Biliary strictures remain a common challenge after LT despite improvements in techniques and learning curves. The incidence of biliary strictures is more common after LDLT (28%–32%) than after DDLT (5%–15%) because anastomoses involving small-caliber peripheral donor ducts are more complex in LDLT. Biliary strictures are less common among the living donor population (0.5%); nonetheless, they are extremely important to recognize and manage appropriately. After LDLT, because living donors sometime require re-hospitalization for further endoscopic treatment or Roux-en-Y biliary reconstruction due to complications, death from liver failure rarely occurs [11,12]. Therefore, this is an important issue for the wider application of LDLT.

These strictures are traditionally categorized into two main morphological types: anastomotic strictures (ASs) and non-anastomotic strictures (NASs); this classification is still accepted and useful for most patients. ASs are single, shorter, localized to the anastomotic site, and have an incidence of 4%–9%; they are thought to develop as a result of (1) localized fibrosis caused by operative techniques; (2) compromised local microcirculation in the bile duct; or (3) associated bile leaks [13]. Conversely, NASs are usually longer, multiple, more proximally located (near the hilum or intrahepatic ducts [IHDs]), and located proximal to the actual anastomosis in the donor’s bile ducts. They have an incidence of 5%–15% in DDLT and are especially common in cases of donation after cardiac death transplant (20%–33%). They usually present late with a mean time to presentation of 3.3–5.9 months post-orthotopic LT (OLT). However, in a significant number of patients, the morphology and complexity of these strictures do not follow such traditional patterns. Secondary complicating factors—such as stasis, sludge, and stones proximal to the strictures—lead to low-grade cholangitis, which in turn has secondary local effects on the classical stricture patterns, causing many more strictures to develop spontaneously in more proximally located IHDs. The classical strictures (ASs and NASs) are usually responsive to traditional endoscopic techniques, while these complex secondary events present challenges including endoscopic failure. Ultimately, such cases require rare radiological or surgical interventions, including new grafts.

We fully agree with the views of previous reviewers that a conscious effort by the attending endoscopist to ascertain the full operative details of the biliary and liver anatomy is important [14]. In this regard, the pre- and peri-operative radiological images of individual donor biliary anatomy as well as the operative details of biliary reconstruction must be fully used. The endoscopist must comprehend and memorize the individual images before starting the endoscopic management of post-LT biliary complications. A detailed review of the pre-intervention magnetic resonance cholangiopancreatography (MRCP) images, if available, makes it possible to know which tools may be necessary for the pathological morphology before the procedure. We believe that this is one of the most important factors improving the outcomes of individual procedures.

In both DDLT and LDLT patients with duct-to-duct anastomosis, the bile duct is usually accessed via traditional endoscopic retrograde cholangiopancreatography (ERCP) using conventional side-viewing endoscopes. However, in some post-LT patients, especially in the post-LDLT context, even reaching the duodenum involves a learning curve. The technique necessitates position changes and intra-abdominal pressure to manipulate the scope into the descending duodenum. Additionally, endoscopic access to patients with surgically altered anatomy—by Roux-en-Y HJ or Roux-en-Y gastric bypass—requires deep ERCP techniques and modifications thereof. Several studies have been published in this regard with variable success rates. Furthermore, each has had different limitations on its use in post-LT settings, focusing on pediatric colonoscopy, double-balloon enteroscopy (EN-450P5; Fujinon, Saitama, Japan), single-balloon enteroscopy (SIF-Q260; Olympus, Tokyo, Japan), spiral enteroscopy overtube (Endo-Ease Vista^®; Spirus Medical, Stoughton, MA, USA), or surgical or endoscopic ultrasonography-guided percutaneous gastrostomy followed by conventional ERCP through the gastrostomy port (in patients with Roux-en-Y gastric bypass anatomy) [15-20].

Percutaneous transhepatic cholangioscopy (PTCS), the most recent upgrade to the endoscopic management of bile duct complications in post-LT patients, provides access to the biliary system through percutaneous ports created by an interventional radiologist. This is an effective and safe intervention that substantially improves outcomes in select patients with (1) previously challenging post-LT bile-duct access problems or (2) complications that previously required more risky radiological or surgical interventions. In addition, the development and routine use of many endoscopic tools and techniques have substantially improved. This has improved bile-duct access and effective management of difficult bile-duct strictures with complex and mixed pathologies.

