Antifibrinolytic therapy with tranexamic acid in cardiac operations

Cardiovascular Surgery, Vol. 7, No. 2, pp. 195–199, 1999  1999 The International Society for Cardiovascular Surgery. Published by Elsevier Science Ltd All rights reserved. Printed in Great Britain 0967–2109/99 $19.00 ⫹ 0.00

PII: S0967-2109(98)00141-0

Antifibrinolytic therapy with tranexamic acid in cardiac operations K. Matsuzaki, K. Matsui, Y. Tanoue, I. Nagano, N. Haraguchi and H. Tatewaki Department of Cardiovascular Surgery, Matsuyama Red Cross Hospital, Bunkyo-cho-1, Matsuyama 790, Japan To demonstrate its antifibrinolytic effects and establish an effective regimen of tranexamic acid for hemostasis, the authors measured ␣2-plasmin inhibitor–plasmin complexes, thrombin–antithrombin III complexes and postoperative blood loss in three groups undergoing different regimens during cardiac operations. Forty-six patients undergoing coronary artery bypass grafting or valve replacement were enrolled in this study. They were divided into three groups of drug administration. A bolus infusion of 50 mg/kg tranexamic acid was given to 17 patients at the end of cardiopulmonary bypass (control group) and to 14 patients at the beginning of cardiopulmonary bypass (group A). In addition to the same bolus infusion at the beginning of cardiopulmonary bypass as group A, a continuous infusion of 10 mg/kg per h, starting at the time of skin incision and maintained for 6 h after cardiopulmonary bypass was given to 15 patients (group B). The marked increase in ␣2-plasmin inhibitor–plasmin complexes at the end of cardiopulmonary bypass in the control group was significantly reduced in group A (P ⬍ 0.01) and a further reduction was observed in group B (P ⬍ 0.001). The difference in postoperative blood loss only reached significant levels between the control group and group B (P ⬍ 0.05). Although a significant increase in thrombin–antithrombin III complexes during cardiopulmonary bypass was similarly observed in all groups, no thromboembolic events occurred in any group, nor was any difference seen in graft patency. From the tranexamic acid therapy regimens tested in this study, a continuous infusion of 10 mg/kg per h starting at the time of skin incision to 6 h after cardiopulmonary bypass, with a bolus infusion of 50 mg/kg at the beginning of cardiopulmonary bypass, proved to be the most effective.  1999 The International Society for Cardiovascular Surgery. Published by Elsevier Science Ltd. All rights reserved. Keywords: ␣2-plasmin inhibitor–plasmin complexes, postoperative blood loss, thrombin– antithrombin III complexes

Introduction The bleeding tendency following cardiopulmonary bypass is a serious complication in open-heart operations. Hyperactivity of the fibrinolytic system has been recognized as one disturbance of the hemostatic system after cardiopulmonary bypass [1–3]. Recently, numerous authors have reported that antifibrinolytic therapy with aprotinin, a protease inhibi-

Correspondence to: Dr Matsuzaki

CARDIOVASCULAR SURGERY

MARCH 1999 VOL 7 NO 2

tor, significantly reduces the postoperative blood loss in open-heart operations [4–7]. Tranexamic acid, a plasmin inhibitor, also has been reported to reduce blood loss in cardiac operations. However, reports about tranexamic acid are relatively few [8–11] and its antifibrinolytic effect has not been clearly demonstrated in cardiac operations. Therefore, this study was carried out to demonstrate the antifibrinolytic effects of tranexamic acid by measuring ␣2-plasmin inhibitor–plasmin complexes during cardiac operations.

195

Antifibrinolytic therapy with tranexamic acid in cardiac operations: K. Matsuzaki et al.

