Consensus Statement on Hemostatic Management, Anticoagulation, and Antiplatelet Therapy in Liver Transplantation.

Anticoagulation and antiplatelet therapies are increasingly used in liver transplant (LT) candidates and recipients due to cardiovascular comorbidities, portal vein thrombosis, or to manage posttransplant complications. The implementation of the new direct-acting oral anticoagulants and the recently developed antiplatelet drugs is a great challenge for transplant teams worldwide, as their activity must be monitored and their complications managed, in the absence of robust scientific evidence. In this changing and clinically heterogeneous scenario, the Spanish Society of Liver Transplantation and the Spanish Society of Thrombosis and Haemostasis aimed to achieve consensus regarding the indications, drugs, dosing, and timing of anticoagulation and antiplatelet therapies initiated from the inclusion of the patient on the waiting list to post-LT surveillance. A multidisciplinary group of experts composed by transplant hepatologists, surgeons, hematologists, transplant-specialized anesthesiologists, and intensivists performed a comprehensive review of the literature and identified 21 clinically relevant questions using the patient-intervention-comparison-outcome format. A preliminary list of recommendations was drafted and further validated using a modified Delphi approach by a panel of 24 transplant delegates, each representing a LT institution in Spain. The present consensus statement contains the key recommendations together with the core supporting scientific evidence, which will provide guidance for improved and more homogeneous clinical decision making.

INTRODUCTION

Anticoagulation or antiplatelet therapy may be required in liver transplant (LT) patients either for prophylactic or therapeutic purposes, and this need will probably increase in the near future as metabolic-associated fatty liver disease becomes one of the leading indications of LT.[1] Indeed, nowadays, both LT donors and candidates are older and have a more adverse cardiovascular risk profile.[2] In addition, recent surgical advances allow LT to be considered in patients with complex splanchnic vein thrombosis receiving anticoagulants. There is growing interest in evaluating the role of anticoagulation or antiplatelet therapy to prevent graft thrombosis.[3] Finally, some vascular complications are initially approached percutaneously, with specific hemostatic requirements.

In parallel to the growing indications of anticoagulation and antiplatelet therapy in the LT setting, the pharmacological armamentarium has become wider with the upcoming of new direct-acting oral anticoagulants, including factor Xa inhibitors (rivoraxaban, apixaban, edoxaban, betrixaban) and the direct thrombin inhibitor dabigatran, and also the recently developed antiplatelet drugs, with which there is limited experience in patients with advanced liver disease or LT. Indeed, the scientific evidence regarding the efficacy and safety of anticoagulation and antiplatelet therapies in the peritransplant setting is scarce and of suboptimal quality.[3] As a result, there is wide heterogeneity in clinical practice.

The Spanish Society of Liver Transplantation and the Spanish Society of Thrombosis and Haemostasis aimed to achieve consensus regarding the indications, drugs, dosing, and timing of anticoagulation and antiplatelet therapies initiated from the inclusion of the patient on the waiting list to post-LT surveillance. In the present document, we summarize the main recommendations and the core supporting scientific evidence facilitating improved and more homogeneous clinical decision making.

MATERIALS AND METHODS

In December 2020, the Spanish Society of Liver Transplantation engaged a multidisciplinary group of experts composed of 2 transplant hepatologists, 2 surgeons, 1 transplant-specialized anesthesiologist, and 1 intensivist. The Spanish Society of Thrombosis and Haemostasis endorsed this initiative and identified 2 expert hematologists who joined the multidisciplinary group. The consensus statement was stratified into 3 sections, each corresponding to a period with inherent peculiarities regarding anticoagulation and antiplatelet therapy: (a) the pretransplant period, which was defined from the inclusion of the patient on the waiting list until the upcoming of a donor; (b) the intraoperative period; and (c) the posttransplant period. Three experts were assigned to each section taking into account their area of expertise and a total of 21 clinically relevant questions were formulated according to the patient-intervention-comparison-outcome (PICO) format.[4] This questionnaire was distributed among 25 transplant delegates, each of whom represented a LT institution in Spain. The inputs received allowed the expert panel to identify the most conflicting clinical scenarios and major sources of clinical heterogeneity to delineate the structure of the consensus document.

