Endoscopic strategy and covered self-expandable metal stents for malignant hilar biliary obstruction
Article information
Abstract
Malignant hilar biliary obstruction (MHO), most commonly caused by cholangiocarcinoma, is an aggressive condition with a poor prognosis. Because most patients with MHO are unsuitable for primary surgical resection at presentation because of advanced age or comorbidities, palliative biliary drainage is essential to relieve obstructive jaundice and improve the quality of life. Endoscopic drainage has become the preferred palliative approach, with the choice between plastic and metal stents depending on subsequent therapeutic plans, such as systemic chemotherapy or local ablative therapies. Among biliary stents, self-expandable metal stents (SEMSs) are widely used, typically in their uncovered form. However, unlike plastic stents, uncovered SEMSs cannot be removed once deployed, and endoscopic revision is technically challenging. To improve stent patency and facilitate removability, covered SEMSs (CSEMSs) were developed, and are now commonly used in distal malignant biliary obstruction. Nevertheless, in advanced MHO, the primary use of CSEMSs remains controversial. This review summarizes recent endoscopic strategies for advanced MHO, the evolution of CSEMSs, their clinical outcomes, current limitations, and future directions.
INTRODUCTION
Malignant hilar biliary obstruction (MHO), most commonly caused by cholangiocarcinoma, gallbladder cancer, or metastatic malignancies, is an aggressive condition associated with a poor prognosis.1 Surgical resection remains the only curative option. However, most patients are ineligible for surgery at diagnosis because of locally advanced or unresectable disease, or poor performance status resulting from age or comorbidities.2,3 Accordingly, palliative biliary drainage is crucial for relieving obstructive jaundice, improving the quality of life, and enabling systemic chemotherapy or radiotherapy.
Endoscopic biliary decompression is currently recommended as the first-line approach, even for advanced MHO. According to recent guidelines and expert consensus, multi-segmental drainage using either plastic stents (PSs) or self-expandable metal stents (SEMSs) is advised to achieve adequate liver volume drainage.4 PSs are generally preferred when the therapeutic plan remains undetermined or when systemic chemotherapy or local ablative therapy is anticipated.4,5 Although PSs are advantageous because of their ease of placement and removal, their relatively short patency and high occlusion rates remain major limitations. In contrast, SEMSs provide longer patency and reduce the need for reintervention; however, they are typically uncovered, making stent revision in cases of dysfunction more technically challenging than that with PSs.
To address these limitations, fully or partially covered SEMSs (CSEMSs) have been developed as alternatives to PSs or uncovered SEMSs. CSEMSs offer several potential advantages, including prolonged patency and removability, but also present limitations that require careful evaluation in the management of advanced MHO, particularly when compared with their well-established use in distal malignant biliary obstruction. A thorough understanding of both their benefits and drawbacks is therefore essential for selecting the optimal drainage strategy.
PRINCIPLES OF ENDOSCOPIC DRAINAGE
For effective drainage in advanced MHO, several preprocedural considerations must be addressed. First, a thorough understanding of biliary anatomy and accurate determination of the drainage site are essential. MHO is morphologically classified according to the extent of biliary invasion using the Bismuth-Corlette classification,6 which is critical not only for determining the scope of surgical resection but also for selecting the appropriate drainage site. High-grade strictures, such as Bismuth-Corlette types III and IV, typically require bilateral or multiple drainage procedures, unlike types I and II.2,3,7-9 Inadequate drainage after contrast injection increases the risk of complications such as cholangitis and hepatic abscesses, which can negatively affect survival. Therefore, in patients with high-level strictures, careful planning of the drainage site and method before endoscopic intervention is crucial.8 Preprocedural assessment using multidetector computed tomography (CT) or magnetic resonance imaging/cholangiopancreatography (MRI/MRCP) provides detailed anatomic information for diagnosis and staging. These imaging modalities are valuable not only for evaluating resectability but also for identifying the optimal site for biliary drainage.2,3,8,9 Three-dimensional reconstruction of CT or MRI images allows predetermination of potential drainage locations. Importantly, if tumor-related hepatic parenchymal atrophy or intrahepatic bile duct involvement without viable parenchyma is present, unnecessary drainage or contrast injection should be avoided or minimized during the procedure.1-4
Regarding drainage volume, draining more than 50% of the liver volume has been associated with improved survival compared with less extensive drainage (119 vs. 59 days, p=0.005).10 Furthermore, Caillol et al.11 reported that drainage of more than 80% of viable liver volume was associated with significantly longer survival compared with less than 80% drainage (hazard ratio [HR], 2.46; 95% confidence interval [CI], 1.16–5.23; p=0.02). Accordingly, in patients with Bismuth type II or higher MHO, drainage of at least 50% of the liver volume is recommended to achieve effective decompression.4,7
In summary, for high-level MHO, preprocedural imaging is essential to determine the appropriate drainage site, prevent unnecessary stenting, and select the optimal approach, either endoscopic retrograde cholangiopancreatography (ERCP) or percutaneous transhepatic biliary drainage (PTBD), to achieve drainage of more than 50% of the liver volume. In addition, future treatment strategies, including surgical resectability, expected survival, and potential adjunctive therapies such as chemotherapy or local treatment, should be considered when planning biliary drainage.