Materials and methods The study was approved by the ethical committee of the hospital and informed consent was obtained from all patients. Forty-six patients, who were undergoing coronary artery bypass grafting or valve replacement, were randomly divided into three groups according to three different regimens of tranexamic acid administration. Seventeen patients (14 coronary artery bypass grafting and three valve replacement) served as controls and received a bolus infusion of 50 mg/kg tranexamic acid at the end of cardiopulmonary bypass (control group), as an antecedent regimen. Fourteen patients (group A) (nine coronary artery bypass grafting and five valve replacement) received the same bolus infusion at the beginning of cardiopulmonary bypass. Fifteen patients (group B) (14 coronary artery bypass grafting and one valve replacement) received the same bolus infusion at the beginning of cardiopulmonary bypass, as group A, but group B were given an additional continuous infusion of 10 mg/kg/h tranexamic acid starting at the time of skin incision, which was maintained until 6 h after cardiopulmonary bypass. The cardiopulmonary bypass circuit consisted of a bubble oxygenator (Bentley 10, Bentley Lab., Irvine, CA) and non-heparinized tubes, and was primed with asanguineous fluids. Before the cardiopulmonary bypass was initiated, 300 IU/kg of body weight of heparin was given intravenously. Activated clotting times were maintained at appropriate levels of > 400 s during cardiopulmonary bypass with additional heparin administration if necessary. At the conclusion of cardiopulmonary bypass, heparin was neutralized with protamine sulfate at a dose of 3 mg/kg. Activated clotting times of ⬍ 125 s were estimated to indicate adequate neutralization of heparin. The residual blood in the extracorporeal circuit was salvaged by a hemofiltration device (Senko, Tokyo, Japan) and returned to the patient within 1 h after the operation. To measure thrombin–antithrombin III complexes for coagulation activity and ␣2-plasmin inhibitor–plasmin complexes for fibrinolytic activity, blood samples were taken before the operation, 30 min after the initiation of cardiopulmonary bypass, at the end of cardiopulmonary bypass prior to protamine sulfate administration, 6 h after the end of cardiopulmonary bypass, and on the morning of the first postoperative day. Thrombin–antithrombin III complexes were measured by an enzyme-linked immunosorbent assay purchased from Behringerwerke AG, Marburg, Germany, and ␣2-plasmin inhibitor–plasmin complexes were also measured by an enzyme-linked immunosorbent assay, which was purchased from Teijin, Osaka, Japan. The amount of chest tube drainage within the first 6 h after the 196

operation was recorded as postoperative blood loss. In all patients who underwent coronary artery bypass grafting, postoperative coronary angiography was performed 1 month after the operation to investigate graft patency. All data were expressed as the mean ⫾ standard deviation of the mean. Statistical analysis was performed with Student’s t-test or ␹2 test. A P-value of ⬍ 0.05 was considered to be significant.

Results Preoperative patients’ characteristics are shown in Table 1. There were no significant differences in gender and body weight between the groups. However, the patients in group A were significantly younger than the patients in the control group. The operative details are shown in Table 2. There were no significant differences in cardiopulmonary bypass time, aortic occlusion time, or operation time between the control group and group A, and between the control group and group B, respectively. No patient undergoing a reoperation was included in this study. The time-course of ␣2-plasmin inhibitor–plasmin complexes levels are shown in Figure 1. During cardiopulmonary bypass, patients in the control group who did not receive tranexamic acid before and during cardiopulmonary bypass, showed a marked increase of ␣2-plasmin inhibitor–plasmin complexes levels up to 4.8 ⫾ 1.7 ␮g/ml at the end of cardiopulmonary bypass, and the levels remained high until 6 h after cardiopulmonary bypass. On the other hand, the increase in ␣2-plasmin inhibitor– plasmin complexes levels at the end of cardiopulmonary bypass was significantly reduced to 2.3 ⫾ 1.0 ␮g/ml in group A (P ⬍ 0.01) and 1.7 ⫾ 0.5 ␮g/ml in group B (P ⬍ 0.001). Thereafter, ␣2plasmin inhibitor–plasmin complexes levels in both groups A and B showed significant increases during the 6 h after cardiopulmonary bypass, and there was no significant difference between the groups in ␣2plasmin inhibitor–plasmin complexes levels 6 h after cardiopulmonary bypass. In all groups, ␣2-plasmin inhibitor–plasmin complexes levels had returned to almost preoperative levels by the first postoperative day. The time course of thrombin–antithrombin III complexes levels are shown in Figure 2. The thrombin–antithrombin III complexes levels increased gradually during cardiopulmonary bypass in all groups. There were no significant differences between the control group and group A or group B in the values at each measurement. The amount of chest tube drainage within the first 6 h after the operation in the control group was 281 ⫾ 197 ml, and those in group A and B were 212 ⫾ 141 ml and 181 ⫾ 89 ml, respectively (Figure 3). The difference in blood loss between the control CARDIOVASCULAR SURGERY

