C Blackwell Munksgaard 2003 Copyright
American Journal of Transplantation 2003; 3: 1604–1607 Blackwell Munksgaard
doi: 10.1046/j.1600-6135.2003.00267.x
Case Report
Biliary-Venous Fistula Complicating Transjugular Intrahepatic Portosystemic Shunt Presenting With Recurrent Bacteremia, Jaundice, Anemia and Fever Qaiser Jawaida,∗ , Zahid A. Saeeda , Adrian M. Di Biscegliea , Elizabeth M. Bruntb , Sanjay Ramrakhiania , Chintalapati R. Varmac and Harvey Solomonc a
Division of Gastroenterology and Hepatology, Division of Pathology, and c Division of Abdominal Transplant Surgery, Saint Louis University School of Medicine, Saint Louis, MO ∗ Corresponding author: Qaiser Jawaid,
[email protected] b
A 50-year-old White man with noncirrhotic portal hypertension presented with bleeding from gastric varices. Bleeding was initially managed with band ligation and subsequent transjugular intrahepatic portosystemic shunt (TIPS). Over the next few months, the patient had recurrent episodes of anemia, jaundice, fever and polymicrobial bacteremia. Computed tomography (CT) of the abdomen and chest, upper and lower endoscopy, endoscopic retrograde cholangiopancreatography (ERCP), and echocardiography failed to explain the bacteremia and anemia. Followup CT scan and Doppler sonography 9 months after placement showed TIPS was occluded. Repeat ERCP showed a bile leak with free run-off of contrast from the left hepatic duct into a vascular structure. The patient’s status was upgraded for liver transplantation with Regional Review Board agreement and subsequently received a liver transplant. Gross examination of the native liver demonstrated a fistula between the left bile duct and the middle hepatic vein. Pathologic evaluation confirmed focal necrosis of the left hepatic duct communicating with an occluded TIPS and nodular regenerative hyperplasia consistent with noncirrhotic portal hypertension. Infection is rarely reported in a totally occluded TIPS. Biliary fistulas in patent TIPS have been treated by endoluminal stent graft and endoscopic sphincterotomy with biliary stent placement. Liver transplantation may be the preferred treatment if TIPS becomes infected following its complete occlusion. Key words: Biliary-venous fistula, orthotopic liver transplant, TIPS Received 27 November 2002, revised 19 February 2003 and accepted for publication 11 July 2003
1604
Introduction Transjugular intrahepatic portosystemic shunt (TIPS) is now a well-recognized treatment for decompressing portal hypertension. It has been used to control bleeding from varices and portal hypertensive gastropathy, management of refractory ascites, hepatic hydrothorax and hepatorenal syndrome (1). Frequent complications of the procedure include hepatic encephalopathy, bleeding, deterioration of liver function and occlusion of the stent. Rare cases of biliary-venous fistula and infection have been reported (2,3), but the optimal treatment for this complication is unknown. We report a case of TIPS complicated by TIPSbiliary fistula and recurrent sepsis, which was treated successfully by orthotopic liver transplantation.
Case Report A 50-year-old White man with noncirrhotic portal hypertension and presumed alcoholic liver disease presented with recurrent episodes of bleeding from gastric varices. Hemostasis for the active bleeding was accomplished by band ligation of gastric varices. Six months later, for recurrent bleeding, TIPS was placed. A 9-cm long Wallstent (Boston Scientific Corporation, MA) was inserted diagonally to connect the right hepatic vein to the left portal vein. The tract was dilated to a 16-mm diameter and resulted in reduction of portal venous pressure from 20 mmHg to 10 mmHg; a 50% reduction in the index gradient. The patient’s past medical history included type 2 diabetes mellitus controlled by oral agents and hypertension controlled by angiotensin-converting enzyme (ACE) inhibitor. He was abstinent from alcohol for approximately a year before presentation. One month after the TIPS, and over the next 11 months, the patient had frequent episodes of anemia, jaundice, fever and bacteremia. During these episodes, multiple microorganisms were isolated from his blood cultures (Table 1). Treatment of each episode was guided by results of bacterial sensitivity and the antibiotics used included amoxicillin/clavulanate, amphotericin B, ceftriaxone, fluconazole, imipenum and vancomycin. On several occasions investigations were carried out to find the source of bacteremia. CT scans of the abdomen, pelvis, and chest failed to show the source of infection. Routine surveillance
Biliary-TIPS Fistula Table 1: Micro-organisms isolated from blood cultures over an 11-month period of time after transjugular intrahepatic portosystemic shunt insertion Weeks after TIPS insertion
Organism/s isolated from blood
6 7
α Streptococcus
8 10 14 28 40 44 48
Candida albicans, Coagulase negative Staphylococcus aureus Candida albicans α Streptococcus, Enterococcus Staphylococcus aureus Staphylococcus aureus Hemophilus parainfluenza Streptococcus intermedius, Streptococcus salivarus Enterococcus fecium, Staphylococcus aureus, Coagulase negative Staphylococcus aureus
TIPS = transjugular intrahepatic portosystemic shunt.
