Introduction

Biliary strictures are one of the most common pathologic processes encountered by therapeutic endoscopists. Understanding the etiology and endoscopic management of biliary strictures is critical. Biliary strictures vary widely in clinical presentation; their location in the biliary tree; and span benign, malignant, and indeterminate etiologies (Table 1). Although biliary strictures can be managed endoscopically, percutaneously, or surgically, the scope of this article will focus on endoscopic management via endoscopic retrograde cholangiopancreatography (ERCP). When an obstruction is identified, initiating a prompt evaluation to determine the cause, relieve the obstruction, and manage the underlying pathology are key steps. Here, we aim to review the ERCP technique and treatment paradigms used to manage a variety of biliary strictures based on location and etiology. 

Clinical History, Symptoms, Laboratory Features, and Non-Invasive Imaging of a Biliary Stricture

While biliary strictures can result in a constellation of signs and symptoms, a thorough patient history often provides significant insight into the etiology of the stricture. For instance, a history of painless jaundice and weight loss in an elderly patient, suggests a malignant etiology. Alternatively, a history of inflammatory bowel disease may suggest primary sclerosing cholangitis (PSC). Other critical information includes a thorough family history, tobacco usage, and personal history of new-onset diabetes or pancreatitis, especially when malignancy is suspected. Additionally, understanding the chronicity of symptoms can, in some cases, help elucidate the cause of a stricture. A chronic presentation >3 months may indicate a benign/fibrotic stricture that has developed over time. If symptoms are rapidly progressive and without a predisposing event, a malignancy should be suspected.¹ An acute presentation with a predisposing factor such as recent biliary surgery or liver transplant suggests a benign etiology such as an anastomotic stricture or post-operative ischemic stricture. Some benign biliary strictures, such as those related to IgG4-related disease or chronic pancreatitis, are inflammatory and may have fluctuating symptoms over time.¹ 

Signs and symptoms that may be related to a biliary obstruction are non-specific and can include jaundice (most commonly), scleral icterus, pruritus, weight loss, anorexia, abdominal discomfort, nausea and vomiting.² In certain cases, with advanced malignancies, patients may present with iron-deficiency anemia due to luminal bleeding. When evaluating liver function tests, biliary obstruction leads to an elevation in alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT), and ultimately conjugated hyperbilirubinemia. Chronic biliary obstruction can lead to vitamin K malabsorption and a prolonged prothrombin time.³ Although non-specific, and typically used for prognostic purposes, other indicators of biliary obstruction include elevated tumor markers carbohydrate antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA). Neither CA19-9 nor CEA are sensitive or specific tests to be used for diagnosis alone and can be elevated in both benign and malignant biliary obstruction.^4,5 However, marked elevations in CA19-9 >1000 IU are typically only seen in cancer or severe cholangitis.⁶ There may be elevations in other liver function tests including aspartate transaminase (AST) and alanine transaminase (ALT). Although non-specific, the R-factor can be used to differentiate hepatocellular from cholestatic liver injury. An R-factor <2 may indicate a cholestatic liver injury related to a biliary stricture.⁷ When determining the etiology of a stricture, in the appropriate clinical setting, the immunoglobulin G subfraction four level (IgG4) may be elevated. IgG4 elevations may be seen in autoimmune pancreatitis and IgG4-related cholangiopathy. 

When initially faced with a jaundiced patient with elevated liver function tests, the recommended first imaging modality is typically a transabdominal ultrasound (US). Although US is highly sensitive for bile duct dilation and cholelithiasis, it is not ideal for identifying the specific etiology of a stricture. After US has demonstrated duct dilation, the next step in noninvasive imaging typically depends on clinical judgement. In some situations, the next imaging modality is computed tomography (CT) of the abdomen. CT, although useful for identifying and staging mass lesions as well as inflammatory processes, is not as sensitive as magnetic resonance cholangiopancreatography (MRCP).⁸ MRCP is the ideal non-invasive imaging modality to define stricture extent and location. MRCP is an excellent diagnostic tool, especially if tissue acquisition or direct invasive therapy is not needed. It provides the cross-sectional detail of CT plus sensitive cholangiographic images as a non-invasive and low-risk modality. Both CT and MRI can play roles in establishing surgical resectability, provide a vascular evaluation, and establish the best modality for tissue sampling and stenting.⁹ Studies have shown that the sensitivity and specificity of MRCP for the diagnosis of malignant strictures may even approach that of ERCP.¹⁰ For instance, MRCP has a sensitivity of 77%-86% and a specificity of 63%-98% for the diagnosis of malignant biliary obstruction caused by cholangiocarcinoma.^10,11 Furthermore, MRCP often serves as a guide for future ERCP with stenting for intrahepatic and hilar strictures as well.⁹ 

Benign Biliary Strictures

Overview and Role of ERCP

ERCP plays a key role in the management of benign biliary strictures (BBS). The most common cause of these strictures is typically post-surgical, usually after cholecystectomy or anastomotic strictures after liver transplant. Other common causes of benign biliary strictures include chronic pancreatitis, IgG4-related cholangiopathy, PSC, in addition to a variety of other rare etiologies. The scope of this section will focus on the endoscopic management of the most common causes of benign strictures due to iatrogenic surgical injury, anastomotic strictures, chronic pancreatitis, and IgG4-related disease. PSC-related strictures will be discussed in a separate section.  

