Blood loss and coagulation management remain serious concerns in patients undergoing orthotopic liver transplantation (OLT). Postoperative infection and mortality correlate well with perioperative transfusion requirements (1,2). Although improvements in operative management, surgical techniques, and graft preservation have contributed to reducing intraoperative blood loss and transfusion requirements in the last 10 yr (3), patients still often require a large volume of red blood cells (RBCs) (4–6). Activation of the fibrinolytic system is one of the most important causes of bleeding during OLT (7–9). Therefore, there is a growing interest in the use of antifibrinolytic drugs to reduce blood product transfusion during OLT (5,10,11).
Studies of the prophylactic administration of aprotinin have produced contradictory and controversial results. Certain nonrandomized and noncontrolled studies have found that the administration of a small or large dose of aprotinin resulted in a reduction of blood loss and transfusion requirements (12). However, in two randomized studies, aprotinin was no more effective than placebo (13,14). ε-Aminocaproic acid (EACA) has been administered mainly for the treatment of fibrinolysis during OLT (15). However, prophylactic administration of this drug has not been demonstrated to significantly reduce blood product use (16). In contrast, tranexamic acid (TA) reduces intraoperative blood loss and perioperative transfusion requirement at a large dose (40 mg · kg−1 · h−1) (17); a small dose (<5 mg · kg−1 · h−1) reduces fibrinolysis (18) and blood loss (19), but not transfusion requirements. However, a large dose of TA may increase the risk of thrombosis; to prevent the development of this complication, administration of dipyridamole-heparin has been suggested (17).
Our objective was to evaluate the efficacy of the prophylactic administration of EACA and a moderate dose of TA for reducing blood transfusion requirements in patients undergoing OLT.
After institutional review board approval, written consent was obtained from all patients. A double-blinded, prospective, randomized, placebo-controlled study was performed in all consecutive patients undergoing OLT from January 1997 to September 1998 in two adult liver transplantation centers. Exclusion criteria were 1) Budd-Chiari syndrome, 2) acute liver failure, 3) early retransplantation (<1 mo), 4) simultaneous kidney and liver transplantation or renal insufficiency with dialysis, and 5) primary familial amyloidotic neuropathy.
Participants were randomly assigned to receive TA, EACA, or placebo (P). Patients in the TA group received a continuous dose infusion of TA (5 g in 450 mL of normal saline) at a rate of 10 mg · kg−1 · h−1. Patients in the EACA group received a continuous infusion of EACA (8 g in 480 mL of normal saline) at a rate of 16 mg · kg−1 · h−1. Patients in the control group received an equal volume of normal saline (10 mL · kg−1 · h−1). Drugs were prepared by using a randomization schedule provided in sealed envelopes. All patients received drugs via an infusion pump, without an initial bolus, from the induction of anesthesia until the portal vein was unclamped.
General anesthesia was induced with 2 mg/kg of propofol, 0.3 mg of fentanyl, and 0.5 mg/kg of atracurium and was maintained by continuous infusion of the same drugs. Mechanical ventilation was begun at 10 mL/kg with a ventilatory rate adjusted to obtain end-expiratory partial pressure of CO2 of 30–35 mm Hg, by using an inspired oxygen fraction (Fio2) of 0.5 in air. Calcium was administered to maintain ionized calcium levels between 1 and 1.2 mmol/L and sodium bicarbonate to maintain arterial pH greater than 7.30. Isotonic saline solution was infused at a rate of 7–10 mL · kg−1 · h−1. All patients were placed on a warm blanket with their lower limbs covered with cotton and aluminum foil; all IV fluids were administered at a temperature of 36°C.
Liver allografts were preserved using University of Wisconsin solution (20). All surgical procedures were performed by using a piggy-back technique that did not cross-clamp or transect the inferior vena cava. Before reperfusion of the graft, the liver was flushed with 1000 mL of Hartmann solution at 38°C and 200 mL of portal blood to remove the air.
