Percutaneous transhepatic portal vein stenting as rescue treatment for recurrent oesophageal variceal bleeding in a 31-year-old woman with haepatocellular carcinoma in a non-cirrhotic liver

Abstract

A 31-year-old woman with hepatocellular carcinoma suffered from recurrent oesophageal variceal bleeding due to portal hypertension, which was caused by severe compression of the portal vein by metastatic lymph nodes. Endoscopic band ligation and pharmacological treatment did not suffice to prevent recurrence of variceal bleeding. Eventually, after the fifth variceal bleeding within 6 months, the patient was admitted to the intensive care unit in a haemodynamic shock. A Sengstaken-Blakemore tube was inserted and all treatment options were discussed, but only percutaneous transhepatic recanalisation of the portal vein with stent placement to reduce portal vein pressure was thought to be feasible with any chance to relieve portal vein pressure. After successful portal vein stenting, our patient did not have any recurrent bleeding in the remaining year of her life. We suggest that percutaneous transhepatic portal vein stenting may be a feasible and adequate last line treatment for complications of portal hypertension.

Background

Portal vein flow obstruction can be caused by a variety of benign and malignant diseases, including liver cirrhosis, postsurgical stenosis, thrombosis and tumour encasement. In most cases, portal vein obstruction results in portal hypertension, characterised by among others ascites and portosystemic varices. Especially haemorrhage of varices, mostly located in the oesophagus, may pose a life-threatening complication of portal hypertension and needs adequate treatment. Endoscopic band ligation of oesophageal varices is often effective, but may be inadequate in some cases, prompting the need for more invasive treatments.

Case presentation

In 2011, a 26-year-old woman visited the surgical outpatient clinic with severe abdominal pain and a palpable mass in the epigastric region. A CT scan confirmed a tumour in the left liver lobe. A hemihepatectomy and cholecystectomy were performed, along with construction of a hepaticojejunostomy. Postoperative pathological examination showed a fibrolamellar haepatocellular carcinoma with lymph node metastasis. Histological assessment of the remaining liver tissue showed no signs of hepatitis, fibrosis or cirrhosis.

In the following years, multiple metastatic lesions were found in the remaining part of the liver and both lungs, for which a rehepatectomy in combination with neo-adjuvant radiotherapy and two pulmonary wedge resections were performed. In 2015, systemic therapy with sorafenib was started, after which a switch to chemotherapy with cisplatin/gemcitabine because of progressive disease was made in 2016. Three months later, a first presentation of oesophageal variceal haemorrhage occurred, characterised by haematemesis and melena. Oesophageal varices were formed, supposedly, because of severe compression of the portal vein due to metastatic lymph node compression. This was confirmed on CT scan, where extensive metastatic abnormalities were seen at the hilum of the remnant liver with compression of the portale vein (figures 1 and 2). Severe variceal haemorrhages recurred on four different occasions during the following 4 months. These bleedings were treated by endoscopic band ligation of the oesophageal varices, and pharmacologically by octreotide and antibiotics (figure 3A). Non-selective beta blocker (carvedilol 6.25 mg two times a day) and octreotide (0.1 mg, three times a day subcutaneous injections) prophylaxes proved to be ineffective, in light of the frequent rebleeding. In between all events, no elective endoscopic band ligation was performed. At the fifth variceal bleeding, the patient was admitted to the intensive care unit in haemodynamic shock and was intubated. The variceal bleeding could temporarily be controlled by a Sengstaken-Blakemore tube. In the subsequent multidisciplinary consultation, endoscopic and/or surgical options for controlling the bleeding were ruled out and a percutaneous approach was suggested to reduce portal vein pressure.

Figure 1.

CT images of tumour mass at the hilus of the remnant liver.

Figure 2.

CT image of a compressed portal vein, indicated by arrows, with hepatic metastasis and a patent superior mesenteric vein.

Figure 3.

(A, B) Endoscopic oesophageal images. Left: Oesophageal varices before procedure; Right: Oesophagus after stent placement.

Treatment

Given the prehepatic location of the underlying tumour manifestation, a classic Transjugular Intrahepatic Posto-Systemic Shunt (TIPSS) procedure would not suffice. Therefore, a percutaneous transhepatic recanalisation of the caudal vena porta with stent placement was proposed. Due to the high risk of perforation or false route, this option was discussed extensively with the patient and her family, after which they consented to this procedure. The procedure was performed under conscious sedation. A portal branch in the remnant liver was punctured under ultrasound guidance after which an 8 French sheath was inserted. Recanalisation of the compressing extrahepatic tumour mass was achieved using a simple vascular catheter and a guidewire after which the latter was exchanged for a stiff wire. An 8–10×6 cm/2 cm Viatorr graft was deployed at the level of compression, which was subsequently dilated with a 10 mm balloon. Post procedural digital subtraction angiography showed a patent stent graft with a brisk, hepatopetal flow (figure 4 A, B). No preprocedural portal vein pressure measurements are available; after graftplacement the the transportal pressuregradient was 5mm Hg.