First, a critical role of ERCP is excluding malignancy as the cause of the stricture, especially in the setting of chronic pancreatitis and PSC where such determinations may be very difficult. Although a variety of approaches can contribute to diagnosing a biliary stricture including non-invasive (MRCP, CT)¹² and invasive (endoscopic ultrasound/fine-needle aspiration) options, ERCP often provides additional diagnostic reassurance via cytology brushings or direct cholangioscopic biopsies. When used as a diagnostic tool, ERCP brushings have been shown to have poor sensitivity of 35% to 69% and a specificity of 90%.¹³ The yield can be increased with direct cholangioscopy or biopsy forceps to target the stricture fluoroscopically. The cumulative diagnostic yield for strictures is increased to 63% when direct biopsies are obtained in addition to brushings.¹⁴ The advent of single-operator digital cholangioscopy (SOC; Spyglass, Boston Scientific^©, Natick MA) has improved yields. In a recent multicenter study, the sensitivity and specificity of SOC visual impression for diagnosis of malignancy were 90% and 95.8%, respectively. The sensitivity and specificity of SOC-guided biopsies for diagnosis of malignancy were 85% and 100%, respectively.¹⁵

Therapeutic management of a benign stricture via stenting involves (1) traversing the stricture with a guidewire, (2) dilating the stricture in certain situations (Image 1), and finally (3) placing one or more biliary stents across the stricture (Image 2). After biliary cannulation, a cholangiogram is necessary to determine the length and location of the stricture. A complete sphincterotomy can be helpful as it allows for easy endoscopic access to the biliary tree, needed for instrument exchanges during the index ERCP and future endoscopic stent exchanges in the management of a BBS.¹⁶ It should be noted that not all patients require sphincterotomy for stricture management.

After biliary access is obtained, a guidewire must be passed across the stricture. Benign strictures are generally short, asymmetric, and fibrotic, sometimes making them difficult to traverse. Utilizing hydrophilic wires of various sizes, wires with angled-tips, and manipulating the sphincterotome may all be necessary to deliver the wire across the stricture.¹⁶ Guidewire passage requires the ability to interpret fluoroscopy, persistence, and technical skill as missteps can lead to false passages and an inability to traverse the stricture. Once the wire is across the stricture, dilation may be necessary to open the bile duct for durable, adequate drainage. Dilation can be accomplished, most commonly, with hydrostatic balloons or, less commonly, mechanically with passage dilation catheters. In rare situations where balloons or catheters cannot dilate a biliary stricture, screw-type stent extractors (Sohendra stent extractor; Cook Endoscopy^©) can be used to facilitate eventual dilation. 

Studies have shown that although balloon dilation is immediately effective, either in single or multiple sessions, this modality alone is insufficient for durable duct patency and associated with a high rate of re-stenosis up to 47%.^17-19 Therefore, stent placement is imperative to allow for stricture patency for a prolonged period to allow scar remodeling.²⁰ A single plastic stent has not been shown to provide durable patency over the long term, although many go this route at the initial ERCP for expediency and this is within the standard of care.²¹ For refractory benign biliary strictures, studies have shown an effective method for long-term duct patency is the gradual placement of up-sized temporary plastic stents, with exchanges every 3-4 months, over the period of a year. Although this “multi-stenting strategy” requires multiple procedures, it is highly effective for benign biliary strictures, especially post-operative strictures.^16,22 Recently, fully covered self-expandable metal stents (FCSEMS) have emerged as an excellent alternative to the “multi-stenting strategy,” potentially accomplishing dilation with a fewer number of ERCP for stent exchanges. A critical point is that uncovered stents should not be placed for benign strictures as tissue in-growth makes these stents irretrievable. Secondly, placing fully covered metal stents across the hilum can potentially obstruct or jail one side of the liver and should be avoided.²² 

Biliary Strictures after Liver Transplantation (LT)

Post-LT strictures can be anastomotic, occurring at the duct-to-duct connection between the donor duct and recipient duct, or non-anastomotic with an overall incidence of 5-32%.^23-25 Anastomotic strictures comprise 80% of all post-LT biliary strictures.²⁶ ERCP is first-line for the management of anastomotic biliary strictures and can also play a role in non-anastomotic strictures. 

