Address reprint requests to Ji Kon Ryu, M.D., Department of Internal Medicine, Seoul National University College of Medicine, 101 Daehak-Ro, Jongno-Gu, Seoul, Korea 110-744. Telephone: 82-2-2072-1962; FAX: 82-2-743-6701; E-mail: email@example.com
Liver transplantation is widely performed for end-stage liver disease, and major improvements in outcomes have been achieved over the last 20 years.[1, 2] However, biliary complications are significant causes of morbidity in liver transplant recipients. Biliary cast syndrome (BCS), which is defined as the presence of casts within the hepatic biliary system, occurs in 3% to 18% after orthotopic liver transplantation (OLT).[4, 5] The pathogenesis of BCS is unclear; however, biliary obstructions, acute cellular rejection, biliary infections, prolonged cold ischemia times, hepatic ischemia, and postoperative biliary drainage tubes are etiological factors proposed in previous reports.[6-8]
Leading to biliary obstructions and ascending cholangitis, BCS is associated with increased morbidity, mortality, and graft rejection after OLT.[9, 10] In the early years of liver transplantation, open surgery and retransplantation were the only treatment modalities for BCS, and the success rates were low.[4, 7, 11] Nowadays, the initial management of BCS involves an endoscopic or percutaneous approach, and the success rates are improving.[5, 12, 13] However, because BCS is an uncommon complication after OLT, the overall clinical outcomes and prognosis of BCS remain poorly defined. Thus, we aimed to study the long-term treatment outcomes and prognosis of BCS in patients undergoing liver transplantation.
PATIENTS AND METHODS
Patients and Clinical Data
The medical records of adult patients who developed biliary casts after liver transplantation at a single tertiary hospital (Seoul National University Hospital) from January 2000 through September 2009 were reviewed retrospectively. The enrolled patients were followed until July 2011. The protocol was reviewed and approved by the institutional review board.
Patient characteristics, including the sex, age, and underlying liver disease at the time of OLT, were collected. Surgical details, including the place of transplantation, the type of biliary reconstruction, and the use of T-tubes, were obtained. Posttransplant data, including information about acute cellular rejection, biliary strictures, vascular strictures, and the diagnosis and treatment of BCS, were also recorded. Acute cellular rejection was diagnosed on the basis of pathological findings in a liver biopsy specimen: portal lymphoid aggregates, subendothelial lymphoid infiltration, and bile duct damage. Portal inflammation and bile duct damage were also seen in patients with biliary obstructions, but the changes were mild, and endotheliitis was not present.
The clinical and laboratory data at the diagnosis of BCS, the treatment methods, and the follow-up results were noted. With these data, the clinical characteristics, treatment outcomes, survival, and relapses of patients with BCS were evaluated.
The diagnoses of biliary casts were made by the radiologist and/or endoscopist after the identification of irregular, linear, and/or castlike debris in the biliary system on endoscopic retrograde cholangiopancreatography (ERCP) or percutaneous transhepatic cholangiography (PTC). The diagnoses of biliary strictures were also based on ERCP or PTC.
The complete removal of biliary casts was defined as no residual casts on imaging with no clinical symptoms of jaundice or cholangitis. The successful treatment of BCS was defined as the complete removal of casts with patient and graft survival. The time of relapse was based on recurrent biliary casts identified during endoscopic or percutaneous interventions or cholangiographic findings representing the recurrence of biliary casts. The relapse frequency was defined as the number of cast relapses during follow-up (ie, relapses divided by follow-up years).
Overall survival was defined as the length of time after the diagnosis of BCS until death due to any cause. The relapse-free time was defined as the length of time after the complete removal of biliary casts until the time of relapse.
All the endoscopic treatments were performed with a therapeutic duodenoscope (TJF-240, JF-240, TJF-260, and JF-260, Olympus, Tokyo, Japan) with fluoroscopic assistance. Biliary casts were removed with a basket catheter. When biliary strictures were detected, balloon dilatation was performed with a Hurricane or CRE balloon catheter (Boston Scientific, Natick, MA), and a plastic or fully covered, self-expandable metal stent (Niti-S, Taewoong Medical, Gimpo-Si, Korea) was inserted if this was necessary.
During percutaneous treatments, the biliary system was visualized with a Chiba needle under fluoroscopic guidance. A 5F puncture needle catheter was introduced into the biliary system, and an 8.5F or 10F external drainage catheter was inserted percutaneously into an appropriate intrahepatic duct. Biliary casts were removed with a basket catheter. When biliary strictures were detected, balloon dilatation was performed, and plastic stents were inserted through the percutaneous transhepatic biliary drainage (PTBD) tract if this was necessary.
