International Journal of Colorectal Disease

, Volume 20, Issue 1, pp 67–71

Palliative portal vein stent placement for lymphatic recurrence of gastric cancer


    • Institute of Diagnostic and Interventional RadiologyFriedrich-Schiller-University Jena
  • S. O. R. Pfleiderer
    • Institute of Diagnostic and Interventional RadiologyFriedrich-Schiller-University Jena
  • G. Trebing
    • Department of General and Visceral SurgeryFriedrich-Schiller-University Jena
  • J. Scheele
    • Department of General and Visceral SurgeryFriedrich-Schiller-University Jena
  • W. A. Kaiser
    • Institute of Diagnostic and Interventional RadiologyFriedrich-Schiller-University Jena
Case Report

DOI: 10.1007/s00384-004-0618-0

Cite this article as:
Heyne, J., Pfleiderer, S.O.R., Trebing, G. et al. Int J Colorectal Dis (2005) 20: 67. doi:10.1007/s00384-004-0618-0



Percutaneous transhepatic stenting of the main portal vein is a rare intervention.

Case report

In the current patient, percutaneous angioplasty and stenting of a main portal vein stenosis due to lymphatic recurrence of gastric cancer ameliorated the progressing therapeutic restriction. The wall stent achieved portal venous patency that enabled ongoing chemotherapy. The stent remained patent for the entire subsequent survival period.


StentPortal veinGastric cancer


Placement of a metallic stent into the biliary tract for management of biliary obstruction due to tumor stenosis is a widely established procedure [1]. In comparison, stenting of the portal vein is still rare. Only a few publications address stent placement for treatment of portal vein thrombosis, stenosis, or occlusion due to biliary and pancreatic neoplasms, hepatocellular carcinoma, and complications after liver transplantation [27]. Furthermore, knowledge about the clinical usefulness of stent placement for malignant portal venous stenosis caused by inoperable lymphatic recurrence of gastric cancer is limited.

Case report

A 40-year-old woman with signet ring cell carcinoma of the stomach underwent an abdominal transhiatal total gastrectomy with resection of the distal esophagus, and cholecystectomy, in December 1999. Histological evaluation of the specimen revealed a signet ring cell carcinoma of the stomach (pT2pN2M0) with clear resection margins (R0resection). Since September 2000, impaired quality of life with tiredness, powerlessness and dysphagia developed. Computed tomography (CT) demonstrated a lymphatic local recurrence (3×2.5 cm) involving the liver hilum and extending to the celiac trunk. Palliative radiochemotherapy was initiated, which was aborted because of increasing serum bilirubin levels and decreasing thromboplastin time as well as thrombocytopenia. In March 2001, CT showed progression of the hilar lymphatic recurrence with biliary stricture and stenosis of the main portal vein, which was surrounded and compressed by the tumor.

After bile duct dilatation and insertion of a metallic stent, bilirubin levels normalized and the patient improved. On ultrasound (US), with use of color Doppler, the main portal vein diameter measured 7 mm prestenotic, 3 mm intrastenotic and up to 15 mm poststenotic (Fig. 1). A segmental branch of the portal vein was punctured percutaneously using an intercostal approach under US guidance (3.75 MHz puncture transducer). A 5-F catheter was placed via a hydrophilic guide wire (0.035 in.) and was pushed forward into the superior mesenteric vein. Portography was performed and the diameter and length of the stenosis were measured precisely (Fig. 2). After the insertion of a 10-F sheath, the gradual dilatation of the tract in the liver parenchyma was carried out. Even after balloon dilatation there was still a high grade stenosis detectable. Therefore, a metallic stent (Easy Wallstent, Schneider, Switzerland, nominally diameter 14 mm, length 5 cm) was placed across the stenosis (Fig. 3).
Fig. 1

Ultrasound image of the stenosis of the main portal vein. The main portal vein is surrounded and is invaded by lymphatic recurrence. The minimal diameter is 3 mm with a poststenotic dilatation up to 15 mm

Fig. 2

Percutaneous transhepatic portography before dilatation and stent placement. The portogram shows the severe stenosis of the main portal vein. The stent in the biliary duct is visible near the stenosis

Fig. 3

Portography after stent placement. The image shows the expanded metallic stent without significant residual stenosis. Notice the thrombi at the end of the stent (arrow) and in some branches of the portal vein

Despite the administration of 5,000 IU heparin during the intervention, a subsequent angiography showed thrombotic material at the end of the stent and in some branches of the intrahepatic portal vein. Finally the needle tract was closed during withdrawal of the sheet with biological tissue-adhesive (Tissucol Duo S, Baxter Hyland Immuno Division, Vienna) to prevent intraperitoneal hemorrhage. Continuous intravenous heparinization was maintained for 24 h (24,000 IU/d) followed by 2×5,700 IU heparin for 7 days and long-term administration of 100 mg acetylsalicylic acid.

