Accessory bile duct: a rare but important anatomical variant

Abstract

An 84-year-old man presented with pancreatic adenocarcinoma. Following neoadjuvant chemoradiation, the patient underwent a pancreaticoduodenectomy, complicated by early bile leak. Re-exploration and intraoperative cholangiogram identified an accessory common bile duct draining segment 5 of the right hepatic lobe, which was then ligated. The patient underwent a complicated postoperative course eventually developing sepsis secondary to biliary stasis. He elected for comfort measures and passed away secondary to complications of sepsis.

Background

Accessory bile ducts are a rare but recognised part of the literature in biliary surgery. Variations in biliary anatomy occur not infrequently. In this case, we present an 84-year-old man with pancreatic adenocarcinoma who underwent a pancreaticoduodenectomy complicated initially by bile leak and later sepsis secondary to bile stasis originating from a ligated accessory duct. This case emphasises the importance of recognising anatomic variants of biliary anatomy as their lack of recognition can lead to major morbidity in routine hepatobiliary and pancreatic surgical procedures.

Case presentation

The patient is an 84-year-old man with a medical history of myocardial infarction s/p coronary artery bypass graft, hypertension and diabetes mellitus, who presented with 5 weeks progressive epigastric abdominal pain radiating to the back. A CT scan of the abdomen at that time revealed a pancreatic head mass, which was further evaluated using endoscopic ultrasound. This demonstrated a 3.6 cm uncinate mass that invaded the portal vein. Needle aspiration of the mass was consistent with pancreatic andenocarcinoma. Further imaging demonstrated that the mass abutted the superior mesenteric artery, superior mesenteric vein and portal vein. At this time, his CA-19-9 was 54. There was no evidence of distant disease.

Treatment

During the procedure, a thin, tubular structure that was approximately 3 mm in diameter was encountered lateral to the common bile duct (CBD) within the hepatoduedenal ligament. At the time, it was felt to be a lymphatic vessel and was divided using a vessel sealing device. The procedure was carried on in the usual fashion. A flat Jackson-Pratt (JP) drain was left posterior to the hepaticojejunostomy.

Outcome and follow-up

Over the course of the next day, the patient remained comfortable with pain well controlled and haemodynamic parameters within normal limits. However, JP drainage during this time transitioned from serous to frank bilious, and the decision was made to take the patient back to the operating room for re-exploration. On entry into the abdomen, bilious fluid was encountered, and the abdomen was irrigated. No obvious defect in the hepaticojejunostomy was found. During our investigation, a small accessory duct, separate from the common hepatic duct, was noted. Intraoperative cholangiogram of this aberrant duct demonstrated drainage of mainly segment V of the liver (figure 1). The duct was friable and would not hold sutures. It was ligated, and a flat JP drain was left in place. The JP drain remained with minimal serous output and a low fluid amylase.

Figure 1.

Intraoperative cholangiogram after contrast injection demonstrating: right-sided biliary drainage after injecting the common hepatic duct (A), and distinct segmental biliary drainage after injecting the ligated accessory bile duct (B). (C) Intraoperative photograph demonstrating the accessory duct (superior probe).

The patient continued to improve, and the drain was removed on postoperative day 5. The patient’s postoperative course was then complicated by Staphylococcus epidermidis bacteraemia. He was successfully treated with vancomycin and discharged to rehabilitation on postoperative day 12.

Two weeks after discharge, he presented to the office for a follow-up appointment. During this visit, he was identified as having poor wound healing secondary to low nutritional status. He was readmitted and underwent an abdominal wound exploration and washout and was placed on supplemental parenteral nutrition. During this hospitalisation, he began to develop fevers and leucocytosis. Workup including cultures revealed enteric flora (bacteraemia), and broad-spectrum antibiotics were started. A CT scan of the abdomen demonstrated an abnormality in segment V that was phlegmonous and partially fluid filled consistent with an abscess. A CT-guided percutaneous drain was placed from which cultures were consistent with the resulted blood cultures. Antibiotics were continued based on infectious disease recommendations. Despite these efforts, the patient’s clinical condition continued to deteriorate. After a long discussion with the family, the patient chose to transition to comfort care and eventually passed away from cardiopulmonary arrest secondary to uncontrolled sepsis.

Discussion

An accessory CBD is an exceedingly rare congenital anomaly of the biliary system in which two separate channels drain bile from the liver in a parallel fashion. The incidence of this anatomical variation is not clear, <50 cases are reported in the literature, most of which originate in Asian populations.¹ Because of its relative rarity, this anatomic variation is difficult to recognise as a source of biliary leak following surgery.

It is thought that this duplication of the biliary tree is the result of non-regression of fetal anatomy.² In the third week of gestation, the biliary tree contains separate drainage for the right and left lobes of the fetal liver. It is believed that at a later stage of development the bile ducts fuse to form one.³ Failure of this process to complete can result in an accessory bile duct which may drain an isolated portion of the liver or result in a branch from intrahepatic ducts which communicate with the CBD.²

Indeed, this embryological origin is consistent with many of the variations described by the classification scheme first proposed by Goor in 1972.⁴ A modification of this classification scheme exists with five possible variants depending on the origin of the accessory duct and the location of the distal drainage.¹ Type 1 is a large CBD with two channels divided by a septum distal to the entry of the cystic duct. Type 2 is a large CBD, which branches into two separate channels distal to the cystic duct. Type 3a describes a parallel duct separate from the CBD, which drains an isolated part of the liver. Type 3b also describes a parallel duct, which originates as a branch of the right or left hepatic duct. Type 4 is a branch of the left hepatic duct, which has connections with the CBD and its own separate drainage into the target organ.

Typically, one duct will drain into the ampulla while the other duct may fuse at the ampulla or have a separate drainage location.⁵ Variations have been described with the accessory duct draining into the stomach as well as into the duodenum and pancreas.^{1 6 7} The anatomical drainage of the accessory duct has been associated with different types of cancers.¹ For example, accessory ducts that drain into the stomach have been associated with cancers of gastric origin, while ducts draining into the duodenum or pancreas are commonly associated with biliary-related neoplasms.¹

These variations in anatomy are significant also for their role in bile duct injury during cholecystectomy. Injuries to these ducts have been documented and are associated with early bile leak following cholecystectomy.⁸ Additionally, the anatomy of the biliary tree must be evaluated for these aberrant ducts to prevent complications associated with liver transplantation.⁹

We suspect that our patient developed sepsis secondary to cholestasis with a biliary-enteric anastomosis. This was likely the result of ligating an accessory bile duct, which was the primary drainage for an isolated hepatic segment. The case also highlights the challenges of performing pancreaticoduodenectomy in octogenarians as by the time it was determined that his sepsis could not be controlled with percutaneous catheter drainage, the patient was too ill to be able to survive an attempt to salvage him by resecting the involved liver segment. We hope that this case may demonstrate the important role that these accessory ducts play in biliary anatomy and that understanding their unique anatomy is critical to preventing significant morbidity and mortality in patients undergoing surgery involving the biliary tract. When an accessory duct needs to be divided, or it is divided incidentally, an attempt should be made to reconstruct it, particularly when a biliary-enteric anastomosis is made to the common hepatic duct which seeds the intrahepatic biliary tree with bacteria.

Learning points.

• Accessory common bile duct is a rare but documented variant in biliary anatomy, which must be considered when performing pancreaticoduodenectomy.

• When biliary leak is identified post operatively, aberrant biliary anatomy should be considered in the differential.

• Variations in biliary anatomy may be associated with malignancy and should be considered in surgical planning.