for TIPS occlusion was performed. He had ultrasound with Doppler 1 month, 2 months and 6 months after the initial insertion of the stent. All of these sonograms showed adequate directional flow with good waveform indicating patent shunt. An abdominal CT scan with intravenous contrast 9 months after placement of TIPS suggested stent occlusion, which was then confirmed by duplex sonography. Upper and lower gastrointestinal endoscopy, endoscopic retrograde cholangiopancreatography (ERCP), and transesophageal echocardiogram had not offered any explanation for the anemia, jaundice, or bacteremia. Endoscopic retrograde cholangiopancreatography was repeated during a hospital admission for fever and worsening jaundice. During this study, a leak of contrast was seen from the left hepatic duct. A ‘run off’ and rapid clearing of the leaking contrast seen under real time fluoroscopy suggested the communication with a hepatic vein (Figures 1 and 2).
Figure 1: Flow of contrast in the transjugular intrahepatic portosystemic shunt (TIPS) stent (arrow) on initial injection of contrast in the biliary tract with endoscopic retrograde cholangiopancreatography (ERCP) cannulation of common bile duct.
American Journal of Transplantation 2003; 3: 1604–1607
Figure 2: Free flow of contrast through the transjugular intrahepatic portosystemic shunt (TIPS) stent and reaching above the right dome of the diaphragm (arrow). Gall bladder is partially filled (arrow head).
Hemobilia was not apparent. A CT scan of the abdomen and chest could not localize this bile leak. A HIDA scan performed within 24 h of this ERCP was normal. The patient was treated with antibiotics and was upgraded to 2A status for orthotopic liver transplantation (OLT) after Regional Review Board agreement. A month later, the patient received successful liver transplant. The explanted organ weighed 2272 g. It contained a TIPS stent occluded with clot; bile was present within the clot. A biliary fistula to the middle hepatic vein was noted on the superior aspect of the liver. Pathologic evaluation confirmed a totally occluded stent with bile staining, focal necrosis of left hepatic duct and nodular regenerative hyperplasia. A probe demonstrated a communication between the left hepatic duct and the clotted TIPS (Figure 3). The patient recovered uneventfully and
Figure 3: Probe in the explant liver showing communication between the left hepatic duct (arrow) and the transjugular intrahepatic portosystemic shunt (TIPS) (arrow head) in the middle hepatic vein.
1605
Jawaid et al.
no episode of bacteremia has occurred during 2 years of follow up.