Prior to endoscopic intervention and when a biliary complication is suspected, cross-sectional imaging and hepatologic workup should be performed to rule out vascular complications (i.e., hepatic artery stenosis) or organ rejection. In patients with an anastomotic stricture (Image 3), standard ERCP treatment includes the multi-stenting strategy as described above with stricture dilation and multiple stent exchanges every 3-4 months. Studies have shown success rates of 70-80% in cases of orthotopic liver transplant and 60% in living-donor liver transplant with this strategy.^27-31 Despite successful response to ERCP, the rate of cholestasis recurrence has been shown to be approximately 18%.³² The alternative approach is the placement of a single FCSEMS across the stricture, which has shown to be effective and shorten the duration of endoscopic treatment in patients with BBS.^33-35 The stent is removed 3-6 months after placement. Although safe and feasible, disadvantages include the risk of stent migration and a stricture recurrence of 9-47% over 5 years of follow-up.^36-38 

Non-anastomotic strictures typically occur as a consequence of ischemia and, less commonly, a recurrence of the underlying hepatic dysfunction, such as PSC.³⁹ Strictures in this scenario can be unifocal or multiple strictures can develop anywhere along the biliary tree. The role of ERCP is to preserve biliary patency until definitive treatment can be performed, such as re-transplantation. Several studies have shown the multi-stenting strategy can be useful with plastic stents, but have low success rates ranging from 50-75%.⁴⁰

Post-Cholecystectomy Benign Biliary Strictures

There are numerous anatomic, vascular, inflammatory, and surgical causes that lead to benign biliary strictures after cholecystectomy, the incidence of which as increased with the transition to a laparoscopic approach.⁴¹ ERCP has replaced surgery due to its safety and high success rate in the management of these strictures as compared to surgery.⁴² Similar to the management of other benign strictures, the multi-stenting strategy with replacement of plastic stents every 3-4 months with possible dilation and up-sizing of stents has been the gold-standard, with increasing use of FCSEMs as well given their ease of use and excellent outcomes. Stent exchanges and dilations should continue until there is no fluoroscopic evidence of stricture. Long term success ranges from 80-100% with long-term durability.²² Although data with FCSEMS in this population is less robust, several studies^33,35 have shown the efficacy of metal stents for BBS due to various etiologies. 

Benign Biliary Strictures in Chronic Pancreatitis

In patients with advanced chronic pancreatitis (CP), up to 33% develop a biliary stricture but surprisingly overt jaundice is only seen in a minority of patients.^43,44 There is no clear relationship between the degree of biliary obstruction with the severity or duration of chronic pancreatitis.⁴⁴ Biliary compression caused by edema from ongoing inflammation or mass-effect by a pseudocyst typically improves with treatment of the underlying CP, but obstruction from a fibrotic stricture requires endoscopic intervention.

A multi-stenting strategy in chronic pancreatitis has a success rate of 44-92%.^45-47 This success rate is lower than post-surgical biliary strictures due to the ongoing inflammation, calcification, and fibrosis associated with CP, particularly in the head of the pancreas.^45-47 However, as opposed to post-cholecystectomy strictures, the data for FCSEMS is quite robust and these strictures may be particularly well-suited to FCEMS because CP-related strictures typically develop in the intrapancreatic portion of the common bile duct, which is easy to span with a short FCSEMS. Success rates of FCSEMS range from 43-77%.^48,49 Anti-migration flaps were developed to these FCSEMS to overcome this disadvantage and these stents are now available commercially.⁵⁰

IgG4-Related Autoimmune Cholangiopathy

Autoimmune, or IgG4-related cholangiopathy, can result in a BBS anywhere along the biliary tree mimicking hilar cholangiocarcinoma, pancreatic ductal adenocarcinoma, or even PSC. This entity is associated with autoimmune pancreatitis. Although it can be associated with an elevation in IgG4, serum levels can be normal as well,⁵¹making the diagnosis quite challenging. Oftentimes, the diagnosis is made in retrospect after a high-clinic suspicion and empiric steroid administration improves the biliary obstruction and imaging findings. When endoscopic intervention is needed, tissue acquisition by the methods described earlier followed by histologic staining for IgG4 plasma cells clinches the diagnosis.⁵¹ Temporary biliary stenting can also be useful to relieve the jaundice prior to initiation of steroid therapy.⁵²

Malignant Biliary Strictures

Overview and Role of ERCP

Malignant strictures of the biliary tree are another common pathologic process encountered by therapeutic endoscopists. The most common presenting symptom is jaundice, and the most common underlying etiology is pancreatic adenocarcinoma. However, other etiologies include ampullary carcinoma, cholangiocarcinoma, gallbladder carcinoma, and metastatic disease (gastric, colon, malignant lymphadenopathy, melanoma, among others). The management of distal and hilar/intrahepatic biliary strictures vary in several key areas. The scope of this section is to discuss the various endoscopic techniques via ERCP and outcomes to provide diagnostic and therapeutic solutions to biliary strictures. As reviewed in the section on benign biliary strictures, the role of ERCP in malignant strictures also includes tissue acquisition and palliative stenting.

