Skip to main content
BMJ Case Reports logoLink to BMJ Case Reports
. 2020 Jan 15;13(1):e233052. doi: 10.1136/bcr-2019-233052

Successful liver transplantation for drug-induced vanishing bile duct syndrome

Ahmed Hashim 1,, Ashley Barnabas 1, Rosa Miquel 1, Kosh Agarwal 1
PMCID: PMC7035849  PMID: 31948982

Abstract

Drug-induced cholestasis has a wide range of clinical presentations, and in a small number of patients, it can progress to severe ductopenia. A 63-year-old woman was referred to our department with progressive cholestasis and hyperbilirubinaemia following a course of flucloxacillin. Despite the comprehensive laboratory, imaging and genetic investigations, no other cause for the cholestasis was demonstrated. Sequential liver biopsies confirmed the development of vanishing bile duct syndrome. She developed significant cachexia and pruritus that did not respond to medical therapy, and hence she was listed for liver transplantation. She underwent liver transplantation 6 months after the initial drug-induced injury. She has remained well with good graft function at 1-year follow-up. The case highlights an extreme form of drug-induced ductopenia and underscores the need for meticulous hepatology input and consideration of liver transplantation in some patients.

Keywords: drugs and medicines, liver disease, transplantation

Background

Vanishing bile duct syndrome (VBDS) is a rare disorder that leads to progressive loss and destruction of intrahepatic bile ducts and severe ductopenia or bile duct paucity. Despite the scarcity of cases, VBDS is a recognised complication of cholestatic drug-induced liver injury (DILI).1–3 Most cases improve with conservative management, and only a very small proportion of patients develop progressive systemic symptoms and chronic liver disease. The underlying mechanism leading to VBDS appears to be multifactorial but is thought to be at least partly immunological in nature through cytotoxic T-lymphocytes causing ductal destruction. Genetic human leukocyte antigen (HLA) susceptibility could be implicated in some patients with HLA DRB1(*04, *11, *15).4

Although orthotopic liver transplantation (OLT) is considered the last resort in severe cases of drug-induced ductopenia, only a few cases of successful transplantation in this situation have been reported.1 5 We present a case of successful liver transplantation following drug-induced VBDS.

Case presentation

A 63-year-old woman who was previously fit and well presented to her local hospital with jaundice and itchy skin. She had started a 10-day course of flucloxacillin for a skin infection 2 weeks prior to the onset of symptoms. She developed conjunctival icterus 4 days following the completion of the antibiotic course. She had no history of liver disease and had only mild psoriasis. She was not taking regular medication. She was referred to our regional liver unit with jaundice and pruritus progressively worsening over 3 months.

Investigations

Her admission laboratory results showed bilirubin of 501 mmol/L, alkaline phosphatase (ALP) 279, aspartate transaminase (AST) 83, alanine transaminase (ALT) 60 and gamma-glutamyltransferase (GGT) of 36 IU/L, and international normalised ratio (INR) of 1.0. Her renal function and full blood count were unremarkable. Her non-invasive liver panel was negative, including a negative extended viral screen (HIV, Hepatitis A, B, C, and E, Cytomegalovirus, Epstein-Barr virus), normal autoimmune profile (including anti-mitochondrial antibody) with normal immunoglobulins (apart from slightly low IgM).

Liver CT and magnetic resonance cholangiopancreatography (MRCP) were performed. Her cross-sectional imaging was reported to show some areas of collapse within the liver, but there was no evidence of chronic liver disease and no significant biliary pathology to account for the marked cholestasis or features of malignancy. She underwent a liver biopsy which showed bland canalicular cholestasis, focal mild portal inflammation with rare bile duct loss; features consistent with a DILI.

Differential diagnosis

Aside from DILI, the differential diagnosis in this case with severe cholestasis included primary biliary pathology such as primary sclerosing cholangitis (PSC) or primary biliary cholangitis (PBC) but this possibility was ruled out given negative serology, normal imaging and lack of typical features on liver biopsy. Haematological conditions that can predispose to a VBDS picture, particularly lymphomas were suspected given the weight loss and cachexia but this was excluded by imaging. The low GGT at presentation prompted a strong suspicion of underlying biliary transporter abnormality. However, her genetic cholestasis screen was negative apart from heterozygosity to ATP-binding cassette transporter B II (ABCBII) and UDP glucuronosyltransferase 1, polypeptide A1 (UGT1A1), which is only a carrier state for Gilbert disease.

