Technically, single stent insertion to alleviate hilar obstruction is a safe and simple method. De Palma et al.18,19 revealed that unilateral metal stent drainage is safe, easy, and achieves adequate drainage in most patients with nonresectable HCCA. Technical success was achieved in 96.7% (59 of 61), with stent malfunction occurring in only three cases (4.9%). Functionally successful drainage was achieved in 96.7% (59 of 61), and complete resolution of jaundice occurred in 86%. Compared with bilateral stents, unilateral stents showed a significantly higher rate of successful endoscopic insertion (76.9% vs. 88.6%, respectively; p=0.041) and a significantly lower rate of complications (26.9% vs. 18.9%, respectively; p=0.026). Therefore, the authors concluded that the insertion of more than one endoprosthesis would not be justified as a routine procedure in patients with hilar tumors. Unilateral stenting with either metal or plastic stents showed a significantly higher rate of technical success, lower complication rate, and higher rate of successful drainage compared with bilateral stent insertion.15,18,19,20,21 Iwano et al.21 also reported that unilateral drainage is associated with a lower incidence of liver abscess, as well as comparable outcomes with regard to stent patency and complication-free survival compared with bilateral drainage. Recently, Mukai et al.22 reported that there was no significant difference in the stent patency between unilateral and bilateral stenting. However, reintervention for stent dysfunction is more complicated in bilateral stenting. The success rate for endoscopic reintervention was significantly higher in unilateral stenting than in bilateral stenting (100% vs. 68%; p=0.0272). They concluded that unilateral stenting is more beneficial than bilateral stenting.

Nevertheless, despite being a simple and safe method, unilateral drainage may not provide adequate drainage physiologically and does not show superior stent patency.

Although bilateral or multiple biliary drainage is controversial, it may be necessary specifically when both hepatic lobes are diseased or opacified, when a nondominant or atrophic lobe has been inadvertently stented without achieving drainage, or if bilateral brachytherapy is scheduled. Chang et al.23 observed the highest survival among HCCA patients who underwent bilateral drainage (225 days), and markedly lower survival among those who showed cholangiographic filling of both lobes (i.e., opacification) but underwent drainage on only one side (80 days, p<0.01). In addition, the authors suggested that inadvertent contrast medium injection into the intrahepatic ducts without adequate drainage was associated with worsened outcomes. In addition to ensuring adequate drainage, bilateral stenting may improve stent patency. Naitoh et al.24 found no significant difference between unilateral and bilateral endoscopic metal stenting with regard to successful insertion (100% vs. 90%), successful drainage (100% vs. 96%), incidence of early complications (0% vs. 10%), or incidence of late complications (65% vs. 54%). However, the cumulative stent patency was higher in the bilateral stent group (median stent patency, 488 days) than in the unilateral stent group (210 days, p=0.009). Liberato and Canena25 reported that endoscopic reintervention for stent occlusion was required more frequently in patients with unilateral rather than bilateral plastic stents (80.9% vs. 34.2%, respectively; p<0.001) and in those with unilateral rather than bilateral self-expanding metal stent (SEMS) (31.4% vs. 11.9%, respectively; p=0.036). The median stent patency period was 17 weeks for unilateral plastic stents, 18 weeks for bilateral plastic stents, 24 weeks for unilateral SEMS, and 29 weeks for bilateral SEMS. Multivariate analysis revealed that SEMS placement and bilateral deployment were the only independent prognostic factors associated with prolonged stent patency.

Accordingly, bilateral or multiple drainage may be more physiologically sound than unilateral drainage. Also, various bilateral drainage techniques and newly developed metal stents are now available.26,27,28,29,30,31 Nevertheless, the deployment of multiple metal stents for bilateral drainage is generally thought to be technically demanding. One commonly encountered surgical error is the inadvertent injection of contrast medium past the hilar stricture and into atrophied and/or unintended multiple hepatic segments, which may lead to postprocedure cholangitis and lower survival rates.23 In addition, the incidence of cholangitis in patients with hilar obstruction was significantly higher than in those with a distal obstruction.32 To reduce the incidence of postendoscopic retrograde cholangiopancreatography (ERCP) cholangitis, several studies have examined the use of selectively targeted and planned endoscopic drainage guided by MRCP or CT imaging, and found a decreased incidence of postprocedure complications.19,31,32,33,34,35 Thus, when performed correctly, endoscopic bilateral deployment of multiple metal stents has been shown to be a feasible and highly useful procedure, particularly in terms of prolonging patency and preserving functional liver volume.36 A multicenter retrospective study reported that successful biliary drainage was significantly higher in the percutaneous group than in the endoscopic group (93% vs. 77%; p=0.049).37 However, institutional expertise in endoscopy may also be an important factor.

When viewed from the perspective of an experienced endoscopist, bilateral drainage by using metal stents may be a technically feasible and safe method. Numerous recent studies also revealed higher technical and functional success rates in bilateral drainage even with metallic stents.14,26,27,28,29,38,39,40,41,42 Accordingly, if appropriate endoscopic expertise is available, endoscopic palliation may be the first choice even for advanced HCCA, Bismuth type III or IV strictures.

