Successful Management of Endotipsitis: After Direct Intrahepatic Portocaval Shunt for Budd–Chiari Syndrome

Abstract

We report successful management of transjugular intrahepatic portosystemic shunt stent infection (endotipsitis) following direct intrahepatic portocaval shunt for chronic Budd–Chiari syndrome. PETCT confirmed FDG-avid vegetation/thrombus within the stent graft, which was removed with a large-bore sheath aspiration technique, and the residual thrombus was excluded by placing an additional stent graft. The thrombus was positive for E. coli, and appropriate antibiotics were given. Patient improved with no recurrent symptoms during a follow-up of 7 months.

 

Scroll to top

Keywords

endotipsitis - DIPS - Budd–Chiari syndrome

Introduction

The transjugular intrahepatic portosystemic shunt (TIPS) procedure is performed to manage the sequelae of portal hypertension. Common complications during and after TIPS include capsular perforation, intraperitoneal hemorrhage, encephalopathy, TIPS dysfunction, and liver failure. Other uncommon complications of TIPS are stent graft migration, misplacement, and occlusion. However, persistent bacteremia post-TIPS is a rare and serious complication of the procedure and is known as “endotipsitis.” Endotipsitis is difficult to diagnose. It is confirmed when other sources of bacteremia are excluded, along with the presence of shunt occlusion or vegetation, or bacteremia in the presence of a patent shunt. Until 2024, only 58 cases have been reported with a mortality of up to 30%.[1] [2] We report a rare case of successful treatment of endotipsitis following direct intrahepatic portocaval shunt (DIPS) for chronic Budd–Chiari syndrome. ([Fig. 1])

Fig. 1 (A–D) Direct intrahepatic portocaval shunt (DIPS) was performed for chronic Budd–Chiari syndrome. CECT coronal image showing hepatomegaly with congestion (A). Portogram showing right portal vein branches during DIPS procedure (B). Prestenting, portogram showing long parenchymal DIPS tract and IVC entry site (arrow) (C). Poststenting, portogram showing good flow across the stent (D).

 

Scroll to top

Case Report

A 55-year-old woman with Budd–Chiari syndrome and chronic immune thrombocytopenic purpura presented with abdominal distension and pain. The medical history was significant for hypertension and type II diabetes mellitus, which were controlled on medications. Contrast-enhanced CT (CECT) and ultrasound Doppler showed occluded hepatic veins. Model for end-stage liver disease-sodium score was 10. An extended workup revealed the presence of paroxysmal nocturnal hemoglobinuria. A DIPS was performed, using a partially covered self-expandable Niti-S (Taewoong, Seoul, Korea) stent of size 10 mm × 120 mm (100 mm covered intrahepatic portion and 20 mm uncovered PV portion). Two weeks later, the patient was admitted with septic shock, high-grade fever (103°F) with chills and rigors. The patient was treated with antibiotics. Two more episodes of sepsis were observed whenever antibiotics were stopped. Laboratory tests revealed a hemoglobin of 7.2 g/dL, platelet count of 6 × 10³/μL, white blood cell count of 4 × 10³/μL with neutrophil predominance, total bilirubin of 1.6 mg/dL, aspartate transaminase 119 U/L, alanine aminotransferase 82 U/L, and alkaline phosphatase of 110 U/L. Blood cultures revealed extended-spectrum beta-lactamase E. coli bacteremia. PETCT showed focal FDG uptake in the liver close to the lateral aspect of the mid portion of the covered stent with intrastent eccentric thrombus or vegetation and a focus of air. Endotipsitis was diagnosed in view of PET-positive eccentric thrombus/vegetation within the stent and recurring bacteremia, and no other detectable source of infection.[1] Patient was given the treatment option of thrombus aspiration and containing the focal infection source by placing another covered stent, with possible risks of repeat stent graft infection, migration of infected thrombus, and lung infection in addition to long-term antibiotics. Surgical removal of the stent was offered if all the treatments failed.

Through a right internal jugular venous access, the DIPS stent catheterized with a 10F, 45 cm Rosch–Uchida TIPS sheath (Cook, Bloomington, Indiana, United States), and a large (15 mm × 4 mm) thrombus fragment was aspirated. Later, a few unsuccessful attempts were made to aspirate the residual thrombus attached to the stent wall; however, the patient developed chills and rigors. Balloon maceration was not performed in view of the risk of thrombus migration to the lungs. A stent graft was placed 10 mm × 80 mm (Fluency, Becton Dickinson, New Jersey, United States) inside the previous stent over the residual adherent thrombus ([Fig. 2A–D]). Postaspiration and repeat stenting, no fresh episodes of bacteremia or sepsis were observed. Aspirated clot ([Fig. 2E]) was positive for E. coli on culture. The patient clinically improved, and antibiotics were stopped after 1 week. The platelet counts increased significantly and reached 70,000/mL at 3 months. During a follow-up of 7 months, the patient remained asymptomatic.