On the first and third post-interventional days, color Doppler US was performed to assess stent patency and to exclude complications. The woman had slight local pain at the puncture site. No postoperative hemorrhage was detectable. The stent was patent with a minimum diameter in the central part of 7 mm. Some thrombotic material as well as a floating thrombus at the end of the stent (approximately 20×5 mm) was still visible (Fig. 4).
Fig. 4

Ultrasound of the main portal vein 1 day after stent insertion. The wall stent is patent. There is no remnant stenosis after angioplasty. The thrombus at the end of the stent is clearly visible

At follow up investigations after 5 weeks and 14 weeks, color Doppler US demonstrated a complete orthograde perfusion of the main portal vein. The closest region inside the stent had widened to a diameter of 9 mm (Fig. 5). No thrombi were seen any more. Apart from a slightly reduced fitness the patient felt well. Chemotherapy (EAP protocol—etoposide, adriblastine, cisplatin) was performed for 5 months. Diarrhea was the only side effect that occurred. Two days after the last chemotherapy treatment the patient suddenly complained about violent abdominal pain. She was referred to her hometown hospital and died under the features of septic shock the following day.
Fig. 5

Ultrasound 5 weeks after stent placement. The closest region in the stent is 9 mm in diameter. Notice the patency of the stent without any thrombi


In contrast to arteries, venous vessels occlude earlier due to tumor compression. In general, venous occlusion does not require PTA because of abundant collateralization [8]. There are only a few reports on stenting of the portal vein [9, 10]. Some case reports show the clinical benefit of portal vein stent placement, even in patients without symptoms [7, 11]. Watanabe et al. [10] applied the stent in one case intraoperatively via the ileal vein.

In the patient presented, lymphatic recurrence of gastric cancer led to stenosis of both the main biliary tract and the main portal vein with severe biliary obstruction and liver dysfunction. Due to the rapid decrease of the diameter of the portal vein to 3 mm total occlusion was expected within a very short time.

Such stenosis or occlusion of the portal vein may lead to portal hypertension, splenomegaly, encephalopathy, and gastrointestinal bleeding due to esophageal varices. Ascites may develop, and is increased by hypalbuminemia due to a reduced liver function followed by decreasing colloid-osmotic pressure. To prevent this vicious circle, the portal vein should be kept patent as part of any palliative regimen. In our patient a metallic stent was placed after percutaneous balloon dilatation to ensure adequate liver function for continuing chemotherapy and to diminish the risk of gastrointestinal bleeding.

In our opinion, stenting of the portal vein is indicated in patients with rapid decrease of lumen or expected occlusion, in particular if the possibilities of chemotherapy are not yet exhausted. The goal of the procedure is to avoid the consequences of the portal hypertension and to improve quality of life and life extension. There remains no other option for patients who are not candidates for surgery to decompress the portal vein, increase the hepatic blood flow and save their lives [10]. Stenting is not (yet) indicated in patients with slow decrease of lumen without signs of acute developing portal hypertension. Independently, the bile duct is stented in cases of biliary obstruction.

Color Doppler US is eminently suited for measuring the stenosis and preselecting the size of the stent preinterventionally and for follow-up examinations postinterventionally. By means of a portogram the whole expansion of compressed and infiltrated vessel can be determined and the stent placement is facilitated.

Periinterventional heparinization and further antiplatelet or anticoagulant therapy seems essential to avoid an in-stent thrombosis. Yamakado et al. [2] administered 3,000 IU heparin after sheath insertion and then 5,000–10,000 IU heparin per day for 3–5 days. The authors reported stent thrombosis and intimal hyperplasia in two patients who did not receive anticoagulant therapy. Nevertheless, in our patient, thrombi developed during the PTA at the down stream end of the stent, in the punctured and two adjacent portal vein branches. This might be due to a thrombocytosis of 504 Gpt/l (normal 177–379 Gpt/l) on the day of the intervention. After 5 weeks all thrombi were resolved, as proven by US. Other authors have observed dissolving of thrombi after the administration of heparin [2, 3].

Even with a remaining thrombus, this approach could still be considered a success in the palliative settings of the procedure. The liver function is not affected due to a compensatory hypertrophy of the other liver segments, as reported in cases of selective portal vein embolization before partial liver resection [11, 12].

Besides portal vein thrombosis, further complications related to stent placement are abdominal pain at the puncture site, hemorrhage, arterioportal or arteriobiliary shunts, transient fever, pleural effusion, and pseudoaneurysma of the hepatic artery [2, 3]. In most cases shunts close spontaneously after some months. The liver parenchymal tract should be closed with fibrin glue or a gelatin sponge inserted through the sheath during withdrawal to avoid acute intraperitoneal hemorrhage.

In the presence of combined stenosis of the biliary tract and the portal vein, the dilatation of the choledochus duct should be performed first. The stenting of the portal vein after the regression of cholestasis reduces the risk of puncture-related bleeding into the biliary tract and the abdominal cavity.

Stents in cases with splanchnic venous involvement have shown poor patency [3, 13]. Another factor associated with stent occlusion is the tumor volume around or in the portal vein [13]. In most cases in the literature, the patency period was equal to the survival period. In our case, in the last 6 weeks of life there were no clinical signs of acute or chronic portal hypertension and no ascites. The stent improves the patient’s quality of life and enables further treatment like chemotherapy or radiation therapy. The indication for the stent placement always remains an individual decision with consideration of patient status, severity of stenosis, and the natural course of the disease.

Copyright information

© Springer-Verlag 2004