Case Reports  |   May 2002
Recurrent Pulmonary Embolism during Liver Transplantation: Possible Role of Hepatitis B Immune Globulin As a Causative Agent
Author Affiliations & Notes
  • Weng H. Wong, M.Med. (Anesth.)
  • Michelle Braunfeld, M.D.
  • Philip Levin, M.D.
  • *Fellow, Department of Anesthesiology, University of California–Los Angeles School of Medicine, Los Angeles, California. Current position: Associate Consultant, Department of Anesthesiology, National University Hospital, Singapore. †Associate Professor, ‡Assistant Professor, Department of Anesthesiology, University of California–Los Angeles School of Medicine.
  • Received from the Department of Anesthesiology, University of California–Los Angeles School of Medicine, Los Angeles, California.
Article Information
Case Reports
Case Reports   |   May 2002
Recurrent Pulmonary Embolism during Liver Transplantation: Possible Role of Hepatitis B Immune Globulin As a Causative Agent
Anesthesiology 5 2002, Vol.96, 1261-1263. doi:
Anesthesiology 5 2002, Vol.96, 1261-1263. doi:
MASSIVE pulmonary embolism (PE), although an uncommon complication in liver transplant surgery is associated with a high mortality rate. Five of nine patients reported to have massive PE in the past 12 yr died from this complication. 1,2 We report a case of PE that occurred in the same patient during two consecutive liver transplant procedures in association with the use of an antifibrinolytic and hepatitis B immune globulin (HBIG).
Case Report
A 52-yr-old man was admitted for orthotopic liver transplantation because of end-stage liver disease secondary to hepatitis B and C and recurrent hepatocellular carcinoma. Two years before admission, he had intractable ascites and severe portal hypertension for which a transjugular intrahepatic portosystemic shunt procedure was performed.
Previous operations for repair of inguinal and umbilical hernias as well as a splenectomy 7 yr before the current admission for thrombocytopenia were without complications. Physical examination revealed a well-nourished patient with no abnormalities of the cardiovascular or respiratory systems. An echocardiogram showed normal cardiac function with an ejection fraction between 55 and 60% and no discrete vegetations, mural thrombi, or intracardiac masses. Laboratory results included prothrombin time of 17.6 s, international normalized ratio (INR) of 1.7, partial thromboplastin time of 31.1 s, fibrinogen of 128 mg/dl, and platelet count of 128 × 109/l.
Anesthesia was induced in a rapid-sequence manner with thiopental and succinylcholine and was maintained with isoflurane and cisatracurium. A loading dose of 5 g ε-aminocaproic acid followed by an infusion of 1 g/h was started as per our usual practice. A CaCl2infusion at 1 g/h was also started. A right radial arterial catheter was placed, and two 9-French ports (Arrow, Reading, PA) were inserted in the right internal jugular vein through which a 7.5-French pulmonary artery (PA) catheter (Baxter, Oakland, CA) was inserted in one without difficulty. Initial blood pressure was 140/70 mmHg; heart rate was 75 beats/min; pulmonary arterial pressure was 37/25 mmHg; cardiac output was 6.4 l/min, and central venous pressure was 16 mmHg. Venovenous bypass with a centrifugal pump (Biomedicus; Medtronic, Eden Prairie, MN) and nonheparinized tubing was started at a flow of 1.8–2 l/min. The surgery proceeded uneventfully until the liver was removed and a 2-ml test dose of HBIG (Nabi, Boca Raton, FL) was administered via  one of the internal jugular ports. Blood pressure decreased from 115/54 mmHg to a low of 58/20 mmHg within 10 min after HBIG administration. Simultaneously, pulmonary arterial pressure increased from 37/25 to 50/32 mmHg, and end-tidal carbon dioxide (ETco2) decreased to 18 mmHg. Cyanosis and venous engorgement of the upper torso were observed. HBIG and ε-aminocaproic acid infusions were discontinued immediately. A transesophageal echocardiographic probe inserted to confirm a clinical diagnosis of PE showed a large thrombus in the right atrium. Resuscitation with intravenous ephedrine and epinephrine was instituted with return to baseline level hemodynamics and resolution of thrombus after 30 min. The new graft was transplanted successfully. The patient was transfused with a total of 16 units of erythrocytes, 18 units of fresh frozen plasma, 20 units of cryoprecipitate, and 10 units of platelets. The coagulation profile at the end of surgery was prothrombin time of 21.3 s, INR of 2.2, platelet count of 94 × 109/l, and fibrinogen of 73 mg/dl.