A modified Delphi approach[5] was used as summarized in Figure 1. A comprehensive literature search was conducted by the expert panel to identify all relevant articles regarding anticoagulation and antiplatelet therapy in patients with end-stage liver disease or with LT. MEDLINE, Google Scholar, PubMed, the Cochrane Library, and resources of international societies of transplantation and hepatology were searched using the following keywords or equivalent free-text terms: [“cirrhosis” OR “Liver Transplantation”] AND [“anticoagulation” OR “antiaggregant therapy” OR “thrombosis”]. Recent reviews and position statements were hand-searched to retrieve additional relevant studies. A preliminary list of recommendations was issued to address each of the PICO questions. In the first Delphi-like round, the list of recommendations was distributed among the 24 transplant delegates who reviewed the list and provided feedback to implement modifications, which were incorporated to the preliminary article upon approval by the expert multidisciplinary panel. Then, an online interactive consensus meeting was organized including the expert panel and the transplant delegates on February 17 and 18, 2021. Recommendations were reviewed individually, and the level of agreement for each statement was obtained using a real-time voting system. If the agreement among the transplant delegates for a given recommendation was lower than 90%, a debate took place to reassess the recommendation. Only recommendations with an agreement higher than 90% entered the final version of the document, which was distributed among transplant delegates for minor remarks and final approval (second Delphi-like round).

FIGURE 1.

Flow chart illustrating the modified Delphi approach to achieve consensus regarding indications of anticoagulation and antiplatelet therapy in patients with end-stage liver disease or liver transplantation. PICO, patient-intervention-comparison-outcome.

The scientific evidence and strength of recommendations was evaluated using the “Grading of Recommendations Assessment, Development and Evaluation” (GRADE) system,[6] which rates 2 dimensions: (a) Strength of the recommendation: classified as “1” (if strong) or as “2” (if weak), and (b) Quality of the evidence: classified as “A” (high-quality evidence coming from well-designed randomized trials or overwhelming evidence from other sources), “B” (moderate-quality evidence from randomized trials with methodological limitations or well-designed observational studies), or “C” (low-quality evidence from observational studies or unsystematic clinical experience). Recommendations with more solid scientific background classified as grade 1A or 1B formed part of the main document, while the recommendations based on weaker evidence are provided as Supplementary Digital Content, http://links.lww.com/TP/C321. The present consensus statement complies with the highest methodological standards according to the Guidelines International Network.[7] This initiative did not involve patients and was exempt from approval from an ethics’ board.

Pretransplant Period

There is a growing number of medical conditions routinely managed with anticoagulant or antiplatelet therapies in LT candidates, which may be classified as related and unrelated to the liver disease. Although the vast majority of patients with advanced liver disease show severe thrombocytopenia and prolonged prothrombin time, liver cirrhosis is associated with a complex and fragile rebalanced hemostasis,[8] often complicated with thrombotic events. The most specific liver-related thrombotic events are the Budd-Chiari syndrome, which is an uncommon indication of LT,[9] and portal vein thrombosis with a prevalence ranging from 5% to 26% of LT candidates.[10] It is paramount to rule out malignant disease underlying both conditions. In patients with Budd-Chiari syndrome, a complete hematological workup is mandatory, while patients with portal vein thrombosis may require dynamic liver imaging techniques (computed tomography or magnetic resonance) reviewed by an expert radiologist within a multidisciplinary team. On the other hand, extrahepatic conditions requiring specific hemostatic management are becoming more frequent due to: older age and more adverse cardiovascular profile of LT candidates,[11] particularly among patients with metabolic-associated fatty liver disease,[2,12] a deeper cardiovascular pretransplant workup,[13] and recent advances in percutaneous management of coronary artery disease.[14] Indeed, patients with significant coronary stenosis, even with 2 or 3 vessels involved, may safely receive a LT nowadays if they undergo prior percutaneous stenting, with subsequent double antiplatelet therapy.[15] The prevalence of atrial fibrillation is expected to rise from 7.8% to 9.5% among patients older than 65 y in Europe in the next decades.[16] Other frequent clinical conditions among LT candidates requiring chronic hemostatic management are heart valve replacement, venous thromboembolism, and ischemic stroke, among others.

