Challenges of Duplicated Portal Vein in Elective Laparoscopic Cholecystectomy: A Case Report

Introduction

Duplication of the portal vein (PV) is an extremely uncommon congenital anomaly of the portal venous system, characterized by the presence of 2 separate PVs coursing toward the porta hepatis and dividing into distinct segmental branches. Literature on this anatomical aberration is limited, and due to its rare occurrence, it has been documented in only a few case reports worldwide. The origin of such aberrations stems from early developmental failures, which in case of the portal system, typically occur within the first 2 months of gestation [1]. The embryological maturation of the PV is closely linked to the fetal vitelline and umbilical veins, which altogether undergo a series of complex developmental stages, classically culminating in a formation of a single PV originating from the confluence of the superior mesenteric vein (SMV) and splenic vein (SV). In cases when the anastomotic network of the vitelline system fails to obliterate, a second PV can develop [1].

There are several presenting types of a duplicated PV, based on its site of origin, topography, and symptomatology. The stratification of the PV duplication can be broadly categorized into 3 types: intrahepatic, extrahepatic, and double sagittal [2]. The course of the second PV relative to other anatomical structures can be extremely variable, usually predominated by the variations in its pancreatic and duodenal course, as documented in the literature [2–5]. Because the underlying cause of PV duplication is attributed to the embryological failures, oftentimes other concomitant anatomical aberrations in these same individuals might arise [1,6–8]. The clinical manifestations that an accessory PV can pose are broad, depending on its course, ranging from an asymptomatic state to fatty infiltration of the liver, portal hypertension with subsequent gastrointestinal hemorrhage, or intestinal obstruction [1,4,9,10]. Clinical manifestations usually prompt the use of imaging studies, which can potentially aid in identifying the anatomical anomalies when carefully studied. In cases when the second PV does not exhibit any clinical signs, the structural aberrations, as described in this case report, can be left unnoticed or be found incidentally on imaging or intraoperatively.

A hepatic biloma is an extra-biliary collection of bile, which can develop following trauma, including surgical procedures. Frequent causes include abdominal trauma and iatrogenic injuries, predominantly during laparoscopic cholecystectomies, both of which can result in biliary leakage, mass effect on surrounding structures, and eventually infections, all of which can increase morbidity and mortality if not treated promptly. Treatment options include percutaneous drainage, endoscopic stent placement, and surgical intervention [11]. Other complications of laparoscopic cholecystectomy include bile duct injury, peritoneal abscess, bowel injury, and postoperative hemorrhage [12].

A duplicated PV due to its rarity and unusual nature, poses a significant challenge for abdominal and general surgeons. Preoperative imaging is essential to identify these anatomical abnormalities and potentially minimize the risk of complications associated with inadvertent intraoperative injury. This report describes the case of a 61-year-old woman with congenital duplication of the PV and elective laparoscopic cholecystectomy complicated by postoperative peritonitis cause by a biloma, requiring relaparotomy, which resulted in PV injury and mass-bleeding and abdominal packing, which then required further hepatobiliary expertise.

Case Report

In December 2022, a 61-year-old woman underwent elective laparoscopic cholecystectomy for gallstones at a regional hospital. The laparoscopic cholecystectomy was converted to laparotomy when the operating surgeon encountered the unfamiliar anatomical anomalies. Nevertheless, the procedure concluded with no reported complications, and the patient was discharged from the hospital 2 days postoperatively. Four days after hospital discharge, she was readmitted to the same unit, reporting right upper quadrant pain, fever, and general fatigue. On computed tomography (CT), a 150-cm biloma was detected, located in the gallbladder bed, superior to the right hepatic lobe, and proximal to the hepatoduodenal ligament, prompting laparotomy and bile collection drainage on the same day (Figure 1). On intraoperative inspection, the biliary injury was not identified. Several hours after laparotomy, she developed hypovolemic shock, due to massive abdominal hemorrhage, necessitating a third laparotomy. At that time, identification of the origin of the abdominal bleeding was not achieved, which led to abdominal packing with 20 sterile gauzes and immediate transfer to our department for further management.

