Periampullary Carcinoma Complicated by a Transpancreatic Hepatomesenteric Trunk—a Case Report of an Extremely Rare Vascular Anomaly

Introduction

Vascular anomalies pertaining to the coeliac and hepatic arterial system are not uncommon, found in around 40% of individuals. Multiple arterial anomalies and aberrations have been described across literature. These anomalies assume importance during surgical interventions in organs supplied by these arteries and a detailed knowledge of such aberrations in the preoperative setting can go a long way in minimising inadvertent injuries and avoiding surgical catastrophes. We hereby present an extremely rare anomaly—a transpancreatic hepatomesenteric trunk supplying the liver discovered preoperatively and managed accordingly during a pancreaticoduodenectomy for a periampullary carcinoma.

Case Report

A 57-year-old female presented to our centre with complaints of generalised itching for 3 months. She also gave history of minimal loss of weight and loss of appetite during this period. There were also complaints of minimal yellowish discoloration of sclera since the past 1 month. She gave history of intermittent waxing and waning of symptoms. Examination revealed no jaundice and presence of scratch marks, predominantly over the trunk and arms. Abdominal examination did not reveal any palpable lumps.

An initial ultrasonogram (USG) of the abdomen was done which showed mild intrahepatic biliary radical dialatation with dilated common bile duct (CBD-13 mm) and main pancreatic duct (MPD-5 mm). Contrast-enhanced computed tomography of the abdomen showed a well-defined focal lesion in the periampullary region of size 14 × 12 mm with mild dilatation of intrahepatic biliary radicles, CBD and PD. No focal lesions were seen in the pancreas, and no adjacent nodes were noted. Multiple vascular anomalies were noticed. Right hepatic artery (RHA) was seen arising from the hepatomesenteric trunk (HMT) running behind the CBD (Figs. 1 and 2). Left hepatic artery was seen arising from the inferior pancreaticoduodenal artery (Figs. 3 and 4). Left gastric artery was seen arising directly from the aorta as a small twig (Fig. 5). Splenic artery was also seen arising directly from the aorta (Fig. 6). No definite coeliac trunk was visible. CA 19-9 was within normal limits. Side viewing endoscopy was done which showed a prominent ulcerated ampulla from which biopsies were taken. Biopsy confirmed adenocarcinoma with IHC showing CK 7 positivity and CK 20 and CDX2 negativity, confirming periampullary adenocarcinoma.

Fig. 1.

Right hepatic artery (RHA) arising from the hepatomesenteric trunk (HMT) running behind the CBD

Fig. 2.

Right hepatic artery (RHA) arising from the hepatomesenteric trunk (HMT) running behind the CBD

Fig. 3.

Left hepatic artery arising from the inferior pancreaticoduodenal artery

Fig. 4.

Left hepatic artery arising from the inferior pancreaticoduodenal artery

Fig. 5.

Left gastric artery arising directly from the aorta as a small twig

Fig. 6.

Splenic artery seen arising directly from the aorta

Liver function tests were within normal limits except for mild elevation of alkaline phosphatase (ALP-115). Treatment options were discussed with the patient and was planned to undergo a pancreaticoduodenectomy.

Diagnostic laparoscopy did not reveal any obvious evidence of intra-abdominal metastases. Hence, it was planned to proceed with the definite procedure.

On laparotomy, a 2 × 2 cm mass lesion was seen arising from the periampullary region. Multiple vascular anomalies were noted. Right hepatic artery was seen arising from the HMT, coursing posterior to the portal vein and was dissected and preserved along its course (Fig. 7). Left hepatic artery was arising from the inferior pancreaticoduodenal artery (IPDA) and then travelling within the pancreatic parenchyma adjacent to the tumor (Fig. 8). Left gastric artery and splenic artery were found to be originating individually from the aorta. Left hepatic artery dissection was attempted. But due to its close proximity to the primary tumor, fearing a suboptimal oncological clearance, it was divided with a provisional plan to reconstruct the LHA. Due to sufficient pulsatile backflow from the cut end of the left hepatic artery, anastomosis was deferred and the left hepatic artery twig was ligated (Fig. 9). Conventional pancreaticoduodenectomy was done with reconstruction. Rest of the procedure was uneventful.