In parallel, the anticoagulant and antiplatelet therapeutic armamentarium has become wider and there is a paucity of high-quality studies evaluating the safety and efficacy of these drugs in patients with cirrhosis.[3] Novel direct oral anticoagulants, which inhibit the active site of thrombin (dabigatran) or coagulation factor Xa (apixaban, betrixaban, edoxaban, and rivaroxaban), offer fixed dose administration, waived routine coagulation monitoring, and reduced bleeding risk, and they are currently the mainstay in nonvalvular atrial fibrillation and venous thromboembolism, including splanchnic vein thrombosis.[17,18] However, patients with chronic liver disease were systematically excluded from the pivotal randomized trials of these drugs, and thus, data regarding safety and efficacy are lacking in this population.[19] Drug regulatory agencies including the European Medicines Agency and the Food and Drug Administration allow the use of direct oral anticoagulants without restrictions in patients with Child-Pugh class A. In patients with Child-Pugh class B, rivaroxaban and edoxaban are contraindicated, while dabigatran and apixaban may be used with caution. In Child-Pugh C patients, direct oral anticoagulants are associated with a high risk of bleeding events[20] and they are contraindicated.[2] Antiplatelet therapies aim to reduce or slow down platelet aggregation, thus precluding thrombus formation. They are considered more effective than anticoagulants to prevent or treat arterial thrombosis and atherothrombotic events. There is a myriad of antiplatelet drugs classified according to their mechanism of action: inhibition of the enzyme cyclooxygenase (ie, aspirin), P2Y12 receptor blockade (ie, clopidogrel, prasugrel, and ticagrelor), and inhibition of glycoprotein receptor IIb/IIIa (ie, abciximab, tirofiban, and eptifibatide), being the latter group mainly indicated in acute coronary syndrome. Among LT candidates, it is frequent to find indications for aspirin with or without P2Y12 inhibitors in bearers of coronary stents or as secondary prevention of noncardioembolic ischemic stroke.[21,22] Again, the scientific evidence supporting antiplatelet therapies in patients with advanced liver disease is scarce and their use is considered of high risk, particularly in patients with severe thrombocytopenia.

In this volatile and uncertain context, patients with advanced liver disease may require invasive procedures while awaiting LT such as paracentesis, thoracocentesis, and loco-regional ablative therapies of hepatocellular carcinoma, among others. The correction of coagulation disorders and reversal of anticoagulant and antiplatelet therapies depends on the bleeding risk of each procedure, which may be classified as low or high. Table 1 shows the most frequent procedures performed in patients with cirrhosis in each category.

TABLE 1.

Most frequent invasive procedures performed in patients with liver cirrhosis according to their risk of bleeding

Low-moderate risk of bleeding procedures
• Thoracentesis
• Paracentesis
• Upper gastrointestinal endoscopy with biopsy
• Colonoscopy with biopsy or polypectomy
• Superficial percutaneous drainage
• Phlebotomy
• Liver biopsy
• Bronchoscopy with or without biopsy
High risk of bleeding procedures
• Transarterial chemoembolization
• Transarterial radioembolization
• Upper gastrointestinal endoscopy with variceal banding or sclerotherapy.
• Renal biopsy
• Dental procedures
• Intraabdominal abscess drainage
• Cholecystostomy
• Transjugular percutaneous portosystemic shunt (TIPS)
• Transhepatic biliary drainage
• Percutaneous tumor ablation

In summary, an increasing number of LT candidates receive anticoagulant or antiplatelet therapies while on the waiting list and there are many issues requiring clinical guidance including indications/contraindications, monitoring, and effect reversal in patients requiring invasive procedures. The consensus panel has identified 6 PICO questions containing 18 recommendations, 5 with weak supporting evidence shown in Supplementary Material, SDC, http://links.lww.com/TP/C321, and 13 supported by strong evidence and summarized later:

(1) Should patients listed for LT receive specific therapy to correct coagulation disorders before undergoing invasive procedures? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(1.1) In procedures with low risk of bleeding (Table 1), prophylactic correction of coagulation disorders is not required[23-26] (Recommendation 1B).

