Abstract
Endotipsitis is a vegetative endovascular infection of a transjugular intrahepatic portosystemic shunt (TIPS). There is currently no uniformly accepted diagnostic criterion, and most cases are diagnosed by clinical diagnosis of recurrent bacteraemia in patients with TIPS and no identifiable source after appropriate investigation. We present a case of 62-year-old man in whom endotipsitis was suspected clinically after emergent TIPS placement complicated by TIPS thrombosis, need for TIPS revision and recurrent bacteraemia. The diagnosis was confirmed using an Indium-111-labelled leucocyte scan (tagged white blood cell scan). This case highlights the potential risks of endotipsitis with TIPS procedures and provides insight into the utilisation of an old diagnostic tool in a new diagnostic role.
Keywords: nosocomial infections, radiology
Background
Transjugular intrahepatic portosystemic shunt (TIPS) insertion is a minimally invasive procedure that is used for the treatment or prevention of complications of elevated portal pressure. Initially, it was used to treat refractory variceal bleeding or ascites, but currently additional indications are being investigated and its use is expanding.
Endotipsitis is a vegetative endovascular infection of a TIPS, which was first described in 1998 by Sanyal et al. Sanyal et al1 proposed diagnostic criteria for endotipsitis with ‘definite’ infection being continuous bacteraemia with vegetation or thrombus inside TIPS and ‘probable’ infection being sustained bacteraemia in a patient with TIPS without other source of infection.
The epidemiology is poorly understood due to difficulty in diagnosis and lack of uniform definition. The most thorough literature about endotipsitis has been the literature review by Navaratnam et al2 which described 50–60 reported cases in the literature and a single centre, retrospective study in 2010 by Kochar et al which found the incidence to be approximately 1%.3
Case presentation
A 62-year-old male patient with non-alcoholic cirrhosis secondary to primary sclerosing cholangitis and non-alcoholic steatohepatitis presented to our medical intensive care unit with hematemesis secondary to acute oesophageal variceal bleeding. He underwent endotracheal intubation followed by upper gastrointestinal endoscopy that showed active variceal bleeding. He underwent endoscopic variceal banding and Minnesota tube placement but had continued bleeding. He subsequently underwent embolisation of coronary vein varices and posterior gastric varices as well as placement of an emergent TIPS. His hospital course was complicated by multiple intubations, recurrent variceal bleeding, multiple red blood cell transfusions, acute kidney injury, aspiration pneumonia, need for TIPS revision due to a TIPS thrombosis, extended-spectrum beta-lactamase (ESBL) Klebsiella pneumoniae bacteraemia and Candida albicans fungemia. His clinical condition was improved, he was extubated, transferred to the general medicine ward and was treated with a 2-week course of intravenous meropenem.
After completion of his antibiotic course, he remained stable for 48 hours. However, on day 3 he developed sudden onset of sepsis characterised by a fever of 39°C, lactic acidosis of 3.0 mEq/L and hypotension. Blood cultures were obtained, and he was empirically started on intravenous meropenem based on previous susceptibilities. Blood cultures again grew ESBL K. pneumoniae, which was susceptible to amikacin, ceftazidime/avibactam, gentamicin, tigecycline, intermediate to tobramycin and resistant to other antibiotics including ceftriaxone, ciprofloxacin, eertapenem, meropenem, piperaciilin/tazobactam, trimethoprim/sulfa.
Investigations
An investigation for the source of recurrent bacteraemia was undertaken. CT of the abdomen and pelvis with contrast revealed a fluid density lesion around the TIPS and small loculated fluid collection at the right lung base (figure 1). A subsequent Indium-111-labelled leucocyte scan showed increased leucocyte uptake around the TIPS consistent with a diagnosis of endotipsitis (figures 2 and 3).
Figure 1.
CT abdomen and pelvis with contrast showing small loculated fluid collection at right lung base (left panel) and fluid density lesion around the TIPS (right panel). TIPS, transjugular intrahepatic portosystemic shunt.
Figure 2.
Sagittal and coronal views of combined CT abdomen and pelvis with Indium-111-labelled leucocyte scan showing positive tracer uptake around TIPS consistent with endotipsitis. TIPS, transjugular intrahepatic portosystemic shunt.
Figure 3.

Indium-111-labelled leucocyte scan showing positive tracer uptake around TIPS consistent with endotipsitis. TIPS, transjugular intrahepatic portosystemic shunt.
Treatment
The patient completed a 4-week course of antibiotics with intravenous tigecycline and oral rifampin with the clearance of bacteraemia without recurrence. Combination therapy was chosen because of concern for low in vivo effect of tigecycline. Prior to this, ceftazidime–avebactam was used but caused drug reaction with eosinophilia and systemic symptoms and polymyxin was trialled but caused acute renal failure. Following completion of antibiotics, he was placed on prophylactic minocycline and ciprofloxacin and subsequently discharged from the hospital.
Outcome and follow-up
On follow-up 3 months postdischarge with his hepatologist, he is doing well, working with physical therapy and is currently being evaluated for liver transplant.
