Expandable Metal Stents for Benign Biliary Disease

Benign biliary diseases include benign biliary strictures (BBS), choledocholithiasis, and leaks. BBS encompass postoperative injury, anastomotic stricture, chronic pancreatitis, primary sclerosing cholangitis, and gallstone-related stricture. Therapeutic options for benign biliary diseases include surgical, percutaneous, and endoscopic interventions. Endoscopic options include placement of plastic stents as well as self-expanding metal stents (SEMS). SEMS can be uncovered, partially covered, and fully covered, and have been used with some success in resolution of strictures and leaks; however, complications limit their use. This article reviews the currently published experience on SEMS and attempts to define their current role in the treatment of benign biliary diseases.

Benign biliary diseases include biliary strictures, leaks, and biliary stones. Benign biliary strictures (BBS) include postoperative injury, anastomotic strictures, chronic pancreatitis, primary sclerosing cholangitis, gallstone-related stricture, and other less common inflammatory conditions. Postoperative strictures are most frequent, and include postcholecystectomy and liver transplantation–associated anastomotic stricture. Chronic pancreatitis is a common cause of BBS, which can be difficult to manage endoscopically. The clinical presentation of benign biliary diseases can be broad, ranging from subclinical (mild abnormalities in liver function tests) to complete biliary obstruction manifesting with jaundice, cholangitis, and ultimately biliary cirrhosis. The location of biliary strictures can be categorized using a classification proposed by Bismuth. Bismuth I strictures are located more than 2 cm distal to the confluence of the left and right hepatic ducts. Type II lesions are located less than 2 cm from the hepatic bifurcation. Bismuth III strictures are present at the bifurcation, whereas Type IV involves the right or left hepatic ducts, and Type V extends into these ducts more proximally.

Before considering treatment, diagnostic evaluation is of utmost importance because malignancy needs to be excluded. Cross-sectional imaging with high-resolution computed tomography and magnetic resonance cholangiopancreatography allow for better visualization of the biliary tree than most other modalities. In addition, these imaging modalities provide valuable information of surrounding structures, hence the stricture itself can be better classified before therapeutic endeavors are undertaken. Endoscopic retrograde cholangiopancreatography (ERCP) is often performed to obtain tissues sample, though is not always very accurate. More recently, adjunctive diagnostic tools such as endoscopic ultrasonography combined with fine-needle aspiration, intraductal ultrasonography, and even cholangioscopy have allowed further characterization and visualization with the goal of improving diagnostic capability.

Biliary leaks can occur after biliary tract surgery and most commonly after cholecystectomy. Postcholecystectomy leaks are usually from the cystic duct or duct of Luschka. Biliary leaks can become more complex when surgery is performed in the setting of severe cholecystitis or significant fibrosis, making cholecystectomy more difficult and prone to complications. Endoscopic placement of biliary stents has become the first step in the management of postoperative biliary leaks. The number of and type of stents used, however, remains controversial.

Choledocholithiasis can be cleared during ERCP, with success rates of up to 95%. However, stone clearance can be incomplete, and temporizing measures such as stent placement can offer continued biliary drainage until repeat procedures and advanced extraction techniques such as mechanical and intraductal lithotripsy are used. In those patients who are at high risk for surgery or too ill, endoprosthesis placement can reduce the risk of biliary obstruction with resultant jaundice and cholangitis.

Treatment options

Therapeutic options for the treatment of symptomatic, benign biliary strictures include surgical, percutaneous, and endoscopic interventions. Surgery has historically been the gold standard. Location of the stricture determines the optimal surgical approach, such as choledochojejunostomy for distal common bile duct strictures and hepaticojejunostomy for more proximal strictures. Although successful with low reintervention rates, surgical morbidity is not insignificant, and can be upwards of 25%. Less invasive procedures such as percutaneous drainage and endoscopic decompression have gained popularity, as they do not preclude surgery and are optimal for poor surgical candidates. A percutaneous transhepatic approach with balloon dilation and placement of catheters has a success rate of upwards of 60%, with most recurrent strictures occurring within 2 to 3 years. Misra and colleagues showed a success rate of 58.8% with a follow-up of 76 months.

