Efficacy and Safety of Tranexamic Acid in Adult Liver Transplantation---ESTA Trial

Overview

We propose a multi-center randomized trial to test the primary hypothesis that tranexamic acid is superior to placebo on blood loss in adult orthotopic liver transplantation within the initial 24 hours and that tranexamic acid is non-inferior to placebo for a composite of thrombotic events within the initial 7 postoperative days. Secondarily, we will determine whether tranexamic acid is superior to placebo on total postoperative drainage volume and blood product transfusion within the initial 3 postoperative days. We propose to randomize patients to 2.0 g of tranexamic acid intravenously at the start of surgery or a comparable volume of 0.9% normal saline placebo. Because demonstrating safety will require more patients, our sample size is based on safety. Randomizing 1546 patients will provide 80% power for detecting a non-inferiority margin of 4% with a baseline incidence of 10% for composite thrombotic events within the initial 7 postoperative days.

BACKGROUND Chronic liver disease and cirrhosis account for around 2 million deaths per year worldwide. Liver transplantation has evolved remarkably over the years to become the standard-of-care for end-stage liver disease\[1\]. Due to the end stage of liver disease, accompanied by metabolic and coagulation dysfunction, coupled with the complex surgical procedures and prolonged duration, the recipients are prone to massive bleeding during perioperative period. Historical records indicate that the first 100 liver transplants were associated with high mortality, with most patients succumbing to uncontrolled bleeding\[2\]. According to reports, massive intraoperative bleeding has been reported in 10%-34% of OLT cases\[3, 4\]. In a USA multi-center large-scale cohort study, the incidence rate of intraoperative packed RBCs transfusion was 81% and 12% patients received ≥10 RBC units\[5\]. With refined surgical techniques and perioperative care, perioperative bleeding for liver-transplant recipients has gradually decreased to the point where many transplant recipients require no transfusions\[6-8\]. Nonetheless, perioperative bleeding remains an important complication in many liver transplantat patients. Tranexamic acid (TXA) is a synthetic analog of lysine that inhibits fibrinolysis by preventing plasminogen from binding to fibrin\[9\]. In the past decades, it has attracted international attention due to the extensive reduction of blood loss in a wide range of surgical procedures and improves survival rates in patients with severe bleeding\[10, 11\]. It is particularly recommended for surgeries with an anticipated blood loss exceeding 500 mL\[12\]. In a series of international large-scale trials studied in patients having cardiac surgery, trauma, cesarean sections, and non-cardiac surgeries, use of tranexamic acid significantly reduces the incidence and severity of perioperative bleeding, decreases the need for blood transfusions, and does not lead to significant thrombotic complications or seizures\[13, 14\]. However, administration of high-dose TXA can provoke seizures and thromboembolism that may increase the risk of perioperative disability or mortality\[12\]. What's more, after many years of experience with TXA in various fields, clinicians are now querying whether the dose, route and interval of administration currently used and the methods used to control and analyze the antifibrinolytic mechanism of TXA are optimal\[15\]. These issues need to be evaluated and reviewed using the latest evidence to improve the safety and effectiveness of TXA in treating hemorrhage or bleeding in surgeries. Postoperative bleeding is a severe threat to patients having liver transplantation because, with rare exceptions (e.g., cancer surgery), they have decompensated liver function and systemic hyper fibrinolysis. However, evidence for the perioperative use of tranexamic acid in this population remains limited and controversial \[16, 17\]. In a RCT study of 45 liver transplant patients published in the journal of Anesthesiology in 1996, TXA significantly reduced intraoperative blood loss without hepatic artery or portal vein thrombosis within 30 days of transplantation\[17\]. And in a trial of 600 advanced liver cirrhosis patients presenting with upper gastrointestinal bleeding, TXA was proven helpful in decreasing bleeding by day 5 and rebleeding by 6 weeks\[18\]. However, a retrospective cohort study of 779 liver transplantation patients and a RCT study of 1245 cancer-related resection patients concluded that intraoperative TXA administration does not lower blood loss or reduce utilization of red blood cell transfusion\[19, 20\]. A lager sample study\[20\] did not prove that tranexamic acid was effective in reducing intraoperative blood transfusion, it also did not increase postoperative thrombotic events related to large vessels and local hepatic vessels either. It is important to consider that the severity of liver disease, surgical trauma, and blood loss in the liver resection patients included in that study were much better than those with end-stage liver disease requiring OLT. Furthermore, patients scheduled for liver transplantation often present thrombocytopenia and/or a deficiency of coagulation factors produced in the liver, as well as the hyperfibrinolysis during the anhepatic and reperfusion phases, these unique pathophysiological characters make the bleeding and hemostasis strategy in OLT surgery differ from that in hepatectomy. For patients with end-stage liver disease, whether intraoperative tranexamic acid reduces blood loss and whether it increases thrombotic events is remains unclear. Despite the lack of clear evidence for increased risk of hypercoagulability consequent to use of antifibrinolytics during liver transplant, postoperative hypercoagulability and thrombotic complications remain common in clinical and provoke adverse outcomes. Whether TXA promotes hypercoagulation and thrombotic events in OLT patients remains