An Early Experience of Liver Transplantation in Portal Vein Thrombosis
VG Shelat, MS Hepatobiliary
& Pancreatic
Surgery and Liver
Transplant Unit,
Department of
Surgery,
National University
Hospital,
Singapore 119074 , Ravishankar K Diddapur, MD Hepatobiliary
& Pancreatic
Surgery and Liver
Transplant Unit,
Department of
Surgery,
National University
Hospital,
Singapore 119074
Abstract
Portal vein thrombosis (PVT) is a recognised complication of end-stage liver disease. As PVT is a relative contraindication to liver transplantation, potential candidates are often rejected due to increased risk and complexity of the surgical procedure. Thromboendovenectomy is a commonly-used technique for this vexing problem. We report a surgeon’s experience of dealing with PVT at the time of liver transplantation. Two patients with liver cirrhosis and PVT underwent orthotopic liver transplantation (OLT) at the National University Hospital. The case notes of these two patients were retrospectively reviewed, along with a brief review of literature. Our early experience shows that with careful case selection and meticulous surgical technique, OLT can be safely done in patients with PVT without significant complications.
Keywords : end-stage liver disease, liver transplantation, portal vein thrombosis, thromboendovenectomy
Singapore Med J 2008; 49(2): e37-e41
INTRODUCTION
Splanchnic vein thrombosis used to be a contraindication for orthotopic liver transplantation (OLT). Preoperative work-up may fail to detect splanchnic vein thrombosis and transplant candidates can develop de novo thrombosis while they are on the waiting list. The modern era transplant surgeon has to be familiar with the current strategies of dealing with splanchnic venous thrombosis. The technique can vary from simple thromboendovenectomy to complex venous bypass and reconstructive procedures. The inability to restore portal venous inflow to nurture the transplanted liver has resulted in a very high mortality.(1) We report a single surgeon’s experience (RKD) of dealing with portal vein thrombosis (PVT) at the time of liver transplantation.
CASE REPORTS
All liver transplant patients at the National University Hospital were retrospectively analysed. Two patients with cirrhosis and PVT underwent OLT at the hospital (Table I).

Case 1
A 47-year-old male diabetic patient with hepatitis B cirrhosis and associated hypersplenism was worked up for liver transplantation. His magnetic resonance (MR) imaging revealed a complete thrombus in the portal vein extending up to the splenomesenteric confluence (Yerdel Grade II) (Fig. 1). MR imaging also showed the common hepatic artery originating from the superior mesenteric artery. He underwent thromboendovenectomy during OLT. The duodenum was kocherised, and the right and left portal veins were dissected and looped. The common portal vein was dissected to beneath the head of pancreas distal to the thrombus. Thrombosis of the portal vein was approached by dissection of the entire length of the portal vein up to the splenomesenteric confluence (Fig. 2). This was achieved by ligating the coronary vein and the distal short pancreatic veins securing meticulous haemostasis. A soft vascular clamp (spoon-shaped DeBakey’s vascular clamp) was applied to the common portal vein towards the splenomesenteric confluence beyond the thrombus. Stay sutures at three o’clock and nine o’clock positions were taken in the portal vein with 5-0 polypropylene. The common portal vein was then controlled with fingers (thumb and index) and thromboendovenectomy was completed using a combination of Fogarty’s balloon catheter, endarterectomy spatula and vascular forceps (Fig. 3).



Thromboendovenectomy was performed, and the organised thrombus, along with intima, was dissected with the endarterectomy spatula. Dissection in the correct plane ensures complete removal of the thrombus with a smoothtapered edge. The intima was then stitched to the adventitia using 6-0 polypropylene suture where it was separated. The donor common portal vein was anastomosed to the recipient common portal vein with polypropylene 6-0 with eversion of edges after taking care for growth factor (Fig. 4). Prior to portal venous anastomosis, we flushd the recipient portal vein with heparinised saline to remove any clots. Intraoperatively, Doppler ultrasonography (US) was used to confirm patency of the portal vein.
Postoperatively, we used Doppler monitoring for assessing the patency and flow characteristics of the portal vein. Intraoperative blood loss was 5 L, and the postoperative course was complicated by arrhythmias that needed repositioning of the pulmonary artery catheter. The patient also developed an infection by Klebsiella pneumoniae and anastomotic bile duct stricture. The stricture needed endoscopic stenting. The biliary stent was removed three months later without any residual stricture demonstrated. His hospital stay was two weeks and he was doing well at 30 months follow-up. The follow-up Doppler US confirmed a good portal venous flow with a velocity of 40 cm/sec (Fig. 5).


Case 2
A 45-year-old male diabetic patient with a past history of alcohol abuse developed alcoholic cirrhosis that was proven by biopsy. He had grade 1 encephalopathy and ascites. He also had past history of haematemesis. After abstinence for six months, he was worked up for liver transplantation. His computed tomography (CT) of the abdomen revealed multiple varices, splenomegaly, gross ascites and PVT. The thrombosis involved the confluence of splenic and superior mesenteric veins along with extension into proximal superior mesenteric vein (Yerdel Grade III). He underwent thromboendovenectomy during OLT, as discussed in the previous case. Immediately postoperatively, his abdominal drain revealed fresh bleeding and so he was re-explored again. The hepatic artery anastomotic site was found to be bleeding and the anastomosis was refashioned. Ascites, bilateral pleural effusions and wound infection complicated his postoperative course. Ascites needed ultrasound-guided tapping with volume-to-volume replacement by salt free albumin. The wound was infected by methicillin-resistant Staphylococcus aureus (MRSA) and needed regular dressing after pus drainage. He was discharged after four weeks, and he is currently doing well at eight months of follow-up.
Thromboendovenectomy was successful in both patients; they were not put on any platelet anti-aggregate agents (like aspirin) or anticoagulants (like heparin/ warfarin) postoperatively. Both patients did not develop rethrombosis after surgery and were well at follow-up.