A novel anastomotic technique reducing biliary complications in small-duct reconstruction during adult living donor liver transplantation

Adult living donor liver transplantation (LDLT) has become a key surgical intervention for rescuing patients with end-stage liver disease, especially in the setting of a severe shortage of deceased donor organs (1-3). While refinements in surgical technique and perioperative management have continued to improve patient prognosis and graft survival in LDLT, biliary complications remain its “Achilles’ heel”, significantly contributing to postoperative morbidity and impaired long-term outcomes (4,5). The reported incidence of biliary complications following LDLT ranges from 15% to 30%, markedly higher than that observed in deceased donor liver transplantation (10–15%) (6). Bile leak (BL) and biliary anastomotic stricture (BAS) represent the two most common complications (7). Several risk factors contribute to the high rate of biliary complications in LDLT, including small bile duct diameter, thin ductal walls, mucosal injury, and impaired blood supply at the anastomotic site (8). The two primary methods of biliary reconstruction are duct-to-duct (DD) anastomosis and Roux-en-Y hepaticojejunostomy (RYHJ) (8). While DD anastomosis is generally preferred due to its physiological advantages—such as preservation of sphincter of Oddi function, maintenance of entero-biliary continuity, and ease of postoperative endoscopic management (8,9). However, DD anastomosis significantly increases the incidence of biliary complications in small-sized bile duct reconstruction (<4 mm) (7,10,11). Recent multicenter data suggest that RYHJ may be a more suitable option in such scenarios (12).

In an effort to retain the benefits of DD reconstruction while addressing the challenges posed by small-sized ducts (<4 mm), we developed a modified technique termed mucosa-to-hilar plate anastomosis (MHA). This technique aims to mitigate the technical challenges and ischemic hazards associated with conventional small-duct anastomosis.

In this study, a total of 24 patients diagnosed with decompensated cirrhosis accompanied by liver failure or primary liver malignancy underwent LDLT at the Liver Transplantation Center of West China Hospital, Sichuan University, between January 2021 and December 2024. Their donor bile duct diameter was less than 4 mm, and MHA anastomosis was performed during bile duct reconstruction. All living donors underwent comprehensive screening to exclude systemic contraindications and provided written informed consent in accordance with institutional and national guidelines. In addition to routine preoperative assessment, magnetic resonance cholangiopancreatography (MRCP) was performed to evaluate the diameter, anatomical configuration and transection level of the donor intrahepatic bile ducts. The MHA technique was indicated when the bile duct diameter was less than 4 mm. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was reviewed and approved by the Ethics Committees of West China Hospital of Sichuan University (Ethics ID: 2020-385). Written informed consent was not required because the study was retrospective analysis.

The MHA procedure was performed as follows (Figure 1). During donor hepatectomy, 10–12 hours before surgery, intravenous injection of indocyanine green (dose: 0.1 mg/kg) is administered to the donor to ensure intraoperative biliary imaging and determination of transection level, while minimizing periductal dissection to preserve bile duct vascularization. In the recipient, the left and right hepatic ducts were carefully separated. The common hepatic duct was transected distally to the confluence, and the recipient’s duct was fashioned into a trumpet-shaped opening. Using microsurgical techniques, the recipient duct was interruptedly anastomosed to the hilar plate tissue surrounding the donor bile duct with 5-0 Prolene sutures, avoiding direct suture engagement of the donor duct mucosa. The main principle of arterial reconstruction is anatomical reconstruction first and avoidance of grafts with complex arterial variations. For multiple hepatic arteries in grafts, all arteries are reconstructed as much as possible. Subsequently, intraoperative Doppler ultrasonography was performed to confirm the intra-graft arterial flow. The 8-0 Prolene sutures were used for arterial reconstruction by continuous anastomosis. After surgery, three plasma drainage tubes were placed in the liver segment, Winslow’s foramen, and posterior to the liver to monitor the volume and bilirubin in the drainage fluid.

Figure 1.

A schematic model of duct-to-hepatic hilar plate anastomosis. (A) In the recipient, the left and right hepatic ducts were carefully separated. The common hepatic duct was transected distally to the confluence, and the recipient’s duct was fashioned into a trumpet-shaped opening. (B) Using microsurgical techniques, the recipient duct was interruptedly anastomosed to the hilar plate tissue surrounding the donor bile duct with 5-0 Prolene sutures, avoiding direct suture engagement of the donor duct mucosa.

All patients received an immunosuppressive regimen based on tacrolimus combined with mycophenolate mofetil. Postoperative follow-up included regular clinical and laboratory evaluations, including monitoring tacrolimus blood concentration, liver and kidney function, and liver transplant specialist Doppler ultrasonography, with once a week in the first month of discharge, once a month to half a year and once every 3 months. BL was defined as the presence of biliary leakage on postoperative cholangiography, persistent bile-stained drainage from the abdominal tube, or biochemical confirmation of bile in abdominal fluid. BAS was suspected in cases of elevated liver enzymes and/or bilirubin, and confirmed using abdominal ultrasound, MRCP, or direct cholangiography [percutaneous transhepatic cholangiography (PTC)/endoscopic retrograde cholangiopancreatography (ERCP)] when indicated.

The baseline clinical characteristics of the 24 patients underwent LDLT with the MHA technique are summarized in Table 1. The cohort included 16 males and 8 females, with a mean age of 46 years (range, 27–61 years). The underlying liver disease was hepatitis B virus-related in 23 patients and autoimmune hepatitis in one patient. According to the Child-Pugh classification, 8 patients were class A, 12 were class B, and 4 were class C, with a mean model for end-stage liver disease (MELD) score of 12.5. Eleven patients had concomitant primary liver cancer and received pretransplant treatments including transarterial chemoembolization and targeted therapy. Preoperative enhanced computed tomography (CT) and positron emission tomography (PET)-CT confirmed the absence of major vascular invasion or extrahepatic metastasis in these patients.