Fig. 7.

Right hepatic artery arising from the HMT, coursing posterior to the portal vein and was dissected and preserved along its course

Fig. 8.

Left hepatic artery arising from the inferior pancreaticoduodenal artery (IPDA) and then travelling within the pancreatic parenchyma adjacent to the tumor

Fig. 9.

Anastomosis was deferred and the left hepatic artery twig was ligated

Immediate postoperative period was uneventful. Liver function tests normalised by day 3, and coagulation parameters were also normal. She was discharged on post-op day 10 consuming full oral diet.

On follow-up, patient was doing fine. Final histopathology revealed well to moderately differentiated adenocarcinoma arising from ampulla. Tumour measured 1 × 1.5 × 1 cm and was infiltrating deep muscularis propria. Lymphovascular emboli were seen. All margins were free. All lymph nodes were negative (0/10). Immunohistochemistry revealed CK-19 to be diffuse positive, favouring pancreato biliary type.

Discussion

Coeliac and hepatic arterial system has been studied to be often prone to vascular anomalies. It has been stated in studies that only in around 60% of individuals, the ‘normal’ coeliac arterial system has been found [1]. Typically, coeliac trunk arises from the abdominal aorta at the level of twelfth thoracic vertebrae. It then courses anteriorly in the lesser sac and at the upper border of the pancreas divides into three branches: left gastric artery (LGA), splenic artery, and common hepatic artery (CHA).

The normal hepatic arterial system has been stated to be present in around 52–80% of cases [2]. The normal anatomy of the hepatic artery is a CHA arising from the coeliac axis and coursing to the point where the gastroduodenal artery (GDA) arises, beyond which it becomes the proper hepatic artery (PHA).

The many deviations are said to be caused due to the congenital persistence of vitelline arteries. Even though there have been historic reports of such anomalies, the first systematic study was published in 1966 [2]. Miki et al. in 1973, stated that the primitive right (RHA) and left (LHA) hepatic arteries arise from the SMA and the LGA, respectively. This hypothesis could explain the development of a replaced/accessory RHA from the SMA and a replaced/accessory LHA from the LGA [3]. It is with the widespread onset of liver transplantations that the real-world distribution of various anomalies could be studied. In 1994, a group from University of California, Los Angeles (UCLA) reclassified hepatic arterial anatomy based on their work of a 1000 liver transplantations [1].

The term ‘anomalous’ or ‘aberrant’ are commonly used synonymously. They include both a replaced as well as an accessory vessel. A replaced vessel is defined when the liver receives its entire supply from the aberrant vessel, while the term ‘accessory’ is used when the liver receives added supply via the aberrant artery in addition to the normal hepatic supply.

The most common anomalies of the hepatic vasculature are the replaced right hepatic artery (11–21%) followed by the replaced left hepatic artery (3.8–10%) [4].

Of all the anomalies stated, the hepatomesenteric trunk is the rarest variety described forming < 1% of all cases, with only a handful of cases reported. Hepatomesenteric trunk is the common trunk arising from the aorta which then gives off the CHA and then continues as the SMA. In our case, the right hepatic artery originated from the proximal part of the HMT while the LHA originated from the inferior pancreaticoduodenal artery, coursing through the pancreatic parenchyma.

The path taken by the replaced CHA is off surgical importance.

