Metal or plastic stents for preoperative biliary drainage in resectable pancreatic cancer

INTRODUCTION

In pancreatic cancer, preoperative biliary drainage (PBD) increases complications compared with surgery without PBD, demonstrated by a recent randomised controlled trial (RCT). This outcome might be related to the plastic endoprosthesis used. Metal stents may reduce the PBD-related complications risk.

METHODS

A prospective multicentre cohort study was performed including patients with obstructive jaundice due to pancreatic cancer, scheduled to undergo PBD before surgery. This cohort was added to the earlier RCT (ISRCTN31939699). The RCT protocol was adhered to, except PBD was performed with a fully covered self-expandable metal stent (FCSEMS). This FCSEMS cohort was compared with the RCT's plastic stent cohort. PBD-related complications were the primary outcome. Three-group comparison of overall complications including early surgery patients was performed.

RESULTS

53 patients underwent PBD with FCSEMS compared with 102 patients treated with plastic stents. Patients' characteristics did not differ. PBD-related complication rates were 24% in the FCSEMS group vs 46% in the plastic stent group (relative risk of plastic stent use 1.9, 95% CI 1.1 to 3.2, p=0.011). Stent-related complications (occlusion and exchange) were 6% vs 31%. Surgical complications did not differ, 40% vs 47%. Overall complication rates for the FCSEMS, plastic stent and early surgery groups were 51% vs 74% vs 39%.

CONCLUSIONS

For PBD in pancreatic cancer, FCSEMS yield a better outcome compared with plastic stents. Although early surgery without PBD remains the treatment of choice, FCSEMS should be preferred over plastic stents whenever PBD is indicated.

TRIAL REGISTRATION NUMBER

Dutch Trial Registry (NTR3142).

Systematic review and meta-analysis of somatostatin analogues for the treatment of pancreatic fistula

BACKGROUND

Somatostatin analogues are used for the treatment of pancreatic fistula, with the aim of achieving fistula closure or reduction of output.

METHOD

MEDLINE, Embase and Cochrane databases were searched systematically for relevant articles followed by hand-searching of reference lists. Data on patient recruitment, intervention and outcome were extracted and meta-analysis performed where reasonable.

RESULTS

Seven randomized clinical trials met the inclusion criteria and included a total of 297 patients with fistulas of the gastrointestinal tract; of these, 102 patients had fistulas of pancreatic origin. Pooling of closure rates showed no significant difference between patients treated with somatostatin analogues compared with controls: odds ratio 1·52 (95 per cent confidence interval 0·88 to 2·61). Owing to inconsistent descriptions, pooling of results was not possible for other endpoints, such as time to fistula closure.

CONCLUSION

There is no solid evidence that somatostatin analogues result in a higher closure rate of pancreatic fistula compared with other treatments.

Multidisciplinary management of hilar cholangiocarcinoma (Klatskin tumor): extended resection is associated with improved survival

BACKGROUND

Effective diagnosis and treatment of patients with hilar cholangiocarcinoma (HCCA) is based on the synergy of endoscopists, interventional radiologists, radiotherapists and surgeons. This report summarizes the multidisciplinary experience in management of HCCA over a period of two decades at the Academic Medical Center in Amsterdam, with emphasis on surgical outcome.

METHODS

From 1988 until 2003, 117 consecutive patients underwent resection on the suspicion of HCCA. Preoperative work-up included staging laparoscopy, preoperative biliary drainage, assessment of volume/function of future remnant liver and radiation therapy to prevent seeding metastases. More aggressive surgical approach combining hilar resection with extended liver resection was applied as of 1998. Outcomes of resection including actuarial 5-year survival were assessed.

RESULTS

Eighteen patients (15.3%) appeared to have a benign lesion on microscopical examination of the specimen, leaving 99 patients with histologically proven HCCA. These 99 patients were analysed according to three 5-year time periods of resection, i.e. period 1 (1988-1993, n=45), 2 (1993-1998, n=25) and 3 (1998-2003, n=29). The rate of R0 resections increased and actuarial five-year survival significantly improved from 20±5% for the periods 1 and 2, to 33±9% in period 3 (p<0.05). Postoperative morbidity and mortality in the last period were 68% and 10%, respectively.

CONCLUSION

Extended surgical resection resulted in increased rate of R0 resections and significantly improved survival. Candidates for resection should be considered by a specialized, multidisciplinary team.