Unfortunately, the liver graft did not function well postoperatively. Despite fresh frozen plasma and platelet infusions, the patient remained coagulopathic. The results of postoperative radiologic imaging and ultrasound studies excluded any clots in the heart, extremities, or the vena cavae. Four days after the first transplant, the patient underwent subsequent transplantation for primary nonfunction.
The patient arrived in the operating room with the same invasive monitors from the previous operation in place. Induction with isoflurane, midazolam, and fentanyl was uneventful. Variables were as follows: blood pressure, 150/70 mmHg; pulmonary arterial pressure, 35/18 mmHg; central venous pressure, 8 mmHg; and cardiac output, 12.6 l/min. The surgical team believed the first PE to be caused by migration of a thrombus adherent to the patient's transjugular intrahepatic portosystemic shunt. After discussion with the surgeon, it was decided that an antifibrinolytic should be used in anticipation of bleeding. A loading dose of 2 g ε-aminocaproic acid and infusion of 1.5 g/h were administered. The dissection phase of the operation was uneventful. The patient was again placed on venovenous bypass and was pretreated with hydrocortisone, famotidine, and diphenhydramine in anticipation of HBIG administration. Ten thousand units of HBIG was administered via  the right internal jugular port over 15 min during the anhepatic period as the suprahepatic vena cava was being reanastomosed. Blood pressure was then noted to decrease precipitously from 110/55 mmHg to 60/33 mmHg soon after HBIG infusion was completed. Pulmonary arterial pressure and central venous pressure were increased at 63/33 mmHg and 40 mmHg, respectively. ETco2was not recordable. PE was again suspected, and a transesophageal echocardiographic probe was inserted, showing large thrombi in the right atrium and right ventricle. The patient was resuscitated with 35 mg epinephrine, 12 mg norepinephrine, 3 g CaCl2, 0.4 mg atropine, 100 ml NaHCO3, and 200 mg lidocaine. Despite an infusion of epinephrine, the patient remained hypotensive, acidotic, and coagulopathic. Laboratory results in the intensive care unit showed prothrombin time greater than 100 s, INR greater than 10, platelet count of 50 × 109/l, pH of 7.17, partial pressure of oxygen (Po2) of 108, partial pressure of carbon dioxide (Pco2) of 40, and base excess of −13. Two hours after arrival in the intensive care unit, pulseless electrical activity developed in the patient, and he died after unsuccessful resuscitation efforts.
We report an unusual case of PE occurring in the same patient during two consecutive liver transplant procedures. Our patient was fairly healthy with normal cardiopulmonary function before the first transplant. An echocardiogram did not reveal any thrombi or vegetations. He had no history of deep vein thrombus or cerebrovascular accidents and was mildly coagulopathic preoperatively. The use of a PA catheter has been documented to be associated with massive thrombosis. 3 Trauma to the endothelium with a dilator and the use of a protective sheath may trigger clot formation. We did not encounter any difficulty when inserting the ports or PA catheter through the right internal jugular vein. Furthermore, ultrasound study performed in the intensive care unit with the PA catheter in place did not reveal any blood clots in the vena cavae or the heart.