Discussion Transjugular intrahepatic portosystemic shunt was placed in our patient to prevent recurrent bleeding from gastric varices due to portal hypertension. It has been shown that a graded reduction in gradient by 25–50% sufficiently prevents rebleeding (4). Hepatic venous pressure gradient was effectively reduced from 20 mmHg to 10 mmHg; a 50% reduction in the index gradient. A TIPS-biliary fistula can form either acutely at the time of initial TIPS stent placement due to direct injury to bile duct or may develop later. However, the mechanism of delayed formation of TIPS biliary fistula is not fully known. It is presumed that radial forces generated by a stent may cause sufficient tissue pressure in the cirrhotic liver to compromise its already tenuous blood supply, producing pressure necrosis (5). Diagonal course of our long stent might have caused direct injury to middle hepatic vein at the time of initial insertion. The larger diameter of this stent (16 mm) may also have contributed to more radial forces in a curved path and fistula formation. It is not known how often a puncture of bile duct occurs during the creation of TIPS. There is no established treatment for completely thrombosed infected TIPS-biliary fistula. Besides liver transplantation, possible options for treating TIPS-biliary fistula include ERCP with endoscopic sphincterotomy and placement of biliary stent to decompress the biliary system, or radiologic placement of polytetrafluoroethylene (PTFE)covered stent to seal-off the bile leak. We excluded the use of a PTFE-covered stent to treat TIPS-biliary fistula in our patient for two reasons. First, use of PTFE-covered stents is investigational and is reported in partially thrombosed stents with TIPS-biliary fistula (6,7). There is a lack of controlled trials and data to assess their efficacy in complete occlusion. Second, infection was suspected in TIPS-fistula, and vascular stents have not been studied in the setting of infection (8,9). To prevent TIPS occlusion, periodic surveillance is suggested with TIPS venogram (10), endoscopy for esophageal varices (10) or ultrasound with Doppler (11,12). Anticoagulation with heparin (13) or antiplatelet therapy (14) and phenprocoumon (15) has not shown to prevent thrombosis or to improve TIPS patency. Platelet-derived growth factor inhibition showed promising results (16) but needs further investigation. Our patient had episodes of anemia with clinical evidence of occult gastrointestinal blood loss whose origin remained obscure. Gastrointestinal blood loss as a complication of TIPS has been reported (17,18). Absence of visible bleed1606
ing or clot in the biliary tract at the time of ERCP does not exclude the possibility of intermittent hemobilia. Bone marrow suppression due to recurrent bacteremia may be another compounding factor in the development of anemia. Infection of the TIPS stent is a recognized but uncommon complication (19). The pathophysiology of TIPS infection is not fully known. It is speculated that local factors, e.g. thrombus formation or sluggish flow, may allow portal bacteremia to seed the pseudointima associated with stent. There is a possibility of ‘walled-off’ infection within the thrombus or pseudointima (20). Willner et al. (18) described a patient similar to ours who presented with recurrent infections after TIPS and required liver transplantation. In both our and their patients the source of infection may have been the intermittent release of bile in the blood stream. No data from randomized controlled trials are available to provide guidelines to prevent TIPS infection and observational data are limited because it is a rare occurrence. Prophylactic use of antibiotics at the time of TIPS insertion to prevent TIPS infection is suggested (2,20). Whether this one-time dose of antibiotics prevents risk of delayed infection is not known. Treatment in our patient posed multiple issues to be considered including treatment of the occluded TIPS, treatment of biliary-venous fistula and treatment of the presumed infected shunt. Logical initial treatment of suspected infected TIPS is use of appropriate antibiotics. If infected TIPS does not respond to prolonged antibiotic treatment, then, as with prosthetic devices implanted in other parts of the body, removal of the infected stent with replacement by new TIPS would be the ideal choice. As TIPS stents are not removable, the remaining option is the removal of the whole organ. Endovascular balloon thrombectomy, placement of parallel stents or balloon dilatation/angioplasty of TIPS can be used if thrombosis or stenosis is suspected. No attempt was made to declot a suspected completely thrombosed and infected stent in our patient due to concern about septic lung embolization (20) and lack of approved therapy for treatment of his unique problems. In summary, use of a longer and larger diameter stent may lead to fistula formation and should be avoided. A high index of suspicion is needed for infective TIPS with TIPSbiliary fistula if a tetrad of fever, jaundice, bacteremia and anemia is present. Orthotopic liver transplantation should be considered in these patients.
References 1. Schiffman M, Jeffers L, Hoofnagle J, Tralka T. The role of trasjugular intrahepatic portosystemic shunt for treatment of portal hypertension and its complications: a conference sponsored by the National Digestive Diseases Advisory Board. Hepatology 1995; 22: 1591–1597. 2. Kamath P, McKusick M. Transjugular intrahepatic portosystemic shunts (TIPS). Baillieres Clin Gastroenterol 1997; 11: 327–349.