Criteria for Replacement Therapy
RBCs were administered to maintain hematocrit levels at 30% and hemoglobin at 100 g/L. Fresh frozen plasma (FFP) was administered only when the prothrombin time (PT) ratio exceeded 1.8 times control. Platelets were given to maintain a platelet count greater than 50 × 109/L, and cryoprecipitate was administered to maintain a fibrinogen level greater than 1.5 g/L. No intraoperative salvage of blood was used.
Criteria for Pharmacologic Therapy
Desmopressin was given prophylactically to all patients who had a bleeding time of >8 min or abnormal maximum amplitude (MA; <35 mm) in the thromboelastogram at a dose of 4 μg/kg at the beginning of the procedure. The drug was also given as a treatment in the case of massive bleeding and the MA value in the thromboelastogram <35 mm at the reperfusion phase.
EACA was administered at a dose of 0.25 g when there was diffuse bleeding associated with lysis in thromboelastogram or a fibrinogen value <1 g/L.
Additionally, protamine sulfate (dose of 25–50 mg) was administered when significant heparin effect was demonstrated by prolonged activated partial thromboplastin time (aPTT) in the reperfusion phase (11).
Blood samples were taken before induction, during dissection, during the anhepatic phase, 5 min after reperfusion of the graft, at the end of the surgical procedure, and in the intensive care unit for 24 h after OLT. All samples were drawn through a central venous catheter not flushed by heparin. The following assays were performed: hemoglobin and hematocrit level, platelet count, PT, aPTT, fibrinogen, and TEG® (Thromboelastograph Coagulation Analyzer®; Haemoscope, Skokie, IL). Measured TEG® variables were: reaction time, clot formation rate (α), MA, blood clot lysis index (A60/MA × 100) at 60 min.
The number of units of RBCs, FFP, platelets, and cryoprecipitate transfused were recorded throughout the procedure and during the first 24 h in the intensive care unit.
Other variables recorded were demographic data, cold ischemia time, the need for reoperation because of intraabdominal bleeding, and early (≤30 days) postoperative vascular thrombotic complications diagnosed by using systematic screening for all patients with color-pulsed Doppler sonography (128 × P/4; Acuson Computed Sonography, Mountain View, CA). Findings on Doppler sonograms were considered abnormal when the signal was absent in either the main hepatic artery or in one of the intrahepatic branches. Arteriography with selective catheterization of the celiac axis was performed in these abnormal cases to confirm thrombosis of the hepatic artery. Mortality was also recorded.
The trial was designed to detect a 30% decrease in RBC requirements for the treated groups, with a one-tailed α-error of 0.05 and a β-error of 0.1, yielding a total sample size of 120 for the purposes of parametric tests.
Continuous variables are presented as means and sd and median only for perioperative replacement therapy. Analysis of variance was used for comparison of continuous variables between the groups. Non-Gaussian blood product usage data were compared by using the nonparametric analysis of Kruskal-Wallis. When the results were significant, we performed comparisons among groups (Mann-Whitney U-test). Discontinuous variables are presented in percentages and the χ2 test was used to compare the groups. Differences with probability values of 0.05 or less were considered significant.
One hundred fifty-three OLTs were performed in the study. Twenty-nine patients were excluded because of acute hepatic failure (3 cases), retransplantation (11 cases), simultaneous kidney and liver transplantations (3 cases), renal insufficiency with dialysis (1 case), and primary amyloidotic neuropathy (3 cases). One hundred thirty-two patients were included in the study. Five patients were excluded because of incomplete data and three because of air embolism, with cardiac arrest successfully treated during the operation. We therefore analyzed data from 124 patients: 42 in the TA group, 42 in the EACA group, and 40 in the P group.
Patients in the treatment groups were demographically similar regarding age, sex, previous upper abdominal surgery, diagnosis and ischemia, and surgical times (Table 1). Fewer patients in the P group had a Child A score than in the treatment groups, but the difference was not significant.