The success rate of endoscopic therapy of the bile duct is 80%–100% in cases of OLT, although some complex hilar and intrahepatic strictures have success rates within the lower end of this range, especially in cases of LDLT [14]. Nonetheless, success rates are consistently well above 80% in most patients, even in the post-LDLT setting, due to the availability of new access approaches and techniques.

Endoscopic management of bile-duct strictures via conventional, deep endoscopy, or percutaneous routes involves standard guide wire access across the stricture, followed by sphincterotomy, balloon dilation, and/or biliary stenting of the stricture [14,21,22]. Most patients with classical ASs require serial procedures that involve balloon dilation, followed by the placement and exchange of plastic stents over a period of at least 1 year. Several reports have claimed that balloon dilation alone is effective. However, there is now a consensus that dilation alone is less effective than balloon dilation with biliary stenting [22-26]. At this point, we must emphasize that balloon dilation should be avoided in early strictures as well as those associated with bile leaks to avoid anastomotic disruption. The frequency and timing of follow-up in serial procedures has been a subject of debate: some have suggested that an accelerated course with more early interventions and fewer longer intervals between procedures achieves equivalent results. However, in our view, such scheduling should be individualized. Under normal circumstances, we would recommend the accepted average of 2–3-month intervals; this way, endoscopists avoid stent occlusion and the risk of cholangitis.

Moreover, there is general agreement that multiple stents are preferable to single stents. A more aggressive strategy of placing the maximum number of stents (three to four 10-Fr stents depending on stricture location and proximal bile duct size) with follow-up evaluations at 3–5-month intervals as well as further dilation and stent exchanges has shown a more than 90% success rate with complete morphological stricture resolution in patients with ASs (Fig. 1) [13,27]. Indeed, an even further modified and aggressive approach has been described with better resolution of the stricture and few side effects when the maximum number of stents (up to nine) are placed following initial balloon dilation; the approach also involves on-demand repeat procedures (subject to evidence of cholangitis or stent occlusion). One study that used this approach reported that a median of three ERCP procedures were performed per patient over a median duration of 15 months and that complete resolution was achieved in 94% of patients at a median follow-up of 11 months [28].

Uncovered metal stents are typically contraindicated in post-transplant settings because of previously described factors [29]. In contrast, covered self-expanding metal stents (CSEMSs) have been a subject of interest in the treatment of post-LT biliary strictures that are refractory to conventional treatment mainly because such stents have a large diameter (30 Fr for a 10-mm stent) and a potential longer patency [29-32]. One study reported a success rate of 87.5% (14/16 patients) at 10 months follow-up, although there was a high rate of stent migration and the technique was of limited use in post-LDLT patients [32]. To overcome the common problem of stent migration while retaining the advantages of larger lumens and patency rates with good outcomes, several reports have used short-length CSEMSs combined with specialized external anchoring assemblies (GORE Viabil^®; ConMed, Utica, NY, USA) and anti-migration waists (Niti-S™ and KAFFES™; Taewoong, Gimpo, Korea) in post-LT settings [31]. However, although these reports suggest that such instruments play a positive role in selected patients at anatomically suitable locations, some researchers have reservations about case selection, suggesting that traditional multiple plastic stents (MPSs) could have achieved the same results [33]. In another study comparing stricture resolution rates between MPSs and self-expanding metal stents (SEMSs), SEMSs did not have a clear advantage over MPSs [34]. Additionally, SEMSs are less likely to be beneficial in complex hilar and more proximal strictures, especially in the LDLT context, because space is limited in proximal bile ducts. Further experience and data are needed to better evaluate the usefulness of metal stents in treating post-LT biliary strictures.

