Considering the multifarious descriptions in the medical literature, we found RHV tributaries particularly challenging to analyze. We determined the hepatic venous confluence as proposed by De Cecchis et al. We thought it was clinically important because a distal confluence may facilitate parenchymal-sparing liver resection because venous outflow can be preserved in the main trunk plus one of the two tributaries. There was less opportunity for parenchymal-sparing resections in our population because a distal confluence was less prevalent (39% vs 60%) than it was in the European population studied by De Cecchis et al.
A well-defined segment VII tributary was fairly consistent (96%) in our population. Thas venous outflow pattern, especially when combined with a distal confluence, may facilitate anterior right sectionectomy. The rationale is that it preserves venous outflow from the future liver remnant (FLR). Venous congestion has been shown to have a deleterious effect on the FLR[10-13]. Kawaguchi et al demonstrated that when they used intraoperative fluorescence during liver resections to demonstrate significant reduction in indocyanine green uptake in veno-occluded FLRs. Reduced perfusion leads to attenuated hepatocyte function and impaired FLR regeneration. For that reason, most authorities recommend venous reconstruction when congested FLRs develop intraoperatively[11-14].
In the model described by Hjortsjo, segment VIII is divided into dorsal and ventral regions draining separately into the RHV and middle hepatic veins, respectively. They drain large parenchymal territories, with 21% of the entire liver volume drained by the dorsal vein of segment VIII and 16% drained by the ventral vein. That was the dominant pattern in our population, with the dorsal vein of segment VIII draining into the SRHV in 72% of cases. In 28% of cases there was anomalous drainage from segment VIII, where both dorsal and ventral segment VIII veins emptied into the middle hepatic vein. Considering the fact that those two veins combined drain an estimated 37% to 56% of right liver volume, it is easy to appreciate the relevance of this anomaly. Patients with this unsuspected variant who undergo an anatomic left hepatectomy would have venous congestion in segment VIII, compromising an additional 37% FLR. In these patients, the impaired FLR function could increase operative morbidity and mortality. This variant is also important in transplantation because graft dysfunction can result from venous occlusion if the segment VIII veins are not reconstructed[3,12,13]. Therefore, we routinely reconstruct any outflow tract from segments V or VIII that is larger than 5 mm during right lobe living donation. That also lends support to the recommendation by Kawaguchi et al for routine intraoperative ICG fluorescence to achieve real time evaluation and accurate FLR estimation during liver transplantation. Kawaguchi et al demonstrated that the portal uptake function in veno-occluded regions was only 40% of that in nonoccluded regions.
Variants at the HCJ
The most common pattern was a single SRHV with no tributaries within 1 cm of the IVC (Nakamura and Tsuzuki type I). We found that the prevalence in our population was comparable to that in other countries across the globe[4,6,16-19], as shown in Table 1. Although relatively wide in diameter (20.4 mm), a type I SRHV is easier to expose and control outside the liver intraoperatively. It also facilitates the hanging maneuver, where an instrument is passed along the avascular space over the retrohepatic IVC and guided between the right and middle hepatic veins. A longer SRHV allows more separation between the liver and the HCJ, making it is easier for the surgeon to navigate the instrument between the hepatic veins. That maneuver can facilitate liver resections by reducing bleeding during parenchymal transection and guide the transection line. A longer extrahepatic vein makes right lobe donation and recipient implantation technically easier.
Only 4% of individuals had the dangerous Nakamura and Tsuzuki type IV anomaly in which there are two SRHVs emptying independently into IVC. It is technically difficult to control those veins during hepatectomy, and there is increased risk of inadvertently tearing the large SRHVs if their presence is not anticipated. In such cases, there is the potential for excessive hemorrhage during liver resection.
In our population, 60 individuals (51%) had a solitary SRHV without accessory hepatic veins, which was significantly lower than the global incidence of a solitary SRHV (71.8%). Accessory RHVs were present in 49% of unselected individuals in our population. Globally, the prevalence of accessory RHVs ranges from 4% in India to 100% in Scotland. Comparisons were difficult because authors have applied many different names to these vessels, such as retrohepatic veins, accessory RHVs, paracaval veins, inferior accessory veins, segment VI accessory veins, posterior or posteroinferior veins[4,19], middle or lower accessory and supernumerary RHVs. In an attempt to analyze the existing variations, we retrospectively examined data, descriptions, and images from existing publications and attempted to conform to the defined nomenclature of IRHV and MRHV[2,7,8]. The results are shown in Table 2.
Consistently, an IRHV is the most common variant reported in the medical literature (Table 2) and it was significantly more common in our population than the global IRHV prevalence (45% vs 29.2%). In our population, the IRHVs were short and wide (8.1 mm mean diameter). Although CT volumetric analysis was not performed in our study, some authors have reported the IRHV is responsible for up to 20% of the venous drainage from the right liver[2,39]. When they are not anticipated and properly controlled, inadvertent avulsion or lacerations may cause significant bleeding during right hepatectomies.
The presence of an IRHV is important in pretransplantation evaluations, especially for adult recipients, often receive right lobe grafts in order to ensure sufficient FLR. Those with large venous outflow territories would need to be re-implanted to prevent venous outflow obstruction and subsequent parenchymal congestion that would threaten the graft. The IRHV anatomic pattern is also important because re-implantation into the IVC becomes more difficult technically as the distance from the HCJ increases[2,8].
The presence of an IRHV may sometimes be beneficial, especially when they are the dominant drainage of segment VI. Tani et al documented that the IRHV drains an average of 70.8% of the venous blood from segment VI (100% in 32% of their cases). That is clinically important because it may allow tailored liver resections in which the FLR drainage is based on the IRHV preserving segment VI venous outflow. Makuuchi et al were the first to report a series of tailored resections of segments VII/VIII while sparing segments V/VI when they identified IRHV preoperatively.
In our population, all the IRHVs encountered were single vessels. Multiple IRHVs have been reported in the medical literature, ranging in incidence from 1% to 7.3%. Other complex patterns include multiple IRHV tributaries forming a short common trunk prior to joining the IVC[34,37] or IRHVs with anomalous drainage patterns. For example, Koc et al reported an IRHV accepting the adrenal veins to form a common trunk emptying into IVC in two of 1120 individuals.
Although an MRHV was less common in our population than in the global population (4% vs 5.5%), the differences did not achieve statistical significance (Table 2). When present, the MRHV was a short and wide channel (4.9 mm) draining directly into the retrohepatic IVC. Unlike the IRHV, which is relatively easy to control intraoperatively at the inferior border of the liver, we have found the MRHV to be more problematic because it is completely concealed behind the liver. When present, we consider MRHV proxies for increased technical difficulty during laparoscopic hepatectomies because they are difficult to reach and control with laparoscopic instruments. In cases where we proceeded with laparoscopic liver resections in the face of a MRHV, the informed consent procedure was adjusted to reflect the technical difficulty and increased risk profile. Fortuitously, MRHVs were only present in 4% of our patients.
Tani et al reported that the MRHV drained 8% of the venous blood from the right liver. Therefore, they may also need to be re-implanted at the time of liver transplantation[2,8,18]. Although we did not encounter any patients with coexistent MRHV and IRHVs, their coexistence has been reported[2,4,6,18,36,37,39] and it would increase the technical complexity of right liver transplantation.