Antifibrinolytic Therapy and Pulmonary Thromboembolism During Orthotopic Liver Transplantation

Abstract

To The Editor: We read with great interest the article by O’Connor et al. (1) regarding pulmonary thromboembolism during orthotopic liver transplantation (OLT). Also, we agree that antifibrinolytic drugs may play a role in unwanted clot formation during OLT; however, we feel some additional factors not mentioned in these case reports may have also contributed to what was observed intraoperatively. In Case 1, the patient did not exhibit a significant coagulopathy as evidenced by her preoperative laboratory data. In addition, no mention was made of obtaining a baseline intraoperative thromboelastogram at the start of OLT. Nevertheless, both fresh-frozen plasma and aprotinin were subsequently administered. Although we share the concern of the authors regarding the potential for blood loss during redo OLT, as in this case, empiric use of any antifibrinolytic agent without significant preexisting coagulopathy or evidence of fibrinolysis through coagulation monitoring may be difficult to justify knowing the potential for thrombotic complications. In Case 2, venovenous bypass (VVB) was used to facilitate removal of the recipient’s diseased liver. Again, the authors make no mention of obtaining a baseline thromboelastogram, although one can assume that with a prothrombin time of 28.4 s and moderate thrombocytopenia, the thromboelastogram would have been abnormal. Subsequently, large dose ε-aminocaproic acid therapy was initiated. In contrast to this, Kang (2) recommended using small dose ε-aminocaproic acid to treat demonstrable fibrinolysis and cautioned against its prophylactic use to avoid potential thrombotic complications. Of interest, this patient had a possible preexisting clot in the right internal jugular vein. With antifibrinolytic therapy combined with presumed fresh-frozen plasma and platelet administration for “persistent intraabdominal bleeding,” the potential for further clot propagation existed. Whether a preexisting clot extended itself into the right heart or was dislodged from the right internal jugular vein or innominate vein during VVB, one can only speculate. More importantly though, the authors do not mention whether once initiated, adequate VVB flow was established and maintained. Although VVB tubing is heparin bonded, in low flow states of <1 L/min, fibrin deposition within the VVB circuit is possible (3–4). This filamentous material can then migrate into the right heart and pulmonary circulation where further clot formation can occur. In the face of right heart dysfunction and increased central venous pressure secondary to an acute rise in pulmonary artery pressures because of thromboembolism, we doubt “normal” VVB flow could be maintained. Thus, a viscous cycle of clot formation, resulting in a low flow state during VVB, followed by additional clot formation, may have been established in this patient. Brian M. Parker MD Samuel A. Irefin MD