Figure 4.

(A, B) Perprocedural radiologic images. Left: Before stent placement; Right: After stent placement.

Outcome and follow-up

Six months after the procedure, no remaining oesophageal varices were found during gastroscopy (figure 3B). No procedure-related complications occurred. Until her death, 1 year after the procedure due to overall tumour progression, no further variceal haemorrhage occurred.

Discussion

Our case report describes a 31-year-old woman patient with metastasised hepatocellular carcinoma, who underwent percutaneous transhepatic portal vein stenting for severe recurrent oesophageal variceal haemorrhages secondary to portal vein tumour encasement. The purpose of this report is to describe a cutting edge, high risk and relatively rarely performed, but feasible procedure with excellent outcome in case of successful portal vein recanalisation. Percutaneous transhepatic portal vein stenting has been described in only a handful reports. Some interventions were performed in patients with non-malignant stenosis,^1–3 while others described procedures in patients with hepatopancreatobiliary neoplasms.^4–10 Reports of procedures in malignant stenosis, some of which solely regard postsurgical procedures or malignancies other than haepatocellular carcinoma, only described patients in either the Japanese or South-Korean population. One study, specifically regarding hepatocellular carcinoma and associated portal vein thrombosis, describes portal vein stenting.¹¹ The few studies on percutaneous transhepatic stenting, similar to our case, showed promising results.^{5 10 12} All these reports stated that portal vein stenting was a feasible solution with good results for portal vein (tumour) thrombosis: in cases of hepatocellular, perihilar of gallbladder carcinoma, medians of survival duration after procedure were up to 14 months, compared with a median of 6 months in cases without portal vein stenting.⁸ Our case was caused by external compression. Survival times after stenting for benign stenosis were, obviously, significantly better, with median survival duration up to 30 months after stenting.^{4 7} Common complications included transient fever (±10%), liver abscess, acute portal vein thrombosis and sepsis (up to 25%).^{4 7} It should be noted that previous reports described patients that underwent semielective interventions,^4–8 while our case describes an emergency situation, without adequate alternatives. Apart from a small subcapsular haematoma of the liver, no (post)procedural complications occurred in our patient and she remained without recurrent haemorrhage for the remainder of her life.

In conclusion, percutaneous transhepatic portal vein stenting may be a feasible and adequate last line treatment for complications of portal hypertension with malignant aetiology, resulting in a longer life expectancy and an improvement of quality of life, making this a procedure that should be considered an additional treatment option. However, prospective studies are needed to provide further evidence.

Patient’s perspective.

Perspective of patient’s spouse

After the fifth internal bleeding in a relative short period of time (approximately 6 months), she had reached a point where the bleeding could no longer be stopped. It then became clear that there was no chance of recovery (also in view of the advanced stage of the disease). She and I thought we had reached the point where the disease would be fatal to her (or at least the complications from the disease). At that moment she was in shock and she could no longer be fully addressed.

When the doctors came up with an experimental application of a stent, I did not have to think for a moment to seize this opportunity in this life-threatening situation.

It was then extraordinary news for us that the doctors managed to close the bleeding (because for us it seemed like a hopeless situation). After the successful placement of the stent, she was no longer bothered by the bleeding (and by the stent itself) until her death.

Before the placement of the stent, she had gained an enormous fear of new bleedings due to the many bleedings that occurred before (and particularly the fear of vomiting huge amounts of blood). The fear disturbed her mentally, impeding her in her daily life and restricting her in activities. Being her husband, I am grateful that after the placement of the stent, we gained some extra important time together until her death a year later. Even though the disease had spread and progressed extensively, she no longer had to worry about the bleedings in the extra time she has been allowed with this stent.

Learning points.

• Tumour encasement and metastatic lymph node compression can cause portal hypertension in patients with hepatic malignancies.

• Percutaneous transhepatic portal vein stenting is an infrequently used but feasible treatment for complications of portal hypertension.

• Portal vein stenting can lead to a longer life expectancy, improvement of quality of life and prevention of recurrent oesophageal variceal bleeding.