RECENT GUIDELINES AND CONSENSUS FOR ADEQUATE DRAINAGE OF MHO
Recent guidelines and consensus statements regarding drainage strategies for MHO are summarized in Table 1.4,5,12,13 In unresectable MHO, primary endoscopic palliation is currently recommended to improve quality of life and facilitate systemic chemotherapy. However, in patients with perihilar strictures of suspected malignant origin, the available evidence remains insufficient to recommend either ERCP or PTBD as the preferred initial approach.12,13 For advanced perihilar obstructions, PTBD or a combined ERCP–PTBD approach may be considered. Although endoscopic drainage is generally preferred, the final decision should be individualized based on the patient’s condition, disease extent, and operator expertise. When endoscopic drainage fails, endoscopic ultrasound-guided biliary drainage (EUS-BD) is increasingly favored over PTBD.4
Comparison of presented guidelines and consensus
The choice between PSs and SEMSs remains controversial because each has distinct advantages and limitations. Current guidelines and expert consensus suggest the following strategy: PSs should be used initially when systemic therapy is planned or when the optimal drainage approach has not yet been determined. SEMSs may be appropriate for patients with limited life expectancy (<3 months) or for those prioritizing fewer repeat interventions.4,5,12,13 Thus, current practice increasingly incorporates both expected survival and overall therapeutic strategy into the decision-making process.
With advances in systemic therapies for MHO, including chemotherapy, immunotherapy, and targeted therapy, survival may now exceed the patency period of uncovered SEMSs, which are nonremovable.14 Although SEMSs provide higher technical and clinical success rates, this advantage is offset by a significantly increased risk of stent-related adverse events such as migration, occlusion, ingrowth, and overgrowth, particularly during prolonged survival (odds ratio [OR], 4.85; 95% CI, 3.23–7.27).15 Therefore, the use of removable PSs is recommended as the initial approach in patients undergoing systemic therapy, given that uncovered SEMSs cannot be removed and are more difficult to revise.