The advantages of using antifibrinolytics must always be weighed against the risk of thrombosis, particularly in patients with known thrombotic tendency, such as in Budd-Chiari syndrome, disseminated intravascular coagulopathy (DIC), and antiphospholipid syndrome. Preoperatively, there was no evidence to suggest that our patient was in a thrombotic state. He had an INR of 1.7 and a prothrombin time of 17.6 s. At the time of the first PE, he had received 6 units of fresh frozen plasma but no platelets or cryoprecipitate. Nonheparinized venovenous bypass with a heat exchanger was used during both surgeries. There were no clots seen through the clear tubing or in the heat exchanger, and flow was well-maintained until the thrombus obstructed right ventricular outflow. PE occurred both times in this patient during the anhepatic phase when a fibrinolytic state was anticipated.
Violi et al.  4 discovered that patients with moderate or severe hepatic insufficiency had higher values of endotoxemia. Endotoxemia promotes tumor necrosis factor synthesis, which activates the extrinsic clotting pathway. This factor may have played a role in the second episode of PE when the patient was severely ill for 4 days with a nonfunctioning graft.
Administration of HBIG has a close temporal association with the onset of PE on both occasions in our patient. HBIG is a solution of immunoglobulin containing antibodies to hepatitis B surface antigen. It is prepared from plasma donated by individuals with high antibody titers and is formulated in sodium chloride, glycine, and polysorbate 80. Adverse effects associated with the use of intravenous immunoglobulin include headache, low-grade fever, transient hypotension, and deep vein thrombosis. 5 A high infusion rate has been correlated with a higher incidence of adverse reactions. HBIG was administered to this patient to reduce the hepatitis B viral load before the donor liver was reperfused. Kang et al.  1 reported a case of PE during orthotopic liver transplantation when HBIG was administered during the anhepatic phase during venovenous bypass. They suggested that antibody–antigen complexes formed may have activated the intrinsic pathway. Results of an experiment conducted in rabbits by Nakamura et al.  6 suggested that antibody–antigen complexes in the microcirculation may initiate activation of platelets and neutrophils with release of mediators responsible for triggering DIC. Fibrin thrombi first appeared in the rabbits’ organs, including the lungs, 2 h after challenge with an antiserum and antigen. The time frame may have been compressed in our patient in the presence of an antifibrinolytic, endotoxemia, and DIC. The ferritin–antiferritin complexes in this animal model were phagocytized by Kupffer cells, splenic macrophages, and neutrophils. The authors postulated that saturation of capacity of the reticuloendothelial system to remove circulating immune complexes promoted the progression of DIC. This may explain the paradoxical occurrence of PE during the fibrinolytic, anhepatic phase of the operation in our patient. Although no thrombi were found at autopsy, the patient's lungs were noted to be heavy, a finding consistent with descriptions of similar patients in whom PE developed during orthotopic liver transplantation and who died. 7 No macroscopic pulmonary emboli were detected in these patients, but the lungs were heavy, rubbery, and congested. Numerous platelet aggregates were also found occluding the capillaries. It is possible that prompt termination of HBIG and antifibrinolytic infusion in the current patient allowed natural fibrinolytics, such as tissue plasminogen activator and plasmin, to act on the clot, resulting in its resolution intraoperatively.
Heparin-induced thrombocytopenia type II, in which immunoglobulin G antibodies bind to platelet factor 4 heparin complexes, is a clinical example of antibody–antigen complex–induced platelet aggregation leading to release of procoagulant mediators from platelets. 8 Previous exposure to heparin leads to formation of immunoglobulin G antibodies, and further administration of heparin may result in this phenomenon. Acute systemic reactions, such as fever, chills, and rigors, to an initial dose of heparin may be an early indicator of heparin-induced thrombocytopenia type II. 9 Similar nonspecific symptoms, such as myalgia and arthralgia, have been described in patients given HBIG and are attributed to immune precipitation and formation of immune complexes. 10 Whether HBIG–hepatitis B antigen complexes are able to initiate platelet activation and aggregation leading to thrombi formation that is similar to the animal model of Nakamura and the clinical example of heparin-induced thrombocytopenia type II is unclear. Heparin-induced thrombocytopenia type II is often associated with thrombosis and embolism of the venous system, arterial system, or both. Stroke, myocardial and mesenteric infarction, and PE have been reported in patients undergoing cardiac surgery when this phenomenon occurs.