Photodynamic therapy and Klatskin tumour: An overview

The prognosis of patients with an unresectable bile duct cancer is poor. In 60-70% of patients, cholangiocarcinoma is located in the hepatic duct bifurcation and known as Klatskin tumour. Surgical resection offers the only chance for 5-year survival, but less than 20% are surgical candidates. Patients with unresectable cholangiocarcinoma are treated with biliary drains, but commonly die of liver failure or cholangitis due to biliary obstruction within 6 to 12 months. Chemotherapy and/or radiotherapy have not been evaluated in randomized, controlled trials. Photodynamic therapy (PDT) is a new and promising locoregional treatment, the aim of which is to destroy tumour cells selectively. PDT involves the injection of a photosensitizer followed by percutaneous or endoscopic direct illumination of the tumour with light of a specific wavelength. In recent non-randomized studies of small numbers of patients with unresectable cholangiocarcinoma, PDT induced a decrease in serum bilirubin levels, improved quality of life and a slightly better survival. Other non-randomized trials failed to show clinical benefits. Recently, the first prospective, randomized controlled study with PDT in a selected group of non-resectable cholangiocarcinoma patients was stopped prematurely. The improvement in survival in the PDT-randomized patients was so impressive that it was considered to be unethical to continue randomization. However, further studies are awaited in unselected patients with unresectable cholangiocarcinoma before PDT can be considered as the standard adjuvant therapy.

Helicobacter pylori eradication and gastric cancer: when is the horse out of the barn?

Helicobacter pylori infection is a major risk factor for gastric cancer development. Therefore, H. pylori eradication may be an important approach in the prevention of gastric cancer. However, long-term data proving the efficacy of this approach are lacking. This report describes two patients who developed gastric cancer at, respectively, 4 and 14 years after H. pylori eradication therapy. These patients were included in a study cohort of H. pylori-infected subjects who received anti-H. pylori therapy during the early years of development of H. pylori eradication therapy and underwent strict endoscopic follow-up for several years. In both patients, gastric ulcer disease and premalignant gastric lesions, i.e., intestinal metaplasia at baseline and dysplasia during follow-up, were diagnosed before gastric cancer development. These case reports demonstrate that H. pylori eradication does not prevent gastric cancer development in all infected patients after long-term follow-up. In patients with premalignant gastric lesions, in particular in patients with a history of gastric ulcer disease, adequate endoscopic follow-up is essential for early detection of gastric neoplasia.

Immunoglobulin G4-related sclerosing cholangitis in patients resected for presumed malignant bile duct strictures

Background

Immunoglobulin (Ig) G4-related lymphoplasmacytic sclerosing pancreatitis has been described in the context of autoimmune pancreatitis mimicking distal cholangiocarcinoma. The aim of this study was to assess the occurrence of this entity in benign bile duct strictures in patients resected for presumed hilar cholangiocarcinoma.

Methods

Of 185 patients who had undergone resection of proximal bile ducts on suspicion of hilar cholangiocarcinoma between January 1984 and June 2005, 32 (17·3 per cent) had a benign bile duct stricture on histopathological examination. After re-evaluation, further immunohistochemical analysis was performed on specimens from patients with features of autoimmune-like disease.

Results

The periductal stroma in 15 patients showed features of autoimmune-like disease (diffuse, moderate to severe lymphoplasmacytic infiltration with marked fibrosis). Abundant IgG4-positive plasma cell infiltration around the bile duct lesions was seen in two of these. Although not significant, patients with features of autoimmune-like disease on histological changes showed a higher incidence of recurrent biliary complications than those without (P = 0·250).

Conclusion

Features of autoimmune-like bile duct disease were seen in almost half (15 of 32) of patients with benign hilar strictures resected for presumed hilar cholangiocarcinoma. Frank IgG4-related sclerosing disease was found in only two of the 15 patients with autoimmune-like bile duct disease.

[Dutch study on the optimal treatment strategy for patients with a first or second occurrence of gastro-oesophageal variceal bleeding: the TIPS-TRUE trial]

The main options for secondary prevention of gastrooesophageal variceal bleeding are endoscopic therapy and treatment with propranolol. Creation ofa transjugular intrahepatic portosystemic shunt (TIPS) is currently considered a valuable secondary 'rescue' treatment when other therapies fail. Recent data suggest that the use of covered stents markedly increases the efficacy of TIPS, compared with conventional uncovered stents. Therefore, a multicentre randomised trial was designed to compare the effects of TIPS using covered stents with those of endoscopic therapy plus propranolol in patients with a first or second episode ofgastro-oesophageal variceal bleeding. TIPS will be performed in 4 university centres with relevant expertise. The trial will hopefully gain nationwide support, and all centres in The Netherlands are cordially invited to participate.