On admission to our hospital, she was intubated and mechanically ventilated, and required amine pressor administration (noradrenaline). She had a blood pressure of 90/59 mmHg and presented with hypovolemia, anemia (hemoglobin 6.9 g/dL), elevated lactate dehydrogenase level of 2156 U/I, and alanine and aspartate aminotransferase levels of 1031 U/I and 1967 U/I, respectively, therefore requiring hospitalization in the Intensive Care Unit (ICU). On the CT scan, a complete portal flow obstruction was identified, which upon metabolic acidosis and a rising trend of liver enzymes, prompted us to proceed with an immediate (fourth) laparotomy. Following depacking, abdominal inspection identified a midgut malrotation, with an unusual duodenal anatomy lacking typical descending and transverse parts. Profuse venous bleeding adjacent to the hepatoduodenal ligament began immediately after removal of the gauzes. The Pringle maneuver was ineffective due to hemorrhage extending 3 to 4 cm inferior from the duodenal line, requiring the en-mass compression by hand of the hepatoduodenal ligament, pancreas, and duodenum with a partial Kocher maneuver, which was temporarily successful in subsiding the bleeding.

A 2-cm long injury to the atypical right portal vein (RPV) originating from the SMV below the splenomesenteric confluence and making way anteriorly to the duodenum was positioned alongside to the right of the common hepatic duct into the right liver segments, making way via a gate in the Rouvier sulcus. The spleno-mesenteric confluence itself was identified, forming the atypical left portal vein (LPV), entering the left liver lobe via the round ligament.

At this stage of the surgery, the patient was hemodynamically unstable and required the administration of the second vasopressor, adrenaline. Despite 2 amine pressors, intraoperative blood pressure was sustained at around 80/50 mmHg. Under these circumstances, it was decided to perform a minimal, but necessary, surgical repair of the ruptured anterior wall of the RPV. It was sutured with 5/0 Prolene suture. On intraoperative ultrasound examination, blood flow via the RPV was barely visible (v=10 cm/s), but neither the portal flow to the left liver segments through the LPV nor the arterial flow to both liver lobes were visualized. On further inspection, the gallbladder bed exhibited signs of inflammation, and a bile leakage was identified, extending from the bulky inflamed tissue and the cystic duct. This was subsequently confirmed with intraoperative cholangiography following catheterization of the cystic duct. Exact identification of the origin of the bile leak from liver segment 5 was not possible without dissectional exploration. Due to the severe hemodynamic instability of the patient, it was decided to not extend the scope of the surgery and to postpone the biliary repair until there was an improvement in the patient’s general condition.

In the ICU, the patient remained intubated, on continuous adrenaline and noradrenaline infusion until postoperative day 3.

A series of repetitive diagnostic ultrasound and CT scans were performed, revealing PV thrombosis 2 cm distal to the splenomesenteric venous confluence on postoperative day 2, which was subsequently treated with therapeutic doses of low-molecular-weight heparin. On CT, a progressive worsening of the PV narrowing was observed (Figure 2); therefore, she was qualified for an urgent liver transplantation. A 200-mL collection of bile was observed in one of the abdominal drains. The diagnostic imaging studies further revealed several other anatomical anomalies, including polysplenia, atypical small intestine and pancreatic location with midgut malrotation, and numerous unspecified vascular variations, as indicated in Video 1. During the subsequent days, her general condition improved, she was extubated on the fifth day after laparotomy, and was eventually discharged from the ICU on postoperative day 8. Following normalization of liver enzymes and stabilization of PV thrombosis, she was removed from the transplant waiting list. Improvement in the patient’s general condition allowed for management of the biliary fistula, starting with the diagnostic magnetic resonance cholangiopancreatography, which confirmed the location of the biliary fistula from a biliary duct in liver segment 5. The patient underwent endoscopic retrograde cholangiopancreatography (ERCP), which failed to identify the ampulla of Vater due to anatomical abnormalities of the duodenum. Hence, a percutaneous transhepatic cholangiography (PTC) was performed, accompanied by concomitant stenting. A secondary ERCP allowed for the external drainage conversion to internal placement of self-expanding metal stents through the rendez-vous technique (Figure 3).

