Arterial Reconstruction Using the Right Gastroepiploic Artery in Living Donor Liver Transplantation: A Single-Center Experience

Introduction

Liver transplantation (LT) is considered the standard treatment for end-stage liver disease, and living donor liver transplantation (LDLT) is an accepted alternative to deceased donor liver transplantation (DDLT) in countries where the number of deceased donors is low. Hepatic artery (HA) reconstruction in LT is crucial because arterial complications such as HA stenosis, intimal dissection, and hepatic artery thrombosis (HAT) can lead to increased recipient morbidity and mortality [1]. In LDLT, the graft HAs are divided distally. Because of the small caliber, shorter stump, and fragile intima, HA reconstruction in LDLT is technically more difficult than that in DDLT [2,3]. To overcome these difficulties, microvascular techniques have been used since 1992 [4]. Recipient HAs are the first choice for arterial reconstruction based on anatomy; however, recipient HAs are not always suitable for arterial anastomosis, because of HA scarring from previous surgeries, intimal damage during hilar dissection, insufficient length, or HA recipient-donor graft size discrepancy. In such cases, the right gastroepiploic artery (RGEA) is preferred as an alternative for recipient HAs. The usefulness of gastric arteries, including the RGEA, has been reported [5–8]; however, postoperative outcomes are not well defined.

We describe our experience with, and the technique for, extra-anatomical arterial reconstruction using the RGEA, with the objective of analyzing patient outcomes.

Material and Methods

PATIENTS:

From August 1997 to December 2023, we performed 328 primary LDLTs. Among these, we excluded 2 patients who underwent extra-anatomical arterial reconstruction using a vascular graft and 2 patients who underwent portal vein arterialization because of intimal dissection of the donor HA. We analyzed the data of 324 patients with initial LDLT. Patients with re-LDLT were not included in this study. All transplantations were approved by the Ethics Committee of Nagasaki University Hospital and met the ethical standards of the 2000 Declaration of Helsinki and Declaration of Istanbul 2008. The Institutional Review Board of Nagasaki University Hospital approved the study (approval number: 20012022-2). Each patient provided informed consent to undergo the surgery. In adult-to-adult LDLT, we selected a left lobe graft with the middle hepatic vein when the ratio of the graft volume to the recipient’s standard liver volume was greater than 30%. The ratio was calculated based on the results of a volumetric study using computed tomography (CT). A right lobe graft is an alternative if the left lobe is not suitable for donation. When both the left and right liver grafts are not suitable for donation, we considered using a right posterior section graft, but only in selected patients.

SURGICAL TECHNIQUES IN HA RECONSTRUCTION:

We performed all arterial reconstructions microscopically using end-to-end anastomosis with interrupted sutures, usually with a nonabsorbable monofilament 8-0 suture. We preferred a backwall-first approach using small vascular clips to avoid vessel twisting during suturing [9]. When the hepatic graft had multiple arteries, the dominant artery was reconstructed first. We then carefully checked for back bleeding from other arteries and evaluated arterial blood supply in the graft using Doppler ultrasound. We attempted complete reconstruction in cases of insufficiency. Recipient’s HAs were the initial choice for arterial reconstruction; we used the RGEA if the recipient had no usable HAs, due to intimal damage, thrombosis, size mismatch, or insufficient length. The steps in RGEA preparation were as follows. After identifying the RGEA, we divided the arcade between the RGEA and the left gastroepiploic artery. Next, we clamped the distal end of the RGEA using vascular clips and dissected the RGEA from the greater omentum and greater curvature of the stomach with sufficient surrounding fatty tissues. The arterial branches toward the stomach were individually ligated and divided. After preparation, the RGEA was positioned at the liver hilum dorsal (left lobe graft) or ventral (right lobe graft) to the stomach to align the RGEA and the donor HA in a straight line (Figure 1A, 1B). Intraoperative Doppler ultrasound was performed to confirm adequate blood flow after anastomosis completion.

