Article Text

Download PDFPDF

Recurrent bacteraemia following variceal haemorrhage
  1. Rooshi Nathwani,
  2. Benjamin H Mullish,
  3. David Kockerling,
  4. Nikil Rajani,
  5. Ameet Dhar
  1. Division of Integrative Systems Medicine and Digestive Disease/Liver Unit, Faculty of Medicine, Imperial College London, London, UK
  1. Correspondence to Dr Rooshi Nathwani, Division of Digestive Diseases, Imperial College, London W2 1NY, UK; rooshi.nathwani08{at}

Statistics from

Clinical presentation

A 43-year-old man presented to the hospital with septic shock. Eight weeks prior, he had experienced recurrent gastric variceal bleeding (figure 1), treated initially with cyanoacrylate glue injection. Abdominal CT scan suggested features of chronic liver disease, with non-alcoholic fatty liver disease being the presumed aetiology. Despite endoscopic therapy, he had further variceal haemorrhage, and was treated successfully with insertion of a transjugular intrahepatic portosystemic stent shunt (TIPSS) (covered stent). Other past medical history included ulcerative colitis, antiphospholipid syndrome (treated with warfarin) and splenectomy.

Figure 1

Gastroscopy demonstrating gastric varices.

On physical examination, there was no obvious source of infection. Laboratory investigations demonstrated a white blood cell count of 18.4×109/l, haemoglobin of 106 g/L, C-reactive protein of 195 mg/L and international normalised ratio of 2.1 (measured while on warfarin). Blood cultures grew Escherichia coli, and intravenous antibiotics were administered; he made a good recovery.

Six weeks later, following a second admission with E. coli sepsis, 6 weeks of oral antibiotic therapy was prescribed. Further investigations - including ultrasonography, whole body CT, transoesophageal echocardiography and colonoscopy – failed to identify an infection source.

After a third similar hospitalisation, a positron emission tomography (PET) scan was performed within 24 hours of admission (figure 2A,B), which helped to diagnose the cause of his sepsis.

Figure 2

(A) PET scan during episode of bacteraemia – axial view. (B) PET scan during episode of bacteraemia – sagittal view. PET, positron emission tomography.


What is the source of this man’s recurrent bacteraemia?


The PET scan revealed increased signal uptake at the site of the TIPSS, indicative of this being the site of infection (figure 2A,B). While infected cyanoacrylate glue had been considered as a differential diagnosis, there was no PET signal uptake at the sites of glue deposition, excluding this as a possibility. TIPSS blockade was carried out with a 14 mm type II Amplatzer plug, as a strategy to prevent further bacterial distribution from the portal to the systemic circulation. Oral antibiotics were prescribed for four further weeks, and the patient made a complete recovery. In the following 12 months, no further infectious episodes occurred, nor any recurrent gastrointestinal bleeding.

Tipsitis (also known as endotipsitis), infection of TIPSS, may occur from weeks to months after insertion, with little known about the pathogenesis.1 It is a rare but serious complication of this procedure2: 55 cases have been reported with 32% associated mortality.3 There are no uniform diagnostic criteria for the condition, but tipsitis should be considered as a possible diagnosis in patients who have undergone TIPSS insertion and who present with unexplained bacteraemia.4 Targeted antibiotic therapy is the mainstay of treatment, although there are high rates of treatment failure and relapse.4 5 Furthermore, antibiotic therapy alone may not fully decolonise the TIPSS; as demonstrated here, TIPSS blockade may play a role in preventing bacteria transiting into the systemic circulation. TIPSS removal and liver transplantation is the only definitive treatment and may be appropriate in refractory cases.4


With thanks to and in memory of Dr Harry Antoniades, who was involved in the care of this patient.


View Abstract


  • Contributors All authors were involved in the case of this patient and contributed to writing this case.

  • Funding BHM is the recipient of an MRC Clinical Research Training Fellowship (grant reference: MR/R000875/1). The Division of Integrative Systems Medicine and Digestive Disease receives financial support from the National Institute of Health Research (NIHR) Imperial Biomedical Research Centre (BRC) based at Imperial College Healthcare NHS Trust and Imperial College London.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Patient consent for publication Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.