Discussion
Most endotipsitis cases are caused by Gram-positive bacteria (Enterococcal spp and Staphylococcus aureus) and fungal infections. In patients who are ineligible for transplantation, aggressive medical therapy and potentially chronic suppressive antibiotics should be considered while acknowledging the prognosis is poor. In general, the role and timing of liver transplantation in the setting of endotipsitis are poorly understood. There have been four reported cases of liver transplantation after clearance of blood cultures.4–6
A study by Mizrahi et al5 identified four risk factors for bacteraemia associated with TIPS shunting: failure to use prophylactic antibiotics, the duration of the TIPS procedure, hepatitis C infection and hypothyroidism. Other risk factors that have been described are mostly related to the TIPS procedure—procedure duration, whether it was done emergently or as a revision and other procedural variables including overlapping stents, longer stents as well as uncovered wall stents.4–7
Treatment remains poorly understood given the small number of cases. Cases with S. aureus and Candida spp are associated with particularly poor outcomes. There has been a previous case report of a patient who was diagnosed with endotipsitis infection with drug-resistant K. pneumoniae8 and ultimately expired while awaiting a donor to be identified for liver transplantation. Transplantation should be considered in the relapsing disease of infections caused by difficult to eradicate pathogens.
Table 1 shows previous case reports of monomicrobial endotipsitis by Klebsiella species. Previous case series2 9 have more extensively reviewed the microbiology of endotipsitis. In one of these series by Armstrong et al, approximately one-third of the cases did not have stent thrombosis or vegetations on imaging and were diagnosed clinically. In 1998, Sanyal et al proposed diagnostic criteria for endotipsitis as above with ‘definite’ and ‘probable’ infections, although the gold standard is isolating the causative bacteria by performing a biopsy from the pseudoepithelium of the TIPS at autopsy or after liver transplantation. To our knowledge, this is the first case published with a diagnosis of endotipsitis made both clinically and by positive-tagged white blood cellscan. Utilisation of this tool in patients with TIPS and recurrent bacteraemia (without thrombosis or vegetation of TIPS on imaging) should be considered to identify the source of infection.
Table 1.
Previous case reports of monomicrobial endotipsitis by Klebsiella species
| Etiological agent | Outcome | Treatment | Duration of treatment | Cirrhosis aetiology | Onset (days) | Reference |
| Klebsiella pneumonia | Resolved | Ceftriaxone | 4 | Alcohol, Hepatitis C | 329 | 1 |
| Multidrug resistant Klebsiella (unspecified) | Resolved | Tigecycline | – | Alcohol | 1095 | 10 |
| Multidrug resistant Klebisiella pneumonia | Death | Multiple* | – | Alcohol | 3 | 8 |
*Imipenem, colistin, tigecycline, chloramphenicol, amikacin, ciprofloxacin and fosfomycin.
Endotipsitis is likely an underdiagnosed infection with one study estimating the incidence to be 1%.3 The increasing utilisation of this procedure and possible expanded indications are likely to cause an increased number of cases.
Learning points.
Endotipsitis is likely an underdiagnosed infection and with increasing indications, the incidence is likely to rise.
Clinical diagnosis of endotipsitis can be challenging, Indium-111-labelled leucocyte scan can be used as a diagnostic tool.
Treatment of endotipsitis can also be challenging and a long course of antibiotics is often needed.
Footnotes
Contributors: Conception of the case report and writing of the manuscript by AS, literature review performed by GS. AG helped with the planning and writing of the manuscript, the project was supervised by JS.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Patient consent for publication: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Sanyal AJ, Reddy KR. Vegetative infection of transjugular intrahepatic portosystemic shunts. Gastroenterology 1998;115:110–5. 10.1016/S0016-5085(98)70371-3 [DOI] [PubMed] [Google Scholar]
- 2.Navaratnam AM, Grant M, Banach DB. Endotipsitis: a case report with a literature review on an emerging prosthetic related infection. World J Hepatol 2015;7:710–6. 10.4254/wjh.v7.i4.710 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kochar N, Tripathi D, Arestis NJ, et al. Tipsitis: incidence and outcome-a single centre experience. Eur J Gastroenterol Hepatol 2010;22:729–35. 10.1097/MEG.0b013e3282fd6917 [DOI] [PubMed] [Google Scholar]
- 4.Jawaid Q, Saeed ZA, Di Bisceglie AM, et al. Biliary-venous fistula complicating transjugular intrahepatic portosystemic shunt presenting with recurrent bacteremia, jaundice, anemia and fever. Am J Transplant 2003;3:1604–7. 10.1046/j.1600-6135.2003.00267.x [DOI] [PubMed] [Google Scholar]
- 5.Mizrahi M, Roemi L, Shouval D, et al. Bacteremia and "Endotipsitis" following transjugular intrahepatic portosystemic shunting. World J Hepatol 2011;3:130–6. 10.4254/wjh.v3.i5.130 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Willner IR, El-Sakr R, Werkman RF, et al. A fistula from the portal vein to the bile duct: an unusual complication of transjugular intrahepatic portosystemic shunt. Am J Gastroenterol 1998;93:1952–5. 10.1111/j.1572-0241.1998.00553.x [DOI] [PubMed] [Google Scholar]
- 7.Brown RS, Brumage L, Yee HF, et al. Enterococcal bacteremia after transjugular intrahepatic portosystemic shunts (tips). Am J Gastroenterol 1998;93:636–9. 10.1111/j.1572-0241.1998.180_b.x [DOI] [PubMed] [Google Scholar]
- 8.Jessani LG, Nayak S, Lingam P, et al. Endotipsitis caused by extremely drug-resistant Klebsiella pneumoniae. Trop Gastroenterol 2015;36:280–2. 10.7869/tg.310 [DOI] [PubMed] [Google Scholar]
- 9.Armstrong PK, MacLeod C. Infection of transjugular intrahepatic portosystemic shunt devices: three cases and a review of the literature. Clin Infect Dis 2003;36:407–12. 10.1086/346156 [DOI] [PubMed] [Google Scholar]
- 10.Aggarwal S, Park J. Endotipsitis: a diagnostic challenge. J Postgrad Med 2011;57:134. 10.4103/0022-3859.81875 [DOI] [PubMed] [Google Scholar]