Incidence and management of bile leakage after partial liver resection

BACKGROUND/AIMS

Bile leakage after partial liver resection still is a common complication and is associated with substantial morbidity and even mortality.

METHODS

A total of 234 consecutive liver resections without biliary reconstruction, performed between January 1992 and December 2004, were analyzed for postoperative bile leakage.

RESULTS

Postoperative bile leakage occurred in 6.8% of patients (16/234). In univariate analysis, male gender (p = 0.037), major liver resection (p = 0.004), right-sided hepatectomy (p = 0.005), prolonged operation time (p = 0.001), intraoperative blood loss >500 ml (p = 0.009), red cell transfusion (p = 0.02), tumor size (p = 0.026), duration of vascular occlusion (p = 0.03) and surgical irradicality (p = 0.001) were risk factors. No independent risk factors were associated with bile leakage after liver resection. Bile leakage originated from the resection plane in 10 patients (63%). Endoscopic biliary decompression was performed in 9 patients as initial treatment, and percutaneous drainage of the bile collection was used in 4 patients. Bile leakage resolved spontaneously in 3 patients.

CONCLUSIONS

Bile leakage is a persisting complication and in this study occurred in 6.8% of patients after partial liver resection. Percutaneous drainage of bile collection with or without endoscopic biliary decompression are effective interventions in the management of most cases of bile leakage.

Staging of cholangiocarcinoma: the role of endoscopy

The main question for staging is resectability, which is reliant on vascular, longitudinal, and metastatic spread. Today, accurate staging of perihilar tumors is achieved by non-invasive diagnostic investigations. Direct cholangiography has been the gold standard as a diagnostic procedure in recent decades. Endoscopic retrograde cholangiopancreaticography (ERCP) often only shows the ducts below the obstruction, and visualization of an obstructed part of the biliary tree is often not possible. Direct cholangiography reveals no information about local tumor extension, lymph nodes, or vascular involvement. Because of the given limitations, potential complications (cholangitis, sepsis) associated with direct cholangiography and reduction of the accuracy of subsequent cross-sectional imaging studies, these invasive techniques should only be used in the case of palliative interventions. Endoscopic ultrasonography-guided fine-needle aspiration (EUS-FNA) can be used to assess the nature of biliary strictures and to derive information about the extent of periductal disease and the presence of lymph node metastases. In a study by Fritscher-Ravens, 44 patients with hilar strictures underwent EUS-FNA. The overall diagnostic accuracy, sensitivity, specificity, positive and negative predictive values were 91% (95% CI, 78.4-96.3%), 89% (95% CI, 73.3-96.8%), 100% (95% CI, 63.1-100%), 100% (95% CI, 88.8-100%), and 67% (95% CI, 34.9-90%), respectively. The planned surgical approach was changed in 27 of 44 patients. In 15-20% of cholangiocarcinoma, patients with unremarkable abdominal imaging studies have metastatic lymph node involvement according to EUS evaluation. Due to the risk of peritoneal seeding, however, EUS with FNA is not recommended in patients still with a potential curative tumor.

Original article: New surgical approaches to the Klatskin tumour

BACKGROUND

Surgical treatment of hilar cholangiocarcinoma (Klatskin tumours) is difficult because of its central location in the liver hilum. Recent developments in surgical techniques have improved the outcome after resection.

AIM

To describe the surgical approaches currently applied in our centre and the impact of these strategies on outcome and criteria for resection.

METHODS

From 1988 to 2003, 99 consecutive patients underwent resection for hilar cholangiocarcinoma. Patients were analysed for rate of R0 resections in relation with Bismuth classification. Morbidity, mortality and survival were assessed.

RESULTS

The rate of hilar resections in combination with (extended) liver resections for type III and IV tumours increased from 24% to 95% in the last 5 years of the study period. Eight patients (8%) had Bismuth type IV tumours. Four of these patients underwent palliative local excisions of the hepatic duct confluence whereas the other four patients underwent hilar resection in combination with partial liver resection, resulting in microscopically radical resections. There was no mortality in this group. Overall postoperative morbidity and mortality were 68% and 10%, respectively.