(1.2) In urgent procedures associated with a high risk of bleeding (Table 1), prophylactic correction of coagulation disorders is recommended if the platelet count is <50 000/µL or serum fibrinogen is <1.3 g/L[27-29] (Recommendation 1B).

(1.3) The use of fresh frozen plasma to correct coagulation disorders before an invasive procedure should be discouraged in patients with decompensated liver cirrhosis (Recommendation 1B).[27-29]

(2) Should patients listed for LT undergo screening of portal vein thrombosis using abdominal imaging techniques?

(2.1) The screening of portal vein thrombosis in patients awaiting LT should be performed every 3 mo, alternating Doppler ultrasound with dynamic radiological techniques (angio-computed tomography or angio-magnetic resonance)[30-32] (Recommendation 1B).

(3) Should patients at risk of portal vein thrombosis receive thromboprophylaxis while awaiting LT? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(3.1) Workup of thrombophilia cannot be universally recommended although it may prove useful in patients with a family history or analytic suspicion[33] (Recommendation 1A).

(4) Should patients with portal vein thrombosis receive anticoagulation or other interventions to prevent progression of thrombosis while awaiting LT? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(4.1) In patients diagnosed with new onset portal vein thrombosis while awaiting LT, a dynamic liver imaging study (computed tomography or magnetic resonance) should be performed as soon as possible to rule out hepatocellular carcinoma[34,35] (Recommendation 1A).

(4.2) A dynamic imaging study (angio-computed tomography or angio-magnetic resonance) is recommended every 3 mo to evaluate potential progression of thrombosis[30-32] (Recommendation 1B).

(4.3) Anticoagulation should be initiated in all patients with portal vein thrombosis awaiting LT unless otherwise contraindicated[36,37] and maintained until LT irrespective of thrombus resolution[36,38,39] (Recommendation 1A).

(4.4) Low molecular weight heparin is the first line anticoagulation therapy in this setting[40] (Recommendation 1C). The dose of low molecular weight heparin should be tailored according to patient weight, renal function, and platelet count, including an individual risk/benefit evaluation (Recommendation 1A).

(4.5) Vitamin K antagonists may be considered second-line therapies and their effect can be reverted using vitamin K or prothrombin complex[40] (Recommendation 1B).

(4.6) The indication of transjugular intrahepatic portosystemic shunt in patients with portal vein thrombosis should be evaluated within a multidisciplinary team on a case-by-case basis.[41-43] This procedure is contraindicated in patients with Child-Pugh class C (Recommendation 1A).

(4.7) The indication of percutaneous thrombolysis in patients with portal vein thrombosis should be evaluated within a multidisciplinary team on a case-by-case basis[44-46] (Recommendation 1A).

(5) Should patients with portal vein thrombosis requiring anticoagulation receive prophylaxis of variceal bleeding?

(5.1) Primary or secondary prophylaxis of variceal bleeding (beta-blockers or band ligation) should not be delayed in patients with portal vein thrombosis requiring anticoagulation. If needed, variceal bleeding prophylaxis and anticoagulation can be initiated simultaneously[28,47] (Recommendation 1A).

Intraoperative Period

Patients with end-stage liver disease admitted for LT require a specific intraoperative anesthetic management.[48] Anemia, thrombopenia, prolonged prothrombin time, hyperfibrinolysis, and complex rebalanced hemostasis are almost universal, and their reversal is challenging, particularly in cirrhotic patients receiving anticoagulant or antiplatelet therapies. Although the supporting scientific evidence is generally weak, it seems that optimal hematological perioperative management has a positive impact in short-term outcomes. Indeed, a liberal policy of red blood cell transfusion is associated with increased mortality rates after LT.[49,50] A restrictive transfusion policy to target hemoglobin levels between 7 g/dL and 8 g/dL in the absence of massive bleeding or structural cardiomyopathy is currently considered the standard of care.[51] Thrombocytopenia certainly increases the risk of bleeding in cirrhotic patients undergoing elective surgery,[52] although the target threshold to consider platelet transfusion is still a matter of debate. It seems that LT could be safely performed in patients with significant thrombocytopenia, even below 50 000/mm3, without requiring transfusions.[53] Coagulation abnormalities has been traditionally reversed with fresh frozen plasma, but recent evidence suggests that this strategy may not be effective,[54] and could even increase the risk of thromboembolic events in patients with advanced liver disease.[55]