Distal Malignant Biliary Obstruction:
Tissue Acquisition and Endoscopic Therapy

The most common cause of a distal or mid-bile duct malignant biliary obstruction (Image 4) is pancreatic adenocarcinoma, accounting for more than 90% of cases.⁵³ Other causes include, ampullary carcinoma, cholangiocarcinoma, gallbladder carcinoma, and metastatic disease. 

When faced with a malignant distal biliary obstruction, the first step is to establish a diagnosis and stage the malignancy. As described previously, CT and MRI can provide valuable data about vascular involvement, location and extent of the primary tumor, and distant metastasis (Image 5). While there are multiple modalities for tissue acquisition including percutaneous biopsy, and ERCP with brush cytology or cholangioscopy, EUS-guided biopsy has taken on a dominant role given its very high diagnostic success rate and low risk. 

Beyond the potential role of cytology brushings, the primary role of therapeutic ERCP is palliative via biliary stent placement for obstructive jaundice. This can oftentimes be done simultaneously at the time of EUS FNA/B. ERCP is definitively indicated for non-surgical candidates with advanced malignancy causing biliary obstruction and has been shown to improve quality of life in this patient population.⁵⁴ Patients with malignancy presenting with cholangitis should also undergo endoscopic stenting. Stenting is indicated in jaundiced patients prior to the initiation of potentially hepatotoxic chemotherapeutic agents or neoadjuvant chemoradiation.⁵⁵ However, in potential surgical candidates with biliary obstruction, biliary stenting is controversial, particularly if surgery is to occur within a short term (weeks). There are some studies showing a benefit of pre-surgical drainage, whereas others demonstrate a higher rate of significant complications up to 4 months post-surgery (74% in stenting group [notably, only plastic stents used] vs. 39% in early surgery group).^56,57 Currently, if surgery is planned within 1-2 weeks in an asymptomatic but jaundiced patient, biliary drainage may not be indicated. However, if the patient develops symptomatic obstruction or surgical intervention is planned for >3 weeks, a stent should be placed.⁵⁸ In modern clinical practice, most patients with pancreatic cancer undergo neoadjuvant therapy and pre-operative stent placement is the norm.

Initially, the placement of a temporary, removable plastic stent is an effective and inexpensive first step. The stent should traverse about 1cm above the stricture proximally and about 1cm into the duodenum. As smaller diameter stents can get occluded or migrate, larger caliber stents (typically 10 French) are used for biliary patency and can be exchanged every 3-4 months.⁵⁹ The primary disadvantages of plastic stents include stent occlusion (median 3-6 months) by the development of a bacterial biofilm⁶⁰ requiring frequent exchanges and stent migration (in 10% of patients).⁶¹ Finally, once a plastic stent has been placed, we favor a scheduled stent exchange as it has shown to decrease the rate of adverse events as compared to as-needed stent exchange.⁶²

SEMS are an excellent long-term option for the stenting of distal malignant strictures. As they are delivered in a sheath and expand after deployment, they can be delivered through the working channel of the duodenoscope and provide a diameter 3-4 times larger than plastic stents. This larger diameter stent results in a much more prolonged duration of stent patency compared to plastic stents, with a median of 9-12 months.⁶³ Currently, SEMS are available as uncovered, partially covered, and fully covered designs. Although, there is no definitive data favoring one type of SEMS over another,^64,65 they are superior to plastic stents for malignant biliary strictures,⁶⁶ especially when longer-term pre-operative biliary drainage is required.⁶⁷ When placed, SEMS seem more likely to cause post-ERCP pancreatitis than plastic stents, but no difference was seen between FCSEMS and uncovered SEMS.⁶⁸ Adverse events related to SEMS include occlusion and migration. More specifically to FCSEMS, occlusion of the cystic duct and pancreatic duct can lead to cholecystitis and pancreatitis, respectively. Although FCSEMS tend to migrate more frequently than uncovered SEMS, two potential advantages of FCSEMS is that they are less likely to become occluded by tissue in-growth and they are easy to remove/exchange, if necessary.⁶⁹ Uncovered SEMS become occluded via tumor in-growth and mucosal hyperplasia. This makes uncovered metal stents very difficult to remove and can lead to duct injury if the stent is pulled with mucosal/tumor in-growth. 