Bile duct strictures involving the hilum and more proximal IHDs, including the proximal ASs and NASs, are more challenging than those located more distally. One major challenge in managing such strictures is access past the stricture, either up-stream into the proximal bile ducts from below the distal bile ducts or into the enteric lumen across tightly stenosed and tortuous strictures via the transhepatic route. In this regard, swing-tip catheters are extremely useful for selective cannulation of the angled, tightly closed anastomosed strictures in the segmental ducts. Similarly, two-in-one ductoplasties or multiple-duct anastomoses are usually difficult to access using a traditional cannula. In such cases, sphincterotomies and balloons are also used for wire access.

Likewise, a rendezvous technique may be used to achieve selected duct access when the traditional route fails, especially in patients with angulated or tortuous strictures. Such a hybrid technique combines the percutaneous transhepatic and endoscopic transpapillary approaches (Fig. 2) [35,36]. Many studies have demonstrated that the rendezvous technique is useful and safe for the management of biliary stricture after LDLT. In addition to this classical method, various modified techniques have been attempted using a metal ball-tip cannula (StarTipV; Olympus, Tokyo, Japan) and Kumpe catheter (Beacon^®; Cook Medical, Bloomington, IN, USA) [37].

Aside from these access-related issues, the small caliber of more proximal extra- or intrahepatic bile ducts may limit the caliber and number of stents that can be placed. Such cases require multiple procedures, and sometimes, permanent indwelling stents and catheters. There have been some encouraging results showing successful endoscopic treatment of such difficult and complex strictures. However, a standardized therapeutic approach is yet to be identified and the outcomes are still less than satisfactory [14]. One study suggested that balloon dilation alone has greater efficacy (91% vs. 31%) and lower rates of cholangitis (12% vs. 25%) than balloon dilatation with stenting in the case of NASs [38]. Conventional polyethylene stents are relatively rigid, do not have side holes, and can occlude secondary branch ducts. Stent migration is also a risk that can result in intestinal perforation. Such limitations are the usual causes of poor results using conventional plastic stents to treat NASs. In this regard, one recent study reported a modified approach to the management of NASs. Specifically, the study authors combined balloon dilation with multiple specialized biliary stents. Large-caliber (8.5–10 Fr), long (12–20 cm), highly flexible, fenestrated stents with multiple side holes (Johlin™ Pancreatic Wedge Stent; Cook Medical, Bloomington, IN, USA) were placed deep into the IHD. The stents were thought to allow adequate bile drainage through their multiple side holes as well as via the inter-stent space. Moreover, the anchoring effect of the upstream wedge end meant that there was very little risk of migration [14].

Single-operator peroral cholangioscopy (POCS) using the SpyGlass^® Direct Visualization System (Boston Scientific, Marlborough, MA, USA) is effective for successful guide wire placement in patients with biliary strictures by allowing direct visualization of the inner wall of bile ducts. In addition, a pinhole orifice can be visualized at the stricture site. Recent studies have reported that this technique is feasible and can be performed successfully in both pre- and post-OLT patients with no complications [39,40]. Similarly, PTCS is an effective treatment option in the case of benign ASs. It allows direct visualization of the biliary system and enables identification and treatment of intraductal lesions; in this way, it has an advantage over ERCP [41]. In cases of post-LT biliary strictures, PTCS has been used when duct access is endoscopically difficult (e.g. in Roux-en-Y HJ patients) as well as in patients with an intractable stricture that cannot be traversed using a retrograde approach or rendezvous access combined with ERCP. One study reported that PTCS treatment in cases of benign bilio-enteric ASs had an initial technical success rate of 100% and that the overall success rate was 81% [42]. Recently, the combination of endoscopic and percutaneous transhepatic radiologic interventions has had an increasing role, especially in patients with multiple intrahepatic biliary strictures. The percutaneous balloon dilatation and catheter maintenance method for post-LT biliary strictures has a 92% technical success rate, and 84% (10/12) of patients preserved bile duct patency for a mean 19 months [43]. Similarly, PTCS and the placement of internal-external stents may play a role in relieving jaundice and improving patients’ clinical conditions. In any case, both are effective and complementary therapies for LT-associated biliary complications [44].