PROS AND CONS OF EACH STENT AND DEPLOYMENT METHOD
PSs and SEMSs each have distinct advantages and limitations. PSs are inexpensive, easy to insert, and can be readily exchanged in the event of occlusion. They are particularly useful in MHO, where extrahepatic bile duct dilation is often minimal, as they allow temporary dilation of the native duct lumen with relatively few adverse events.2 However, their smaller diameter limits long-term patency. Typically measuring at least 10 to 12 cm in length, PSs are placed across the duodenal papilla, which may promote biofilm formation through reflux of duodenal contents. This predisposes the stent to occlusion and increases the risk of spontaneous migration.3,8 While insertion is technically straightforward, efficacy does not appear to differ substantially among PS types. However, one retrospective study reported that straight-type PSs were associated with significantly more stent-related adverse events than double-pigtail PSs (OR, 6.74; 95% CI, 3.95–11.45; p<0.001).15
In contrast, SEMSs are deployed using delivery catheters and can be advanced across severe strictures more easily than PSs. Because of their open wire mesh, SEMSs do not obstruct intrahepatic secondary branches, provide longer patency, and reduce the need for reintervention, thereby improving cost-effectiveness.8 The 2013 Asia-Pacific consensus and the 2018 European Society of Gastrointestinal Endoscopy guidelines recommend SEMSs for patients with Bismuth type II–IV hilar cholangiocarcinoma (HCCA) and an expected survival exceeding three months.7,12 Similarly, the 2021 American Society for Gastrointestinal Endoscopy guideline supports SEMS placement in unresectable HCCA with a prognosis of less than three months to minimize repeat procedures.5 Nonetheless, when treatment planning or resectability remains uncertain, initial placement with PSs is generally preferred.5
For bilateral SEMS placement, two main techniques are used: stent-in-stent (SIS) and side-by-side (SBS).2 Although each has unique advantages and drawbacks, recent meta-analyses found no significant differences in complication rates, stent patency, or survival outcomes, although SIS may achieve higher technical success.16 Interestingly, this finding contrasts with earlier assumptions that SBS placement is technically easier than SIS. The discrepancy may reflect limited data and the high level of operator expertise in recent studies.
Regarding stent positioning, devices can be deployed either transpapillary (across the papilla into the duodenum) or intraductally (above the papilla within the bile duct). Transpapillary placement facilitates reintervention in cases of occlusion but increases the risk of duodenal reflux, leading to biofilm formation, sludge, stone development, and cholangitis. Intraductal placement may theoretically reduce these risks; however, clinical evidence remains inconclusive. One study comparing both methods in MHO found no significant differences in technical success or stent patency, although transpapillary placement was associated with a higher incidence of pancreatitis and other adverse events.17 While intraductal placement may be theoretically advantageous, transpapillary deployment could be more practical for long-term survivors, given the higher likelihood of stent occlusion and the need for reintervention.
EFFICACY OF CSEMS
Types of CSEMS
CSEMS can be categorized into fully covered and partially covered types, with multihole fully CSEMSs also reported for hilar drainage.18-24 However, the existing literature primarily consists of case reports and retrospective studies, with few well-designed, large-scale comparative studies assessing the efficacy of different CSEMS types. Consequently, the design that provides the greatest benefit and the superior insertion method remain unclear. Regarding stent diameter, both 6-mm and 8-mm models have been used, and neither has been associated with serious adverse events related to overexpansion. Despite reports suggesting effectiveness, current guidelines do not recommend CSEMSs as the primary drainage modality.4,5,12,13
Clinical outcomes
Unlike distal biliary obstruction, data on the efficacy and safety of CSEMSs in advanced MHO remain limited. The available evidence reflects theoretical concerns that CSEMS deployment in the perihilar region may inadvertently occlude peripheral intrahepatic ducts, predisposing patients to serious infectious complications such as liver abscesses and cholangitis. These risks have contributed to the cautious adoption of CSEMSs in this setting.
A recent prospective randomized study compared CSEMSs and PSs in bilateral SBS deployment for MHO. The study found that CSEMSs had a significantly longer mean time to recurrent biliary obstruction (RBO) than PSs, despite a higher stent migration rate (48%).19 Overall adverse event rates were similar between groups, suggesting that CSEMSs may be preferable to PSs, with complications such as liver abscess not considered clinically significant.
According to a recent meta-analysis, the technical success rates of CSEMSs ranged from 93.3% to 100%, while clinical success rates ranged from 82.4% to 100%, showing no significant difference compared with uncovered SEMSs or PSs. The overall incidence of complications, including cholangitis, cholecystitis, pancreatitis, and liver abscess, was reported in approximately 5.9% to 32% of cases.25 The main concern regarding CSEMSs has been peripheral bile duct obstruction–related events, particularly liver abscess and cholangitis. Although the reported frequencies varied, the highest rates of liver abscess and cholangitis were 10% and 20%, respectively. Notably, liver abscess occurred less frequently than initially expected (0%–10%), whereas cholangitis was observed in up to 20% of cases, likely due to longer follow-up durations associated with prolonged survival.25 As survival improves, RBO becomes an inevitable outcome.