Fig. 2 (A–E) Direct intrahepatic portocaval shunt (DIPS) was performed for chronic Budd–Chiari syndrome. PET CT axial image showing FDG uptake in the liver (arrowhead) near the stent (A). PET CECT oblique coronal image showing intrastent thrombus and air (open arrow) (C). DIPS venogram, showing intrastent filling defect/thrombus (arrow; C). Postthrombus aspiration and covered stent graft placement (D) with part of the stent beyond the previous stent graft (bracket). Chronic clot aspirated from the stent (E).

 

Scroll to top

Discussion

Sanyal et al proposed the term “endotipsitis” and offered the diagnostic criteria. They defined “definite endotipsitis” as clinically significant bacteraemia with vegetations/thrombi inside the TIPS and “probable” if sustained bacteraemia with no other source of infection. Since this proposal, there have been several suggestions to improve the diagnostic accuracy and sensitivity of the disease.[3]

The incidence of endotipsitis varies between 0.7 and 1.7%. Early endotipsitis (<120 days) occurs due to bacterial seeding during the procedure, and late endotipsitis (>120 days) occurs mostly following TIPS revision. Gram-positive bacteria are the common cause (72%) in early endotipsitis, and fungal infections are common in late cases. Mortality is higher with Staphylococcus aureus (63%) and Candida infections (67%).[2] [3] There was no difference in mortality rate between early and late-onset infections.[3]

Treatment of endotipsitis mainly involves antimicrobial therapy, as the TIPS stent can not be removed without liver transplantation. Bacteremia resolved after appropriate antibiotic therapy with a median duration of 2 weeks in a study; however, there are no clear recommendations on the duration of antibiotics. Usually, a prolonged course of antibiotics of at least 6 weeks, similar to that offered for endocardial infection, is necessary to prevent relapse. The choice of antibiotics should cover both gram-positive and gram-negative organisms. However, in our patient, E. coli was isolated in blood culture, and an appropriate antibiotic was instituted.[2] [3] [4] [5]

Vizzutti et al reported a case of successful management of endotipsitis post-TIPS for chronic extra-hepatic portal vein obstruction due to chronic alcoholic pancreatitis. Authors have used a rotational thrombectomy device within the covered portion of TIPS to clear repeat thrombosis postmechanical thrombectomy. They also noticed a biliary fistula located between the covered and uncovered portion of the TIPS stent, communicating with the right intra-hepatic biliary system, and excluded the biliary fistula with another 10 mm covered stent. In our case, the PETCT scan showed a vegetation/floating thrombus with air attached to the mid part of the covered portion of the TIPS stent. We didn't find any biliary fistula or communication.[6]

In conclusion, endotipsitis is a rare, serious complication of the TIPS/DIPS procedure, and most cases require medical management with long-term antibiotics. Endovascular management may be considered for cases of endotipsitis with vegetation or stent thrombosis.

 

Scroll to top

 

Scroll to top

Conflict of Interest

None declared.

• References

• 1 Sanyal AJ, Reddy KR. Vegetative infection of transjugular intrahepatic portosystemic shunts. Gastroenterology 1998; 115 (01) 110-115
• 2 Niyas VKM, Keri VC, Kumar P. Endotipsitis: an underdiagnosed complication of transjugular intrahepatic portosystemic shunts. J Clin Exp Hepatol 2022; 12 (01) 222-224
• 3 Navaratnam AM, Grant M, Banach DB. Endotipsitis: a case report with a literature review on an emerging prosthetic related infection. World J Hepatol 2015; 7 (04) 710-716
• 4 Garcia-Zamalloa A, Gomez JT, Egusquiza AC. New case of endotipsitis: urgent need for clinical practice guidelines. Eurasian J Med 2017; 49 (03) 214-216
• 5 Mizrahi M, Adar T, Shouval D, Bloom AI, Shibolet O. Endotipsitis-persistent infection of transjugular intrahepatic portosystemic shunt: pathogenesis, clinical features and management. Liver Int 2010; 30 (02) 175-183
• 6 Vizzutti F, Casamassima E, Falcone G. et al. Management of a complex transjugular intrahepatic portosystemic shunt dysfunction with endotipsitis through rotational thrombectomy. BJR Case Rep 2024; 10 (01) uaae005

Publication History

Article published online:
29 September 2025