Treatment

She started taking ursodeoxycholic acid (UDCA) 750 mg once daily, and her bilirubin declined to 276 mmol/L. A month later, the team reviewed her in the clinic, and her liver function tests showed a persistent cholestatic picture. She began taking rifampicin 150 mg. Although the combination of this and UDCA improved her bilirubin and pruritus initially, her hyperbilirubinaemia then worsened (figure 1).

Figure 1.

Figure 1

Serial liver function tests of the patient since admission to our regional liver unit. OLT, orthotopic liver transplantation; UDCA, ursodeoxycholic acid.

She was readmitted due to worsening hyperbilirubinaemia, pruritus, loss of appetite, lethargy and significant weight loss of 8 kg over a short period of time. An endoscopic retrograde cholangiopancreatography (ERCP) was undertaken and a naso-biliary tube (NB) was inserted, but this was removed within 2 days, as no bile was drained. As she continued to lose weight, a decision was made to commence nasogastric feeding to meet her nutritional requirements. A repeat liver biopsy showed that the histological changes were similar to those seen previously, with persistent severe acute bilirubinostasis, no significant inflammation (no cholestatic hepatitis injury pattern) but progression of the bile duct loss. There was no evidence of significant fibrosis (figure 2).

Figure 2.

Figure 2

(A) First liver biopsy performed at the referring hospital. Portal tracts showed variable inflammatory component at the portal tracts, overall mild, predominantly lymphocytic, which may be indicative of immune-mediated type of injury. Portal vein (PV) and hepatic arterioles (*) were present but bile ducts were lost in few portal areas. Marked acute cholestasis can be observed at the lobule (arrows mark canalicular bile plugs), with no significant lobular inflammation, hallmark of Bland cholestasis. (H&E, ×200). (B) Second liver biopsy at our regional liver unit. CK7 immunostaining demonstrates the lack of bile duct at the portal tract present at the centre of the image and lack of ductular reaction. Hepatobiliary phenotype (aberrant CK7 expression) is observed in periportal hepatocytes as well as in many lobular hepatocytes.

Outcome and follow-up

In view of the progressive cholestasis, she was assessed and then listed for OLT. She was discharged on home enteral feeding and then received a liver transplant a few weeks later (6 months following her initial presentation). The liver explant showed severe cholestasis and ductopenia (>50% bile duct loss), with no fibrosis (figure 3). Of interest was the lack of ductular reaction which may reflect regenerative biliary impairment or severe damage of the biliary progenitor cells. Rare copper-associated protein granules were identified with orcein stain as only indicative evidence of early chronic cholestasis. She was discharged on prednisolone and tacrolimus, and remains well with good graft function at 1-year follow-up post-OLT.

Figure 3.

Figure 3

Representative fresh tissue slice of the explanted liver. Smooth liver capsule and patent porta hepatis structures. Of note is the green colour of the tissue secondary to acute cholestasis. Inset: close view of the parenchyma after formalin fixation, showing a ‘nutmeg’ appearance with cholestasis, more pronounced at the centrilobular areas (green areas). Microscopic examination confirmed the diagnosis of severe ductopenia, minimal degree of portal inflammation and no significant fibrosis.