When deciding which type of stent to place, there are pros and cons between covered SEMS, uncovered SEMS, and plastic stents. Although there are a multitude of individual variables including endoscopist preference, stent efficacy/durability, cost considerations, potential need for re-intervention, stricture location, and patient factors such as life expectancy that go into this decision, we aim to provide broad guidance. Foremost, SEMS provide long term patency as compared to plastic stents.^70,71 One meta-analysis showed that although SEMS showed longer patency, there was no difference between metal and plastic stents when it came to therapeutic success, technical success, adverse events, and 30-day mortality.⁷² In the current financial climate, it must be noted that despite their longer patency, SEMS are significantly more expensive than plastic stents, however recent randomized controlled trials have demonstrated the cost-effectiveness of SEMS regardless of patient survival.^72,73 When comparing covered versus uncovered SEMS for a distal biliary stricture, FCSEMS demonstrated longer patency but was found to have higher rates of stent migration and sludge formation.^72,74 In general, most patients with malignant strictures will ultimately receive a SEMS of some type.

Although hilar and intrahepatic cholangiocarcinoma (CCA) will be discussed in the next section, for distal CCA survival is quite poor and surgical resection in patients without metastatic spread is the only curative option. If the patient is a poor surgical candidate, ERCP can be utilized for palliative relief of biliary obstruction with plastic or metal stents.⁷⁵

Hilar and Intrahepatic Malignant Biliary Obstruction:
Tissue Acquisition and Endoscopic Therapy

Cholangiocarcinoma is the most common cause of malignant biliary obstruction of the hilum and intrahepatic ducts. Cholangiocarcinoma can be divided into proximal (intrahepatic; incidence: 5-10% of CCAs), hilar (Klatskin tumors; incidence: 60-70% of CCAs), and distal (extrahepatic; incidence: 20-30%) subsets.⁷⁶ Hilar CCA is classified according to the Bismuth-Corlette classification (Table 2) as described later in this section. As discussed in the next section, the most important risk factor for CCA is PSC.

Other etiologies of hilar and intrahepatic malignant strictures include gallbladder carcinoma, primary liver tumors, portal lymphadenopathy, and metastatic disease. ERCP plays a key role in relieving malignant biliary stenosis of the hilum and survival has been shown to correlate with the percentage of liver segments drained.⁷⁷ When evaluating and endoscopically treating tumors of the intrahepatic ducts or hilum, it is important to understand the segmental liver anatomy, the Bismuth-Corlette classification of hilar tumors, and normal variants. 

In the liver, segments II, III, and IV make up the left hepatic lobe and ERCP typically drains segment II and III of the liver (large left intrahepatic duct). The right hepatic lobe, divided into the right anterior and right posterior intrahepatic ducts, is divided into segments V-VIII. The caudate lobe (segment I) is not typically drained by ERCP.⁷⁸ Cancers in the perihilar region are classified by the Bismuth-Corlette classification which can help guide ERCP-guided palliative stent placement. The Bismuth classification has four types; Type 1 (tumors below the confluence of the left and right hepatic ducts), type 2 (tumors reaching the confluence of the right and left hepatic ducts), type 3 (tumors occluding the common hepatic duct and either the first radicals on the right [type 3a] or left [type 3b – Image 6] intrahepatic ducts), and type 4 (tumors that involve the major ducts and radicals of the both right and left intrahepatic ducts).⁷⁹

As discussed previously, once typical symptoms and lab values indicate biliary obstruction, the next step is non-invasive radiographic imaging. MRCP is the ideal imaging modality for most hilar and intrahepatic malignancies. MRCP can recreate a three-dimensional view of the biliary system and note the location/extent of structuring to direct stent placement during ERCP.⁸⁰ The next step is tissue acquisition. 

