Intraoperative transesophageal echocardiography reveals thrombotic stenosis of inferior vena cava during orthotopic liver transplantation


Address reprint requests to Ivan Kangrga, M.D., Ph.D., Department of Anesthesiology, Washington University School of Medicine, 660 South Euclid Avenue, Box 8054, St. Louis, MO 63110. Telephone: 314-747-2858; FAX: 314-362-1185; E-mail:


hepatic vein


international normalized ratio


inferior vena cava


orthotopic liver transplantation


right atrium


right ventricle


transesophageal echocardiography


Transesophageal echocardiography (TEE) is increasingly being used for hemodynamic monitoring in orthotopic liver transplantation (OLT).[1] Importantly, intraoperative TEE may provide additional critical information such as the identification of intracardiac thrombi[2] or complications related to transjugular intrahepatic portosystemic shunts.[3] We present a patient with postreperfusion graft congestion in whom TEE revealed hemodynamically significant thrombotic stenosis of the inferior vena cava (IVC). This finding influenced further patient management.

A 34-year-old woman with a large polycystic liver (estimated volume = 10 L; Fig. 1A,B) underwent OLT via the piggyback technique with a temporary portocaval shunt. The initial intraoperative international normalized ratio (INR) was 1.6, but thromboelastography showed an elevated maximal amplitude of 76.5 mm (normal = 55-74 mm) that was consistent with a hypercoagulable tendency.[4] The dissection and anhepatic stage proceeded uneventfully. Compression of the right heart by the enlarged liver was relieved after the completion of the native hepatectomy (Fig. 1C,D). After reperfusion, the surgical team reported persistent graft congestion. To determine the etiology of the congestion, we first assessed the intravascular volume status and cardiac function. TEE revealed a normal chamber size and hyperdynamic biventricular systolic function. The central venous pressure was 9 mm Hg. A possible infracardiac source of congestion was then sought. A TEE probe was redirected to visualize the IVC via a modified bicaval view.[5] The IVC was grossly dilated and tapered acutely toward an area of marked stenosis just inferior to the right atrium (RA; Fig. 2A). Directing the probe superiorly revealed a protruding intracaval mass at the cavoatrial junction (Fig. 2B). The irregular borders were most consistent with a mural thrombus. Highly turbulent flow on color Doppler imaging was indicative of a hemodynamically significant lesion immediately superior to hepatic veins (HVs; Fig. 2C). To confirm this observation, sterile pressure tubing was passed onto the surgical field, and the pressure in the infrahepatic IVC was measured directly; this demonstrated a high pressure gradient (9 mm Hg) between the IVC and superior vena cava.

Figure 1.

Preoperative computed tomography of the abdomen with contrast showed an enlarged polycystic liver (A) occupying the abdominal cavity and (B) compressing the right heart. (C) Intraoperative TEE showed RA and RV compression by the enlarged liver (the arrow indicates a liver cyst compressing the RA). The left heart appeared normal. (D) After hepatectomy, the right heart was re-expanded.

Figure 2.

(A) A grossly dilated retrohepatic IVC tapered acutely toward the RA. (B) A mass consistent with a mural thrombus (star) was seen at the cavoatrial junction and protruded into the RA. (C) Color Doppler imaging of turbulent flow through the IVC narrowing indicated a pressure gradient between the IVC and the RA. The stenosis was immediately superior to an HV. (D) Postoperative contrast venography showed a narrowing of the IVC (short arrow; diameter = 1.1 cm) that was cranial to an area of dilatation (long arrow; diameter = 2 cm, normal ≤ 1.7 cm) and immediately caudal to the cavoatrial junction (star).

On the basis of these findings, heparin anticoagulation was instituted and maintained throughout the surgery and postoperatively. We considered an endovascular intervention to relieve the IVC stenosis but abandoned this approach because of a fear of dislodging the thrombus into the right heart or covering HVs on account of their proximity to the lesion. We proceeded with anticoagulation and close follow-up.

The intraoperative course was complicated by thrombosis of the hepatic artery, which necessitated takedown of the anastomosis and revision. The remainder of the surgery was uneventful. No blood products were transfused, and the patient was discharged to the intensive care unit in stable condition.