unknown. The delicate balance between bleeding and thrombosis after TXA administration therefore requires additional exploration. A robust trial to assess the effectiveness and safety of TXA in liver transplantation is thus needed. We therefore propose to test the primary efficacy hypothesis that TXA is superior to placebo on blood loss in adult orthotopic liver transplantation within the initial postoperative day, and the primary safety hypothesis that TXA is non-inferior to placebo on a composite of thrombotic events within the initial 7 postoperative days. Secondarily, we propose that TXA is superior to placebo for the total amount of postoperative drainage volume and blood product transfusion within the initial 3 days after liver transplantation. METHODS Trial design and ethics Our multi-center randomized trial will enroll 1546 patients scheduled for allogeneic liver transplantation. The trial will be performed in 24 centers. Participating patients will be randomized to TXA or placebo. The trial will be approved by the ethics committee of Renji Hospital of the Shanghai Jiao Tong University School of Medicine and by each participating hospital. The trial will be registered at ClinicalTrial.gov before the first patient is enrolled. Written informed consent will be obtained from all patients or their authorized representatives. We expect enrollment to take about 2 years. Preparation, Anesthesia, surgery, intervention and Perioperative Management Organ donation or transplantation in the study will be implemented under regulation of Local Organ Transplant Committee and the Declaration of Helsinki. Liver grafts will be from brain dead donors or living donors. Brain dead grafts will be preserved in conventional static cold preservation for transport to transplant centers while liver donor grafts will be was injected with 0-4℃ lactated Ringer's solution and UW solution through the portal vein immediately after resection . All the surgical procedures will be performed by specialists with experience in OLTs. Qualifying and consented patients will receive routine surgical, anesthesia and perioperative care per local practice. General anesthesia will be induced and maintained with propofol, sufentanil/remifentanil, rocuronium/cisatracurium and inhaled anesthetics. Intraoperative monitoring will include electrocardiogram, pulse oxygen saturation, intra-arterial blood pressure, airway pressure, end-tidal concentration of carbon dioxide, central venous pressure, nasopharyngeal temperature, and urine output. Thermodilution cardiac output monitoring (measured via a pulmonary artery catheter or PICCO system) or perioperative transesophageal echocardiography, bispectral index may be used when necessary. Intraoperative fluid will be given per the attending anesthesiologist's routine practice. Blood transfusion will be based on clinical need, but generally initiated when intraoperative hemoglobin is less than 70 g/L\[40\]. Plasma, platelets, and anti-fibrinolytics will be used to manage coagulopathic states per routine. Randomization and blinding Randomization will occur shortly before surgery after consenting patients are deemed eligible and use random-sized blocks stratified by trial site and MELD scores ≥16 points. Patients will be randomized in a 1:1 ratio to receive either TXA or matching placebo. Allocations will be generated by REDCap which can be operated via computer or smartphone. The unblinded investigator will enter the eligible patient's data into the randomization application on their computer or smartphone. Once the data is submitted, the patient's treatment allocation will be immediately acquired. The trial will use a double-blind design, with the surgical team, anesthesiologists, postoperative interviewers and patients blinded to the group assignment. The unblinded investigator will prepare the drugs, ensuring the trial drug's outer packaging, color, and texture are completely identical. In patients assigned to tranexamic acid, 2.0 g dissolved in 50 ml of 0.9% saline）will be given intravenously within 30 mins after skin incision. In patients assigned to placebo, a comparable volume of 0.9% saline will be given. Data collection Preoperative laboratory data (most recent within 3 days of surgery) will be recorded. Baseline data will include: 1) Demographic characteristics: sex, date of birth, height, weight, and body mass index (BMI); 2) Medical data: diagnosis (reason for LT), comorbidities, concomitant medication, non-drug therapies, history of smoking and drinking, history of food or drug allergy, and history of anesthesia and surgery; 3) Medical examinations including arrythmia detected by electrocardiography, heart conditions examined by echocardiography and liver diseases detected by ultrasound, computed tomography (CT) or magnetic resonance imaging (MRI); 4) The most recent laboratory investigations that reflects liver, kidney, heart, coagulation function, complete blood count, arterial blood gas (ABG) analysis including albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine, blood urea nitrogen (BUN), troponin, international normalized ratio (INR), hemoglobin, white blood cell, pH, partial pressure of oxygen (PaO2), partial pressure of carbon dioxide (PaCO2); 5) Donor characteristics: age, height, weight, BMI, cause of death, and virology or hepatic fat deposition status. All these test results including diagnostic results of CTs or MRIs will be converted to database fields with defined values in REDCap. Intraoperative characteristics to be recorded will include: 1). Duration of surgery, anhepatic phase, cold and warm ischemia times; 2). Fluids given (including intravenous fluids, colloids, blood products) and output (including blood loss, urine). Intraoperative bleeding (estimated by weighing surgical sponges and suction canisters before and after use), type and volume of blood products transfused, and type and volume of fluid infusion; 3). Central venous pressure, hemoglobin, arterial blood gas, electrolytes, lactic acid, and blood glucose.