Bilateral side-by-side stent placement for HCCA is frequently performed by using an endoscopic approach and a parallel arrangement, as reported by Cheng et al.45 and Dumas et al.46 If both guidewires are inserted into the right and left intrahepatic ducts, sequential stent insertion may be easy. In addition, bilateral revision when stents become occluded may be easier because endoscopic revision can be performed through each stent. The issues surrounding this technique include the potential entanglement of the two guidewires, the difficulty of precisely deploying the stent to ensure adequate drainage, and the need for endoscopic revision should the stents become occluded.11 After the deployment of the first SEMS, insertion of the delivery system of the second stent can be disturbed by the resistance against the first deployed SEMS. Insertion after preloading on the guidewire may speed up the procedure, and small-diameter delivery systems with good pushability are needed. Both SEMS should be placed with their distal ends in the duodenum or at the same level in the common bile duct to facilitate SEMS revision when the stents occlude. However, precise deployment of both stents on the same level may be difficult in cases of different levels of stricture length. To overcome these difficulties, Chennat and Waxman30 introduced a simultaneous side-by-side deployment method by using a thin 6-Fr delivery system. This system showed a high rate of technical success, and it may be feasible and ideal for use in simultaneous side-by-side deployment. However, the handling of a side-by-side simultaneous delivery catheter within the same endoscopic working channel and biliary duct resulted in additional frictional forces, which may increase the tendency of the delivery system to buckle in the distal duct. Also, the resulting median stent patency was relatively short (130 days), which might be due to the short follow-up period and the small diameter of stents. Additional large-scale studies with long-term follow-up are needed.

Technically, the use of a guidewire to cannulate to the desired contralateral bile duct through the previously inserted SEMS may be difficult in bilateral stent-in-stent procedures. In addition, during endoscopic revision for stent occlusion due to tumor ingrowth, the previously inserted wire mesh may preclude the insertion of a revisionary stent. To enhance secondary SEMS insertion, balloon dilation of the contralateral hepatic duct immediately before the first stenting, or dilation of the first deployed SEMS with a balloon before inserting the contralateral second SEMS, can be useful.29,40,42 In cases of severe stricture, following the expansion of the first stent, delayed insertion of the second stent by 2 to 3 days may also be helpful.

Structurally, large open-celled wire mesh stents can improve the technical success rates in cases of severe stricture. Kogure et al.31 (Niti-S large cell D-type; Taewoong Corp., Seoul, Korea) and Lee et al.27 and Hwang et al.28 (Niti-S Biliary Y-stent; Taewoong) showed high technical success rates with large-cell stents in the bilateral stent-in-stent method, although the number of cases in each study was small. However, the expanding radial force may decrease in stricture sites and can create an inherent susceptibility to tumor ingrowth because of their large open-mesh design. In addition, full expansion of the second stent may be hindered owing to the space limitation moving through the large open mesh. Alternatively, closed-cell, cross-wired metal stents, such as the M-hilar Bonastent (Standard Sci Tech Inc., Seoul, Korea) also showed high technical success and revision efficacy for endoscopic bilateral stent-in-stent placement in HCCA patients. The primary technical success rate of endoscopic bilateral stent-in-stent placement was 95.2%, and the median patency was 238 days. The technical and clinical success rates of planned bilateral endoscopic revision for occluded stents were 83.3% (20 of 24) and 79.2% (19 of 24), respectively.29 Endoscopic bilateral revision with plastic or metal stents was not restricted technically.

The reported success rate of endoscopic revisions in bilateral metal stents varied from 44.4% to 100%, as shown in Table 1. However, the reported studies used various combinations of unilateral or bilateral stent insertion by means of side-by-side or stent-in-stent insertion methods. Furthermore, the limited number of patients and conflicting data among these studies prohibit a direct comparison of the results. Nevertheless, newly developed or modified stents and devices recently showed higher technical feasibility without increased complications. These results may encourage ERCP endoscopists to use these techniques and stents as soon as possible to prevent difficulties.

Well-designed, large-scaled comparative studies comparing stent materials and deployment methods in HCCA are now limited. Direct comparison of reported studies is made difficult by the variety of stents and methods used. As long as experienced endoscopists are available, bilateral drainage by using SEMS may be the preferred option, whether placed by means of the stent-in-stent or the side-by-side deployment method. However, it is still controversial which method is better. Also, for optimal drainage, multisectoral drainage rather than unilateral or bilateral drainage should be considered regardless of the use of the endoscopic or percutaneous approach in advanced hilar obstruction. Endoscopists' preferences and institutional capabilities should also be considered before the procedure. Also, besides endoscopic palliation, additional therapies such as chemotherapy, radiotherapy, photodynamic therapy, and the recently reported radiofrequency ablation for the local treatment of tumors should also be considered for improving patency and patient survival. Future prospective multicenter studies are needed to establish the optimal therapeutic protocol.