The endoscopic approach to benign biliary strictures has evolved over time to become an alternative to surgery. Endoscopic therapy typically includes balloon dilation and endoprosthesis placement. Several studies have addressed the effectiveness of plastic stents for treatment of benign biliary strictures. The use of plastic stents typically requires placement of multiple stents with frequent endoscopic sessions, given the short duration in patency. The current trend for the endoscopic treatment of BBS includes placement of multiple plastic stents with exchange every 3 months to avoid obstruction.

Therapeutic options for biliary leaks focus on relieving transpapillary high-pressure gradient, causing bile to flow distally into the duodenum and away from the leak site. This goal can be achieved endoscopically using stent placement with or without biliary sphincterotomy, which is quite successful for simple leaks but less so with complex leaks. Incomplete extraction of bile duct stones can be managed with biliary stent placement using conventional plastic (straight or double pigtail), which allows for adequate biliary drainage. The objective is not only to provide a bridge to further endoscopic attempts or elective surgery, but also to potentially reduce the stone burden by facilitating fragmentation.

Self-expandable metal stents (SEMS) were initially developed for palliation of malignant biliary strictures because their larger diameter results in an increase in duration of stent patency. More recently they have been used to treat a variety of benign biliary disorders.

Uncovered SEMS

Uncovered SEMS have been proved to allow sustained biliary drainage of malignant biliary strictures. The limitations of plastic stents in patients with benign biliary strictures includes suboptimal stricture resolution as well as the need for frequent procedures to exchange stents to prevent recurrent cholangitis. Uncovered SEMS were the first available SEMS and thus the first to be used for benign disease. In one of the earliest studies, Foerster and colleagues endoscopically placed uncovered SEMS (Wallstent; Boston Scientific, Natick, MA, USA) in 7 patients with benign bile duct stenosis (n = 6) or bilioduodenal fistula (n = 1) who failed dilation and plastic stent placement. There were no stent occlusions after a short mean follow-up of 8 months. An important study gave insights to the tissue response of metal stents in benign disease. Choledochoscopic examination at 1 year showed complete epithelialization of the metal stent in all subjects. Median patency was 35 months (range 7–72 months), with 9 symptomatic stent occlusions occurring in 5 of 8 patients (62.5%).

There have been several studies on the use of uncovered SEMS in patients with BBS related to chronic pancreatitis. Deviere and colleagues placed uncovered SEMS in patients (n = 20) with chronic pancreatitis, and initially demonstrated relief of cholestasis for up to 33 months for 18 patients. However, repeat ERCP demonstrated that the stent was embedded within the bile duct and covered with tissue within 3 months. Similar studies confirmed these findings and the development of recurrent obstruction. Table 1 summarizes these studies.

Table 1

Studies reporting endoscopic placement of uncovered metal stents

Study	No. of Patients	Etiology	Type of Stent	Median Stent Patency (Range)/Clinical Success	Median Follow-Up (Range)	Complications
Foerster et al	7	Postoperative	Wallstent	8 mo/100%	8 mo	Sepsis/liver abscess (related to known bilioduodenal fistula) (1)
O’Brien et al	8	Postoperative (5), (2), idiopathic (1)	Wallstent	35 mo (7–72 mo)/unknown	64.5 mo (26–81 mo)	9 symptomatic stent occlusions in 5 patients
Deviere et al	20	CP	Wallstent	33 mo/90%	33 mo	Symptomatic occlusion with tissue hyperplasia (2)
van Berkel et al	13	CP	Wallstent	50 mo/69%	50 mo	Stent occlusion (3); migration (1)
Eickoff et al	6	CP	Wallstent	20 mo/33%	58 mo (22–29 mo)	Stent occlusion (4)
Dumonceau et al	6	Postoperative	Wallstent	17 mo/0%	17 mo	All stents occluded
Tringali et al	18	CP	Wallstent	39 mo (5 mo–1.8 y)/44%	52 mo (26–121)	Stent occlusion (8)

Abbreviation: CP, chronic pancreatitis.