On postoperative day 1, visceral Doppler sonography revealed antegrade flows and normal resistive indices in hepatic arteries, portal veins, and HVs and antegrade flow in the IVC. Hepatic and renal echogenicity was reported to be normal, and this ruled out significant liver congestion and hydronephrosis. Declining trends in transaminases and INR (from 2.3 on admission to the intensive care unit to 1.6) were consistent with improving graft function. Anticoagulation and close follow-up were continued.

On postoperative day 5, repeat Doppler sonography demonstrated decreased resistive indices and blunted waveforms in hepatic arteries that were a concern for arterial thrombosis or kinking. Flows in portal veins and HVs and in the IVC were appropriately directional, but contrast venography of the IVC confirmed stenosis (Fig. 2D). The pressure gradient was not reported. Transaminases were still trending downward, and the INR was 1.07. The patient was brought to the operating room, and a patent but kinked main hepatic artery was reconstructed without complications.

The remainder of the hospitalization was uneventful. The patient was discharged on postoperative day 8 on warfarin and aspirin. One year after transplantation, the graft function was normal, the Doppler resistive indices were normal, and contrast-enhanced magnetic resonance imaging showed remodeling of the thrombus with no residual IVC stenosis.

IVC stenosis has been reported as a rare but serious surgical complication of OLT.[6] In the present report, we describe an IVC stenosis of a different etiology: chronic thrombotic suprahepatic IVC stenosis in a liver transplant recipient associated with polycystic hepatomegaly and diagnosed intraoperatively by TEE.

The implications of the TEE findings were 2-fold. First, a moderate IVC/right heart pressure gradient was impeding hepatic venous outflow and causing graft congestion. Second, located at the cavoatrial junction, the lesion had thromboembolic potential. In retrospect, poor flows and thrombosis of the hepatic artery may have been in part another result of increased intrahepatic venous pressure. This was not appreciated at the time of transplantation.

An intraoperative IVC stent procedure was considered because its effectiveness in treating post-OLT IVC stenosis had been demonstrated.[6, 8] Because the risks of dislodging the mural thrombus or migrating the stent into the RA and covering the HVs were estimated to be high, we decided against intervening. Angioplasty and open surgical repair were not considered because of the nonlasting benefits of the former approach and the high risk of complications and likely need for cardiopulmonary bypass with the latter approach.

We reasoned that because the patient remained hemodynamically stable, the removal of the enlarged polycystic liver would over time result in IVC remodeling and improvement in the pressure gradient. We instituted anticoagulation and followed the patient closely with vascular ultrasound and with clinical and laboratory assessments.

This case illustrates 3 interesting concepts. First, TEE can provide critical and timely intraoperative information: in this case, a hemodynamically significant intracaval lesion with thromboembolic potential was causing postreperfusion graft congestion. The diagnosis of IVC stenosis is traditionally based on imaging and clinical presentation and may take hours to days.[7, 8] In that sense, TEE is a uniquely valuable diagnostic tool because no other routine intraoperative monitor could have detected this lesion in such a timely fashion. The rapid diagnosis had an immediate impact on patient management and prompted the institution of anticoagulation, close follow-up, and consideration of interventions.

Second, a hemostatic imbalance in a patient with end-stage liver disease may lead to a hypercoagulable state.[9] This concept is still commonly underappreciated. The elevated maximal amplitude on thromboelastography was consistent with hypercoagulability despite the abnormal INR, and this emphasizes the advantage of a whole-blood clotting assay.

Finally, our findings support the classic concept of Virchow's triad[10]: venous stasis caused by the external IVC compression by the polycystic liver, a tendency toward hypercoagulability, and likely endothelial dysfunction, which is commonly present in end-stage liver disease,[9] all may have contributed to chronic IVC thrombosis.

  • Jacob Aubuchon, M.D.1

  • Erin Maynard, M.D.2

  • Anand Lakshminarasimhachar, M.D.1

  • William Chapman, M.D., F.A.C.S.2

  • Ivan Kangrga, M.D., Ph.D.1

  • Departments of 1Anesthesiology and 2Surgery Washington University School of Medicine, St. Louis, MO