CONCLUSIONS

An aggressive surgical approach consisting of hilar resections combined with partial liver resections including segments 1 and 4, resulted in a higher rate of R0 resections. Even Bismuth type IV tumours may be resectable depending on the biliary anatomy of the hepatic duct confluence.

[Surgical drainage of the pancreatic duct in patients with chronic pancreatitis is more effective than endoscopic drainage: randomized trial]

OBJECTIVE

To compare endoscopic and surgical drainage of the pancreatic duct for ductal decompression in patients with severe pain due to chronic pancreatitis and a dilated pancreatic duct.

DESIGN

Randomized clinical trial.

METHOD

All symptomatic patients with chronic pancreatitis and a distal obstruction of the pancreatic duct, but without an inflammatory mass, were eligible for this study. Patients were randomized to endoscopic transampullary pancreatic duct drainage or to operative pancreaticojejunostomy. The primary end point was the average Izbicki pain score, measured during 2 years of follow-up. The secondary endpoints were pain relief at the end of follow-up, physical and mental health, morbidity, mortality, hospital stay and number of procedures performed.

RESULTS

Of 118 patients who were evaluated between January 2000-October 2004 39 patients were randomized; 19 were treated endoscopically (16 of whom underwent lithotripsy) and 20 by operative pancreaticojejunostomy. During 24 months of follow-up, compared with endoscopic drainage, surgery was associated with lower Izbicki pain scores (51 versus 25; p < 0.001) and better SF-36 physical health summary scores (p = 0.003). Furthermore, at the end of follow-up, pain relief was achieved in 32% of patients randomized to endoscopic drainage and 75% of patients randomized to surgical drainage (p = 0.007). Complication rates and hospital stay were similar, but endoscopic treatment required more procedures (median 8 versus 3; p < 0.001).

[Claims for damages as a result of bile-duct injury during (laparoscopic) cholecystectomy]

OBJECTIVE

To evaluate the frequency of claims for damages initiated by patients referred to a tertiary centre for the treatment of bile-duct injury after a (laparoscopic) cholecystectomy. To determine the relationship between patient characteristics and the initiation of a claim procedure.

DESIGN

Descriptive.

METHOD

Between 1 January 1990 and 31 December 2005, 500 patients with a bile-duct injury were referred to the Academic Medical Centre, Amsterdam, 454 of whom in the period up to 31 December 2004. Of these, 403 received a mailed questionnaire about the initiation of legal claims for damages.

RESULTS

The questionnaire was completed and returned by 278 patients (69%), a representative cohort ofthe 500. Of these, 53 (19%) had submitted a claim for damages. The percentage of claims did not increase over the periods 1991-1995 (19%), 1996-2000 (18%) and 2001-2005 (20%). In the univariate analysis, factors associated with the initiation of a claim procedure were: younger age, the severity of the injury, surgical treatment, being employed at the time of the initial cholecystectomy, and having been placed on sick leave. A complete transection of the common bile duct was the only independent predictive factor for starting a claim procedure (odds ratio: 7.5; 95% CI: 1.9-30.6).

[Increasing role for angiographic embolisation in the treatment of gastrointestinal haemorrhage]

Endoscopy is the primary diagnostic and therapeutic modality for the vast majority of patients with haemorrhage of the upper or lower digestive tract. In many hospitals, surgery is the therapy of choice when endoscopy fails or is impossible. In patients who have considerable co-morbidity and who are actively bleeding from the digestive tract, surgery is associated with a relatively high morbidity and mortality. Angiographic embolisation for haemorrhage from the upper or lower digestive tract is effective, with success rates varying from 50 to 90%. The risk of ischaemic complications of the procedure is acceptably low (< 5%). Angiography is not very time-consuming and does not preclude subsequent surgical treatment ifangiographic embolisation does not succeed. However, performing embolisation requires skill and experience and the procedure is not available everywhere. Angiographic embolisation is a valuable alternative to surgery and should be considered in all patients with haemorrhage of the digestive tract who cannot be treated by means of endoscopy.

Improved Outcome of Resection of Hilar Cholangiocarcinoma (Klatskin Tumor)

BACKGROUND

Treatment of hilar cholangiocarcinoma (Klatskin tumors) has changed in many aspects. A more extensive surgical approach, as proposed by Japanese surgeons, has been applied in our center over the last 5 years; it combines hilar resection with partial hepatectomy for most tumors. The aim of this study was to assess the outcome of a 15-year evolution in the surgical treatment of Klatskin tumors.

