Anastomoses between the bile duct and jejunum are frequently performed in hepatobiliary and pancreatic procedures involving malignant diseases such as pancreatic cancer, cholangiocarcinoma, or palliative surgeries; and non-malignant conditions, such as bile duct injuries after cholecystectomy, choledochal cysts, or primary bile duct disease. . The purpose of this study was to describe the experience of a single HPB surgeon and an HPB surgery group with HJ for any condition in a cohort in Colombia, South America, postoperative complications, and long-term follow-up, as well as to establish potential risk factors between clinical conditions and postoperative outcomes.
From the multiple described techniques for bilioenteric reconstruction, hepaticojejunostomy is the most frequently performed procedure . Postoperative complications in these conditions can vary between 3–33% [16–20]. Biloma, anastomotic leak, stricture, and cholangitis are considered the most relevant. Anand et al. and Sultan et al. reported biloma/intra-abdominal collections at a rate between 2–22% respectively [16, 21], In our study, biloma was documented in 11.36% of the patients. In addition, anastomotic leakage could vary in a range between 14.6–19% according to reports in the literature. We observed this complication in 4.02% (n = 9) of the cases.
The incidence of cholangitis following HJ is truly unknown, however, according to Okabayasi et al. , cholangitis could be present in 7.7% of the patients, similar results to those observed in our population (8.5%, n = 17).
The preoperative diagnosis of malignancy shows increased morbidity and mortality rates after HJ. In our population, most malignant conditions were related to periampullary tumors, with a rate of bile leak, (5.16% n = 8), anastomotic stricture (1.29% n = 2) and mortality (5.29% n = 8) [14, 15]. Our data shows comparable rates to the ones described in the literature for morbidities. Antolovic et al.  reported an increased rate of bile leak in cases of oncologic conditions, and associated procedures such as liver resection, or pancreatoduodenectomy.
Bile leak is one of the most relevant complications after HJ, several studies highlight its incidence, and Anastomotic leak (AL) is mainly reported after PD in 2 to 8% of all patients. Several risk factors have been described, among which BMI appears to be consistently present . De Castro et al.  reported a statistical relationship between increased BMI (grade II obesity), increasing 11.32-fold the risk of AL. Supporting these results, this study shows a statistical relationship between AL and increased BMI, however, based in a minor cut-off value (> 27 kg/m2), this could be explained by differences in cultural and economic issues. [24, 26]
Long-term follow-up must be performed in all patients following HJ independent of the cause, owing to the possibility of anastomotic stricture. The study population had a median follow-up period of 54 months. In the literature larger studies were performed analyzing this complication; Dimou et al. and Tochi et al. [27, 28] reported a 12% stricture rate in 1,883 patients, and 84 patients respectively. Lombardo et al. , analyzed anastomotic strictures following the Whipple procedure due to benign pathologies and reported an 8% rate. Seeking risk factors for long-term complications such as anastomotic stricture, our data identified that patients who underwent HJ due to benign conditions such as bile duct injury and choledochal cyst have an increased risk of presenting with this complication, similar to those reported by Dimou et al. . Follow-up in this group of patients showed that in most of the cases, stricture was present 2 years after the procedure. In our population overall anastomotic stricture rate was 4.52% (n = 9); with an increased incidence in benign conditions (n = 7). This could be explained by the increased diameter of the bile duct in oncologic conditions. Moreover, due to poor prognosis in hepatobiliary tumors, these complications could not be present possibly because the patient died before the stricture may appear. [27, 30–31]
The limitations of our study include the retrospective nature of the study design and the increased number of patients with a malignant disease compared to benign conditions. However, our study achieved a long-term follow-up; and showed improved outcomes with a standardized technique with a multidisciplinary approach, and management of patients who underwent hepaticojejunostomy independent of the nature of the disease. In addition, we controlled the exposure in terms of the surgical technique, because in all the cases the procedure was performed by the same surgeon with a standardized technique.