Endoscopic retrograde cholangiopancreatography (ERCP) has been substitute for surgery to treat most common bile duct (CBD) stones since it was introduced four decades ago.¹ About 85% to 95% of CBD stones can be successfully removed by conventional ERCP with endoscopic sphincterotomy (EST).^2–4 However, difficult-to-remove stones need stone-fragmentation methods, such as, mechanical lithotripsy (ML), intraductal shock wave lithotripsy, extracorporeal shock wave lithotripsy, or biliary stenting with chemical dissolution.¹

Multiple factors, such as, stone number (>10), size (>15 mm), shape (barrel-shaped), and location (intrahepatic, cystic duct) can cause difficulties during stone removal.⁵ Recently, Kim et al.⁶ identified anatomical factors of the bile duct that contribute to technical difficulties during the endoscopic clearance of CBD stones in patients with acute distal CBD angulation (≤135°) and a short distal CBD arm (≤36 mm). However, no report has associated a stemware-shape and the difficulty of stone removal. In our clinical experience, even a stone diameter of <1 cm in a stemware-shaped CBD can cause post-ERCP complications, such as, basket impaction or bile duct perforation during ERCP.

Accordingly, in the current study, we evaluated the clinical characteristics and outcomes of the endoscopic treatment of stones in stemware-shaped CBDs.

Between January 2008 and December 2012, 34 CBD stone patients who visited eight tertiary referral centers in Korea were enrolled in this study. ERCP images were searched for stemware-shaped CBDs, which were defined as CBDs that met one or more of the following criteria: (1) an intrapancreatic extrahepatic duct (EHD) diameter of <1/2 of the maximum diameter of extrapancreatic portion of the EHD; (2) an intrapancreatic narrow segment of relatively constant diameter with a difference between the smallest and largest portion of <5 mm, and must meet the following criteria at the same time; (3) a proximal biliary tree dilation degree, including the intrahepatic duct (IHD), proportional to that of the extrapancreatic CBD (Fig. 1).

Medical records and imaging studies, including abdominal computed tomography, magnetic resonance cholangiopancreatography (MRCP), and ERCP, were retrospectively reviewed. The study protocol was approved by the Institutional Review Board of all participating centers (IUH-IRB 14-058).

Thirty-four patients with stones in a stemware-shaped CBD were retrospectively enrolled. Baseline characteristics of the study subjects and the characteristics of stones are summarized in Table 1. Median patient age was 56.8 years (range, 26 to 91 years), median total bilirubin was 5.9 mg/dL (range, 0.4 to 22.3 mg/dL), and the median number of CBD stones per patient was 1 (range, 1 to 10). Median diameter of the extrapancreatic CBD portion was 15.4 mm (range, 8.5 to 32.1 mm), median diameter of the intrapancreatic CBD portion was 5.4 mm (range, 2.6 to 13.4 mm), and the median transverse diameter of stones was 12.3 mm (range, 1.8 to 30 mm).

A post-ERCP complication occurred in five patients (14.7%); perforation in two, and post-ERCP pancreatitis, basket impaction, and hyperamylasemia in one patient apiece. No mortality occurred (Table 2).

Initial procedural success and overall success rates were 5.9% and 41.2%, respectively, for first-line procedures, demonstrating the poor efficacy of conventional ERCP for the treatment of stones in a stemware-shaped CBD. Twenty of the 34 patients (58.8%) were converted to a second-line procedure (Table 3). One patient who failed the first-line procedure underwent choledocholithotomy as a second-line of procedure and succumbed to acute renal failure 8 days after surgery.

Eight patients had a maximum stone size of <1 cm and 26 had one of >1 cm. Overall success rates for first-line procedures in these two groups were 62.5% and 34.6%, respectively. ML was needed in 60% and 66.7% of patients, respectively, in each group in whom stone retrieval by conventional ERCP succeeded, and 37.5% and 65.4%, respectively, were converted to a second-line procedure (Table 4). Twenty-two patients were followed up for at least 6 months after stone removal, and only two (9.1%) developed stone recurrence. Median follow-up was 24 months (range, 1 to 64 months).