Uncovered SEMS

Uncovered SEMS have been proved to allow sustained biliary drainage of malignant biliary strictures. The limitations of plastic stents in patients with benign biliary strictures includes suboptimal stricture resolution as well as the need for frequent procedures to exchange stents to prevent recurrent cholangitis. Uncovered SEMS were the first available SEMS and thus the first to be used for benign disease. In one of the earliest studies, Foerster and colleagues endoscopically placed uncovered SEMS (Wallstent; Boston Scientific, Natick, MA, USA) in 7 patients with benign bile duct stenosis (n = 6) or bilioduodenal fistula (n = 1) who failed dilation and plastic stent placement. There were no stent occlusions after a short mean follow-up of 8 months. An important study gave insights to the tissue response of metal stents in benign disease. Choledochoscopic examination at 1 year showed complete epithelialization of the metal stent in all subjects. Median patency was 35 months (range 7–72 months), with 9 symptomatic stent occlusions occurring in 5 of 8 patients (62.5%).

There have been several studies on the use of uncovered SEMS in patients with BBS related to chronic pancreatitis. Deviere and colleagues placed uncovered SEMS in patients (n = 20) with chronic pancreatitis, and initially demonstrated relief of cholestasis for up to 33 months for 18 patients. However, repeat ERCP demonstrated that the stent was embedded within the bile duct and covered with tissue within 3 months. Similar studies confirmed these findings and the development of recurrent obstruction. Table 1 summarizes these studies.

Table 1

Studies reporting endoscopic placement of uncovered metal stents

Study	No. of Patients	Etiology	Type of Stent	Median Stent Patency (Range)/Clinical Success	Median Follow-Up (Range)	Complications
Foerster et al	7	Postoperative	Wallstent	8 mo/100%	8 mo	Sepsis/liver abscess (related to known bilioduodenal fistula) (1)
O’Brien et al	8	Postoperative (5), (2), idiopathic (1)	Wallstent	35 mo (7–72 mo)/unknown	64.5 mo (26–81 mo)	9 symptomatic stent occlusions in 5 patients
Deviere et al	20	CP	Wallstent	33 mo/90%	33 mo	Symptomatic occlusion with tissue hyperplasia (2)
van Berkel et al	13	CP	Wallstent	50 mo/69%	50 mo	Stent occlusion (3); migration (1)
Eickoff et al	6	CP	Wallstent	20 mo/33%	58 mo (22–29 mo)	Stent occlusion (4)
Dumonceau et al	6	Postoperative	Wallstent	17 mo/0%	17 mo	All stents occluded
Tringali et al	18	CP	Wallstent	39 mo (5 mo–1.8 y)/44%	52 mo (26–121)	Stent occlusion (8)

Abbreviation: CP, chronic pancreatitis.

Partially covered SEMS

In an effort to prolong the duration of patency in patients with malignant biliary obstruction, partially covered self-expanding metal stents (PCSEMS) were introduced. PCSEMS were found to be removable when placed across the papilla, offering the option of temporary placement. Cantu and colleagues placed PCSEMS (Wallstent; Boston Scientific) in 14 patients with common duct strictures due to chronic pancreatitis who failed prior plastic stent therapy. Strictures initially responded in all patients, but at a median follow-up of 22 months (range 12–33 months) 7 patients developed stent dysfunction, requiring reintervention. Of note, stent patency decreased over time, to 37.5% at 36 months, demonstrating that PCSEMS cannot remain in place long term.

