Endoscopic retrograde cholangiopancreatography (ERCP) is an invasive procedure for the diagnosis and treatment of benign and malignant biliary tract and pancreatic duct diseases. The use of biliary stents plays a key role in the treatment of biliary obstruction and drainage. There are different types of stents: plastic stents made of polyethylene, polyurethane, or Teflon; coated metal stents or bare metal stents; stainless steel alloy stents (Wallstent^®) and nitinol or platinol stents, among others¹.

Plastic stents are cheaper but are prone to obstruction and migration compared with metal stents. These devices are more commonly used for short-term biliary drainage (3 to 6 months)² and have a success rate of about 100%. Short term morbidity is low but reaches 33-40% if the stent is left in situ too long due to stent obstruction and cholangitis. Mean duration of stent patency is approximately 12 months³.

Most plastic biliary stents are of the “Amsterdam” type with a slightly curved shaft and flaps near each end to prevent migration. Overall migration rate is 8.58% (proximal migration 4.58% and distal migration 4.00%)⁴. In case of proximal migration, endoscopic removal is the preferred approach; in certain situations, watchful waiting or surgical removal may be considered. Distal migration may be asymptomatic, or present as intra-abdominal abscess or perforation; this may be due to the side flaps which cause entrapment of the stent in the bowel wall, pressure necrosis and perforation.

Case 1. An 80-year-old female patient presented with a painful tumor in the abdominal wall with a fistulous orifice and purulent discharge (Fig. 1). She had a history of ERCP 6 months before, which revealed a common bile duct of 25 mm in diameter with a 20-mm stone inside that was impossible to remove, requiring large papillotomy and placement of a 10 Fr plastic stent. The erect abdominal X-ray demonstrated atypical location of the biliary stent. A fistulography was performed through the abdominal orifice, showing passage of the contrast agent into the duodenum and absence of contrast agent passage in the bile ducts (Fig. 1).

Figure 1 A: fistulous orifice in the right hypochondriac region (arrow). B: local removal of biliary stent (arrow). C: stent removed (arrow). D: CT section showing biliary stent migration (arrow). 

Case 2. A 66-year-old female patient presented with a fistulous orifice in the right hypochondriac region with serous discharge and induration of the area around the lesion (Fig. 2). She had a history of mild cholangitis with required ERCP in another center, which revealed multiple common bile duct lithiasis that was impossible to remove. A 10 Fr plastic stent was placed through a large papillotomy. Seventytwo hours later, the patient underwent laparoscopic cholecystectomy with exploration of the biliary tract by choledochotomy, removal of gallstones and bile duct repair, without removing the biliary stent. The patient evolved with biliary leak which spontaneously solved on postoperative day 7. She did not attend the outpatient clinic for postoperative follow-up for two years and was admitted to our hospital thereafter. Several complementary tests were ordered. The laboratory tests were within normal ranges; the ultrasound showed that the intrahepatic and extrahepatic biliary ducts were not dilated, and a tubular hyperechoic structure was observed in the subcutaneous tissue with projection to the hepatic region. The Computed tomography (CT) scan of the abdomen revealed mild intrahepatic bile duct dilation with a 100 mm-long tubular image in close contact with the bowel wall with projection towards the cutaneous plane (Fig. 2).

Figure 2 A: fistulous orifice in the right hypochondriac region (short arrow). B: fistulography showing atypical location of the biliary stent (arrow head). C: local removal of biliary stent (long arrow). D: stent removed (white arrow). 

Biliary stent migration was suspected in both patients. On physical examination, the structure was palpated in the subcutaneous layer; local exploration guided by ultrasound was performed through the fistulous orifices and the stents were successfully removed (Fig. 1 and 2). During follow-up, spontaneous closure of the fistula occurred in both patients. In case 1, common bile duct stones were successfully removed by ERCP; this approach was decided due to comorbidities (elder patient with diabetes, hypertension and COPD) and ASA grade 3. Five months later, the patient in case 2 reported episodes of mild epigastric pain with spontaneous relief. The magnetic resonance cholangiopancreatography (MRCP) demonstrated common bile duct lithiasis at the level of the common hepatic duct and the right hepatic duct. The stones were successfully removed through ERCP. During long-term follow-up, MRCP evidenced bile duct patency. In this case, we asked ourselves whether this common bile duct lithiasis was primary or a corresponded to a stentolith. Stents retained in the biliary tract can lead to de novo stone formation (stentolith), with an incidence of 18% in cases of forgotten stents for more than 2 years. Stentoliths are radiolucent stones formed in the proximal or distal ends of the stent⁵. While reviewing these cases, we wondered whether local removal was the solution or if we simply preempted the natural history of stent migration which did not present any associated complications. In addition, other questions arose. Which factors determine the proximal or distal migration of stents? Which factors determine the development of complications due to this migration?

There are still questions to elucidate on the pathophysiology of stent migration through the bile duct and its associated complications. In 1992, Johanson et al.⁶ described the statistically significant factors for stent migration, as malignant or iatrogenic strictures, distal or proximal diseases, larger diameter stents, shorter stents, absence of inflammation and edema after biliary tract diseases, large papillotomies, and bile stasis generated by migration predisposing to the formation of choledocholithiasis and cholangitis.

In both cases presented here, biliary stents were found in atypical locations, and we tried to analyze the possible trajectories followed by the stents until they reached their final location. In case 1, we may think of distal migration, with the formation of a fistulous tract through the duodenum, and probable retroperitoneal perforation of the duodenum due to the absence of symptoms. In case 2, although proximal migration is more common and the stent generally remains retained in the liver with the different complications of bile duct obstruction, the postoperative self-limited biliary leak could have generated a pathway through which the stent could have possibly migrated. Also, distal migration into the large bowel, with stent impact and exteriorization through the large bowel would be another possibility even though it is even less common as perforation usually occurs in fixed sectors. The patient’s unawareness of the procedure and the lack of follow-up by the intervening medical staff predisposed to this complication. It is important to emphasize the implementation or reinforcement of protocols to document patients intervened and their follow-up in surgery and endoscopy departments where the use of biliary stents is common to avoid these late complications.