Monitoring hemostasis more closely immediately before and during LT would allow for a more accurate diagnosis and a more rational indication of transfusions. A combination of viscoelastic tests, whenever available, and fibrinogen levels performed at critical intraoperative time points, could add to conventional hematological tests such as prothrombin time, activated partial thromboplastin time, international normalized ratio (INR), and platelet count. Two small randomized trials and several observational single-center studies have reported that viscoelastic test-guided management is able to reduce the need for fresh frozen plasma transfusions, but these were replaced by prothrombin complex and fibrinogen concentrate transfusions as appropriate.[56] Although further high-quality studies are needed, the European Society of anesthesiology guidelines recommend the use of viscoelastic tests to monitor patients undergoing LT, particularly in cases with severe bleeding.[53]

Based on the best available evidence, the consensus group has identified 5 PICO questions and have issued key recommendations including critical aspects such as correction of coagulation abnormalities, reversal of anticoagulant and antiplatelet therapies, intraoperative hemostatic monitoring, and blood product intraoperative transfusion. Five PICO questions containing 17 recommendations were included in this section. Seven recommendations with weak supporting evidence are shown in Supplementary Material, SDC, http://links.lww.com/TP/C321, and 10 recommendations with strong supporting evidence are shown later:

(1) Should cirrhotic patients admitted to the hospital for LT undergo prophylactic correction of altered standard coagulation tests? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(1.1) If thromboelastography is not available, the prophylactic use of tranexamic acid is recommended in patients with Child-Pugh class B or C, unless otherwise contraindicated[57,58] (Recommendation 1B).

(1.2) The prophylactic use of tranexamic acid should be balanced against the risk of thrombosis in the following situations: hypercoagulability, thrombotic events within the previous 6 mo, acute liver failure, grade III-IV portal vein thrombosis and uses of liver disease associated with increased risk of thrombosis (autoimmune hepatitis, primary sclerosing cholangitis…).[58,59] (Recommendation 1B).

(1.3) Therapeutic administration of tranexamic acid should be considered if there is clinical suspicion of fibrinolysis (coagulopathy bleeding with decreasing fibrinogen) or there are compatible changes in the thromboelastography[58,60] (Recommendation 1B).

(2) Should patients admitted to the hospital for LT revert the effect of anticoagulant or antiplatelet therapy before surgery? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(2.1) In patients receiving vitamin K antagonists with an INR ≤3.5, a single dose of intravenous vitamin K (10 mg) should be administered as soon as possible[61-63] since its reversal effect requires time (Recommendation 1C).

(2.2) The use of fresh frozen plasma to reverse the activity of vitamin K antagonists before LT should be discouraged[54,63,64] (Recommendation 1B).

(2.3) In patients receiving direct-acting oral anticoagulants, the specific antidote should be administered before surgery whenever available (idarizumab or andexanet-alpha)[63,65] (Recommendation 1B).

(2.4) In patients receiving antiplatelet therapy, with the exception of aspirin, the drug should be withdrawn as soon as possible[63,66] (Recommendation 1B). Systematic platelet transfusion is not recommended[63,67] (Recommendation 1B).

(3) Should patients undergo hemostatic monitoring intraoperatively during LT?

(3.1) Thromboelastography is recommended for hemostatic management and can be helpful for transfusion guidance[48,53,68,69] (Recommendation 1B). Whenever available, thromboelastography should be performed at baseline (after anesthetic induction), within the first 20–30 min after reperfusion, and at surgical wound closure. Additional determinations may also be required if coagulopathic bleeding is observed (ie, diffuse hemorrhage in the absence of macroscopic clots) or after any intraoperative hemostatic intervention[70] (Recommendation 1B).

(4) Should patients with intraoperative hemostatic abnormalities receive replacement of coagulation factors or platelets? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(4.1) If thromboelastography is available, transfusion should be tailored according to the established algorithms and taking into account the presence of coagulopathic bleeding (Table 2). While normal thromboelastography values are well defined for the general population, normality values in patients with liver cirrhosis may be less strict[69, 71, 72] (Recommendation 1B).

