5 ± 1.3 U) (2). The use of bone cement is an important clinical factor to influence blood loss after total knee arthroplasty in a positive way (1,21–23). In our study all implanted components were cemented tibially and femorally. In Hiippala’s studies the majority of implants were not fixed with cement. This may explain that the blood loss in our study was at the lower end of the observed values of 650–1500 mL (21–24). This cannot be the only explanation, however. In the study by Jansen et al. (12) all prosthetic components were fixed with bone cement as in our study, and the total blood loss in the Control group still amounted to 1419 ± 607 mL. Unlike patients in the majority of the other studies, Jansen’s patients received general anesthesia. This should not be the reason for the increased blood loss, however, because the fibrinolytic activity in cases of total knee arthroplasty was observed to be comparable under spinal-epidural anesthesia and under general anesthesia (5). A further difference was the fact that after implantation of the prosthesis no surgical hemostasis was performed after deflating the tourniquet. Small, diffuse hemorrhages may be stopped by applying a tight bandage. Bigger, opened blood vessels, however, may bleed more by using this method.

In our study the baseline hemoglobin value had essential impact on the blood transfusion requirements. Of the patients who received blood transfusions 87.5% had a baseline hemoglobin value of <13 g/dL. None of the patients with a hemoglobin value higher than 13.9 g/dL received a blood transfusion. Other authors also regarded a baseline hemoglobin value of 13 g/dL as borderline value for blood transfusion requirements in total knee arthroplasty (25).

The study by Thorpe et al. (4) showed that the use of antifibrinolytics in knee prosthesis surgery should be well reasoned. A patient lost his leg as a result of massive thrombosis. It remained unclear, however, whether aprotinin treatment or a peripheral occlusive arteriovenous disease had caused this complication. Even in times of heparin prophylaxis, thromboses constitute a major risk for orthopedic patients. There are continuing discussions about whether the administration of antifibrinolytics increase thromboembolic incidence. In our study, the rate of venous thromboses was comparable in all groups. The incidence was 8% compared with the level of other studies (10). The hemostatic markers did not give evidence of increased coagulation after administration of antifibrinolytics.

In summary, the implantation of a knee prosthesis in ischemic conditions induces a significant activation of coagulation and fibrinolysis on a regional level. Tranexamic acid and aprotinin do not cause a significant modulation of fibrinolysis or a reduction of postoperative blood loss. One of the differences in comparison to other studies was the significantly decreased blood loss in our study. The use of bone cement and surgical hemostasis before wound closure may be regarded as the reasons for this decreased blood loss. If total blood loss may be kept lower than 850 mL, the use of tranexamic acid and aprotinin does not seem to be very efficient. The conclusion of our study is to primarily use alternatives such as cemented knee prosthesis and surgical hemostasis to reduce blood loss because these methods do not increase the risk of adverse drug effects. The results should be verified in a larger group of patients.

	References

Top Abstract Introduction Patients and Methods Results Discussion References

 