POSTOPERATIVE FOLLOW-UP:

We evaluated vascular patency and blood flow twice daily using Doppler ultrasound during the first 2 weeks after LT. We also performed a scheduled dynamic CT 7 days after LT, to confirm vascular patency and blood supply for the implanted graft. A diagnosis of vascular obstruction was initially made using ultrasound and confirmed using 3-dimensional CT angiography. We did not routinely prescribe postoperative anticoagulant therapy.

DEFINITION OF BILIARY STRICTURES:

Biliary anastomotic strictures were defined as those requiring endoscopic or percutaneous transhepatic intervention secondary to symptomatic dilatation of the intrahepatic bile ducts. Non-anastomotic biliary strictures were defined as single or multiple strictures of the intrahepatic bile ducts not related to the anastomotic region.

STATISTICAL ANALYSIS:

All statistical analyses were performed using IBM SPSS Statistics 29 software (IBM Corp, Armonk, NY, USA). The Mann-Whitney U test was used to analyze continuous data, and the Fisher exact test was used for categorical data. Overall survival was calculated using the Kaplan-Meier method, and the data were compared using the log-rank test. Statistical significance was set at P value <0.05.

Results

RGEA USE IN LDLT:

The RGEA was used as a primary option in 18 patients, with the following indications: recipient HA intimal dissection or weakness (n=11, 61.1%), recipient-donor HA size discrepancy (n=4, 22.2%), and inadequate length (n=3, 16.7%). In the RGEA group, the operative time was longer due to additional dissection and repeated arterial reconstruction (921 (852–1075) vs 802 (721–895) min; P<0.001). Sixteen patients experienced arterial complications after the primary LDLT (4.9%), including HAT in 11 patients (3.4%), bleeding from the arterial anastomosis in 3 patients (0.9%), and intimal dissection in 2 patients (0.6%). Of these, 3 patients with HAT and 1 patient with bleeding from the arterial anastomosis were from the RGEA group. The rate of arterial complications after primary LDLT using the RGEA was 22.2%, which was significantly higher than that in the HA group (3.9%). All arterial complications were evident during posttransplant hospitalization. There were no cases of arterial anastomotic stenosis. The frequency of biliary strictures in the RGEA group tended to be higher than that in the HA group (27.8% vs 17.6%, P=0.21). Biliary anastomotic leakage was observed in 11 patients (3.6%) in the HA group. The duration of posttransplant hospitalization tended to be longer in the RGEA group than in the HA group (57 [46–77] vs 47 [36–70] days; P=0.12; Table 1).

OUTCOMES AFTER LDLT WITH TYPES OF HA RECONSTRUCTION:

We compared the patient and graft survival with types of HA reconstruction. Cases with left lateral section graft and right posterior section graft were excluded because they were not in the RGEA group. Since the proportion of ABO-incompatible cases was comparable in both groups, and there were no cases of antibody-mediated rejection, these patients were not excluded. The 1-, 3-, and 5-year graft survival rates were 83.7%, 77.2%, and 70.1%, respectively, in the HA group and 83.3%,77.8%, and 77.8%, respectively, in the RGEA group (Figure 2A). The 1-, 3-, and 5-year patient survival rates were 84.7%, 78.6%, and 71.6%, respectively, in the HA group and 88.9%, 83.3%, and 83.3%, respectively, in the RGEA group (Figure 2B). Graft and patient survival rates were not significantly different between the groups.

RGEA USE IN HA RE-RECONSTRUCTION AFTER ARTERIAL COMPLICATIONS:

Among the 324 patients undergoing initial LDLT, 11 patients underwent HA re-reconstruction because of arterial complications: 8 with HAT, 2 with intimal dissection, and 1 with arterial bleeding (Table 2). The median time from the initial LDLT to HA re-reconstruction was 6 (3–9) days. The same artery was used in 4 patients (1 HA and 3 RGEAs). The RGEA was used for the first time in 7 patients because the HAs were not reusable. Arterial complications after HA re-reconstruction occurred in 4 patients (36.4%), and 3 of these patients with HAT died from graft failure. The remaining patient with anastomotic bleeding was treated with relaparotomy.