Table 1. The clinical characteristics of patients who underwent LDLT with the MHA.

Characteristics	N=24
Donors and grafts
Age (years)	30 (19.0–57.0)
Graft type
Right hemiliver graft	11
Left hemiliver graft	13
Actual graft weight (g)	590.3 (355.0–1,050.7)
Diameter of graft’s bile duct (mm)	2.11 (1.3–3.2)
Number of graft’s duct (1/2/3)	23/1/0
Cold ischemic time (min)	80 (43.0–107.0)
Warm ischemic time (min)	40 (23.0–47.0)
Macro-Fatty change (>10%)	1
Recipients
Male sex	16 (66.7)
Age (years)	46 (27.0–61.0)
Underlying liver disease
HBV	23
Autoimmune hepatitis	1
Presence of hepatocellular carcinoma	11
Graft-recipient weight ratio (%)	1.06 (0.61–1.42)
Child-Pugh score
A	8
B	12
C	4
MELD score	12.5 (4.0–23.0)
ABO-identical/compatible	24
Number of MHA anastomosis (1/2/3)	24/0/0
Total operation time (min)	502.5 (354.0–632.0)
Operative p-RBC transfusion (unit)	6.5 (0–17.5)
Rejection	2
Follow-up periods (months)	27 (11.0–41.0)

Data are presented as number (%) or median (interquartile range). HBV, hepatitis B virus; LDLT, living donor liver transplantation; MELD, model for end-stage liver disease; MHA, mucosa-to-hilar plate anastomosis; p-RBC, packed red blood cell.

All patients received a single-duct reconstruction, without the use of transhepatic biliary drainage tube or biliary stent. Among them, 13 patients received left hemiliver grafts and 11 patients received right hemiliver grafts. The mean bile duct diameter of the graft was 2.11 mm, and the graft-to-recipient weight ratio (GRWR) ranged from 0.61% to 1.42%. The postoperative follow-up period ranged from 11 to 41 months, with 27 months of median.

During follow-up, one patient (4.2%) developed a BAS, which was initially suggested by ultrasonography and subsequently confirmed by MRCP. Two patients experienced acute rejection episodes, which were successfully controlled with corticosteroid pulse therapy. One patient had a recurrence of hepatocellular carcinoma after transplantation.

Discussion

Biliary complications remain the “Achilles’ heel” of liver transplantation, with a notably higher incidence in LDLT compared to deceased donor transplantation. Despite continued refinements in surgical technique in LDLT (8,13), lacking the specialized reports exploring the anastomosis techniques to reduce postoperative biliary complications in small-duct reconstruction (diameter <4 mm). In this study, we introduce an MHA technique designed to address the particular challenges of biliary reconstruction in LDLT. Our preliminary experience in 24 patients demonstrates that MHA is associated with a low rate of biliary complications (4.2% vs. 15–30%), suggesting its potential value in cases involving small-caliber ducts.

The persistent high rate of biliary complications—particularly in small ducts—underscores the need for continued technical innovation. Numerous strategies have been explored in LDLT, including microsurgical reconstruction, stent insertion, and meticulous anatomical dissection (8). Nevertheless, factors such as limited duct diameter, mucosal fragility, and tenuous blood supply continue to pose significant challenges (8). The DD anastomosis on the posterior wall and Glissonian sheath to duct anastomosis on anterior wall used in LDLT reduced BAS (14). In this study, the MHA technique was developed to systematically mitigate these risks by minimizing dissection around the donor bile duct, preserving peri-biliary vasculature, and avoiding mucosal trauma through a novel anastomotic approach that attaches the recipient duct to the hilar plate tissue.

Accumulated evidence suggests that type of living donor liver grafts does not increase the incidence of biliary complications (15). Bile duct diameter is a well-established risk factor for biliary complications, with a significant increase in risk observed when the duct measures less than 4 mm (10). While DD anastomosis is generally preferred for its physiological benefits, RYHJ is often recommended for very small ducts due to the technical limitations of conventional DD reconstruction (12). The MHA technique offers a promising alternative that retains the advantages of DD anastomosis—such as sphincter preservation and endoscopic accessibility—while reducing the technical difficulty and ischemic risk associated with small duct reconstruction. In our series, MHA was not associated with an increased risk of biliary stricture or leak, supporting its safety and feasibility in selected cases.

Adequate blood supply is critical to biliary healing (16). The peribiliary vascular plexus, which is easily compromised during dissection, is essential for anastomotic viability (17). By anastomosing the recipient duct to the hilar plate tissue and avoiding direct suturing of the donor duct mucosa, the MHA technique minimizes devascularization and mechanical injury, potentially reducing the risk of ischemic strictures and leaks, suggesting that this anastomotic concept might be applied to RYHJ.

This study has several limitations, including its single-center design, small sample size, and relatively short follow-up period. In addition, the MHA technique in its current form is applicable primarily to single-duct reconstructions. Despite these limitations, our early outcomes are encouraging and justify further evaluation in larger, prospective studies and in comparison with conventional approaches.

In conclusion, the MHA technique represents a technically feasible and potentially useful option for biliary reconstruction in LDLT, particularly in cases involving small-sized bile ducts. By emphasizing mucosal and vascular preservation, this approach may help reduce the incidence of biliary complications without sacrificing the benefits of DD reconstruction.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was reviewed and approved by the Ethics Committees of West China Hospital of Sichuan University (Ethics ID: 2020-385). Written informed consent was not required because the study was retrospective analysis.