When used as a diagnostic tool, ERCP-guided cytology brushings have been shown to have sensitivity of 35% to 69% and a specificity of 90%.¹³ These are widely employed given their very low cost to obtain. Given the suboptimal yield of brush cytology, newer tests to assess DNA proliferation have been developed. These include fluorescence in-situ hybridization (FISH), digital image analysis, and molecular profiling (MP) by flow cytometry. The combination of all 3 tests (cytology, FISH, and MP) had the highest sensitivity for malignancy (66%) and all three individually have been shown to improve the specificity of cytology.^81-83 FISH is the test most commonly incorporated advanced analytic test with biliary brushings to enhance sensitivity. FISH uses fluorescent probes to label certain areas of chromosomes to determine cellular ploidy. Detection of more than five cells with polysomy is considered evidence of malignancy. These tests are further discussed in the section on indeterminate strictures. Obtaining direct tissue from the stricture can be accomplished via direct visualization with cholangioscopy or fluoroscopic-guided sampling with biopsy forceps. Although stricture biopsies can increase the diagnostic yield to 63%,¹⁴ directly visualized biopsy samples via cholangioscopy are better than those without cholangioscopy.⁸⁴ Cholangioscopy is the most frequently utilized method for direct biopsy of hilar strictures and has demonstrated improved diagnostic accuracy.⁸⁵ Visual impression accuracy with single-operator cholangioscopy for diagnosing malignancy has been reported to be as high as 95.1% (with 100% sensitivity and 89.5% specificity). Cholangioscope-guided biopsy accuracy was 80.5% (63.6% sensitivity and 100% specificity).⁸⁶ Finally, studies have shown that cholangioscopy for evaluating intraductal spread in potentially resectable perihilar CCA can detect more extensive disease and change surgical management.⁸⁶

Other methods of endoscopic sampling include endoscopic ultrasound-guided fine needle aspiration (EUS-FNA). Although EUS-FNA is commonly used for distal pancreatobiliary masses/strictures, EUS-FNA for hilar/intrahepatic lesions raises the concern for potential peritoneal tumor seeding from the needle tract.⁸⁷ This is a particular concern for patients who are candidates for curative surgical resection or liver transplantation. In general, EUS-FNA sampling of a hilar stricture should not be performed due to this risk of seeding. However, EUS-guided lymph node biopsy can change management if positive for metastasis. Intraductal ultrasonography (IDUS) has fallen out of favor in recent years with improvements in cholangioscopy. These probes can be passed over a wire into the biliary system. IDUS can determine longitudinal tumor extent and vascular involvement.⁸⁸ Finally, confocal laser endomicroscopy (CLE) probe can be advanced through a cholangioscope. CLE allows for in-vivo microscopic, subcellular evaluation of the biliary mucosa through the utilization of a lower-power laser detecting reflected fluorescent light from tissue. One study found significantly higher accuracy with CLE plus ERCP brushings versus ERCP brushings and tissue acquisition (90% vs. 73%).⁸⁹ The learning curve and specialized equipment required for CLE have limited its utility. 

Surgery (resection or liver transplantation) is the only curative option for hilar cholangiocarcinoma. Preoperative biliary drainage in a jaundiced patient with hilar cholangiocarcinoma has remained controversial as it may delay surgery and increase the risk of adverse events. One large study found no difference in mortality and a higher rate of adverse events/infection. However, in practice, pre-operative biliary drainage is often performed at the discretion of the surgeon.⁹⁰

In nonoperative candidates with symptomatic biliary obstruction or to lower the bilirubin for chemotherapy, ERCP plays a role in palliative drainage. To resolve jaundice, only ~50% of the liver parenchyma needs to be drained.⁹¹ When preparing for ERCP for hilar/intrahepatic strictures, several important factors should be noted throughout the steps of the procedure. First, non-invasive imaging can help provide a roadmap for drainage and scans should be reviewed. Next, a sphincterotomy is important during the sentinel procedure to allow room for multiple stents, simplify endoscopic instrument exchanges, and make subsequent cannulations easier. When the stricture is encountered and there is a suspicion for malignancy, tissue sampling should be performed, ideally with more than one modality, but at minimum with cytology brushings. Antibiotics should be given at the time of procedure and possibly post-procedurally, at the discretion of the endoscopist. If possible, stricture dilation due to the methods described earlier in this article should be pursued followed by stent placement. Finally, it is critical to note that only viable segments of the hepatic parenchyma with dilated ducts should be drained with stenting. Stenting should not be attempted in diminutive, atrophic, or occluded ducts and contrast used judiciously in segments that cannot be drained.^75,92 

Stent choice for hilar and intrahepatic duct malignancy has been an extensively studied topic. First, in non-operative candidates, SEMS should be used over plastic stents. In patients where pre-operative stenting is pursued, plastic stents should be pursued.⁹³ Similar to the advantages discussed earlier for SEMS, in hilar malignancies, SEMS have been shown to have longer patency, decreased risk of cholangitis, and higher success rates.^66,94 Finally, as discussed previously, uncovered SEMS should be used when crossing the hilum due to the risk of covered stents obstructing adjacent biliary ducts. The placement of multiple stents can often be accomplished via a multiple guidewire technique wherein a wire is left in each segment to be drained followed by sequential stent placement. Several studies have evaluated bilateral versus unilateral stenting. When unilateral plastic, bilateral plastic, unilateral metal, and bilateral metal stenting was compared for malignant hilar strictures, Hu et al. found that, if technically possible, dual metal stent placement is a preferred palliation strategy, and unilateral metal stent placement is the second option.⁹⁵ Alternatively, a recent systematic review and meta-analysis has demonstrated that unilateral and bilateral SEMS stenting techniques are comparable in terms of efficacy and safety for unresectable hilar malignancies.⁹⁶ Overall, due to this mixed data of unilateral versus bilateral stenting (SEMS favored over plastic stents), endoscopist discretion in regard to the technical feasibility, resource availability, and patient prognosis often dictates which option is pursued. 