TABLE 2.

Thromboelastography-guided transfusion in liver transplantation

Thromboelastography findings	Clinical interpretation	Therapeutic intervention if diffuse bleeding
Lysis at 30 min < 85%	Hyperfibrinolysis	Tranexamic acid
Maximum clot firmness decreased Low FIBTEMa Normal FIBTEMa	Fibrinogen and platelet deficiency Fibrinogen deficiency Platelet deficiency	Fibrinogenb Platelet
Clotting time prolonged	Coagulation factor deficiency	Fresh frozen plasma vs coagulation factors concentratec

Coagulation abnormalities may be corrected if there is diffuse bleeding.

Thomboelastography test which informs about fibrinogen levels.

Concentrate of fibrinogen or cryoprecipitates, depending on the availability.

Depending on the clinical context.

(4.2) If thromboelastography is not available, transfusion should be tailored to maintain the platelet count >30 000/µL and fibrinogen >1g/L. In patients with active bleeding, thresholds for platelet count and fibrinogen should be set higher (ie, >50 000/µL and >1.3 g/L, respectively)[48,73] (Recommendation 1C)

Posttransplant Period

The immediate postoperative period after LT requires a personalized hemostatic management after a careful balance of the individual risk of bleeding and thrombosis.[74] Thrombotic events after LT may be classified as systemic, which are associated with perioperative general conditions and rebalanced hemostasis, and related to the liver graft, which are strongly influenced by the surgical technique and anatomic variants of donor and recipients. Other external factors may modulate the thrombotic risk, including hydration status, immobilization, cytomegalovirus infection previous transfusions, or administration of prothrombotic agents[75] Noteworthy, the use of mammalian target of rapamycin inhibitors early after LT, particularly sirolimus, was associated with increased risk of hepatic artery thrombosis in a single randomized trial,[76] thus motivating a warning issued by the Food And Drug Administration, and their authorization only beyond day 30 after LT.

The prognostic impact of thrombotic events early after LT is critical. Hepatic artery thrombosis may occur in 3% to 5% of adult patients, usually within the first week after LT, and provokes a rapid deterioration of graft function, thus motivating urgent retransplantation in most cases.[74] Late hepatic artery stenosis or thrombosis is infrequent but can lead to diffuse ischemic cholangiopathy. The incidence of portal vein thrombosis is approximately 2% of LT patients and may require surgery or even retransplantation.[74] Therefore, it is paramount to implement screening strategies to promptly detect these complications and to consider the use of preventive anticoagulant or antiplatelet therapies in patients with known risk factors of portal vein thrombosis (Table 3) or hepatic artery thrombosis (Table 4), respectively. In addition, patients receiving anticoagulant or antiplatelet therapies before LT may need to resume these therapies in a timely manner depending on their original indication.

TABLE 3.

Risk factors of portal vein thrombosis

Risk factors before liver transplantation
• Low platelet count (<70 000/mm3).
• History of variceal bleeding.
• Hepatofugal portal flow or slow portal flow (<15 cm/s) on Doppler ultrasound.
• Thrombophilic disorders in the recipient
Risk factors after liver transplantation
• History of portal vein thrombosis before liver transplantation.
• Slow portal flow (after reperfusion) defined as <1300 mL/min or < 65 mL/min/100g.
• Partial thrombectomy or vein intimal layer lesion during thrombectomy
• Nonphysiological portal vein inflow reconstruction
• Thrombophilic disorders in the recipient

TABLE 4.

Risk factors of hepatic artery thrombosis

• Complex anastomosis (reduced artery diameter, discordant diameters between arteries of donor and recipient).

• Bench arterial reconstruction/use of vascular grafts.

• Arterial flow < 100 mL/min (after inflow modulation).

• Endothelial injury or thrombectomy.

• Reoperation including hepatic artery anastomosis reconstruction.

• Prolonged cold ischemia/operative times.

• Increased blood transfusion requirements.

• Old donor (>70 y old) with atheromatosis.

• Thombophilia condition of the recipient.

• Familial amyloid polyneuropathy.

