Intravascular thrombosis and thromboembolism during liver transplantation: Antifibrinolytic therapy implicated?


  • Michael A.E. Ramsay,

    Corresponding author
    1. Department of Anesthesiology and Pain Management, Baylor Regional Transplant Institute, Baylor University Medical Center, Dallas, TX
    • Department of Anesthesiology and Pain Management, Baylor University Medical Center, 3500 Gaston Avenue, Dallas, Tex., 75246
    Search for more papers by this author
    • Telephone: 214-820-4090; fax: 214-820-6612

  • Henry B. Randall,

    1. Department of Transplant Surgery, Baylor Regional Transplant Institute, Baylor University Medical Center, Dallas, TX
    Search for more papers by this author
  • Elizabeth C. Burton

    1. Department of Pathology and Laboratory Medicine, Baylor Regional Transplant Institute, Baylor University Medical Center, Dallas, TX
    Search for more papers by this author


This case report describes a patient who underwent orthotopic liver transplantation and developed extensive hyperacute venous and arterial intravascular thromboses and thromboemboli intraoperatively. The patient was receiving antifibrinolytic therapy with aprotinin. The safety of routine aprotinin therapy in liver transplantation is examined. The value of the thrombelastograph (TEG) as a qualitative assessment of the coagulation system is emphasized. (Liver Transpl 2004;10:310–314.)

The routine use of the antifibrinolytic agent aprotinin has been advocated during liver transplantation following the successful outcome of a multicenter trial.1 We describe a patient who during an orthotopic liver transplant, while being administered an infusion of aprotinin, developed extensive hyperacute venous and systemic thromboses and thromboemboli. The etiology and prevention of this catastrophic event are discussed.


TEG, thrombelastograph; KIU, kallikrein inhibiting units.

Case Report

The patient was a 48-year-old man, 69 kg, 71 in., with end-stage liver disease secondary to hepatitis C cirrhosis. He had a history of esophageal variceal bleeding, coagulopathy, and mild renal insufficiency. At the time of transplantation his serum creatinine was 1.6 mg/L, serum potassium was 5.0 mEq/L, and the intraoperative laboratory coagulation studies indicated evidence of coagulopathy and fibrinolysis (Table 1). Initial thrombelastograph (TEG) data demonstrated a normal clot formation with no evidence of coagulopathy or fibrinolysis.

Table 1. Laboratory Coagulation Data Obtained During the Liver Transplant
 Normal RangeStart (0 min)Anhepatic (75 min)Reperfusion (135 min)
  1. NOTE: Coagulation data were accumulated at the start of surgery, at the time the patient was anhepatic, and again following reperfusion of the new graft. Timeframes were measured from the start of the procedure.

  2. Abbreviations: PT, prothrombin time; aPTT, activated partial thromboplastin time; FSP, fibrin split products; D-dimer, dimerized plasmin fragment D; sec, seconds.

Platelets, 103/uL(140–440)773939
PT, sec(9.5–12.0)14.320.317.1
aPTT, sec(22–33)48.8> 212127
Fibrinogen, mg/dL(200–400)1907487
Thrombin, sec(16–22)31.540.840.8
FSP, mg/dL(< 5)> 20> 20> 20
D-dimer, mg/L(< 0.2)

To prevent hyperfibrinolysis and reduce blood transfusion requirements, at induction of anesthesia a “regular dose” aprotinin infusion was started at 2×106 kallikrein inhibiting units (KIU) as a loading dose given intravenously over 20 minutes followed by a continuous infusion of 0.5×106 KIU/h as described by Porte et al.1 The liver transplant surgery progressed uneventfully during explantation of the diseased liver. Venovenous bypass, without heparin, was utilized during the anhepatic phase with bypass flow rates recorded at 2–2.3 L/min during portal and femoral bypass and 1–1.5 L/min after portal bypass was discontinued. Reperfusion of the new graft was also uneventful. Venovenous bypass was discontinued after a total time of 97 minutes, and the blood contained in the circuit was transfused back into the patient. The bypass circuit, including the centrifugal pump head, was observed to be free of blood clot or any obvious fibrin deposit.

Coagulation studies were performed at the start of surgery a second time when the patient was anhepatic, and again following reperfusion of the new graft (0, 75, and 135 minutes from the start of the procedure). All the results were compatible with a severe coagulopathy together with fibrinolysis (Table 1). TEG data, however, still demonstrated a normal coagulation pattern.

Six minutes after reperfusion, the patient became hypotensive and bradycardic, and the right heart filling pressures rose acutely. The patient rapidly went into asystole and could not be resuscitated despite full therapy.

Autopsy findings were significant for multiple arterial and venous hyperacute thromboses. Extensive bilateral pulmonary thromboemboli were found, together with intracardiac clot in the right atrium, right ventricle, and inferior vena cava. Much of this clot appeared fresh and certainly some of this clot appeared to be seeding around the pulmonary artery catheter. No septal defects were found in the heart, but thromboses were also noted in the aorta, the right and posterior descending coronaries, bilateral iliac, internal carotid, renal, and cerebral arteries. See Figures 1 through 6. No evidence of a remote source of clot formation was demonstrable.

Figure 1.

Extensive clot forming around a pulmonary artery catheter.