Discussion

The prevalence of VBDS is not well documented but in a report of 2000 cases of small bile duct disease, 1.2% had idiopathic VBDS and 0.5% had drug-induced biliary ductopenia.6 The aetiological factors that predispose individuals to VBDS vary from congenital causes to acquired conditions. In children, the condition is well described and can be associated with genetic disorders such as progressive familial intrahepatic cholestasis (PFIC), cystic fibrosis, Alagille’s syndrome and alpha-1 antitrypsin deficiency. In adults, VBDS occurs to a lesser degree and is often seen as a paraneoplastic phenomenon. It is classically linked to haematological malignancies, and in particular to Hodgkin’s lymphoma where up to 50% of cases have some degree of liver infiltration with lymphoma cells.1 2 7 A few of these reports described VBDS and severe ductopenic cholestatic liver damage as the initial and primary presentation of Hodgkin’s lymphoma.1 2 7 Moreover, VBDS in the context of Hodgkin’s lymphoma carries a high mortality.2 Other conditions associated with VBDS include macrophage activation syndrome and histiocytosis X. Immune-mediated aetiologies in the context of organ transplantation often result in severe ductopenia and this includes chronic graft versus host disease in the context of allogenic bone marrow transplantation, acute T-cell-mediated rejection and the rare chronic ductopenic rejection.1–3

In our case, the VBDS was presumed to be secondary to flucloxacillin as other possibilities were ruled out through the aforementioned extensive imaging and laboratory investigations, which included a cholestatic genetic panel. In fact, she was found to be heterozygous to ABCB11, a gene implicated in bile salt export bump (BSEP) deficiency.8 Mutations of BSEP (ABCB11) are linked to PFIC type 2, benign recurrent intrahepatic cholestasis type 2 and occasionally DILI. However, the association between genetic variations of ABCB11 and risk of DILI is still unclear and the conclusions in this respect remain inconsistent. Our patient had a very common variant in the general population, with a small increase in the risk of cholestasis and low evidence of pathogenicity or phenotypic effect. Nonetheless, the presence of this abnormality may have aggravated the cholestasis induced by flucloxacillin, and thus precipitated a VBDS complication. While drug-induced cholestasis is relatively common, severe ductopenia and VBDS in the context of drug toxicity are rare.1 3 5 Drug-induced VBDS is more frequently reported with augmentin and ciprofloxacin9 10; however, it has been linked to flucloxacillin.11 Moreover, our patient had a rather protracted course of DILI with progressive deterioration in cholestasis and general condition over a 6-month period. Typically, the majority of VBDS cases associated with DILI occur within a few months of the onset of severe cholestatic hepatitis and tend to progress over months or years.1 9–12 Some cases could be associated with an allergic reaction involving rash, fever, facial oedema, lymphadenopathy and/or eosinophilia; and in more dramatic cases may present as part of Stevens Johnson syndrome or toxic epidermal necrolysis.5 Our case did not display any of these immuno-allergic features as eosinophils were not seen in the biopsy while focal portal inflammation was present. Hence treatment options such as steroids were not considered.

The typical diagnostic features for drug-induced VBDS were present in this patient as demonstrated by the persistent elevation of ALP and bilirubin for greater than 6 months and the lack of evidence of biliary disorders such as PBC, PSC or malignancy and indeed paucity of intralobular bile ducts on the second biopsy. The standard diagnostic histopathological criterion for VBDS is loss of 50% or more of the intrahepatic bile ducts on a specimen containing at least 10 portal tracts.1–5 Nevertheless, a milder form can be diagnosed when the duct loss is significant but still less than 50%. The diagnosis can be further aided by immunostaining with cytokeratin 7 and 19, which may help to highlight the bile duct presence or absence. The role of imaging in VBDS is mainly to exclude neoplastic conditions or primary biliary pathologies.

The cases reported in the literature fail to identify a specific clinical pattern or predictors for the development of VBDS following DILI. The outcomes appear to be largely unpredictable. Commonly, especially when the degree of bile duct loss is partial, a reversal of the injury occurs, although after a relatively prolonged period, with clinical recovery and resolution of jaundice and pruritus. In one report, idiopathic biliary ductopenia progressed slowly over 9 years but with no apparentclinical deterioration or significant symptoms of biliary disease.12 In other cases, incomplete clinical recovery may take place but with persistent cholestatic markers and evidence of hepatic fibrosis and rarely progression to advanced chronic liver disease.1–3 13 Those patients may also develop complications of cholestasis such as gallstone formation, hyperlipidaemia, xanthelasmas and vitamin deficiencies. Nevertheless, in a few instances, such as in this case, the ductopenia is dramatic and progressive with concomitant liver synthetic dysfunction and systemic manifestation, such as weight loss and anorexia in which death or liver transplantation are the only two outcomes.13 While our patient had initial improvement with nutritional support, UDCA and rifampicin, cholestasis eventually worsened and her overall symptoms became debilitating with progressive cachexia.