The endoscopic approach was initially preferred because it was less invasive and required less time for treatment than the percutaneous approach. However, the percutaneous approach was initially chosen when the vital signs of a patient were unstable, when a T-tube was in situ, and when casts were in intrahepatic ducts from which endoscopic removal was difficult.
The total bilirubin and alkaline phosphatase levels of patients with BCS before and after biliary cast removal were compared with the Wilcoxon signed-rank test. The relationships between clinical factors and the BCS relapse frequency were analyzed with the Mann-Whitney U test. The survival time was calculated via the Kaplan-Meier method with the log-rank test. The relapse-free times and associated variables were calculated via the Kaplan-Meier method with the log-rank test. To identify independent factors associated with the early relapse of BCS, a multivariate analysis was performed according to a Cox proportional hazards model. A P value < 0.05 was considered to be statistically significant. All statistical analyses were performed with SPSS 17.0 for Windows (SPSS, Inc., Chicago, IL).
Overall Characteristics of Patients With BCS at the Time of OLT
In all, 1097 patients underwent liver transplantation between January 2000 and September 2009. Five hundred sixty-three patients underwent OLT at Seoul National University Hospital, 511 patients underwent OLT abroad, and 23 patients underwent OLT at other domestic hospitals. BCS developed in 23 patients (2.1%) within a median follow-up of 10.2 months (range = 0.9–54.9 months) after liver transplantation. The incidence of BCS was 1.2% at our hospital and 3.1% overseas. At the time of OLT, the median age was 54.5 years (range = 42–64 years), and 17 patients (74%) were male. The reasons for liver transplantation were hepatocellular carcinoma in 11 patients and liver cirrhosis in 12 patients (Table 1).
Table 1. Demographic Characteristics of Patients With BCS at the Time of OLT and Perioperative Details (n = 23)
Seven patients underwent liver transplantation at Seoul National University Hospital: 5 patients received livers from living donors, and 2 patients received livers from heart-beating cadaveric donors. Sixteen patients underwent liver transplantation from an institute abroad, and all of them received livers from cadaveric donors: 7 non–heart-beating donors, 1 heart-beating donor, and 8 donors of an unknown status. Three patients experienced acute cellular rejection after OLT, and 2 of these patients who underwent liver biopsy at Seoul National University Hospital experienced mildly active rejection. However, the rejection severity for the patient who underwent liver biopsy abroad was not reported.
All the patients had benign biliary strictures: 13 patients had anastomosis site strictures, and 10 patients had nonanastomosis site strictures.
Diagnosis and Management of BCS
BCS was confirmed by ERCP in 10 patients (43%), by PTC in 9 patients (39%), and by T-tube cholangiography in 4 patients (17%). At the diagnosis of BCS, the median age of the patients was 56 years (range = 4265 years), and the median levels of total bilirubin and alkaline phosphatase were 2.7 mg/dL (range = 0.720.8 mg/dL) and 321.0 U/L (range = 871173 U/L), respectively. Seven patients (30%) had biliary casts in common bile ducts, whereas 8 patients (35%) had casts in intrahepatic ducts, and 8 patients (35%) had casts in both common bile ducts and intrahepatic ducts.
Primary biliary cast removal was attempted with an endoscopic approach in 11 patients and with a percutaneous approach in 12 patients (Fig. 1). One patient who initially tried ERCP was switched to percutaneous cast removal because of failed bile duct cannulation, and another patient who initially tried PTBD was switched to endoscopic cast removal because of intrahepatic duct variation. Casts were completely removed from 22 patients, and the initial success rate was 96% (Fig. 2A,B). However, cast removal for 1 patient who initially underwent a percutaneous approach failed despite repetitive procedures, and the patient died of intractable biliary sepsis 3.7 months after the diagnosis of BCS. After 1 month of biliary cast removal, the median levels of total bilirubin and alkaline phosphatase for the 22 patients for whom cast removal was successful had significantly declined to 1.4 mg/dL (range = 0.713.0 mg/dL, P < 0.001) and 234 U/L (range = 90606 U/L, P = 0.01), respectively. There were 5 treatment-related complications during the initial treatment of BCS [3 cases of cholangitis (2 by ERCP and 1 by PTBD) and 2 cases of pancreatitis (1 by ERCP and 1 by PTBD)], and the complication rates showed no significant differences between the 2 treatment methods. All the patients repeatedly underwent balloon dilatation by an endoscopic or percutaneous approach because of biliary strictures.
