Liver transplantation is the best treatment option for patients with end-stage liver disease and early unresectable hepatocellular carcinoma. Unfortunately, biliary complication after liver transplantation is still the Achilles' heel, especially in living donor liver transplantation. Early treatment with endoscopy can achieve satisfactory outcomes. Most of the time, biliary anastomotic stricture can be treated by endoscopic retrograde cholangiopancreatography with balloon dilatation with or without plastic stent insertion. Although endoscopic treatment has been reported to have a successful rate of over 70%, multiple sessions of endoscopic treatment, typically 4 to 5 sessions, are frequently required before adequate stricture dilatation is achieved. This is likely secondary to suboptimal post-dilatation splintage. The most common and popular form of splintage is plastic stent insertion. Unfortunately, plastic biliary stent has a small calibre, and therefore even with multiple stents the configuration of buttressing would not provide a circumferential, evenly distributed buttressing effect at the dilated stricture site. Moreover, given the small calibre of the plastic stent, there is higher resistance on the inner surface of the stent, leading to a higher chance of stent blockage. Many studies have suggested that self-expandable metallic stent (SEMS) is superior to plastic stent in terms of patency rate. However, SEMS is generally reserved for malignant stricture due to its permanent nature, as the traditional SEMS is not removable. Recently, retrievable SEMS (r-SEMS) has been developed, and its indications have been extended to include benign disease conditions. It has been reported that a series of 29 patients with biliary anastomotic stricture treated by r-SEMS, and they concluded that r-SEMS was safe and efficacious. Results of the preliminary study on 5 patients at our centre were favourable; all of the patients had no stricture after retrievable metallic stenting for at least 3 months and no complication was encountered.

Biliary anastomotic stricture (BAS) is one of the most common complications after liver transplantation (LT). It happens more often after living donor liver transplantation (LDLT) than deceased donor liver transplantation (DDLT). The reported incidence was 20% in LDLT and 12% in DDLT. Although BAS seldom affects graft survival, it is associated with significant morbidity and affects quality of life. Clinical manifestation of BAS can be highly variable, ranging from low-grade cholangitis with slightly deranged liver function to life-threatening septic shock to graft and multi-organ failure. Up to 30% of the cases of BAS require surgical intervention at some point. Revision hepaticojejunostomy - a major undertaking judging from the magnitude of the operation - is sometimes required as a remedial procedure. Most of the time BAS can be treated by endoscopic retrograde cholangiopancreatography (ERCP) with balloon dilatation with or without plastic stent insertion. Although endoscopic treatment has been reported to have a successful rate of over 70%, multiple sessions of endoscopic treatment, typically 4 to 5 sessions, are frequently required before adequate stricture dilatation is achieved. This is likely secondary to suboptimal post-dilatation splintage. Since stricturoplasty features breaking up the fibrous ring at the anastomotic site and hence widening the calibre of the lumen, any new wound created by dilatation injury is susceptible to the formation of new scar. Therefore, some form of buttressing device is needed to keep the anastomotic site open. This underscores the importance of post-dilatation splintage. The most common and popular form of splintage is plastic stent insertion. Unfortunately, plastic biliary stents have a small calibre, with the largest size being Fr11.5 only. Even if multiple stents are inserted, the configuration of buttressing would not provide a circumferential, evenly distributed buttressing effect at the dilated stricture site. Moreover, given the small calibre of the plastic stent, there is higher resistance on the inner surface of the stent, leading to a higher chance of stent blockage. Frequent admissions for repeated dilatation and stent exchange (not to mention emergency admission for a cholangitic episode secondary to stent blockage) significantly disrupt the patient's normal daily activities and form a clinical and financial burden to the community. Many studies have suggested that self-expandable metallic stent (SEMS) is superior to plastic stent in terms of patency rate. However, SEMS is generally reserved for malignant stricture due to its permanent nature, as the traditional SEMS is not removable. Recently, retrievable SEMS (r-SEMS) has been developed, and its indications have been extended to include benign disease condition. It has been reported that a series of 29 BAS patients treated by r-SEMS, and they concluded that r-SEMS was safe and efficacious. Results of the preliminary study on 5 patients at our centre were favourable; all of them had no stricture for at least 4 months after r-SEMS treatment and no complication was encountered. The median number of session for success was 2, which is significantly fewer than that in the ordinary approach (median session: 4).

Up till this moment, there is no randomized controlled trial comparing the performance of r-SEMS with that of the conventional approach. In this study, the null hypothesis is that there is no difference in performance between r-SEMS and the conventional approach in endoscopic treatment of BAS.

Inclusion Criteria:

• Patients who give informed consent

Exclusion Criteria:

• Patients who refuse to give consent
• Patients who have previously hepaticojejunostomy as biliary re-construction
• Patients who have previous upper gastrointestinal surgery making endoscopic treatment not posssible