1. Cushner FD, Friedman RJ. Blood loss in total knee arthroplasty. Clin Orthop 1991; 269: 98–101.
2. Hiippala ST, Strid LJ, Wennerstrand MI. Tranexamic acid (Cyklokapron®) reduces perioperative blood loss associated with total knee arthroplasty. Br J Anaesth 1995; 74: 534–7.[Abstract/Free Full Text]
3. Hiippala ST, 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.[Abstract]
4. Thorpe CM, Murphy WG, Logan M. Use of aprotinin in knee replacement surgery. Br J Anaesth 1994; 73: 408–10.[Abstract/Free Full Text]
5. Sharrock NE, Go G, Williams-Russo P, et al. Comparison of extradural and general anaesthesia on the fibrinolytic response to total knee arthroplasty. Br J Anaesth 1997; 79: 29–34.[Abstract/Free Full Text]
6. Nakara M, Sakahashi H. Effect of application of a tourniquet on bleeding factors in dogs. J Bone Joint Surg Am 1967; 49: 1345–51.[Abstract/Free Full Text]
7. Klenerman L, Mackie I, Chakrabarti R. Changes in haemostatic system after application of a tourniquet. Lancet 1977; 1: 970–2.[Web of Science][Medline]
8. Petäja J, Myllynen P, Myllylä G, Vahtera E. Fibrinolysis after application of a pneumatic tourniquet. Acta Chir Scand 1987; 153: 647–51.[Web of Science][Medline]
9. Benoni G, Carlsson A, Petersson C, Fredin H. Does tranexamic acid reduce blood loss in knee arthroplasty? Am J Knee Surg 1995; 8: 88–92.[Medline]
10. Benoni G, Fredin H. Fibrinolytic inhibition with tranexamic acid reduces blood loss and blood transfusion after knee arthroplasty. J Bone Joint Surg Am 1996; 78: 434–40.
11. Benoni G, Lethagen S, Fredin H. The effect of tranexamic acid on local and plasma fibrinolysis during total knee arthroplasty. Thromb Res 1997; 85: 195–206.[Web of Science][Medline]
12. Jansen AJ, Andreica S, Claeys M, et al. Use of tranexamic acid for an effective blood conservation strategy after total knee arthroplasty. Br J Anaesth 1999; 83: 596–601.[Abstract/Free Full Text]
13. 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 after total knee replacement. Anesth Analg 1999; 89: 1382–7.[Abstract/Free Full Text]
14. Baker WF, Bick RL. Deep venous thrombosis: diagnosis and management. Med Clin North Am 1995; 78: 685–713.
15. Dunn CJ, Goa KL. Tranexamic acid: a review of its use in surgery and other indications. Drugs 1999; 57: 1005–32.[Web of Science][Medline]
16. Grosse H, Lobbes W, Frambach M, et al. The use of high dose aprotinin in liver transplantation: The influence on fibrinolysis and blood loss. Thromb Res 1991; 63: 287–97.[Web of Science][Medline]
17. Fuhrer G, Gallimore MJ, Heller W, Hoffmeister HE. Aprotinin in cardiopulmonary bypass–effects on the Hageman factor (FXII)-Kallikrein system and blood loss. Blood Coagul Fibrinolysis 1992; 3: 99–104.[Web of Science][Medline]
18. Segal HC, Hunt BJ, Cottam S, et al. Changes in the contact system during orthotopic liver transplantation with and without aprotinin. Transplantation 1995; 59: 366–70.[Web of Science][Medline]
19. Kätzel R, Keuper H, Wiedemann B, et al. Effects of intraoperative aprotinin on perioperative changes in hemostasis and fibrinolysis and their consequences on blood loss and transfusion requirements in thoracic surgery [in German]. Infusionsther Transfusionsmed 1998; 25: 236–45.
20. Miric A, Lombardi P, Sculco TP. Deep vein thrombosis prophylaxis: a comprehensive approach for total hip and total knee arthroplasty patient populations. Am J Orthop 2000; 29: 269–74.[Medline]
21. Mylod AG, France MP, Muser DE, Parsons JR. Perioperative blood loss associated with total knee arthroplasty. J Bone Joint Surg Am 1990; 72: 1010–2.[Abstract/Free Full Text]
22. Burkart BC, Bourne RB, Rorabeck CH, et al. The efficacy of tourniquet release in blood conservation after total knee arthroplasty. Clin Orthop 1994; 299: 147–52.
23. Healy WL, Seidman J, Pfeifer BA, Brown DG. Cold compressive dressing after total knee arthroplasty. Clin Orthop 1994; 299: 143–6.
24. Goodnough LT, Verbrugge D, Marcus RE. The relationship between hematocrit, blood loss, and blood transfused in total knee replacement. Am J Knee Surg 1995; 8: 83–7.[Medline]
25. Keating EM, Meding JB, Faris PM, Ritter MA. Predictors of transfusion risk in elective knee surgery. Clin Orthop 1998; 357: 50–9.

This article has been cited by other articles:

				T. Shiga, Z. Wajima, T. Inoue, and A. Sakamoto Aprotinin in Major Orthopedic Surgery: A Systematic Review of Randomized Controlled Trials Anesth. Analg., December 1, 2005; 101(6): 1602 - 1607. [Abstract] [Full Text] [PDF]

				A. Kokoszka, P. Kuflik, F. Bitan, A. Casden, and M. Neuwirth Evidence-Based Review of the Role of Aprotinin in Blood Conservation During Orthopaedic Surgery J. Bone Joint Surg. Am., May 1, 2005; 87(5): 1129 - 1136. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Engel, J. M. Articles by Hempelmann, G. Search for Related Content PubMed PubMed Citation Articles by Engel, J. M. Articles by Hempelmann, G. Related Collections Blood Surgery