The authors’ group published a heterogeneous series of patients (N = 79) with BBS receiving PCSEMS (Wallstent) ( Fig. 1 ). In 65 patients the stent remained in place for a median of 4 months (range 1–28 months) and were subsequently removed. Median follow-up after stent removal was 12 months (range 3–26 months). Three patients developed a stricture at the re-uncovered proximal portion, 3 failed primary therapy, and 2 developed duodenal obstruction preventing SEMS reinsertion, resulting in a success rate of 90% (59/65). Successful stricture resolution was noted to be lowest with strictures related to chronic pancreatitis (17/22, 77%). Stent migration occurred in 11 patients (14%), and 6 (8%) developed mucosal hyperplasia at the uncovered portion. In a follow-up study, Sauer and colleagues further analyzed long-term response in these patients after a mean follow-up of 910 ± 687 days. Sustained stricture resolution was seen in 59%. Of note, 15 stent migrations occurred and 7 patients developed tissue hyperplasia.

Most recently, Chaput and colleagues evaluated placement and removal of PCSEMS (Wallstent) in patients with anastomotic biliary strictures after liver transplantation. In this prospective, multicenter study of 22 patients, SEMS remained in place for 2 months, with subsequent removal in 20 (2 had distal migration). Temporary placement was not problematic, but the investigators did find extraction challenging. Complications with stent removal were minor and included self-limited hemorrhage and fever. Of interest, only those with migration or dislocation had persistent strictures after 3-month follow-up. Sustained stricture resolution was observed in only 45.6% of patients analyzed on an intent-to-treat basis. Table 2 summarizes the studies using PCSEMS.

Table 2

Studies reporting endoscopic placement of partially covered metal stents

Study	No. of Patients	Etiology	Type of Stent	Median Stent Patency (Range)/Clinical Success	Median Follow-Up (Range)	Complications
Cantu et al	14	CP	Wallstent	30 mo/37.5%	21 mo (12–33 mo)	Distal migration (2) Hyperplasia (5) Cholestasis (7) Cholangitis (5) Cholecystitis (2)
Tringali et al	6	CP	Wallstent	20 mo (16–24 mo)/33%	35 mo (33–37 mo)	Stent occlusion (4)
Kahaleh et al	79	CP, OLT, BDS, inflammatory, surgical	Wallstent	4 mo (1–28 mo)/90%	12 mo (3–26 mo) after removal	Migration (11) Stricture (6) Pain (2)
Sauer et al	66	CP, OLT, BDS, inflammatory, surgical	Wallstent	198 d, 265 d/59%	909.8 d, 687.4 d	Migration (15) Hyperplasia (7)
Chaput et al	22	Anastomotic strictures (OLT)	Wallstent	3 mo, 2.1 mo/45.6%	12 mo after removal	Minor pancreatitis (3) Pain (1) Cholangitis (1) Distal migration (2) Partial stent dislocation (5)

Abbreviations: BDS, bile duct stones; CP, chronic pancreatitis; OLT, orthotic liver transplantation.

Fully covered SEMS

Because of limitations related to partially covered SEMS, particularly epithelial hyperplasia at the uncovered portions, fully covered metal stents (FCSEMS) were introduced. Cahen and colleagues published a series of 6 patients with strictures resulting from chronic pancreatitis treated with a 10-mm FCSEMS (Hanaro; M.I. Tech Co, Ltd, Seoul, South Korea). A 66% resolution rate was reported; however, 2 stents could not be removed and required plastic stents to be placed through the metal stent. More recently, the authors’ group analyzed a FCSEMS with anchoring fins (Viabil; Conmed, Utica, NY, USA) to treat BBS of varying etiology. A total of 44 patients were included. Median time of FCSEMS placement was 3.3 months (interquartile range [IQR] 3.0–4.8 months), with resolution noted in 34 of 41 (83%) ( P = .01) after median follow-up of 3.8 months (IQR 1.2–7.7 months) following removal. Complications were observed in 6 (14%) patients after placement, and 4 (9%) patients after removal, mainly pain and post-ERCP pancreatitis. A lower rate of resolution was seen with chronic pancreatitis (58%), and there was moderate difficulty in removing the stent, due to its anchoring fins. Ulceration and bleeding were seen after stent extraction ( Fig. 2 ).