In conclusion, we find that a combination of factors, including the use of a PA catheter, endotoxemia and DIC, the use of an antifibrinolytic, and HBIG may have induced a thrombotic state in this patient. Further studies on the use of HBIG in liver transplant patients with similar predisposing factors may ensure its safer use in this group of patients.
The authors thank C. Philip Larson, M.D. (Professor, Department of Anesthesiology, University of California–Los Angeles, Los Angeles, California), for reviewing the manuscript and Daisy Valdez (Administrative Assistant, Department of Anesthesiology, University of California–Los Angeles) for her assistance in the preparation of this manuscript.
Navalgund AA, Kang Y, Sarner JB, Jahr JS, Gieraerts R: Massive pulmonary thromboembolism during liver transplantation. Anesth Analg 1988; 67: 400–2Navalgund, AA Kang, Y Sarner, JB Jahr, JS Gieraerts, R
Sopher M, Braunfeld M, Shackleton C, Busuttil RW, Sangwan S, Csete M: Fatal pulmonary embolism during liver transplantation. A nesthesiology 1997; 87: 429–32Sopher, M Braunfeld, M Shackleton, C Busuttil, RW Sangwan, S Csete, M
Chastre J, Cornud F, Bouchama A, Viau F, Benacerraf R, Gibert C: Thrombosis as a complication of pulmonary artery catheterization via the internal jugular vein. Prospective evaluation by phlebography. N Engl J Med 1982; 306: 278–81Chastre, J Cornud, F Bouchama, A Viau, F Benacerraf, R Gibert, C
Violi F, Ferro D, Basili S, Saliola M, Quintarelli C, Alessandri C, Cordova C: Association between low-grade disseminated intravascular coagulation and endotoxemia in patients with liver cirrhosis. Gastroenterology 1995; 109: 531–9Violi, F Ferro, D Basili, S Saliola, M Quintarelli, C Alessandri, C Cordova, C
Sherer Y, Levy Y, Langevitz P, Rauova L, Fabrizzi F, Shoenfeld Y: Adverse effects of intravenous immunoglobulin therapy in 56 patients with autoimmune diseases. Pharmacology 2001; 62: 133–7Sherer, Y Levy, Y Langevitz, P Rauova, L Fabrizzi, F Shoenfeld, Y
Nakamura S, Shimokama T, Yoshikawa Y, Watanabe T: Immune complex-induced disseminated intravascular coagulation (DIC): An experimental study. Acta Pathol Jpn 1990; 40: 476–85Nakamura, S Shimokama, T Yoshikawa, Y Watanabe, T
Sankey EA, Crow J, Mallett SV, Alcock RJ, More L, Burroughs AK, Rolles K: Pulmonary platelet aggregates: Possible cause of sudden perioperative death in adults undergoing liver transplantation. J Clin Pathol 1993; 46: 222–7Sankey, EA Crow, J Mallett, SV Alcock, RJ More, L Burroughs, AK Rolles, K
Suh J-S, Malik MI, Aster RH, Visentin GP: Characterization of the humoral immune response in heparin-induced thrombocytopenia. Am J Hematol 1997; 54: 196–201Suh, J-S Malik, MI Aster, RH Visentin, GP
Warkentin TE, Soutar RL, Panju A, Ginsberg JS: Acute systemic reactions to intravenous bolus heparin therapy: Characterization and relationship to heparin-induced thrombocytopenia. Immunohematology 1997; 160a: 80Warkentin, TE Soutar, RL Panju, A Ginsberg, JS
Shouval D, Samuel D: Hepatitis B immune globulin to prevent hepatitis B virus graft reinfection following liver transplantation: A concise review. Hepatology 2000; 32: 1189–95Shouval, D Samuel, D