In the treatment groups, the mean dose of TA was 2.5 ± 0.8 g (range 1–4.5 g), and the mean dose of EACA was 4.2 ± 1.5 g (range 1.7–7.6 g).
Hemogram data and coagulation profiles during surgery are shown in Table 2. There was only a significant reduction in α in TEG® in the P group during the reperfusion phase.
Intraoperative RBC usage was significantly reduced in the TA group, but not after the first 24 h (Table 3). There was also a reduction of RBC and FFP usage in the EACA group, but it did not achieve statistical significance. Group by group comparison revealed differences only in the TA group (Table 3). RBC transfusion was not required by 13 patients in the TA group, 6 patients in the EACA group, and 3 patients in the P group (P = 0.016). Although the difference was statistically significant, the small number of patients does not allow us to draw firm conclusions.
Prophylactic desmopressin at the beginning of the procedure was administered in 9 TA patients (9 of 42), in 12 EACA patients (12 of 42) patients, and in 9 patients (9 of 40) in the P group (P = 0.713). Desmopressin as a treatment at the time of reperfusion was administered to no TA patients (0 of 42), to 6 EACA patients (6 of 42), and to 4 patients (4 of 40) in the P group (P = 0.047).
EACA (for treatment of clinical fibrinolysis) was administered in no TA patients (0 of 42), in 6 EACA patients (6 of 42), and in 7 patients (7 of 40) in the P group (P = 0.021).
Two patients in the P group only received protamine.
At the end of the operation, the hemoglobin, platelet count, PT, aPTT, fibrinogen, and TEG® values were comparable among groups, as were hemoglobin, platelet count, PT, aPTT, and fibrinogen on the first postoperative day.
Four patients in the TA group had postoperative arterial thrombosis; in three patients, the complication may have been caused by technical difficulties in the arterial anastomosis during OLT, such as aortoiliac grafting, splenic artery anastomosis, reconstruction during bench surgery, and arterial reanastomosis during OLT. One patient required retransplantation and another died while waiting for retransplantation. In the EACA group, two patients had arterial thrombosis, but there were no technical problems in any of the vascular anastomosis performed; one patient required retransplantation. In the P group, two patients presented thrombosis, which may have been the result of technical difficulties in the anastomosis during OLT. Neither required retransplantation.
Reoperations for either thrombosis or bleeding were performed in 3 patients in the TA group, in 4 patients in the EACA group, and in 4 patients in the P group.
Mortality at 5 mo postoperatively was 3 patients (3 of 42) in the TA group, 3 patients (3 of 42) in the EACA group, and in 4 patients (4 of 42) in the P group (P = 0.861).
In this study, prophylaxis with TA was associated with some improvement in fibrinolysis, significant intraoperative decreases in RBC use, and a greater number of patients without any RBC transfusion. During the first 24 hours postoperatively, the transfusion requirement was not different among groups. EACA prophylaxis did not result in these benefits. Although fewer patients in the P group were preoperatively defined as Child A than in the other two groups, the percentage of high-risk patients (Child C) was similar in all three groups; furthermore, the preoperative hemostatic status was similar throughout the sample, and so this variation in Child class would not have influenced the results.
The aim of our therapy was to reduce transfusion requirements during OLT by decreasing fibrinolysis. We demonstrated the reduction in intraoperative RBC usage in the TA group (Table 3). In addition, the thromboelastographic profile in the reperfusion phase [one of the phases with the highest increase of fibrinolysis activity (8,21)] was better in the TA and EACA groups than in the P group (Table 2). When compared with the P group, the TA group presented more patients with no RBC usage.