Magnetic compression anastomosis (MCA) was recently developed as another hybrid technique for resolving post-LDLT biliary duct-to-duct anastomosis strictures that cannot be resolved using conventional methods such as ERCP and PTCS. This technique can be applied in cases of completely obstructed or disconnected biliary strictures (Fig. 3) [45]. One study involving this technique achieved magnet approximation to the anastomosis stricture in 84% (10/12) of patients. With regard to MCA-related complications during the follow-up period (mean, 331 days; range, 148–581 days), there was one case of mild cholangitis and one case of ASs recurrence [46]. Furthermore, two bilio-enteric MCA procedures have been used to treat ASs after LT that could not be crossed using the conventional guidewire method. During the follow-up period (54 months and 6 months, respectively), restenosis occurred in a single case 6 days after drainage tube removal, but this patient remained asymptomatic after additional balloon dilatation [47]. The advantages of MCA are that it is less traumatic than surgical therapy and has a very low stricture recurrence rate. Therefore, MCA safely and effectively resolves difficult and completely obstructed post-LDLT biliary strictures.

Most of the literature on bile leaks after LT surgery is old and revolves around the T-tube and its management. With the increasing prevalence of LDLT and the many modifications of DDLT, the pathogenesis of bile leaks has shifted to more technical factors; in this regard, T-tube related problems are now reported less frequently. The main factors contributing to bile leakage are partial or complete anastomosis site disruption, ischemic injury due to hepatic artery ischemia, and cut surface leakage at a bile duct tributary or T-tube insertion site. The prevalence of bile leaks does not differ significantly between DDLT and LDLT [3], nor does anastomosis type (CC or HJ) [6]. The overall incidence of bile leakage is 2%–25% [3]. Moreover, it is also important to recognize the association between bile leaks and bile duct strictures because bile leaks are an important factor in the development of biliary strictures in post-LDLT recipients—particularly older patients (>50 years) [10]. A comprehensive review of 55 articles on the subject noted that the onset of biliary leak ranged from 1 day to 6 months after LT. Meanwhile, the incidence of the complication was 7.8% (668/8585) among DDLT patients and 9.5% (268/2812) among LDLT patients [4].

Bile leaks can be classified as early (within 4 weeks of LT) and late (>4 weeks after LT) [3]. Early leaks are typically associated with anastomotic leaks, local ischemic injury due to hepatic artery thrombosis, leakage around a T-tube insertion site (if present), or the cut surface of the liver after LDLT. Late bile leaks tend follow T-tube removal—up to 33% of such leaks occur at this point [9,48].

Leaks at the T-tube insertion site can be treated conservatively and symptomatically. Indeed, one-third to one-half of such leaks close spontaneously within 24 hours [49]. The majority of the remaining patients with bile leaks require non-operative diversion of biliary flow by the following measures: (1) unclamping of the T-tube; (2) endoscopic sphincterotomy with or without biliary stenting; (3) placement of a nasobiliary or percutaneous transhepatic biliary drainage (PTBD) catheter; or (4) a combination of any of these methods with high success rates. Most studies have reported symptom resolution in >85% of patients [22,50,51]. Historically, nasobiliary catheters were used successfully to treat post-OLT bile leaks (Fig. 4) [44,51]. In one study, among 551 reported cases of bile leakage (283 after DDLT and 268 after LDLT) with a clearly delineated treatment modality, ERCP was used in 38% of cases, surgical revision in 28% of cases, PTBD in 10% of cases, and observation in 34% of cases. Re-transplantation was not required in any cases, but 13 patients (1%; 10 after DDLT and 3 after LDLT) died of secondary sepsis [4].

ERCP, along with sphincterotomy and biliary stent placement, has become the preferred method to diagnose and treat bile leakage with high success rates and low complication rates [52,53]. The stent is usually kept in place for several weeks to ensure adequate leak healing. After successful conservative management and in the absence of any pathology, the stent is removed. Small leaks can also be managed with sphincterotomy alone [54].