ERCP is one of the most widespread endoscopic methods, and is used to remove CBD stones in most cholelithiasis patients. CBD stones can usually be retrieved endoscopically by conventional ERCP. However, multiple factors can affect the success or failure of endoscopic removal. The reasons why treatment fail are as follows; difficulties accessing the bile duct (periampullary diverticulum, sigmoid-shaped CBD, and altered anatomy due to gastric surgery), a large stone (diameter >15 mm), a large number of stones (≥10), an unusual stone shape (barrel-shaped), or stone location (intrahepatic, cystic duct, proximal to a stricture).⁷

Some studies have been conducted on relations between anatomical CBD condition and CBD stone recurrence or the difficulty of endoscopic stone retrieval. Keizman et al.⁸ demonstrated that more acute angulation of the CBD (≤145°) influences symptomatic stone recurrence after successful endoscopic treatment. Recently, Kim et al.⁶ first described anatomical bile duct features that increase the technical difficulty of endoscopic stone removal in patients with acute distal CBD angulation and a short distal CBD arm. However, no previous study has addressed the effect of a stemware-shaped CBD on difficult CBD stone removal. In this study, we devised a definition for stemware-shaped CBD and excluded other anatomical anomalies, such as, choledochal cyst and biliary stricture.

In practice, we have experienced many technical difficulties removing stones from a stemware-shaped CBD by ERCP. Even when stones are small, ERCP is laborious, often requires mechanical lithotripsy, and sometimes fails. Furthermore, in such cases, adverse events, such as, basket impaction or perforation, occur frequently. Initially, we considered that failures are caused by a narrow, long, and rigid intrapancreatic portion. Post-ERCP perforation occurred in two of our patients (5.9%); one during biliary insertion of guidewire and the other after balloon dilation of the papilla. We chose a large diameter balloon catheter for EPBD because extrapancreatic CBD portions were relatively wide and because we failed to recognize stemware-shaped anatomies before and during procedures. Accordingly, this study demonstrates the importance of assessing distal CBD shape by imaging study before the endoscopic procedure, including balloon dilatation, to prevent those avoidable complications.

In the present study, overall success rates of complete bile duct clearance at initial and after first-line procedures were only 5.9% and 41.2%, respectively. In addition, more than 70% of patients that underwent successful complete stone removal at first-line procedure needed more than two sessions of ERCP, and for stones of <1 cm, the overall success rate of first-line procedures was only 62.5% and ML was needed in 60% of patients.

The recurrence rate of CBD stones has been reported to range from 4% to 24%.⁹ Recurrence is defined as the development of a stone at least 6 months after complete stone clearance.⁹ The risk factors for recurrence are a bile duct diameter of >13 mm, peri-ampullary diverticulum (type I or II), CBD angulation, previous history of cholecystectomy, and an age exceeding 80 years.^8–11 Stemware-shaped CBDs exhibit proximal dilatation and distal stenosis, which could cause bile stasis, and thus, increase the risk of stone recurrence. However, in the current study, the recurrence rate was not too high. Furthermore, comparing our findings with those of previous studies, it would appear stone recurrence is not higher in patients with a stemware-shaped CBD, but the number of patients recruited was too small to determine the real recurrence rate accurately.

Although the present study is limited by its small sample size and retrospective nature, it was conducted on a multicenter basis and describes for the first time the clinical features and outcomes of stones located in stemware-shaped CBDs.

The endoscopic treatment of stones in stemware-shaped CBDs is challenging and requires careful assessment before endoscopic procedure. Endoscopists may need to prepare for mechanical lithotripsy or other treatment options before commencing ERCP even for small stones in stemware-shaped CBDs.