Two main courses of the replaced CHA have been established:

1. The extra-parenchymal path (outside the pancreas head). CHA arises from the SMA and is moves towards the liver on the posterior surface of the pancreas head; in this case, the dissection of the CHA from the pancreas can be achieved relatively with ease

2. The intra-parenchymal path (transpancreatic). Here, the CHA arises from the SMA and is moves towards the liver hilum through the parenchyma of the pancreatic head; safeguarding the CHA can be an issue in this case. When not possible to safeguard, options include resection and anastomosis of the remaining ends of the CHA. This is the rarest variety possible [5].

The issue of arterial anomalies is crucial for many procedures including liver transplantation and resection, hepatic artery chemotherapy, gastrectomy, biliary reconstruction, and especially for pancreatoduodenectomy [6]. It takes importance while resection, as due caution has to be taken to prevent injury to the vessel to avoid post-op hepatic ischemia [7]. Injury to the vessel can also be detrimental to the vascularity of the biliary tree, as most of the blood supply of the biliary tree arrives from the right hepatic artery. Ischemia here can result in adverse events ranging from post-op biliary leak to long-term biliary stricture. Another matter of concern is the adjacent nature of the aberrant vessel to the pancreatic anastomosis, thereby causing chances for a post-op pseudoaneurysm and fatal haemorrhage, especially in the untoward event of a pancreatic leak [8]. Hence, having a sound knowledge of the anatomy and its variations is of paramount importance for ensuring a safe and effective surgery.

Pre-operative imaging can usually detect around 60–80% of most arterial anomalies in the pre-operative setting [9]. Cross-sectional imaging is the cornerstone of pre-operative imaging protocol in pancreatic and periampullary lesions, especially to assess local extent of lesion, presence of distant metastasis and the assessment of vascular anatomy. Detecting anomalous anatomy in the pre-op setting can help the surgeon in better planning of the surgery and avoiding inadvertent vascular injury during surgery. CT angiography has also been used for better delineation of vascular anatomy as a problem solving tool. Though not routinely used in the pre-op setting, it can help in assessing the coeliac and hepatic arterial anatomy and can help in enhanced planning, especially when complex vascular reconstruction is being anticipated.

When a replaced common hepatic artery is encountered intra-op, the origin must be identified, mostly being from the SMA [2]. During pancreaticoduodenectomy, ligation of the gastroduodenal artery should be delayed until the entire course of the CHA has been identified and dissected from the pancreatic parenchyma. The GDA can also be the sole arterial supply of the liver at times. The pancreatic neck should only be divided after securing the aberrant vessel, as failure to do so may increase risk of injury to the aberrant CHA.

In our case, we had an even rarer presentation, where the RHA originated from the proximal part of the HMT and coursed directly to the liver while the LHA originated from the IPDA coursing through the pancreatic parenchyma to the liver (Fig. 10). There was no definite CHA.

Fig. 10.

RHA originated from the proximal part of the HMT and coursed directly to the liver while the LHA originated from the IPDA coursing through the pancreatic parenchyma to the liver

Preserving the LHA was not possible in our case, due to fear of close margins with the tumor; hence, it was divided. But due to adequate backflow, via arterial supply from the dominant RHA, reconstruction was not needed. This may be the only ever such case reported with an extremely rare anomaly.

If adequate arterial supply is not obtained, anastomosis has to be done either between the cut ends or using a vascular graft [8, 10].

A conscious attempt to define the vascular anatomy has to be done during the initial steps of a pancreaticoduodenectomy. After a complete kocherization, the porta hepatis should be evaluated to determine the location of the arterial pulsation and to define the vascular anatomy [11]. Only after having a detailed picture of the vascular anatomy should resection proceed. Even though its known that anomalous arteries cause longer operative time, higher transfusion rate and more post-operative complications, better pre-op planning can decrease said complications [12].

Conclusion

The most crucial stage in the management of hepatic vascular anomalies is its recognition. Once discovered, the surgeon can take necessary measures to tackle it—including dissection or resection and anastomosis. Adequate pre-op imaging, planning and systematic surgical design can help in dealing successfully with these scenarios and also in avoiding adverse events.