Percutaneous approaches were preferred for the patients with intrahepatic duct casts, but there was no significant difference (81% versus 57%, P = 0.32; Table 2). Also, endoscopic approaches were preferred for the patients without intrahepatic duct casts (86% versus 44%, P = 0.09).
Table 2. Clinical Characteristics According to the Presence of Intrahepatic Duct Casts
Intrahepatic Duct Cast
Yes (n = 16)
No (n = 7)
Percutaneous approach [n (%)]
Endoscopic approach [n (%)]
Initial success rate for cast removal [n (%)]
Relapse rate for casts [n (%)]
Moreover, 18 patients underwent plastic stent insertion for the treatment of biliary strictures (7 patients with an endoscopic approach, 8 patients with a percutaneous approach, and 3 patients with both approaches). To resolve biliary strictures, we inserted as many plastic stents as possible. A maximum of 5 plastic stents were inserted into 2 patients (Fig. 2C). However, despite undergoing endoscopic plastic stent insertion 4 times, 1 patient still had a biliary stricture, so a fully covered, self-expandable metal stent was inserted (Fig. 2D).
BCS relapsed in 12 of the 22 patients (55%), and the mean number of relapses was 2.1 (range = 13) during a median follow-up of 63.7 months. Recurrent casts were successfully removed from all patients; however, multiple sessions of endoscopic cast removal (mean = 3.8, range = 112) and insertions of percutaneous biliary tubes (mean = 1.3, range = 13) were needed. The mean time interval for an endoscopic procedure was 7.0 months (range = 0.129.0 months). Percutaneous tubes were maintained for a long period as well (mean = 24.8 months, range = 0.365.7 months).
Long-Term Treatment Outcome of BCS
The mean relapse-free time and the mean survival time for patients with BCS were 45.5 months (95% confidence interval = 29.561.5 months) and 113.9 months (95% confidence interval = 93.1134.7 months), respectively. At the end of follow-up, 12 patients still had biliary strictures and were receiving ongoing treatment: 4 patients had biliary stents, 1 patient had a percutaneous transhepatic tube, and 7 patients had both biliary stents and percutaneous transhepatic tubes. Persistent biliary strictures were significantly associated with the BCS relapse frequency (0.37 versus 0.05 times per year, P = 0.006; Table 3). Three patients remained listed for repeat OLT because 2 had persistent biliary strictures with recurrent cholangitis and 1 had progressed to worsening liver cirrhosis. While waiting for retransplantation, the patient with worsening liver cirrhosis died of biliary sepsis.
Table 3. Clinical Factors Associated With the Relapse Frequency of BCS
Relapse Frequency (Times/Year)
NOTE: The data are presented as means and standard deviations.
During follow-up, 4 patients died, and the 5-year overall survival rate for BCS patients after liver transplantation was 87% (Fig. 3A). Three patients died from biliary sepsis, and 1 patient died from duodenal ulcer perforation; BCS relapse was not significantly associated with the mean survival time (80 versus 60 months, P = 0.23). The overall success rate for the treatment of BCS was 78%.
The relapse-free times of the 22 patients who initially underwent successful cast removal were analyzed (Fig. 3B and Table 4). Six of the 12 patients who relapsed did so less than 1 year after the procedure, and the other 6 patients relapsed after 1 year. The relapse frequencies were similar for the 2 groups (0.46 versus 0.46 times per year, P = 0.87). However, there were significantly more ERCP sessions for the treatment of BCS in the patients with BCS relapse within 1 year (8.5 versus 3.8, P = 0.03); also, PTBD sessions tended to be more frequent in these patients (1.3 versus 1.0, P = 0.28). The risk factors associated with early relapse were female sex, hepatic artery stenosis, and acute cellular rejection. In a multivariate analysis, acute cellular rejection was the only significant risk factor for early relapse (hazard ratio = 11.1, 95% confidence interval = 1.963.2, P = 0.007).
Table 4. Clinical Factors Associated With the Relapse-Free Time for BCS
NOTE: The data are presented as means and 95% confidence intervals.
Hepatic artery stenosis
Acute cellular rejection
Intrahepatic duct casts
Insertion of biliary stents
The aim of this study was to investigate the long-term clinical outcomes of BCS, and the current study indicates that patients with BCS frequently relapse and that an endoscopic or percutaneous approach can be used as the first treatment method for cast removal. On the basis of these prolonged follow-up data, we can state that multiple interventions are required for the management of BCS, and the treatment of combined biliary strictures is important.