Figure 2.

Bilateral pulmonary emboli.

Figure 3.

Cerebral arterial thrombosis.

Figure 4.

Vena cava with both the superior and inferior anastomoses within the liver.

Figure 5.

Hematoxylin and eosin stain from a section of the posterior descending coronary artery with thrombosis taken at a 4x magnification.

Figure 6.

Hematoxylin and eosin stain from a section of the right middle cerebral artery taken at a 4x magnification.

During the procedure, the following blood components had been transfused: 4 units of packed erythrocytes, 3 units of conventional fresh frozen plasma, and 3 units of salvaged and washed erythrocytes. At the time of the patient's demise, a unit of platelets was being transfused, as was a transfusion of 18 units of cryoprecipitate through a separate intravenous cannula. Serum ionized calcium levels were maintained by the intermittent administration of calcium chloride in 1-gram aliquots as necessary.

Paraffin-embedded liver tissue from both the donor liver and native liver was submitted for molecular analysis. For the Factor V Leiden gene molecular analysis, the results were negative for the Factor V gene mutation on both the donor and native liver tissue. For the prothrombin gene 20210 mutation, again, the results were negative (homozygous normal) on both the donor and native liver tissue.


Liver cirrhosis is often associated with hyperfibrinolysis, and this may be more severe after reperfusion of the graft.2 Epsilon-aminocaproic acid has been used frequently during liver transplantation and has been demonstrated to reverse TEG evidence of hyperfibrinolysis.3 The usual custom in this center is to administer 5 grams of epsilon-aminocaproic acid following reperfusion only if fibrinolysis is noted. The routine use of antifibrinolytic drugs is not included in our management protocol. This patient had very large varices and was considered to be at high risk for a massive blood transfusion. Therefore, in this particular patient aprotinin was selected as the antifibrinolytic therapy because the laboratory data indicated that significant fibrinolysis was taking place. In addition, with elevated serum potassium levels at the start of the procedure, a more aggressive attempt to reduce blood transfusion requirements was instigated. In this manner, the potassium load from the blood transfusions could be avoided. A regular-dose aprotinin infusion has been reported to reduce both blood loss and the requirement for blood transfusion in patients undergoing liver transplantation and to be free of complications related to hypercoagulation.1, 4, 5 It is noteworthy that the TEG recording did not reveal significant fibrinolysis, hypercoagulation, or coagulopathy despite the abnormal laboratory data.

Hemostasis is the result of the fine balance between the enzymic coagulation cascade and the opposing fibrinolytic cascade. Tilting that balance to profound coagulation is a concern when using powerful antifibrinolytic agents. Nevertheless, the routine use of antifibrinolytic agents has demonstrated reduced blood loss and a reduced need for erythrocyte transfusion during liver transplant surgery.2, 3, 4, 5 However, many case reports have associated thrombosis and thromboembolism with antifibrinolytic administration during liver transplantation and open-heart surgery.6–17 Conversely, studies performed to date do not substantiate this concern.18–22 The anticoagulant, as opposed to a procoagulant, effect of aprotinin has been demonstrated in patients undergoing liver transplantation.22 Clearly, a controversy exists regarding the safety of routine aprotinin use in liver transplant surgery.23 Consequently, aprotinin is rarely used in this center's liver transplantation program. That rare case materializes when a patient presents a high expectation of needing a massive blood transfusion.

The association of this devastating complication with the use of antifibrinolytic agents cannot be proven but must be questioned. More than a century ago, Rudolph Virchow identified the three factors responsible for vascular thrombosis: vessel injury, alterations in blood flow, and the coagulation state. These conditions exist during liver transplantation but rarely lead to pathological thrombosis. These reports of thrombotic episodes raise an important question: the clinician should contemplate whether individual patients have an increased risk for developing a hypercoagulable state. The hypercoagulable condition might exist because of inherited or acquired defects that predispose individuals to clot if coagulation is weighted in one direction by using antifibrinolytics.

Inherited hypercoagulability defects, including protein C deficiency, have been well described and may have been a factor in this patient. However, aprotinin has so far not been shown to be associated with a hypercoagulable state in patients with any of these defects.24, 25 Platelets also play a key role in the coagulation process; genetic variations may occur, creating platelet receptor polymorphisms.26, 27

In this report, despite severely deranged laboratory data, the TEG did not reveal any problems with the coagulation system. The TEG provides a qualitative expression of the status of the coagulation system and perhaps should have been relied on more heavily in the management of this patient's coagulation state.

The use of powerful antifibrinolytic agents is brought into question by this dreadful outcome. Although the risk of thrombosis and thromboembolism has not been shown to be significant in controlled trials, the number of published case reports of this complication continues to rise. This case report describes the spontaneous formation of fresh clot in both the venous and arterial circulations in a patient undergoing liver transplantation. Venous thromboembolism and hepatic artery thrombosis have been noted rarely in our practice, but massive arterial and venous hyperacute thrombosis have never been observed in more than 2,500 patients undergoing liver transplantation. Although the administration of aprotinin in this patient cannot be confirmed as the causative agent, it may be prudent to base the decision to use this drug on the careful analysis of TEG data rather than only on sources from laboratory data or as part of a routine protocol.28