In patients with unresectable cholangiocarcinoma, ERCP-guided tumor ablation therapies have been developed to improve the quality of life by tumor debulking and improving stent patency. Although the data is mixed and cost is significant, these two therapies include wire-guided radiofrequency ablation (RFA) and photodynamic therapy. RFA, although commonly used for Barrett’s esophagus, provides local heat therapy to CCA via endobiliary probes. Endobiliary RFA can be used for tumor debulking within the biliary tree⁹⁷ and to open up occluded uncovered SEMS.⁹⁸ Recent data suggests they may even improve survival prior to stent placement.⁹⁷ Although, the heat applied by RFA leads to adverse events and limits its use, newer temperature-controlled RFA devices are being studied.⁹⁹ Photodynamic therapy (PDT) is another ablation therapy which utilizes the administration of a photosensitizing agent activated by light to kill cancer cells. Although it had promising data with longer survival times, improved biliary drainage, and improved quality of life on initial studies¹⁰⁰ or as neoadjuvant therapy for unresectable CCA,¹⁰¹ its use is limited due to adverse effects such as severe light sensitivity, which significantly impacted quality of life. 

Biliary Strictures in Special Populations

Indeterminate Biliary Strictures

A crucial point when evaluating a biliary stricture is to determine whether it is benign or malignant as the treatment paradigms and prognosis differ significantly based on the underlying etiology. As described previously, the immediate next steps include obtaining an accurate history, laboratory testing, non-invasive imaging, tissue sampling, and ultimately relief of the obstruction. Although each of these steps can be accomplished in a variety of ways, ERCP plays a crucial role in tissue acquisition and the relief of obstruction. 

As described earlier, there are certain signs and symptoms indicating biliary obstruction, most common of which is jaundice. Laboratory markers, specifically elevated total bilirubin and alkaline phosphatase further suggest obstruction. Serologic markers such as CA19-9, although non-specific, can contribute to characterization of an indeterminate stricture as benign or malignant. Non-invasive imaging again provides a roadmap for further, more invasive investigation. Ultimately, ERCP is the gold-standard for the characterization of biliary strictures, particularly extrahepatic strictures. Optimal evaluation of a stricture requires expert use of contrast to opacify the length and duct involvement strictures using pressure cholangiograms, a rotatable C-arm, and wire manipulation. Of note, cholangiographic interpretation during ERCP is inadequate to determine malignancy and strictures often interpreted to be benign are actually malignant.¹⁰² Cholangiographic predictors of malignancy include stricture length greater than 14mm, progressive structuring over time, abrupt “shelf-like” strictures, and intrahepatic duct dilation.^102,103 After wire-guided access and stricture dilation, tissue acquisition can be pursued. As mentioned before, a combination of multiple modalities has been shown to improve yield. These methods have been discussed previously including brush cytology (at minimum), intraductal forceps biopsy, and confocal laser endomicroscopy. Flow cytometry, FISH, and digital image analysis (DIA) are additional methods of tissue evaluation. DIA uses a computerized assessment to identify and analyze the DNA content of cells. Although data is mixed¹⁰⁴ and the analysis costly, recent data demonstrates that DIA provides a higher sensitivity identifying malignancy than standard cytology.¹⁰⁵ Studies have demonstrated that FISH provides higher sensitivity (42.9% vs. 20.1%) than routine cytology with equivalent specificity.^106,107 IDUS has also been extensively studied. Data has demonstrated its superior sensitivity and accuracy compared to cytology and biopsy sampling.¹⁰⁸ However, IDUS has significant limitations including an inability to acquire tissue, its steep learning curve/difficult interpretation, and inability to evaluate distant disease hindering its routine use. Finally, as extensively discussed previously, direct sampling techniques with cholangioscopy provides an additional method of sampling. 

Primary Sclerosing Cholangitis (PSC)

Primary sclerosing cholangitis (PSC) is a chronic inflammatory condition of the biliary tree characterized by structuring and dilation of the bile ducts with biliary obstruction. Over time PSC leads to chronic cholestasis and biliary cirrhosis, making liver transplantation the only cure. PSC also carries a risk of developing cholangiocarcinoma in 10-20% of patients.¹⁰⁹ PSC is most commonly diagnosed by MRCP (Image 7) due to its excellent safety profile, high sensitivity, and specificity.¹¹⁰ The role of ERCP in PSC largely consists of tissue sampling and biliary intervention including stricture dilation and stenting. ERCP is largely indicated in PSC when there is evidence of symptomatic cholestasis, cholangitis, and to distinguish a dominant stricture from cholangiocarcinoma.