METHODS

A total of 99 consecutive patients underwent resection for hilar cholangiocarcinoma in three 5-year time periods: periods 1 (1988-1993; n=45), 2 (1993-1998; n=25), and 3 (1998-2003; n=29). Outcome was evaluated by assessment of completeness of resection, postoperative morbidity and mortality, and survival.

RESULTS

The proportion of margin negative resections increased significantly from 13% in period 1 to 59% in period 3 (P<.05). Two-year survival increased significantly from 33%+/-7% and 39%+/-10% in periods 1 and 2 to 60%+/-11% in period 3 (P<.05). Postoperative morbidity and mortality were considerable but did not increase with this changed surgical strategy (68% and 10%, respectively, in period 3). Lymph node metastasis was, next to period of resection, also associated with survival in univariate analysis.

CONCLUSIONS

Mainly in the last 5-year period (1998-2003), when the Japanese surgical approach was followed, more hilar resections were combined with partial liver resections that included segments 1 and 4, thus leading to more R0 resections. This, together with a decrease in lymph node metastases, resulted in improved survival without significantly affecting postoperative morbidity or mortality.

Endoscopic and surgical management of bile duct injury after laparoscopic cholecystectomy

Laparoscopic cholecystectomy has become the first choice of management for symptomatic cholecystolithiasis. While it is associated with decreased postoperative morbidity and mortality, bile duct injuries are reported to be more severe and more common (0-2.7%), when compared to open cholecystectomy (0.2-0.5%) [New Engl. J. Med. 234 (1991) 1073; Am. J. Surg. 165 (1993) 9; Surg. Clin. N Am. 80 (2000) 1127]. These bile duct injuries include leaks, strictures, transection and removal of (part of) the duct, with or without vascular damage. Bile duct injury might be due to misidentification of the biliary tract anatomy due to acute cholecystitis, large impacted stones, short cystic duct, anatomical variations, but also due to technical errors leading to bleeding with subsequent clipping and coagulation trauma [Ann. Surg. 237 (2003) 460]. Early recognition and adequate multidisciplinary approach is the cornerstone for the optimal final outcome. Suboptimal management of injuries often leads to more extensive damage to the biliary tree and its vasculature with as consequences biliary peritonitis, sepsis, abscesses, multiple organ failure, a more difficult (proximal) reconstruction and in the long run, secondary biliary cirrhosis, and liver failure. Despite increasing experience in performing laparoscopic cholecystectomy, the frequency of bile duct injuries has not decreased [Ann. Surg. 234 (2001) 549]. Therapy encompasses endoscopic stenting, percutaneous transhepatic dilatation (PTCD) and surgical reconstruction.

Acute upper GI bleeding: did anything change? Time trend analysis of incidence and outcome of acute upper GI bleeding between 1993/1994 and 2000

OBJECTIVE

The aim of this study was to examine recent time trends in incidence and outcome of upper GI bleeding.

METHODS

Prospective data collection on all patients presenting with acute upper GI bleeding from a defined geographical area in the period 1993/1994 and 2000.

RESULTS

Incidence decreased from 61.7/100,000 in 1993/94 to 47.7/100,000 persons annually in 2000, corresponding to a 23% decrease in incidence after age adjustment (95% CI = 15-30%). The incidence was higher among patients of more advanced age. Rebleeding (16% vs 15%) and mortality (14% vs 13%) did not differ between the two time periods. Ulcer bleeding was the most frequent cause of bleeding, at 40% (1993/94) and 46% (2000). Incidence remained stable for both duodenal and gastric ulcer bleeding. Almost one half of all patients with peptic ulcer bleeding were using nonsteroidal anti-inflammatory drugs or aspirin. Also, among patients with ulcer bleeding, rebleeding (22% vs 20%) and mortality (15% vs 14%) did not differ between the two time periods. Increasing age, presence of severe and life-threatening comorbidity, and rebleeding were associated with higher mortality.

CONCLUSIONS

Between 1993/1994 and 2000, among patients with acute upper GI bleeding, the incidence rate of upper GI bleeding significantly decreased, but no improvement was seen in the risk of rebleeding or mortality in these patients. The incidence rate of ulcer bleeding remained stable. Prevention of ulcer bleeding is important.

