Both plastic stents (PS) and self-expanding metal stents (SEMS) have been used for biliary drainage (Figure 2). PS have smaller diameters (10-12 Fr), resulting in faster occlusions and a median patency time of only 1.4 to 3 mo[13]. SEMS have a wider diameter (8-10 mm), resulting in longer patency of 6-10 mo[13]. SEMS, which are used for HC, are uncovered with an open mesh, allowing the drainage of side branches. Perdue et al[14], in a multicenter study involving 62 patients, compared metal stents with plastic stents in HC for 30 d outcomes. Adverse effects, including cholangitis, stent occlusion, migration, perforation and the need for reinterventions, occurred in a significantly higher proportion of patients in the plastic stent group (39.3%), compared to metal stents (11.8%). These results were revalidated in a recent study from Thailand[15]. In this randomized study involving 108 patients, Sangchan et al[15] demonstrated that metal stents were better in terms of success rate (70% vs 46%) and patient survival (126 d vs 49 d). In spite of its high initial cost, SEMS are considered more cost-effective than PS. This result could be due to the greater success rate, shorter hospital stays, fewer blockages, fewer re-interventions and lower antibiotic needs associated with SEMS. The superior cost-effectiveness of SEMS compared to PS is particularly apparent if patient survival is expected to be more than 4-6 mo. However, in situations where a decision of palliation has not been made, plastic stents may be preferred because uncovered SEMS used for hilar blocks are not easily removable.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Stenting for hilar cholangiocarcinoma. A: Unilateral plastic stent; B: Bilateral plastic stents; C: Unilateral metal stent; D: Bilateral metal stents (previously placed percutaneous drains are still in place).

For Bismuth type I HC, it is obvious that one stent is sufficient to drain both lobes of the liver because the confluence is patent. However, the placement of single stent for Bismuth type II to type IV HC would drain only a proportion of liver; instead, multiple stents are often placed in these advanced HC patients. However, no consensus has been reached in terms of the need for multiple stents.

De Palma et al[16] reported better results with unilateral stenting (vs bilateral stenting) in terms of successful stent placement, effective drainage and complication rates (Table 2). However, approximately one-third of the patients in this series had Bismuth type I HC, leading to a bias in the study result. Contrary to this study, Naitoh et al[17] demonstrated a trend towards longer survival and lower cholangitis with bilateral, compared to unilateral, drainage. An important conclusion was drawn in a study by Chang et al[18], which showed that patients with HC faired very poorly, with high cholangitis rates (32%), high 30 d mortality (30%) and poor survival (45 d) if only one stent was placed after opacifying both sides of the liver. This was in contrast to the other two groups, in which the opacified lobes were drained by one or two stents (Table 3). It therefore seems that any contrast placed in an obstructed system must be drained. An experienced endoscopist dealing with hilar blocks is always reluctant to inject any contrast until a guide wire has been negotiated well beyond the site of the biliary obstruction. A study in India[19] demonstrated the importance of contrast-free unilateral endoscopic palliation in Bismuth type II HC. In 18 patients submitting to this technique, successful drainage was achieved in all, with no cholangitis or 30 d mortality.

Conventional teaching suggests that, in the absence of cholangitis, it is only necessary to drain 25% of the liver volume for the adequate palliation of jaundice[20]. However, a recent study by Vienne et al[21] showed better results with the drainage of more than half of the liver volume. More than a 50% drainage of the liver volume was associated with a greater decrease in bilirubin levels, a lower incidence of early cholangitis and longer patient survival compared to patients with less than a 50% drainage of the liver volume. It is known that the right lobe, left lobe and caudate lobes of liver constitute 55%-60%, 30%-35% and approximately 10% of the liver volume, respectively[22]. Incorporating this information in the results of the study by Vienne et al[21], a significant proportion of patients should undergo the draining of at least two segments and the placement of more than one stent to achieve good long-term palliation. However, more data are needed in this respect. It is important to note that, in the presence of cholangitis, all infected ductal systems need to be drained.