UDCA is the most widely used treatment in this scenario. However, the evidence behind its use in VBDS is lacking and mainly comes from case reports. In one case, it was found to be successful for treating augmentin-induced VBDS, but significantly high doses had to be used to achieve a good clinical, biochemical and histological response (45 mg/kg/day).10 Cholestyramine, rifampicin and naloxone can be used to alleviate pruritus. Ultimately, identification, withdrawal and avoidance of the offending drugs is key in the management of such cases. Some reports indicated the use of immunosuppression, tacrolimus, infliximab or even plasmapheresis may also be helpful.14–16 This could be an acceptable approach if the immune-allergic features dominate, but the evidence is scarce. In our case, UDCA was not successful, and neither was the insertion of an NB tube. NB tubes have been used with success in severe cholestasis but mainly in the context of congenital aetiologies and in particular in those with low GGT cholestasis.17 Our patient had a low normal GGT level initially, a typical feature of congenital cholestasis, hence an NB tube insertion was attempted while the genetic results were pending. It is also interesting that the GGT started to rise months later and increased significantly. Ondansetron, a serotonin (5-HT3) receptor antagonist, has been anecdotally used as an intervention in patients with cholestatic pruritus. However, one systematic review showed that its role in this situation is negligible.18

While liver transplantation in those with debilitating systemic symptoms and aggressive cholestasis is a good option, there are only a few cases in the literature describing liver transplantation in this context.5 19 A case of idiopathic adulthood cholestasis was reported to have a successful OLT with no recurrence for years.19 A review of our database from the year 2000 revealed only three other cases transplanted for VBDS: one in the context of end-stage PBC, the second was due to Hodgkin’s lymphoma and the last one was secondary to an unclear aetiology. Our case continued to demonstrate good graft function at 1 year following OLT with only one episode of cellular rejection in the immediate postoperative phase.

Learning points.

  • Clinicians should be aware of vanishing bile duct syndrome (VBDS) as a potential severe and serious complication of drug-induced liver injury secondary to commonly used antibiotics.

  • Our case illustrates that consideration of liver transplantation (when this option is available) in patients with drug-induced VBDS should be made early, particularly in those who show poor initial response to medical management following drug withdrawal and ursodeoxycholic acid.

  • Attention should be paid to the nutritional status and dietary support, as drug-induced VBDS often runs a chronic course with significant weight loss and anorexia. This is particularly important when liver transplantation is planned or considered.

Footnotes

Contributors: AH: writing of the main body of the case. AB and KA: revision of the case study and further changes to manuscript. RM: selection of histology slides and their description.