Among reported complications, stent migration occurred in up to 48% of cases,19 representing a notable limitation of CSEMSs. The meta-analysis also demonstrated that clinical success, overall adverse events, and migration rates did not differ significantly according to stent type (fully vs. partially covered), diameter (6 mm vs. 8 mm), or distal tip position (above vs. across the papilla).25 The RBO rate varied widely (8.6%–88%), but reintervention was technically successful in 84.6% to 100% of cases. The median time to RBO ranged from 79 to 190 days.25
Of the six studies focusing on unresectable MHO, only one was a prospective comparative study with a limited sample size. Considerable heterogeneity in methodology, stent characteristics, and follow-up duration, along with small patient numbers, represented major limitations.25 These factors likely contributed to statistical and publication bias. Nevertheless, several studies using CSEMSs have demonstrated promising results in advanced MHO. Based on these findings, large-scale, well-controlled comparative trials and tailored CSEMS designs are warranted to further define their clinical utility.
Limitations of CSEMS
The limitations of CSEMSs in the management of biliary obstruction are multifactorial. First, the anatomical complexity of the biliary tree, particularly in the perihilar region, poses substantial challenges. This area is characterized by the confluence of multiple intrahepatic bile ducts, and placement of a fully CSEMS in this region carries the risk of occluding peripheral ducts. Such occlusion can increase the incidence of liver abscesses and cholangitis.
Second, the covered membrane may predispose patients to sludge formation, biofilm development, and bile stasis, thereby increasing the frequency of cholangitis and RBO. Sludge formation remains a key drawback of CSEMSs, particularly in patients with prior cholangitis or RBO. Currently, no functional CSEMSs with sludge-inhibitory or anti-reflux properties have shown superior outcomes, and these specialized designs are not widely available for managing MHO. Fully CSEMSs are typically positioned across the papilla to allow removal in case of stent malfunction; however, this placement may facilitate sludge accumulation on the covered membrane, underscoring the need for improved stent designs.
Third, because the covered surface of a CSEMS does not firmly adhere to the bile duct wall, the risk of stent migration is relatively high, with reported migration rates reaching 48%.
Fourth, obstruction of the cystic duct or pancreatic duct by a CSEMS can lead to complications such as cholecystitis or pancreatitis. This risk is particularly relevant in cases of hilar obstruction when the stent is deployed near the cystic duct orifice. Although mild pancreatitis has been reported in some patients, the removability of fully CSEMSs remains a major advantage.
Finally, technical difficulties may arise during reintervention. Although CSEMSs are theoretically removable, reintervention may be hindered by tumor progression, stent ingrowth or overgrowth, or sludge accumulation, complicating subsequent procedures.
The future of CSEMS in MHO
The primary use of CSEMSs in advanced MHO remains controversial. However, recent studies indicate that complication rates are not significantly higher than expected and that outcomes are comparable to those with uncovered SEMSs or PSs. The ability of CSEMSs to overcome the limitations of PSs while maintaining the advantages of uncovered SEMSs is particularly promising. Nonetheless, confirming their clinical value will require the development of stents specifically optimized for advanced MHO, as well as large-scale comparative studies evaluating long-term safety and efficacy.
CONCLUSIONS
Although current guidelines do not recommend CSEMSs as the primary drainage modality for MHO, recent evidence demonstrates increasing clinical interest, particularly in patients receiving systemic or local therapies, for whom PSs are typically used first. Emerging but limited data suggest that CSEMSs may achieve comparable complication rates to PSs while providing longer patency. These findings suggest a potential role for CSEMSs either as an alternative after PS dysfunction or, in selected cases, as an initial drainage option. However, prospective multicenter randomized controlled trials comparing fully and partially CSEMSs with uncovered SEMSs or PSs are needed to clarify their clinical utility.
Notes
Conflicts of Interest
The authors have no potential conflicts of interest.
Funding
This work was supported by the Soonchunhyang University Research Fund.
Author Contributions
Supervision: all authors; Writing–original draft: THL; Writing–review & editing: all authors.