MARCH 1999 VOL 7 NO 2

Antifibrinolytic therapy with tranexamic acid in cardiac operations: K. Matsuzaki et al. Table 1 Preoperative characteristics

No. of patients Gender (male) Age (years) Body weight (kg)

Control

Group A

P-value

Group B

P-value

17 16 64.3 ⫾ 6.9 65.7 ⫾ 10.8

14 12 58.2 ⫾ 9.1 59.1 ⫾ 7.3

NS ⬍ 0.05 NS

15 10 63.8 ⫾ 7.6 59.9 ⫾ 7.0

NS NS NS

Control

Group A

P-value

Group B

P-value

358 ⫾ 55 155 ⫾ 30 82 ⫾ 21 14 3

383 ⫾ 68 161 ⫾ 19 88 ⫾ 14 9 5

NS NS NS NS NS

396 ⫾ 75 159 ⫾ 32 84 ⫾ 21 14 1

NS NS NS NS NS

NS, not significant

Table 2 Operative details

Operation time (min) Pump time (min) Ischemic time (min) CABG Valve replacement

CABG, coronary artery bypass grafting; NS, not significant

Figure 1 Levels of ␣2-plasmin inhibitor-plasmin complexes (PIC). At the end of cardiopulmonary bypass (CPB), there were significant differences between control and group A (*P ⬍ 0.01), and between control and group B (†P ⬍ 0.001), respectively. POD1, postoperative day one; Pre-Op, preoperation

group and group B reached statistical significance (P ⬍ 0.05). No thromboembolic episodes nor perioperative myocardial infarctions occurred in any group. Postoperative graft patency rates were good in all groups and did not differ significantly between the control group and group A or group B (Table 3).

Discussion Many investigators have proven that the fibrinolytic system is activated during cardiopulmonary bypass [1–3]. Recently, some studies have shown that antiCARDIOVASCULAR SURGERY

MARCH 1999 VOL 7 NO 2

Figure 2 Levels of thrombin-antithrombin III complexes (TAT). There were no significant differences between control and group A, and between control and group B at each measurement, respectively. CPB, cardiopulmonary bypass; POD1, postoperative day one; Pre-Op, preoperation

fibrinolytic therapy with tranexamic acid, a selective plasmin inhibitor, reduces postoperative blood loss in open-heart operations. However, each study significantly varied in the dose and mode of tranexamic acid administration. Øvrum and associates [11] reported a slight but significant reduction (14%) in postoperative chest tube drainage when a single dose of 40 mg/kg tranexamic acid was given only after heparin neutralization. Yau and associates [8] used a huge bolus infusion of 10 g tranexamic acid at the induction of anesthesia. Horrow et al. [9] administrated a relatively small dose of 10 mg/kg tranexamic acid intravenously after the induction of anesthesia, followed by a continuous infusion of 1 mg/kg for 197

Antifibrinolytic therapy with tranexamic acid in cardiac operations: K. Matsuzaki et al.