In this section, the consensus panel has identified 10 PICO questions, and have issued 33 key recommendations to provide guidance on screening and treatment of graft-related vascular complications, management of bleeding, and indications of anticoagulant and antiplatelet therapies after LT either with prophylactic or therapeutic purposes, including indications as adjuvant therapies after surgical or radiological procedures. Twenty-two recommendations with weak supporting evidence are shown in Supplementary Material, SDC, http://links.lww.com/TP/C321, and 11 recommendations with strong scientific evidence are shown later:

(1) Should patients undergo screening of vascular complications of the liver graft immediately after transplantation?

(1.1) Screening of liver graft vascular complications after transplantation is mandatory. A Doppler ultrasound should be performed by a trained specialist within the first 24 h after surgery and at any time point afterwards if there is graft dysfunction or an otherwise unexplained alteration of blood liver tests[77-79] (Recommendation 1B).

(1.2) Hemostatic surveillance with routine coagulation tests is sufficient for most patients after LT. However, in critically ill patients with surgical complications, the use of thromboelastography could be helpful[80] (Recommendation 1B).

(2) Should patients receive thromboprophylaxis after LT to prevent venous thromboembolism? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321)

(2.1) Perioperative venous thromboembolism prophylaxis with early walking and compression devices is universally recommended[81-84] (Recommendation 1A).

(2.2) Pharmacological thromboprophylaxis should be prolonged between 10 and 24 d after hospital discharge according to the individual risk of thrombosis in each patient[82,83,85,86] (Recommendation 1B).

(3) Should patients receive specific therapy to correct altered coagulation tests immediately after LT?

(3.1) In the absence of active bleeding, an increased INR or a reduced platelet count may not require any hemostatic intervention, not even before procedures associated with low risk of bleeding[28,29,66] (Recommendation 1B).

(3.2) The universal prescription of vitamin K after LT does not provide any clinical benefit. However, it can be considered in patients with chronic cholestasis or malabsorptive conditions. The initial intravenous dose in such cases would be 10 mg every other day[87] (Recommendation 1B).

(4) Should patients receive specific therapy to prevent portal vein or hepatic vein thrombosis after LT? (Continued in Supplementary Material, SDC, http://links.lww.com/TP/C321).

(4.1) In the absence of coagulopathy, liver graft dysfunction or low platelet count (<30 000–50 000/µL), patients with risk factors of portal vein thrombosis (Table 2 from the main document) should receive therapeutic low molecular weight heparin (ie, 1 mg/kg) started within the first 24 h after surgery[88,89] (Recommendation 1B).

(5) Should patients with postoperative bleeding receive replacement of coagulation factors or platelets to facilitate hemostasis?

(5.1) Whole blood transfusion should aim to maintain hemoglobin around 8 g/dL. Platelets should be maintained at >50 000/µL and serum fibrinogen >1.3 g/L[90,91] (Recommendation 1B).

(5.2) Identification and correction of the cause of bleeding is paramount and massive transfusion should be discouraged, as it is associated with increased mortality[92,93] (Recommendation 1B).

(6) Should patients with pretransplant anticoagulation resume this therapy after LT?

(6.1) In patients at high risk of thrombosis, anticoagulant therapy should be resumed within the first 24 h after LT using the same drug whenever possible. Bridging therapy with low molecular weight heparin can alternatively be considered[81,94-96] (Recommendation 1B).

(6.2) In patients at intermediate or low risk of thrombosis, anticoagulant therapy can be delayed 48–72 h after LT without bridging therapy with low molecular weight heparin[81,94-96] (Recommendation 1B).

CONCLUSIONS

The present multidisciplinary consensus statement addresses 21 clinically relevant PICO questions with key recommendations to allow more objective and homogeneous clinical decision making regarding the use of anticoagulant and antiplatelet therapies in patients with advanced liver disease before and after LT. Although agreement among the members of the expert panel was high for all recommendations, the quality of the evidence was moderate or low on average, thus highlighting the need for randomized controlled trials focused on patients with cirrhosis and LT receiving antiplatelet or anticoagulant therapies.

ACKNOWLEDGMENTS

The consensus group greatly appreciates the continuous support of Paloma Bellés from AOPC for her contribution to secretariat and logistic aspects. We also acknowledge Ms. Donna Pringle for professional English language polishing.