American Journal of Transplantation 2003; 3: 1604–1607
Biliary-TIPS Fistula 3. Freedman A, Sanyal A, Tisnado J et al. Complications of transjugular intrahepatic portosystemic shunt: a comprehensive review. Radiographics 1993; 13: 1185– 1210. 4. Rossle M, Siegerstetter V, Olschewski M, Ochs A, Berger E, Haag K. How much reduction in portal pressure is necessary to prevent variceal rebleeding? A longitudinal study in 225 patients with transjugular intrahepatic portosystemic shunts. Am J Gastroenterol 2001; 96: 3379–3383. 5. Mallery S, Freeman ML, Peine CJ, Miller RP, Stanchfield WR. Biliary-shunt fistula following transjugular intrahepatic portosystemic shunt placement. Gastroenterology 1996; 111: 1353– 1357. 6. Spahr L, Sahai A, Lahaie R et al. Transient healing of TIPS-induced biliovenous fistula by PTFE-covered stent graft. Dig Dis Sci 1996; 41: 2229–2232. 7. Sze DY, Vestring T, Liddell RP et al. Recurrent TIPS failure associated with biliary fistulae: treatment with PTFE-covered stents. Cardiovasc Intervent Radiol 1999; 22: 298–304. 8. Baker M, Uflacker R, Robison JG. Stent graft infection after abdominal aortic aneurysm repair: a case report. J Vasc Surg 2002; 36: 180–183. 9. Jackson MR, Joiner DR, Clagett GP. Excision and autogenous revascularization of an infected aortic stent graft resulting from a urinary tract infection. J Vasc Surg 2002; 36: 622– 624. 10. Sanyal AJ, Freedman AM, Luketic VA et al. The natural history of portal hypertension after transjugular intrahepatic portosystemic shunts. Gastroenterology 1997; 112: 889–898. 11. Feldstein VA, Patel MD, LaBerge JM. Transjugular intrahepatic portosystemic shunts: accuracy of Doppler US in determination of patency and detection of stenoses. Radiology 1996; 201: 141– 147.
American Journal of Transplantation 2003; 3: 1604–1607
12. Feldstein VA, Patel MD. Doppler ultrasonography of transjugular intrahepatic portosystemic shunts. West J Med 1996; 165: 56– 57. 13. Rossle M, Haag K, Ochs A et al. The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med 1994; 330: 165–171. 14. Sauer P, Theilmann L, Herrmann S et al. Phenprocoumon for prevention of shunt occlusion after transjugular intrahepatic portosystemic stent shunt: a randomized trial. Hepatology 1996; 24: 1433–1436. 15. Theilmann L, Sauer P, Roeren T et al. Acetylsalicylic acid in the prevention of early stenosis and occlusion of transjugular intrahepatic portal-systemic stent shunts: a controlled study. Hepatology 1994; 20: 592–597. 16. Siegerstetter V, Huber M, Ochs A, Blum HE, Rossle M. Platelet aggregation and platelet-derived growth factor inhibition for prevention of insufficiency of the transjugular intrahepatic portosystemic shunt: a randomized study comparing trapidil plus ticlopidine with heparin treatment. Hepatology 1999; 29: 33–38. 17. Menzel J, Vestring T, Foerster EC, Haag K, Roessle M, Domschke W. Arterio-biliary fistula after transjugular intrahepatic portosystemic shunt: a life-threatening complication of the new technique for therapy of portal hypertension. Z Gastroenterol 1995; 33: 255– 259. 18. Willner IR, El-Sakr R, Werkman RF, Taylor WZ, Riely CA. A fistula from the portal vein to the bile duct: an unusual complication of transjugular intrahepatic portosystemic shunt. Am J Gastroenterol 1998; 93: 1952–1955. 19. Kamath P, McKusick M. Transvenous intrahepatic portosystemic shunts. Gastroenterology 1996; 111: 1700–1705. 20. Sanyal AJ, Reddy KR. Vegetative infection of transjugular intrahepatic portosystemic shunts. Gastroenterology 1998; 115: 110– 115.
1607