It is important to select the appropriate segments of the liver that need to be drained when unilateral or incomplete drainage is planned in patients with advanced HC (Bismuth type II to type IV). It is advisable to select segments with dilated ductal systems and to avoid atrophic segments (Figure 3). MDCT or MRI can provide this useful information prior to palliative stenting[23,24]. A number of studies with both PS and SEMS have shown the usefulness of CT/MRCP-guided, selective biliary drainage (Table 4)[16,25,26]. This CT/MRCP approach can reduce the need for further intervention and has been found to be cost-effective compared to routine bilateral stenting[27]. A recent study[28] with a small number of patients used air cholangiography rather than MRCP to act as a lower cost road map and found air cholangiography to be safe and effective with no cholangitis and no 30 d mortality or morbidity.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Hilar stricture: Unilateral metal stent placed in dilated left lobe duct. A: Left duct selectively cannulated and showing dilated system; B: Metal stent being deployed in the left ductal system.

Stents in HC are mostly introduced at endoscopy[23]. In terms of the complexity of endoscopic procedures, stenting a hilar block is considered one of the most difficult. In a recent consensus, the American Society of Gastrointestinal Endoscopy graded endoscopic hilar stenting as a level 3 ERCP procedure in terms of complexity, with level 1 being the simplest and level 4 being most complex[29]. In general, a higher level of complexity is associated with a lower success rate and a higher complication rate[30]. Therefore, hilar stenting should be practiced only be experienced therapeutic endoscopists.

A variety of plastic and SEMS are available and have been used for the stenting in HC. In general, 10 Fr plastic stent and uncovered SEMS are preferred. The distal end of stents may be left in the duodenum or in distal bile duct, but the later situation may make reintervention more difficult. When more than one stent (plastic or SEMS) is to be placed, the stents are usually placed side by side (Figure 4). However, in the last few years, a new dual stent design called “stent-in-stent” has been developed for metal stenting[31,32] (Figure 5). In this device, the first stent has an open-cell design, allowing the second stent to pass easily through the first stent. In a recent study by Lee et al[33] from South Korea, stent-in-stent technology for bilateral stenting was evaluated in 84 patients with inoperable, high-grade HC. Technical and clinical success was achieved in 95.2% and 92.9% of patients, respectively. The median survival and patency were noted to be 256 d and 239 d, respectively. Still, this new stent design can be problematic if the first stent becomes occluded. In the study by Lee et al[33], 30.8% patients had an obstruction of the primary biliary stents. For revision stenting, bilateral metal stents could be placed in 55%, while plastic stents were placed in the remaining patients[33].

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Two metal stents placed side by side in a Type II Bismuth hilar cholangiocarcinoma. A: Magnetic resonance cholangiopancreaticography showing the stricture; B: Endoscopic view showing two metal stents placed side by side; C: X-ray showing two parallel stents with proximal ends in the right or left ductal systems.

Open in New Tab   Full Size Figure   Download Figure

Figure 5 Y stent (stent-in-stent) from Tae Woong Medical (South Korea) for Hilar Cholangiocarcinoma. A: Two stents one with open mesh in the middle (arrow) and other with normal mesh; B: Y stent system in which the normal mesh stent passes through the open mesh of the second stent.

It is important to institute antibiotic prophylaxis in patients with anticipated incomplete biliary drainage by any technique[34]. Antibiotics should be continued in cases of incomplete biliary drainage. The choice of antibiotics should be based on local hospital data.

The occlusion of metal stents, while more common and earlier with plastic stents, can still occur[35]. Blocked plastic stents should always be removed, and the bile ducts should be cleaned of all sludge by a balloon. Then, a new plastic stent can be positioned. Metal stents, however, may be the better choice, particularly in the presence of thick bile or purulent material. It may also be a good idea to temporarily place a nasobiliary drain for repeated flushing prior to restenting[35]. Uncovered SEMS cannot be removed and therefore should be cleaned if they become clogged. However, it is advisable to place another stent through the blocked metal stent. The new stent could be either plastic or SEMS, depending upon the patient’s life expectancy.