Discussion

Meticulous arterial reconstruction is essential for successful LT. Recipient HAs are the first choice for reconstruction; however, the quality of recipient HAs is sometimes unacceptable for arterial anastomosis, and alternative arteries are required. In DDLT, aortohepatic conduits, celiac trunk, and the splenic artery are alternatives [10,11]. In LDLT, donor HAs are narrow and short, and interpositional grafts cannot be easily obtained. Therefore, previous reports have indicated the advantages of using gastric arteries [5–9,12,13].

RGEA, 1 of the 4 main gastric arteries, has been used safely in coronary bypass surgery [14,15] and can be used without causing gastric symptoms [6]. Our results also showed the feasibility of extra-anatomical HA reconstruction using the RGEA. In primary LDLT, although the incidence of arterial complications was rather high in the RGEA group, the graft and patient outcomes were similar in the HA and RGEA groups. The rate of biliary strictures was comparable between the groups. We also successfully used the RGEA as an alternative for recipient HAs in 7 patients undergoing arterial re-reconstruction because of arterial complications. This was possible because the RGEA is away from the hepatic hilum and thus protected from severe adhesions, and sufficient length can be harvested from the greater curvature of the stomach [5,12].

HAT is considered a serious complication after LT, and in particular, early HAT occurring within 30 days after LT is associated with increased posttransplant morbidity and mortality [16]. Recent reports indicate an incidence of HAT after LDLT of 0.3% to 4.2% [3,17], and in our study, the incidence of HAT was 3.4% after the initial LDLT. Of the 3 patients experiencing HAT in the RGEA group, 2 died from graft failure even though HA re-reconstruction was performed. In another patient, interventional endovascular treatment via the celiac axis and common HA with intra-arterial thrombolysis was performed first, followed by HA re-reconstruction, which successfully rescued the patient without further complications. The availability of the endovascular approach is another advantage of using RGEA for extra-anatomical HA reconstruction.

An association between HA complications and biliary strictures has been reported [18], and a recent study indicated that extra-anatomical HA reconstruction was a risk factor for biliary strictures after LT [6], with a reported biliary strictures incidence of >50% in the extra-anatomical group. In our series, the incidence of biliary strictures in the RGEA group was 27.8%, which was less than that previously reported. This difference may be a consequence of the different reconstruction techniques used in our study, compared with that of the previous study, including the RGEA usage rate (100% vs 55%, respectively).

Our study has several limitations, including its retrospective design in a single center. Because posttransplant arterial complications were uncommon, the number of patients who underwent reconstruction using the RGEA was limited. All arterial anastomoses were performed by the same surgeon or under his supervision, which provided consistent quality and the same technical approach over time. However, this consistency meant that we could not compare outcomes in the RGEA group with outcomes in groups receiving other reconstruction methods, because we did not have sufficient experience using other methods.

We emphasize important points for successful arterial reconstruction using RGEAs. First, the RGEA should be harvested with sufficient length and surrounding fatty tissue. Damage to the artery during the dissection of small branches and excessive tension at the anastomotic site must be avoided. Second, it is better to position the RGEA in the liver hilum dorsal to the stomach in the left liver grafts. This better approximates the RGEA and donor HA, and aligns the vessels in the same direction. This approach also minimizes the gap in surgical depth when performing the anastomosis in a limited microscopic visual field. Although the RGEA is usually smaller than other donor HAs, simple dilatation of the RGEA stump with vascular forceps before anastomosis is sufficient in most cases, and we rarely obliquely cut the RGEA. Finally, during arterial reconstruction, a back-wall-first technique without rotating the anastomotic site may reduce the intimal injury [9].

Conclusions

RGEA can be harvested simply and safely with sufficient length, even in rescue surgeries after arterial complications. Despite the increased risk of arterial and biliary complications, arterial reconstruction with RGEA is a valid option when HA is not an appropriate option, as it provides graft outcomes comparable to those of HA reconstruction.