In the control CT examination, hepatic hyperarteralization and compensational left lobe hypertrophy were observed. The patient was discharged on postoperative day 47 in good general condition, with normal laboratory test results. At the time of this report, she was in good condition, with satisfactory hepatic function. At the 1-year follow-up, she remained in good general condition (Table 1), with a single episode of fever and hyperbilirubinemia (bilirubin 1.8 mg/dL), which was treated successfully with antibiotics. A minor, asymptomatic postoperative 2-cm hernia was identified.

For the sake of this publication, a reconstruction of the portal venous system from the initial CT scans following the cholecystectomy was performed (Figure 4, Video 2), displaying the RPV and the main PV, without evidence of thrombosis requiring liver transplantation.

Discussion

This case underlines the critical importance of comprehensive preoperative imaging other than common ultrasound, to detect anatomical variations and mitigate associated intraoperative risk. Furthermore, it highlights that in case of biliary and vascular complications after cholecystectomy, patients’ condition should be stabilized, and they should be transferred as soon as possible to specialized centers in hepato-pancreatico-biliary care. In patients after abdominal packing, the limitation of time to consecutive surgery is of the utmost importance [13]. The risk of liver failure through PV thrombosis is highly probable and requires advanced surgical care, with the possibility of rescue transplantation if necessary. Our patient, although she developed portal thrombosis both in the RPV and LPV, which greatly limited blood flow through the liver, at the same time, it provided a period long enough to develop collaterals through which visceral and portal flow was sufficient and prevented liver failure. In another turn of events, if collateralization would not take place, acute liver failure through PV thrombosis would lead to the necessity of an urgent liver transplantation [14]. In the coming years, it is unknown whether the patient will develop secondary biliary cirrhosis, if episodes of cholangitis will reoccur [15].

The anatomical variations of the portal venous system are of significant interest to anatomists and surgeons, due to the wide range of possible morphological presentations, which have critical implications for surgical interventions. Portal anomalies frequently occur within hepato-pancreatico-biliary units, with studies reporting an incidence of up to 25% [10]. Despite extensive studies and numerous attempts to classify these aberrations, the literature continues to document novel and previously unreported variations, with some presumably remaining undiscovered. Duplication of the PV exemplifies this concept, given its limited incidence, as reflected by the scant number of case reports. The variability of the RPV in its course, symptoms, and points of origin can be extremely diverse, making it imperative to report such anatomical variants when encountered in clinical practice.

The genesis of any vascular anomaly typically originates from developmental failures early in the embryonic stage. The PV develops within week 4 to 12 of gestation in close association with the vitelline system and 2 umbilical veins. The vitelline system forms a pre-hepatic anastomotic network around the duodenum, consisting of cranial-ventral, dorsal, and caudalventral anastomoses, terminating in the sinus venosus. Over time, selective obliteration of the anastomotic networks occurs, typically resulting in the persistent patency of only the dorsal connection. Concurrently, the major part of the left vitelline vein undergoes involution. Ultimately, the non-obliterated distal left umbilical vein, proximal part of the right umbilical vein, and the dorsal anastomotic vitelline connection form the main PV, which arises from the confluence of the splenic vein (SV) and superior mesenteric vein (SMV), coursing posterior to the first part of the duodenum and draining blood from the visceral structures to the liver [1,8]. Developmental anomalies at the embryological level can result in deviations of the portal architecture from expected branching patterns. In the case of our patient, we propose that the persistence of the right umbilical vein could have contributed to the formation of the RPV, which connected with the SMV during the fetal period. However, the exact embryological explanation cannot be ascertained. The main PV was in a typical retro-pancreatic position, excluding the possibility of the atypical RPV originating from a failure of obliteration of the anastomotic vitel-line networks. One case report of a duplicated PV supports another hypothesis explaining the etiology of portal variations, conceptualized by the reverse rotation theory of the duodenum. This mechanism attempts to explain the course of the pre-pancreatic mesenteric PV and an accessory retro-pancreatic splenic PV in relation to the posterolateral common bile duct, presumed to be a consequence of abnormal clockwise duodenal rotation [3,11].