CSEMSs have recently been used to treat biliary leaks that are refractory to conventional treatment. In one study, 35 patients with bile leaks were treated with fully and partially covered 8–10 mm diameter SEMSs that were subsequently removed. The median duration of stenting was 9 weeks, and 33 (94%) of the bile leaks were resolved [55]. PTCS is used in cases where ERCP cannot be performed or in patients who have undergone Roux-en-Y reconstruction, in which the biliary orifice cannot be reached using a regular side-view endoscope. Occasionally, PTCS may require internal-external biliary stent placement via interventional radiology, and patients may eventually require surgery to control the leaks [44,50].

Post-LT bile duct stones, sludge, and casts pose a formidable challenge to both patients and their treating physician because they have multiple pathogeneses, comorbidities, and management complexities. A multidisciplinary approach is usually the only way to improve success rates and outcomes, and multiple interventional resources are usually required to manage individual patients.

Biliary stones and sludge can occur at any time after LT, with a reported incidence of 5%–10% [23]. Furthermore, biliary strictures have been associated with stones and may be observed in up to 90% of patients with bile duct stones [56]. In addition to biliary obstruction, kinking of the bile duct, mucosal damage, ischemia, infection, foreign bodies, cholesterol supersaturation, and bile acid pool depletion may also result in lithogenesis [57,58]. Medications used in transplant patients may play a role in this regard since cyclosporine inhibits bile acid synthesis, thereby promoting stone formation [59]. Elevated cholesterol levels (>200 mg/dL) and triglyceride levels (>150 mg/dL) lead to a marked increase in stone formation after LT, suggesting that hyperlipidemia should be aggressively treated in LT recipients [60]. Occasionally, severe complications such as cholangitis and secondary biliary cirrhosis may occur as a result of stones and sludge. Furthermore, biliary cast syndrome requires a special mention; this is a distinct condition that is characterized by multiple, hard,pigmented stone casts within the bile ducts. The condition has been observed in post-transplant recipients, and it affects morbidity, graft failure, re-transplantation, and mortality rates. The reported incidence is 2.5%–18% [25,61]. The pathogenesis of this condition is thought to be multifactorial: acute cellular rejection, ischemia, infection, and biliary obstruction due to stasis in patients with or without bile duct strictures are contributing factors. It is extremely important that physicians caring for patients with biliary cast syndrome remain vigilant, particularly because immunosuppressants and catastrophic late presentation tend to mask symptoms. As such, the syndrome often fails to manifest its typical clinical course and can lead to graft loss and mortality.

ERCP is the investigation of choice for most patients with post-LT bile duct stones, sludge, and casts. In this regard, several studies have reported success rates as high as 90%–100% for stone clearance and that most patients experience symptomatic, clinical, and liver biochemical improvements [44,62]. Indeed, in patients with non-altered anatomy, a standard sphincterotomy followed by extraction balloon or extraction basket clearance of stones and sludge can be achieved without any major limitations using the conventional side-viewing ERCP scopes [36]. In patients with an anatomy that has been altered by bilio-enteric anastomosis, the same can be achieved using modified longer enteroscopic versions. In cases of severe coagulopathy or thrombocytopenia, sphincterotomy may be contraindicated. Therefore, balloon dilation of the intact sphincter may be indicated. The removal of large bile duct stones and casts can be achieved using a combination of biliary sphincterotomy and large-balloon dilation of the biliary orifice. Such a combined procedure has improved efficacy and minimized complications [14]. If biliary stricture occurs with stones, it should be treated simultaneously (Fig. 5) [36].

