Since the description of the Hepp and Couinaud biliary-enteric anastomosis using the extrahepatic left hepatic duct [14], HJ has remained the standard procedure in the surgical management of most postcholecystectomy bile duct injuries. In this setting, HJ anastomotic strictures nevertheless occur in 5–22 % [15–19] of the patients. The absence of bile duct dilatation has long been incriminated as the most prominent risk factor for the development of HJ stricture following bile duct injury repair [16], leading some authors to either postpone the intervention until bile duct dilatation was obtained or to routinely use transanastomotic stents on non-dilated bile ducts [20]. However, several studies have highlighted that early repair achieved similar results as delayed biliary reconstruction provided that the procedure was performed by a specialist hepatobiliary surgeon [21]. This clearly emphasizes the need for early diagnosis and referral to a specialized HPB unit. Other risk factors for the development of HJ stricture include the presence of biliary peritonitis at repair [22] and postoperative complications following the repair, especially biliary leakage [16], which both intuitively increase the risk of postoperative inflammatory stenosis. Finally, the impact of an associated right hepatic arterial injury in the occurrence of anastomotic stricture is still a matter of ongoing debate [15, 21, 22], with several arguments suggesting a role of arterial injury in favoring ischemia and retraction of the bile ducts and others supporting a rapid revascularization through the anastomosis. In our experience, routine CT scan with vascular reconstruction is always performed to preoperatively assess the arterial vascularization. Similarly, we believe that both operative evaluation of the biliary vascularization and confection of high anastomoses may help to prevent postoperative strictures. Finally, we consider that the existence of an associated vascular injury should probably lead to considering early repair with the utmost caution.

In deceased donor liver transplantation (LT), HJ is more and more restricted to a limited number of situations including large disparity in size between the recipient’s bile duct and the donor’s bile duct, liver retransplantation, LT for primary sclerosing cholangitis, and biliary atresia. In these situations, the incidence of HJ strictures ranges from 2 %, in the case of liver retransplantation [23], to 21 % in patients with PSC [24]. In this latter setting, it has been recently suggested that duct-to-duct anastomosis (DDA) provided better long-term functional results than HJ [24] without increasing the risk of disease recurrence [25, 26], supporting that it should be probably preferred over HJ. In living donor LT, no randomized study has yet documented the superiority of DDA over HJ. Hence, HJ is still performed in 20–40 % of the cases [27, 28]. In this latter setting, biliary anastomotic strictures represent the Achilles heel of these procedures with reported rates ranging from 6 to 22 % [29, 30]. The occurrence of HJ strictures may be the consequence of: (1) impaired graft quality as evidenced by increased rates of HJ strictures with significant steatotic grafts [31, 32] with prolonged cold ischemia time [33] or grafts from donors aged > 50 years [30]; (2) technical factors including biliary anatomical variations [31] requiring > 1 biliary anastomosis [30] and small donor right or left bile ducts; [31]; and (3) postoperative complications, mainly biliary leakage [30, 34], hepatic artery thrombosis [35], CMV infection [35], and acute cellular rejection [33]. Obviously, prevention of HJ strictures in patients undergoing LDLT may essentially be achieved by improving the selection of the grafts with the systematic use of preoperative donor liver biopsy but also with refinements in surgical technique. In this latter setting, Lin et al. have emphasized the value of routine microsurgical biliary reconstruction in decreasing the number of anastomotic strictures regardless of both types and number of ducts [27].

Biliary complications following pancreatic head resection are often neglected and overlooked by those involving the pancreatic anastomosis. Hence, only one study has to date specifically examined the incidence of biliary strictures after pancreaticoduodenectomy (PD) [3]. In this large single-center study analyzing 1595 patients undergoing PD over 8 years, 42 (2.6 %) patients experienced HJ stricture and median time for stricture occurrence was 13 (median: 1-98) months. No significant risk factor for the development of strictures was observed with only marginal influence of preoperative biliary drainage and no impact of either common bile duct size or postoperative biliary leakage. Interestingly, the rates of strictures were also strictly similar in patients operated on for benign and malignant disease. In this latter context, less than 10 % of the patients were found to have recurrent neoplastic disease involving the bilioenteric anastomosis. Of these, none were operated on for pancreatic or ampullary carcinoma supporting that development of a biliary stricture in these patients is usually benign. However, all patients with malignant anastomotic strictures carried a diagnosis of cholangiocarcinoma. This result, which is in line with the reported 16 % rate of tumor recurrence at the proximal stump in patients operated on for extrahepatic cholangiocarcinoma [36], suggests that anastomotic tumor recurrence should be systematically ruled out in this subset of patients.

In the long-term follow-up of patients undergoing choledochal cyst excision, the development of postoperative HJ anastomotic stricture widely varies according to the age of the patient at the time of surgery. Indeed, the rates of HJ strictures range from virtually 0 to 6 % in children [37, 38] when the anastomosis is performed on the hepatic hilum, while it may reach up to 24 % in adults [39]. This finding is probably related to the fact that inflammation of the cyst wall is mild in children under 10 years of age and more severe in older children and adults, likely resulting from severe histological damage to the common hepatic duct used for a bilioenteric anastomosis [40]. Other risk factors for the development of HJ after choledochal cyst excision include shorter duration of symptoms [39], increased size of the cyst [39], and type IVa cysts [40] where inflammation is associated with histological damage of the common hepatic duct after HJ and may lead to severe scaring at the bilioenteric anastomosis [40]. Altogether, these results suggest that the balance between the risks of malignant transformation and the risks of invalidating symptoms following HJ stricture, especially in adults with type IVa cysts, should lead to cautious consideration of surgery on a case-by-case basis rather than on a systematic operative approach basis. In this situation, definition of a subgroup of patients at low risk of malignant transformation would probably allow avoiding unnecessary procedures at extremely high risk of postoperative complications.