pulmonary bypass. Therefore, that regimen served as a control in this study. ␣2-Plasmin inhibitor–plasmin complexes levels in the control group showed a marked increase at the end of cardiopulmonary bypass. In this group, tranexamic acid was only administered after cardiopulmonary bypass. This finding confirms the results of other studies, which indicate that the fibrinolytic system is activated during cardiopulmonary bypass [1, 3]. In addition, the single dose administration of 50 mg/kg tranexamic acid at the beginning of cardiopulmonary bypass (group A) significantly suppressed the elevation of ␣2-plasmin inhibitor–plasmin complexes during cardiopulmonary bypass. Furthermore, an injection of 50 mg/kg tranexamic acid at the beginning of cardiopulmonary bypass, along with the additional continuous infusion of 10 mg/kg tranexamic acid starting at the time of skin incision and maintained during cardiopulmonary bypass (group B), resulted in further significant reduction in ␣2-plasmin inhibitor–plasmin complexes levels. Because the antiplasmin action of tranexamic acid is to block the lysin binding sites of plasmin(ogen) and make plasmin(ogen) unable to bind to fibrin [12], tranexamic acid is more effective when given before the activation of the fibrinolytic system. For ␣2-Plasmin inhibitor–plasmin complexes levels at the end of cardiopulmonary bypass, a single dose administration of 50 mg/kg tranexamic acid before cardiopulmonary bypass (group A) was insufficient to suppress fibrinolytic activity. Especially in cases of longer cardiopulmonary bypass, additional doses of tranexamic acid could be deemed necessary because the half-life of tranexamic acid in vivo is only ~80 min [12]. Postoperative blood loss in group B (181 ⫾ 89 ml) showed a significant 36% reduction compared with that of the control group (281 ⫾ 197 ml). This reduction rate and blood loss volume during the 6 h after the operation were matched in the report of Kawasuji et al. whose patients were on low-dose aprotinin therapy [7]. However, Kawasuji et al. reported that they administered 3 million kallikreininhibiting units of aprotinin per patient, which cost

Figure 3 Postoperative blood loss within 6 h after the operation. A significant difference (P ⬍ 0.05) was found between control and group B

10 h. Nakashima and associates [10] used an intermediate dose of 50 mg/kg tranexamic acid, just before and after cardiopulmonary bypass, and intermittently every 90 min during cardiopulmonary bypass. The latter three authors reported a 30–34% significant reduction in postoperative blood loss as compared with non-tranexamic-acid-treated groups. Previously, the authors had administered a single dose of 50 mg/kg tranexamic acid after cardioTable 3 Coronary artery bypass grafting

No. of patients Grafts/patient No. of SV grafts No. of ITA grafts Graft patency rate (%) SV ITA

Control

Group A

14 2.9 ⫾ 0.9 29 12

9 2.8 ⫾ 0.5 17 9

96.6% 100

94.1% 100

P-value

NS

NS NS

Group B 14 2.8 ⫾ 0.9 27 13 100 100

P-value

NS

NS NS

ITA, internal thoracic artery; NS, not significant; SV, saphenous vein

198

CARDIOVASCULAR SURGERY

MARCH 1999 VOL 7 NO 2

Antifibrinolytic therapy with tranexamic acid in cardiac operations: K. Matsuzaki et al.

⫽ Y78,500 in Japan. On the other hand, the average dosage of tranexamic acid in group B described here was ~10 g in each patient, and this only cost ⫽ Y1,190. Moreover, the fact that tranexamic acid has no anaphylactic effect is an advantage over aprotinin, which has been reported to have allergic side-effects [13]. A significant increase in thrombin–antithrombin III complexes at the end of cardiopulmonary bypass, which indicates activation of the clotting system during cardiopulmonary bypass, was observed as reported in other studies [7, 14]. After heparin neutralization, hypercoagulability may cause thrombosis under oversuppression of the fibrinolytic system. Cosgrove and associates [15] pointed out the risk of graft thrombosis in aprotinin therapy. In this study, we found marked increases in ␣2-plasmin inhibitor– plasmin complexes levels 6 h after cardiopulmonary bypass in groups A and B. This fact suggests that a sufficient degree of fibrinolysis is going on, even under the infusion of tranexamic acid; and there were no thromboembolic events nor adverse effects on graft patency observed. However, a large scale study will be required to conclude the safety of tranexamic acid therapy. In conclusion, tranexamic acid infusion during cardiac operations enables the suppression of fibrino(geno)lysis, which is activated during cardiopulmonary bypass. Although antifibrinolytic therapy is usually performed with aprotinin, tranexamic acid achieves the same hemostatic effect at much lower cost and with no allergic side-effects. From the tranexamic acid therapy regimens tested in this study, a continuous infusion of 10 mg/kg starting at the time of skin incision to 6 h after cardiopulmonary bypass, with a bolus infusion of 50 mg/kg at the beginning of cardiopulmonary bypass, proved to be the most effective.