Recognizing PV duplication, despite its rarity, is crucial as early identification can prevent adverse sequelae and poor patient outcomes. Therefore, the topography of the portal venous system should be meticulously examined in every case. Several types of accessory PV relative to proximal structures have been described in the literature [2–5]. The main PV typically runs within the hepatoduodenal ligament and, when duplicated, can be accompanied by a second PV. The atypical RPV can also arise independently outside the ligament, forming a third inflow to the liver [16]. In cases in which both PVs are found within the ligament, the RPV can vary in its course, with possibilities of a retro-pancreatic or pre-pancreatic course, with pre-duodenal or post-duodenal locations [6,9,17].

In our case, the main PV was formed by the classical retro-pancreatic confluence of the SMV and SV. The atypical RPV originated separately as an individual branch of the SMV, running inside the hepatoduodenal mesentery, posterior to the duodenum and anterior to the pancreas, splitting into 2 branches to form a vein fenestration around the common bile duct, and finally entering the right hepatic lobe. Du et al reported the incidence of portal venous fenestration of the accessory PV, leading to choledochal stenosis and dilation, with thrombus formation at the confluence [5]. Unlike in their case, these complications did not occur in our patient. Another report described an accessory PV similar to our case, with a pre-pancreatic and post-duodenal course originating from the SMV but draining into hepatic segment II, attributed to the obliteration failure of the cranial part of the left vitelline vein [5]. Several authors also reported accessory PVs originating from the SV, with normal development of the main PV [4,9,18]. In 3 case reports, the main PV entered the liver through the porta hepatis, while the accessory PVs coursed outside it and directly drained into the hepatic parenchyma [4,17,17].

A duplicated PV can present a wide spectrum of clinical manifestations, making diagnosis challenging. If not asymptomatic, it can present with duodenal obstruction, inhomogeneous hepatic steatosis, or portal hypertension, potentially progressing to variceal bleeding and fatal hemorrhage [4,9]. Choi et al postulated that the presence of an additional hepatic venous flow, as seen in PV duplication, can be detected by a focal difference in hepatic fat accumulation, providing a useful differential diagnosis when observed on imaging [18]. Our patient did not exhibit clinical signs of a portal anomaly, nor were hepatic or hemodynamic abnormalities suspected. Despite concomitant anatomical abnormalities, such as midgut malrotation, no gastrointestinal symptoms were reported. This case underscores the importance of detailed anatomical knowledge of the portal venous system, particularly in patients without characteristic or relevant symptoms. Prioritizing imaging studies aids in identifying atypically located structures, and, in the evaluation of portal flow, preoperative ultrasonography constitutes a quick and easy assessment tool. In cases of PV duplication, ultrasonography should depict accessory venous flow within the liver hilum but is highly dependent on the experience of the examining physician. In our patient, preoperative vascular CT visualization was impeded due to abdominal packing, preventing thorough vascular evaluation.

Patients with similar portal anomalies are more likely to possess variations affecting other systems and organs, such as polysplenia, cardiac defects, situs inversus, or biliary atresia [10]. Ichikawa et al described a case of incomplete pre-duodenal PV duplication presenting with multiple congenital anomalies, including doubled inferior vena cava, splenic lobulation, and an accessory liver lobe [6]. Mimatsu et al suggested a correlation between anteriorly located PV relative to the duodenum and a predisposition to other congenital anomalies [7]. Although our case included both PVs in a post-duodenal location, additional anatomical anomalies were also observed, such as polysplenia, atypical small intestine and pancreatic location with midgut malrotation, and numerous unspecified vascular variations. Ziegler et al reported an incidental cadaveric encounter of duplicated PV with concurrent left renal vein duplication, although the underlying explanation for both variations remains unclear [8]. These cases highlight the critical importance of comprehensive preoperative imaging assessments, which can significantly influence patient management strategies and guide optimal treatment plans. Although the question still remains if it is justified to perform a pre-cholecystectomy CT examination, since these anatomical variations are so rare, the availability and cost of abdominal CT by far surpass those of ultrasound.

Conclusions

This case underlines the importance of comprehensive pre-operative imaging in identifying anatomical anomalies and mitigating potential complications. The findings prompt a reevaluation of whether ultrasonography alone is adequate for preoperative screening in elective cholecystectomy. Given the potential for complex vascular anomalies, a more robust imaging protocol may be warranted to optimize surgical outcomes and ensure patient safety.