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. Reau NS, Jensen DM. Vanishing bile duct syndrome. Clin Liver Dis 2008;12:203–17. 10.1016/j.cld.2007.11.007 [DOI] [PubMed] [Google Scholar]
  • 2. Bakhit M, McCarty TR, Park S, et al. Vanishing bile duct syndrome in Hodgkin’s lymphoma: A case report and literature review. World J Gastroenterol 2017;23:366 10.3748/wjg.v23.i2.366 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Dancygier H. Biliary ductopenia (vanishing bile duct syndrome). in: clinical hepatology. Berlin, Heidelberg: Springer, 2010. [Google Scholar]
  • 4. García-Jiménez ME, Quiroga JA, Gutiérrez ML, et al. Association of HLA-DR genes with mild idiopathic adulthood biliary ductopenia. Am J Gastroenterol 2001;96:1178–82. 10.1111/j.1572-0241.2001.03717.x [DOI] [PubMed] [Google Scholar]
  • 5. Harimoto N, Wang H, Ikegami T, et al. Hepatology: rare Stevens-Johnson syndrome and vanishing bile duct syndrome induced by acetaminophen, requiring liver transplantation. J Gastroenterol Hepatol 2015;30:656 10.1111/jgh.12849 [DOI] [PubMed] [Google Scholar]
  • 6. Ludwig J. Idiopathic adulthood ductopenia: an update. Mayo Clin Proc 1998;73:285–91. 10.4065/73.3.285 [DOI] [PubMed] [Google Scholar]
  • 7. Wong K-M, Chang C-S, Wu C-C, et al. Hodgkin's lymphoma-related vanishing bile duct syndrome: a case report and literature review. Kaohsiung J Med Sci 2013;29:636–41. 10.1016/j.kjms.2013.05.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Kubitz R, Keitel V, Scheuring S, et al. Benign recurrent intrahepatic cholestasis associated with mutations of the bile salt export pump. J Clin Gastroenterol 2006;40:171–5. 10.1097/01.mcg.0000196406.15110.60 [DOI] [PubMed] [Google Scholar]
  • 9. Davies MH, Harrison RF, Elias E, et al. Antibiotic-associated acute vanishing bile duct syndrome: a pattern associated with severe, prolonged, intrahepatic cholestasis. J Hepatol 1994;20:112–6. 10.1016/S0168-8278(05)80476-3 [DOI] [PubMed] [Google Scholar]
  • 10. Smith LA, Ignacio JRA, Winesett MP, et al. Vanishing bile duct syndrome: amoxicillin-clavulanic acid associated intra-hepatic cholestasis responsive to ursodeoxycholic acid. J Pediatr Gastroenterol Nutr 2005;41:469–73. 10.1097/01.mpg.0000178086.44155.73 [DOI] [PubMed] [Google Scholar]
  • 11. Turner IB, Eckstein RP, Riley JW, et al. Prolonged hepatic cholestasis after flucloxacillin therapy. Med J Aust 1989;151:701–5. 10.5694/j.1326-5377.1989.tb139652.x [DOI] [PubMed] [Google Scholar]
  • 12. Moreno A, Vicente Carreño V, Cano A, et al. Idiopathic biliary ductopenia in adults without symptoms of liver disease. N Engl J Med 1997;336:835–8. 10.1056/NEJM199703203361204 [DOI] [PubMed] [Google Scholar]
  • 13. Bonkovsky HL, Kleiner DE, Gu J, et al. Clinical presentations and outcomes of bile duct loss caused by drugs and herbal and dietary supplements. Hepatology 2017;65:1267–77. 10.1002/hep.28967 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Okan G, Yaylaci S, Peker O, et al. Vanishing bile duct and Stevens-Johnson syndrome associated with ciprofloxacin treated with tacrolimus. World J Gastroenterol 2008;14:4697–700. 10.3748/wjg.14.4697 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Kawasaki Y, Matsubara K, Hashimoto K, et al. Nonsteroidal anti-inflammatory drug-induced vanishing bile duct syndrome treated with plasmapheresis. J Pediatr Gastroenterol Nutr 2013;57:e30–1. 10.1097/MPG.0b013e3182a95951 [DOI] [PubMed] [Google Scholar]
  • 16. White JC, Appleman S. Infliximab/Plasmapheresis in vanishing bile duct syndrome secondary to toxic epidermal necrolysis. Pediatrics 2014;134:e1194–8. 10.1542/peds.2013-2239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Bhalerao A, Mannu GS. Management of pruritus in chronic liver disease. Dermatol Res Pract 2015;2015:1–5. 10.1155/2015/295891 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. To THM, Clark K, Lam L, et al. The role of ondansetron in the management of cholestatic or uremic pruritus--a systematic review.. J Pain Symptom Manage 2012;44:725–30. 10.1016/j.jpainsymman.2011.11.007 [DOI] [PubMed] [Google Scholar]
  • 19. Rios R, Herrero JI, Quiroga J, et al. Idiopathic adulthood ductopenia: long-term follow-up after liver transplantation. Dig Dis Sci 2001;46:1420–3. 10.1023/A:1010679502324 [DOI] [PubMed] [Google Scholar]

Articles from BMJ Case Reports are provided here courtesy of BMJ Publishing Group

RESOURCES