Proximally located stones and associated tortuous and long multiple strictures have always been a major limitation of the endoscopic management of post-LT bile duct complications. When direct cholangioscopy previously permitted access to the biliary tree through the conventional route, it allowed the management of difficult or complex proximal pathological stones, sludge, and associated strictures. This method was combined with advanced intraductal techniques that break down large stones, such as electrohydraulic lithotripsy (EHL) or holmium laser lithotripsy [14]. Relatedly, since bile duct stones and sludge are associated with bile duct strictures, PTCS is considered the treatment of choice in patients for whom duct access is endoscopically difficult as well as in those with intractable strictures that cannot be traversed using a retrograde approach or rendezvous access combined with ERCP. In this regard, PTCS is especially successful in improving graft functioning by clearing sludge and stones, particularly when combined with basket and EHL stone fragmentation. Biliary stones can be removed using (1) mechanical basket lithotripsy through the PTBD tract or (2) the scope push method to the common bile duct (CBD). EHL can be used to treat large stones that are not successfully removed by the mechanical basket. In one recent study, 15 patients underwent PTCS for the following post-LT biliary complications: IHD stone (10 patients), CBD stone (one patient), biliary cast (one patient), and biliary stricture (three patients). The study found that the prevalence and severity of these complications in PTCS patients are significantly lower than those in patients who have undergone traditional radiological percutaneous transhepatic interventions. This obviates the need for additional radiological or surgical interventions [63]. PTCS and internal-external stents play a role in patients with multiple intrahepatic biliary strictures as well as in those with stones, sludge, and casts. Indeed, they may be extremely useful for relieving jaundice and improving the clinical condition of such patients [44]. The most common limitation of PTCS and percutaneous approaches is significant patient discomfort; furthermore, the procedure is inconvenient because it requires frequent percutaneous catheter changes to enable remodeling.

Hemobilia is an uncommon LT-related biliary complication with a frequency of 1.2% (33/2701) [64]. It is usually associated with percutaneous liver biopsy or PTCS through the creation of an arterio-biliary fistula. The pathogenesis of spontaneous hemobilia in the post-LT setting has not yet been elucidated. Nonetheless, blood clots in the bile duct can cause recurrent obstructive cholangitis and result in irreversible graft damage. The commonly described triad of right-upper-quadrant pain, jaundice, and gastrointestinal bleeding is only seen in a small proportion of patients. In LT patients, “United Network for Organ Sharing” statuses I and IIa, alcoholic liver cirrhosis, and a body mass index <24.5 kg/m² were significant risk factors for hemobilia [64]. In rare cases, hemobilia has been associated with biliary stricture and multiple IHD stones have formed above the obstruction level [65].

Treatment of hemobilia requires both hemostasis and the treatment of any associated biliary obstruction by clots. In some cases, the bleeding stops spontaneously after supportive therapy and the correction of coagulopathy. Alternatively, if the bleeding is persistent or severe, embolization of the bleeding vessel by interventional radiology is required [66,67]. Clinicians may prefer to manage spontaneous hemobilia using non-surgical intervention (via either the endoscopic transpapillary or percutaneous approach) in LT patients. Endoscopic sphincterotomy and blood clot extraction for relieving obstruction is usually performed via ERCP. Nasobiliary drainage (NBD) catheters may promote blood clot lysis, slow bleeding, and prevent cholangitis by irrigation. Therefore, ERCP combined with sphincterotomy and an NBD catheter should be considered a primary choice in cases of hemobilia (Fig. 6). In one study, endoscopic NBD was performed in all patients with hemobilia, and 87.9% (29/33) improved [64]. On a different note, several case reports have stated that CSEMSs promote endoscopic hemostasis in cases of bleeding that cannot be controlled using conventional therapy [68]. Indeed, CSEMSs may apply stronger pressure to the bile duct wall to achieve effective hemostasis; however, they may increase the risk of post-procedure pancreatitis. Recently, the placement of both an endoscopic biliary plastic stent (EBPS) and an NBD catheter has been shown to be an alternative hemostatic tool to SEMSs. The EBPS, combined with coagulation in the lumen, increases intrabiliary pressure. Therefore, biliary obstruction causes cholangitis. The combination of EBPS and an NBD catheter to prevent cholangitis is an alternative hemostatic tool to the CSEMS that reduces the risk of post-procedural pancreatitis [69].

In summary, post-LT hemobilia is not yet well defined as a disease entity, and most reported incidents are case studies only. Therefore, an additional large study is needed.

As mentioned above, PTCS has an advantage over ERCP in that it allows direct visualization of the biliary system for the identification and treatment of intraductal lesions [41]. Foreign body material that has not clearly been seen in image studies—such as non-absorbable suture material, which that can act as a nidus for IHD stones—can be removed using PTCS. We successfully used PTCS to remove foreign bodies from the bile ducts of post-LT patients with bile duct stones, strictures, and sludge (Fig. 7).