Photodynamic therapy (PDT) is a relatively new technique that has evolved over the last 10 years[55]. PDT uses a photosensitive agent that concentrates preferentially in malignant tumors. Subsequent photoactivation with red laser lights of a specific wavelength[56] creates reactive oxygen, leading to selective tumor cell death. Table 5 summarizes some of the studies[56-58] that have shown improved patient survival by the addition of PDT to biliary stenting. Studies have also shown improved quality of life after PDT[56,58]. In HC patients who have been treated with plastic stents, the stents should be removed temporarily for light delivery during PDT. In cases of SEMS, PDT can be performed through the stent by adjusted the required light dose to compensate for the decreased transmission of light[59]. A further moot point that should be addressed is whether these results can be reproduced in certain Asian countries, in which the sun does not affect skin photosensitivity to the same extent as in the Western white population[57,60].

In general, PDT is preferred through the percutaneous route [compared to the endoscopic] because of the lower chance of laser tip fracture, the easy repeatability of the procedure and the feasibility of obtaining cholangiogram. Shim et al[61] showed that PDT using percutaneous cholangioscopy is safe and effective, improving quality of life with a good median survival time (i.e., 558 ± 178.8 d). Cholangitis, hemobilia and photosensitivity are known complications of PDT.

A recent study by Wagner et al[62] used temoporfin rather than conventional porfimer for PDT in 10 patients with HC. Temoporfin PDT was highly tumoricidal and had double the depth of local tumor ablative effect, compared to porfimer PDT. Infectious complications and skin photosensitivity were similar with both agents[62].

Radiofrequency ablation (RFA) is an ablative procedure that is well established for treating small liver cancers. With the introduction of an endoscopic probe for RFA and the use of a lower power setting with the existing generators, it is now possible to treat pancreatico-biliary cancers with this modality[63]. Within the bile duct, it has the potential of improving stent patency by decreasing tumor ingrowth and benign epithelial hyperplasias. Steel et al[64] were the first to demonstrate its safety and efficacy in 22 patients with biliary malignancy (pancreatic/cholangiocarcinoma). A recent study involving 20 patients with pancreatico-biliary malignancy (11 with cholangiocarcinoma) showed a significant increase in bile duct diameter after RFA[65]. Further studies are required to revalidate these results, but the preliminary data are interesting.

Radiotherapy has a very limited role in HC. Intraluminal brachytherapy has been delivered to cholangiocarcinoma using Ir-S192 seeds, with a total dose of 30-50 Gy through the percutaneous route. Good short-term effects were demonstrated, showing prolonged stent patency and improved survival[66]. A number of studies from Japan have shown better SEMS patency (10-18 mo vs 4-12 mo) after external beam radiotherapy[67,68].

Chemotherapy has been used in HC as a palliative therapy for both locally advanced cancers and metastatic disease[69]. However, most of these studies involved case series with a heterogeneous mixture of patients with HC or distal bile duct cancers, as well as pancreatic or gall bladder cancers. A retrospective study comparing a gemcitabine-based regimen with a cisplatin-based therapy and a fluoropyrimidine-based therapy clearly showed the gemcitabine-based regimen to be superior in terms of the lowest fatality[70]. Gemcitabine has been used in combination with cisplatin, oxaliplatin or capecitabine to improve its efficacy[69]. Valle et al[71,72] compared gemcitabine with a combination of gemcitabine with cisplatine in their two ABC trials, which were published sequentially. ABC-02 trial by Valle et al[72], which involved 410 patients with advanced biliary cancers in a multicenter randomized phase III study, clearly showed the superiority of the combination over gemcitabine alone. The combination improved progression-free survival and overall survival by 30% over gemcitabine alone. Another multicenter trial from Japan[73] also demonstrated the superiority of gemcitabine combined with cisplatin over gemcitabine alone in terms of one year survival, median survival time, median progression-free survival time and overall response rate. These data have resulted in this combination being used as standard of care chemotherapy for advanced cholangiocarcinoma, including HC. Gemcitabine-free combinations, such as capecitabine with oxaliplatin, capecitabine with cisplatin and 5 fluorouracil with cisplatin, have also been used for bile duct cancers with modest results[69]. However, newer drugs are being investigated for bile duct cancers; these include erlotinib, cetuximab and bevacizumab[69].