Our data confirm some of the findings in OLT of Boylan et al. (17) who used 40 mg · kg−1 · h−1, considered to be a large dose of TA. However, OLT was performed in all our patients with vena cava preservation, whereas in Boylan’s study a veno-venous bypass was used in some patients. Boylan et al. (17) did not report the percentage of patients who did not need any transfusion. In our study, 31% of the patients of the TA group did not receive any RBC. Kaspar et al. (18), using a small dose of TA (2 mg · kg−1 · h−1) was able to demonstrate that TA administration inhibits fibrinolysis, but could not show any improvement in blood transfusion, indicating that at a small dose no clinical advantage of using TA during OLT was observed. In summary, a medium and large dose of TA, but not a small dose, are able to produce a significant reduction of RBC transfusion during OLT.
Thrombosis caused by antifibrinolytic therapy is a potential complication of great concern. We did not find differences among groups. However it would be necessary to study a larger number of patients to determine conclusively whether the incidence of thrombosis increases with the prophylactic administration of antifibrinolytics. We were unable to establish whether the incidence of thrombosis increases with the prophylactic administration of TA. Boylan et al. (17) did not report any case of thrombosis in their study; this may have been because of the dipyridamole-heparin perfusion administered 24 hours after completion of the hepatic arterial anastomosis or alternatively to the small number of patients studied.
EACA has been reported to be effective in the treatment of fibrinolysis (15). Although the prophylactic administration of EACA in our study tended to reduce fibrinolysis and blood product usage, it was not as effective as TA. The EACA dose used in this trial was not as potent as the TA dose, which may explain EACA results. Although EACA is 6–10 times less potent than TA (22), we chose the dose of 16 mg · kg−1 · h−1 of EACA to maintain a relation of 1.5:1 to TA dose, which is the ratio most often applied in the studies of cardiac surgery (23). In addition, Kang (11) demonstrated that a dose of 250–500 mg is sufficient to treat severe fibrinolysis with a second dose needed in the most serious cases during OLT. Furthermore, other cases of thrombosis have been described elsewhere, with a large dose of EACA (24).
Other drugs have been shown capable of reducing blood loss and exposure to allogenic blood products. Studies with aprotinin therapy in OLT have produced conflicting results (12–14). Desmopressin has proven useful for preventing bleeding in patients with acquired platelet dysfunction, especially in cases of abnormally low MA in TEG® and a prolonged bleeding time.
We conclude that the prophylactic administration of TA at 10 mg · kg−1 · h−1 from anesthetic induction until portal vein unclamping can reduce fibrinolysis and intraoperative RBC transfusion in patients undergoing OLT. This effect was not maintained during the first 24 hours postoperatively. EACA at 16 mg · kg−1 · h−1 does not produce such beneficial effects. More patients need to be studied to establish the risks involved in administration of this dose of TA, especially in relation to arterial thrombosis.
The authors thank C. Bartolomé, C. Benito, I. Camprubí, M. J. Castro, C. Martin, V. Mayoral, M. Mercadal, V. Perela, R. Sopena, R. Sanzol, T. Garcia, N. Villar, M. Garriga, M. Blasco, E. de la Riba, P. Ruiz, J. Fabregat, J. Torres, E. Ramos, A. Ribó, M. Reche, R. Beltrán, V. Roqués, R. Robles, F. S. Bueno, P. Ramirez, J. Rodriguez, J. Lujan, J. Pastor, M. H. Palazón, and M. Estan for participating in this clinical trial. The authors thank E. Figueras for helping with the translation of this manuscript. We have had the unconditional support of both hospitals, especially the liver transplantation units.
1. Bontempo FA, Lewis JH, Van Thiel DH, et al. The relation of preoperative coagulation findings to diagnosis, blood usage, and survival in adult liver transplantation. Transplantation 1985; 39:532–6.
2. Mor E, Jennings L, Gonwa TA, et al. The impact of operative bleeding on outcome in transplantation of the liver. Surg Gynecol Obstet 1993; 176:219–27.
