Ultrasonography (USG) i s a non-invasive imaging modality that enables observation of the intrahepatic bile ducts and confirmation of the surrounding vascular patency. However, USG is not suitable for confirmation of the presence of biliary stricture. Sharma et al. [29] reported that only 38% to 66% of biliary strictures occurring after LT are diagnosed successfully by USG. Therefore, diagnosis and screening for biliary stricture cannot be performed using USG alone.

In recent years, multidetector computed tomography (CT) has been developed, and its resolution has improved. As a result, CT is used widely as a follow-up examination for patients after LT. It is also useful for the identification of non-biliary complications, such as fluid collection around the operation site. However, the utility of contrast-enhanced CT may be limited by the occurrence of adverse events related to the intravenously administered contrast agent [12]. Thus, development of a more stable and safer contrast agent is needed.

Magnetic resonance cholangiography (MRC) can be used to obtain detailed images of the whole biliary tract, and its sensitivity and specificity for the diagnosis of AS are both ≥ 90% [30,31]. In addition, the performance of unnecessary interventions can be reduced due to the high negative predictive value of MRC findings [32]. Thus, detailed observation of the biliary tract by MRC not only increases the success rate of interventions to treat biliary stricture but also reduces intervention-related complications. However, MRC has no therapeutic capability and is expensive.

Hepatobiliary scintigraphy using 99m-technetium-labeled iminodiacetic acid has shown excellent accuracy for the diagnosis of bile leaks. However, studies have yielded inconsistent findings regarding its performance in the diagnosis of biliary stricture. Some authors have reported that the sensitivity and specificity of scintigraphy for the diagnosis of post-LT biliary stricture were 75% and 100%, respectively, whereas others have reported a sensitivity of only 60% [33,34]. These results suggest that the use of this imaging modality for the diagnosis of post-LT biliary stricture cannot be strongly recommended [35,36].

Biliary reconstruction in LDLT has been achieved using Roux-en-Y hepaticojejunostomy. In recent years, however, duct-to-duct anastomosis has been used in most cases. As a result, endoscopic management is feasible and can be attempted as first-line therapy for biliary stenosis occurring after LDLT [37,38]. New endoscopic accessories have been developed and endoscopic techniques improved during the past two decades, increasing the success rate of endoscopic management [39]. When NAS occurs post-LDLT, endoscopic management at all affected duct sites is practically impossible. Therefore, endoscopic management should be performed preferentially in patients with AS.

In general, selective cannulation of the bile duct is performed as in endoscopic retrograde cholangiography (ERC), and a guidewire is passed to the proximal side of the AS. Next, endoscopic sphincterotomy is performed. Subsequently, balloon dilation is used at the stricture site, and the AS is expanded using a Sohendra biliary dilation catheter or Sohendra stent retriever, as needed. Finally, a plastic stent is inserted into the stricture site. After 2 to 3 months, the stent is replaced by a stent of larger diameter [40-42].

Balloon dilation is commonly used before insertion of a plastic stent into a stricture site. However, some endoscopists prefer to insert the stent directly, without dilation, in the first ERC procedure, and to use balloon dilation during the second ERC procedure, prior to stent exchange [12]. No study has shown that this method results in a better outcome compared with the conventional protocol, and further research is needed.

Endoscopic sphincterotomy may prevent post-ERC pancreatitis due to the blockage of pancreatic juice after the insertion of a plastic stent of large diameter [40]. However, the function of the sphincter of Oddi is lost irreversibly after this procedure, potentially resulting in duodenobiliary reflux, stent occlusion, and bacterial colonization in the biliary system [43-47]. For this reason, sphincterotomy may reduce the advantage of biliary reconstruction with duct-to-duct anastomosis in patients undergoing LDLT.

Recently, Isayama et al. [46] described the “inside stent method” for AS management. In this method, a plastic stent is placed across the AS without performing sphincterotomy, and the distal tip of the stent is positioned inside the bile duct. Using this method, Isayama et al. [46] inserted stents into the desired locations in 80% of 118 patients with post-LDLT AS and confirmed stricture resolution in 69% of the patients.

The performance of sphincterotomy may reduce the overall procedure time, as it facilitates the use of accessories and cannulation in subsequent ERC procedures. A prospective randomized controlled trial comparing the “inside stent method” with the conventional protocol is needed to establish the efficacy of this approach.

When stent exchange is performed every 2 to 3 months, according to the conventional protocol, approximately 1 year is needed to resolve an AS [22,47-51]. Therefore, several attempts have been made to shorten the duration of treatment.

Morelli et al. [52] performed stent exchange every 2 weeks, with a mean treatment duration of 3.6 months and a high (87%) success rate. In another study, stents with maximal diameters were inserted during the initial ERC procedure, and stent exchange was used only when stent occlusion occurred [53]. Neither of these studies involved follow-up of a large number of patients, and these methods cannot yet be recommended as standards of care.

Because self-expanding metal stents (SEMSs) have larger diameters (up to 30 Fr) than those of plastic stents, their use is expected to achieve excellent resolution of various benign biliary strictures, including post-LDLT AS [54]. When uncovered SEMSs are used, occlusion due to the ingrowth of granulation tissue and other complications, such as pseudoaneurysm in surrounding vascular strictures, may occur [54-57]. The use of fully covered SEMSs (cSEMSs) can overcome these drawbacks but may result in occlusion of secondary branches of the bile duct and subsequent bile stasis [58]. In addition, cSEMSs have a migration rate of 16% to 33%, resulting in failure to expand the AS for a sufficient period of time and an increased number of ERC procedures [59,60]. In a recent study involving the use of a new cSEMS (Kaffes stent, Taewoong Medical, Seoul, Korea) with a central waist, a high treatment success rate (83%) and low migration rate (6%) were achieved [61]. Furthermore, the Kaffes stent can be moved into the endoscopy channel by grasping and pulling of the lasso (Fig. 1).

The incidence of stricture recurrence during long-term follow-up after AS treatment using the conventional protocol is 12% to 30% (Table 1) [15,62-65]. However, the use of ERC to treat AS was introduced relatively recently, and longer-term follow-up studies are needed. In addition, existing studies have been criticized due to the heterogeneity of parameters such as donor age and endoscopic management protocols. Thus, no generalization can be made about the long-term outcomes of AS treatment post-LDLT.

Several studies have shown that the occurrence of AS after LDLT does not affect overall survival [62,66]. Recently, Chok et al. [67] reported that the 1-, 3-, and 5-year survival rates did not differ between patients with AS (n = 55) and those without AS (n = 210; 97.3% vs. 91.4%, 90% vs. 85.3%, and 84.5% vs. 82.5%, respectively; p = 0.574).