Certain critical points should be considered when treating a patient with PSC and a dominant stricture endoscopically. Data has shown that repeat endoscopy to maintain biliary patency may improve the survival of patients with PSC.¹¹¹ A dominant stricture has been defined as a stricture >1.5mm diameter in the common bile duct, or >1mm in the left or right main hepatic ducts, but this definition is not universally accepted and others exist.¹¹² About 36-57% of PSC patients that undergo ERCP have dominant strictures and patients can have multiple dominant strictures.¹¹³ Patients with a dominant stricture have poorer survival compared those without and the stricture may harbor CCA.^113,114 Therefore, ERCP plays a key role in treating the stricture and ruling out malignancy. As always, biliary access is key via sphincterotomy and wire-placement. Prophylactic antibiotics are recommended at the time of the procedure and a few days post-procedurally.¹¹² Rectal indomethacin should also be considered in PSC patients without contraindications to avoid post-ERCP pancreatitis.¹¹²

Dominant strictures should always be sampled at each ERCP procedure to rule out malignancy prior to endoscopic intervention. In patients with PSC with dominant strictures, malignancy is more likely if the stricture is greater than 1cm in length, located at the hilum, and have irregular borders.¹¹⁵ As discussed previously, numerous methods have been utilized to diagnose CCA in PSC including tumor markers (CA19-9, CEA), brush cytology, fine-needle aspiration, and forceps biopsy. These all carry a poor sensitivity despite a high specificity. Therefore, FISH is most commonly added to brush cytology specimens and increased specificity for CCA.¹¹⁶ In PSC, FISH has been shown to be superior to DIA and cytology for indeterminate strictures.¹⁰⁶ Studies report that FISH combined with cytology can improve the sensitivity for malignant lesions to 45-59%, while keeping the specificity near 100%.¹¹⁷ One meta-analysis of the utility of FISH in PSC encompassing 8 studies noted sensitivity and specificity of 68% and 70%, respectively.¹¹⁸

Finally, numerous studies have evaluated stricture dilation versus stenting in PSC and the American Association for the Study of Liver Diseases (AASLD) recommends that biliary stricture dilation be the initial treatment for dominant strictures.¹¹⁹ However, the European Association for the Study of the Liver (EASL) leave the decision to the discretion of the endoscopist.¹¹² Balloon dilation should be performed serially with gradual dilation 1-4 weeks apart. Studies have shown improvement in both biliary obstruction and transplant-free survival.^112,120 Stent placement has been reserved for when biliary dilation is inadequate or not durable. Stent placement has been associated with higher rates of adverse events including cholangitis and bile duct perforation in some older studies, but in practice is widely performed.^111,121 One study comparing dilation versus dilation plus stenting demonstrated no benefit to stenting after dilation, an increased need for procedures, and a higher adverse event rate in the stent group.¹²² Finally, the multicenter, randomized DILSTENT trial demonstrated that short-term stents were not superior to balloon dilation and associated with a significantly higher occurrence of complications.¹²³ When stenting is indicated (Image 8), some authors have advocated for short-term stent duration of 1-2 weeks with either a 10-French stent for an extrahepatic duct stricture or two 7-French stents for hilar strictures.^112,124 No benefit has been shown for longer duration stenting in some studies, but in practice many PSC patients do become stent dependent and require long term stenting to good effect.¹¹³

Conclusion

ERCP has been used to manage biliary pathology since the 1970s. However, over the last 20 years, the diagnosis and treatment of benign, malignant, and indeterminate biliary strictures has evolved significantly. There has been a refinement of techniques and novel treatments developed. ERCP provides a minimally invasive, safe, and feasible treatment option for the management of biliary strictures, one of the most common pathologic processes encountered by therapeutic endoscopists. The paradigm shifts in management that have occurred over this time frame parallel advances in endoscopic technology and tools. The improvements made in biliary wires, adjunct devices such as cytology brushes and dilating balloons, cholangioscopy, and stent development have made ERCP a first-line treatment for the diagnosis and treatment of biliary strictures. As endoscopic technology evolves with the impending arrival of biodegradable stents, artificial intelligence, and endoscopic robotics, the management of biliary strictures will continue to progress. Continued efforts in training highly skilled and capable endoscopists are key to ensure that our patients receive the highest-quality care. 

References

Download Tables, Images & References