3. Porte RJ, Knot EAR, Bontempo FA. Hemostasis in liver transplantation. Gastroenterology 1989; 97:488–501.
4. Reyle-Hahn M, Rossaint R. Coagulation techniques are not important in directing blood product transfusion during liver transplantation. Liver Transplant Surg 1997; 3:659–63.
5. Kufner RP. Antifibrinolytics do not reduce transfusion requirements in patients undergoing orthotopic liver transplantation. Liver Transplant Surg 1997; 3:668–74.
6. Busuttil RW, Shaked A, Millis M, et al. One thousand liver transplants—the lessons learned. Ann Surg 1994; 219:490–9.
7. Dzik WH, Arkin CF, Jenkins RL, Stump DC. Fibrinolysis during liver transplantation in humans: role of tissue-type plasminogen activator. Blood 1988; 71:1090–5.
8. Porte RJ, Bontempo FA, Knot EAR, et al. Systemic effects of tissue plasminogen activator associated fibrinolysis and its relation to thrombin generation in orthotopic liver transplantation. Transplantation 1989; 47:978–84.
9. Steib A, Gengenwin N, Freys G, et al. Predictive factors of hyperfibrinolytic activity during liver transplantation in cirrhotic patients. Br J Anaesth 1994; 73:645–8.
10. Ramsay MAE. The use of antifibrinolytic agents results in a reduction in transfused blood products during liver transplantation. Liver Transplant Surg 1997; 3:665–8.
11. Kang Y. Coagulation and liver transplantation: current concepts. Liver Transplant Surg 1997; 3:465–7.
12. Mallet SV, Cox D, Burrougoughs AK, Rolles K. Aprotinin and reduction of blood loss and transfusion requirements in orthotopic liver transplantation. Lancet 1990; 6:886–7.
13. Groh J, Welte M, Azad SC, et al. Does aprotinin affect blood loss in liver transplantation? Lancet 1992; 340:173.
14. Garcia-Huete L, Domenech P, Sabaté A, et al. The prophylactic effect of aprotinin in intraoperative bleeding in liver transplantation: a randomized clinical study. Hepatology 1997; 26:1143–8.
15. Kang YG, Lewis JH, Navalgund A, et al. Epsilon-aminocaproic acid for treatment of fibrinolysis during liver transplantation. Anesthesiology 1987; 66:766–73.
16. McSorley MW, Taraporewalla KJ. Does prophylactic epsilon-aminocaproic acid improve blood loss and coagulation in liver transplantation? Transplant Proc 1991; 23:1941.
17. Boylan JF, Klinck JR, Sandler AN, et al. Tranexamic acid reduces blood loss, transfusion requirements, and coagulation factor use in primary orthotopic liver transplantation. Anesthesiology 1996; 85:1043–8.
18. Kaspar M, Ramsay MAE, Nguyen A-T, et al. Continuous small-dose tranexamic acid reduces fibrinolysis but not transfusion requirements during orthotopic liver transplantation. Anesth Analg 1997; 85:281–5.
19. Yassen K, Bellamy MC, Sadek SA, Webster NR. Tranexamic reduces blood loss during orthotopic liver transplantation. Clin Transplant 1993; 7:453–8.
20. Beltzer FO, Kalayoglu M, D’Alessandro AM, et al. Organ preservation: experience with University of Wisconsin solution and plans for future. Clin Transplant 1990; 2:73–7.
21. Kang Y. Coagulation and liver transplantation. Transplant Proc 1993; 25:2001–5.
22. Verstraete M. Clinical application of inhibitors of fibrinolysis. Drugs 1985; 29:236–61.
23. Hardy J-F, Bélisle S. Natural and synthetic antifibrinolytics in adult cardiac surgery: efficacy, effectiveness and efficiency. Can J Anaesth 1994; 41:1104–12.
24. Prah GN, Lisman SR, Maslow AD, et al. Transesophageal echocardiography reveals an unusual cause of hemodynamic collapse during orthotopic liver transplantation—two case reports. Transplantation 1995; 59:921–5.