Acknowledgements We thank Arthur Jack Arends for editing the manuscript and Akemi Hirano for her assistance in the preparation of this manuscript.

CARDIOVASCULAR SURGERY

MARCH 1999 VOL 7 NO 2

References 1. Stibbe, J., Kluft, C., Brommer, E. J. P. et al., Enhanced fibrinolytic activity during cardiopulmonary bypass in open-heart surgery in man is caused by extrinsic (tissue-type) plasminogen activator. European Journal of Clinical Investigations, 1984, 14, 375–382. 2. De Haan, J., Scho¨nberger, J., Haan, J. et al., Tissue-type plasminogen activator and fibrin monomers synergistically cause platelet dysfunction during retransfusion of shed blood after cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery, 1993, 106, 1017–1023. 3. Nakajima, H., A clinical study on hemostatic fluctuation during and after cardiopulmonary bypass using hemostatic molecular markers. Nippon Kyobu Geka Gakkai Zasshi [English Abstract], 1992, 40, 1987–1997. 4. Royston, D., Bidstrup, B. P., Taylor, K. M. and Sapsford, R. N., Effect of aprotinin on need for blood transfusion after repeat open-heart surgery. Lancet, 1987, 2, 1289–1291. 5. Van Oeveren, W., Jansen, N. J. G., Bidstrup, B. P. et al., Effects of aprotinin on hemostatic mechanisms during cardiopulmonary bypass. Annals of Thoracic Surgery, 1987, 44, 640–645. 6. Bidstrup, B. P., Royston, D., Sapsford, R. N. et al., Reduction in blood loss and blood use after cardiopulmonary bypass with high dose aprotinin (Trasylol). Journal of Thoracic and Cardiovascular Surgery, 1989, 97, 364–372. 7. Kawasuji, M., Ueyama, K., Sakakibara, N. et al., Effect of lowdose aprotinin on coagulation and fibrinolysis in cardiopulmonary bypass. Annlas of Thoracic Surgery, 1993, 55, 1205– 1209. 8. Yau, T. M., Carson, S., Weisel, R. D. et al., The effect of warm heart surgery on postoperative bleeding. Journal of Thoracic and Cardiovascular Surgery, 1992, 103, 1155–1163. 9. Horrow, J. C., Hlavacek, J., Strong, M. D. et al., Prophylactic tranexamic acid decreases bleeding after cardiac operations. Journal of Thoracic and Cardiovascular Surgery, 1990, 99, 70–74. 10. Nakashima, A., Matsuzaki, K., Fukumura, F. et al., Tranexamic acid reduces blood loss after cardiopulmonary bypass. ASAIO Journal, 1993, 39(M1), 85–89. 11. Øvrum, E., Holen, E. Å., Abdelnoor, M. et al., Tranexamic acid (Cyklokapron) is not necessary to reduce blood loss after coronary artery bypass operations. Journal of Thoracic and Cardiovascular Surgery, 1993, 105, 78–83. 12. Verstraete, M., Clinical application of inhibitors of fibrinolysis. Drug, 1985, 29, 236–261. 13. Bidstrup, B. P., Harrison, J., Royston, D. et al., Aprotinin therapy in cardiac operations: a report on use in 41 cardiac centres in the United Kingdom. Annals of Thoracic Surgery, 1993, 55, 971–976. 14. Havel, M., Teufelsbauer, H., Kno¨bl, P. et al., Effect of intraoperative aprotinin administration on postoperative bleeding in patients undergoing cardiopulmonary bypass operation. Journal of Thoracic and Cardiovascular Surgery, 1991, 101, 968–972. 15. Cosgrove, D. M., Heric, B., Lytle, B. W. et al., Aprotinin therapy for reoperative myocardial revascularization: a placebo-controlled study. Annals of Thoracic Surgery, 1992, 54, 1031–1038.

199