Biliary casts have developed in up to 18% of liver transplant recipients in previous studies.[14, 15] However, previous values for the incidence of BCS were overestimated because some patients with biliary sludge were diagnosed with BCS. In addition, the progress of liver transplant techniques has diminished postoperative biliary complications. In a recent study that enrolled 9 patients with BCS after liver transplantation, BCS occurred in 2.5% of liver transplant recipients during a mean overall follow-up of 14.5 months. In the present study, BCS developed in 2.1% of liver transplant recipients during a mean overall follow-up of 59.9 months. The rates of BCS were similar in the 2 studies despite differences in the follow-up periods. This could be explained by prior data showing that BCS develops within the first year after liver transplantation.[5, 10] In the present study, the median time from liver transplantation to the diagnosis of BCS was 10.2 months, and this is in agreement with previous studies.
The incidence of BCS was higher among the patients who underwent OLT abroad versus the patients who underwent OLT at our institute (3.1% versus 1.2%). Because patients with complications after OLT from abroad were more likely to be referred to our institute, the incidence of BCS in patients from abroad could have been overestimated.
Biliary cast formation is significantly associated with biliary strictures.[5, 7] We found that all the patients with BCS had biliary strictures after OLT. This finding strongly supports a relationship between biliary casts and biliary strictures. Also, this means that ischemic insults during liver transplantation might be precursors to BCS because biliary strictures have ischemic etiologies.[16-18] At the end of follow-up, 52% of the patients with BCS still had biliary strictures and were undergoing ongoing treatment for strictures. Because unsolved biliary strictures were associated with the BCS relapse frequency, treatments for combined biliary strictures in patients with BCS are important.
Patients with BCS tended to relapse earlier in combination with acute cellular rejection after OLT in this study. Acute cellular rejection is associated with many complications after OLT and indicates a poor prognosis. Furthermore, acute cellular rejection is a significant risk factor for biliary anastomotic strictures after OLT. Acute cellular rejection causes pathological changes, including portal lymphoid aggregation, subendothelial lymphoid infiltration, and bile duct damage. This postoperative damage would result in BCS.
Because patients with BCS easily relapse, repetitive sessions of cast removal are required; also, percutaneous tubes should be placed for a long time. BCS will debase the quality of life and induce more complications. Moreover, BCS has been thought to negatively affect graft and patient survival. However, the overall survival rate of the BCS patients in this study was not inferior to the overall survival rate of general liver transplant recipients. Appropriate management for biliary casts and combined biliary infections would compensate for reduced survival.
Endoscopic and percutaneous therapies are known to be less effective in treating BCS, and the success rate for BCS therapy has been reported to be 25% to 60%.[5, 12] Although casts have been removed by endoscopic or percutaneous interventions, surgical treatments have been necessary for most patients. In our series, the overall success rate for the treatment of BCS by endoscopic and percutaneous approaches was 78% at the end of follow-up. This supports the use of an endoscopic or percutaneous intervention as the primary treatment for BCS. Neither endoscopic approaches nor percutaneous approaches were superior for the removal of casts. Also, it was hard to compare the 2 methods in this study because many patients alternately underwent endoscopic and percutaneous procedures during follow-up. However, percutaneous approaches were preferred for the patients with intrahepatic duct casts, whereas endoscopic approaches were preferred for the patients without intrahepatic duct casts.
A limitation of our investigation is the retrospective nature of the study. Several important factors for BCS such as the warm ischemia time were not noted because approximately two-thirds of the patients underwent OLT at institutes abroad. In addition, because 1 diagnosis of acute rejection was based only on a pathological report without any mention of the severity, we should be cautious about interpreting the association of acute rejection with relapse. Moreover, the number of enrolled patients was relatively small because of the rarity of BCS. Therefore, prospective and multicenter studies would be required to clarify the long-term clinical outcomes of BCS.
Despite these limitations, we believe that this study provides useful information about the treatment and clinical course of BCS because, to the best of our knowledge, this study includes the largest number of patients with BCS over a lengthy follow-up period. We think that this study can provide some guidance to clinicians when they are making decisions about the treatment of BCS.
In summary, patients with BCS frequently relapse, and multiple interventions and treatments for combined biliary strictures are necessary. However, with respect to treatment, endoscopic or percutaneous approaches can be quite successful, and the overall survival of BCS patients can be improved if appropriate management is provided. Acute cellular rejection after liver transplantation is a significant risk factor for early relapse.