Lack of accuracy of the noninvasive Helicobacter pylori stool antigen test in patients with gastroduodenal ulcer bleeding

OBJECTIVE

Oral presentation at Digestive Diseases Week, San Francisco, California, May 2002. The antigen-based stool assay has proven to be accurate in diagnosing Helicobacter pylori infection in dyspeptic patients. We evaluated the H. pylori antigen-based stool assay (HpSA) in patients with peptic ulcer bleeding (PUB).

METHODS

Thirty-six patients with PUB were endoscoped, and antral and corpus biopsy specimens were taken for rapid urease test (RUT), histology, and culture. The first stool sample after admission was collected for the HpSA test. The gold standard was defined as either positive culture or positive RUT and histology. If only RUT or histology was positive, this was defined as indeterminate. To evaluate cross-reaction with blood constituents, citrated blood samples from 10 healthy volunteers (nine H. pylori serology negative and one H. pylori serology positive) were assessed by the HpSA test.

RESULTS

A total of 36 consecutive patients with PUB (21 male) with a mean age of 69.5 yr were included in the study. Using the gold standard, the sensitivity and specificity of the HpSA test were 100% and 52%, respectively. Citrated blood samples of three H. pylori negative and one H. pylori positive volunteer gave a positive result in the HpSA test, suggesting cross-reaction with blood con stituents.

CONCLUSIONS

The HpSA test gave a high number of false- positive results in patients with PUB, probably because of blood constituents cross-reacting in the enzyme immunoassay. The HpSA test is not accurate for testing H. pylori infection in patients with PUB.

Choledochal cysts: age of presentation, symptoms, and late complications related to Todani's classification

PURPOSE

The aim of this study was to compare presentation, complications, diagnosis, and treatment of choledochal cysts in pediatric and adult patients.

METHODS

Forty-two patients were analyzed after subdivision into 3 groups: group A, less than 2 years (n = 10); group B, 2 to 16 years (n = 11); group C, greater than 16 years (n = 21).

RESULTS

The cysts were classified as extrahepatic (n = 33), intrahepatic (n = 5), and combined (n = 4). Seventy-six percent of patients presented with abdominal pain, (20 of 21 group C), and 57% with jaundice, (10 of 10 group A). Cholangiocarcinoma occurred in 6 patients, 4 of whom had previously undergone internal drainage procedures. Excision of the extrahepatic cyst was performed in 27 of 37 patients. Five patients, of whom, 4 had cholangiocarcinoma, were beyond curative treatment at the time of diagnosis. Six patients had died at the closure of this study, 5 of them had carcinoma.

CONCLUSIONS

Presenting symptoms are age dependent with jaundice prevailing in children and abdominal pain in adults. In view of the high risk of cholangiocarcinoma, early resection and not internal drainage is the appropriate treatment of extrahepatic cysts. Patients who had undergone internal drainage in the past still should undergo resection of the cyst.

Natural history of primary sclerosing cholangitis and prognostic value of cholangiography in a Dutch population

BACKGROUND

Median survival of patients with primary sclerosing cholangitis (PSC) has been estimated to be 12 years. Cholangiography is the gold standard for diagnosis but is rarely used in estimating prognosis.

AIMS

To assess the natural history of Dutch PSC patients and to evaluate the prognostic value of a cholangiographic classification system.

PATIENTS

A total of 174 patients with established PSC attending a university hospital and three teaching hospitals from 1970 to 1999.

METHODS

Charts were reviewed for validity and time of diagnosis, concurrent inflammatory bowel disease, interventions, liver transplantation, occurrence of cholangiocarcinoma, and death. Follow up data were obtained from the charts and from the attending clinician or family physician. Median follow up was 76 months (range 1-300). The earliest available cholangiography was scored using a radiological classification system for the severity of sclerosis, developed in our institution. Survival curves were computed by the Kaplan-Meier method. Cholangiographic staging was used to construct a prognostic model, applying Cox proportional hazards analysis.

RESULTS

The estimated median survival from time of diagnosis to death from liver disease or liver transplantation was 18 years. Cholangiocarcinoma was found in 18 (10%) patients. Fourteen patients (8%) underwent liver transplantation. Cholangiographic scoring was inversely correlated with survival. A combination of intrahepatic and extrahepatic scoring, together with age at endoscopic retrograde cholangiopancreatography, proved strongly predictive of survival.

CONCLUSIONS

The observed survival was considerably better than reported in earlier series from Sweden, the UK, and the USA. Classification and staging of cholangiographic abnormalities has prognostic value.