Consecutive disruption of intrahepatic bile ducts after ABO-incompatible living-donor re-liver transplantation: A case report

Abstract

Introduction

Hyperacute rejection leading to hepatic necrosis or intrahepatic bile duct stricture in ABO incompatible living-donor liver transplant (ABO-i LDLT) has been reported many times. With the advent of rituximab, the incidence of these complications has decreased significantly. However, consecutive biliary disruption after ABO-i LDLT has rarely been reported.

Presentation of case

A female in her 50s with blood type A was admitted to our hospital for ABO-i LDLT due to failure of a graft (refractory ascites [Child-Pugh C(10), MELD 9]) that had been primarily transplanted 20 years ago from her ABO-identical father. Since the living donor was her husband with blood type B, rituximab was administered for ABO-i re-LDLT. After the LDLT, the patient recovered quickly despite bile leakage at the biliary anastomosis. Subsequently, the bile duct of the graft liver was serially disrupted with a bile lake, which required multiple instances of biliary drainage. A liver biopsy was performed and did not show any C4d staining on 195 post-transplant days. The patient ultimately developed sepsis due to cholangitis and expired at 11 months after the re-LDLT and finally C4d was positive on post-mortem biopsy.

Discussion

Advances in ABO-i LDLT, particularly with rituximab, have reduced complications, but consecutive bile duct disruption remains challenging. Despite positive donor-specific antibody, early rejection markers were absent, suggesting complex mechanisms of complication.

Conclusion

We herein report a rare case as an important observation that may aid in preventing and treating potentially fatal complications after ABO-i LDLT.

Highlights

• •Graft failure after ABO incompatible living donor liver transplantation is common.
• •Antibody-mediated rejection diagnosed by the donor-specific antibody, C4d deposition
• •Rituximab has reduced rejection rates, but biliary disruptions may still happen.
• •Bile duct disruption can occur without clear signs of antibody-mediated rejection.

1. Introduction

ABO incompatible living-donor liver transplantation (ABO-i LDLT) is especially important in Eastern countries, where the scarcity of donors for deceased-donor liver transplantation has adversely affected the treatment of end-stage liver disease [1]. However, even in Western countries, ABO-i LDLT has been an important approach for overcoming donor shortages [2]. Hyperacute rejection leading to hepatic necrosis or intrahepatic bile duct stricture (BDS) in ABO-i LDLT has been reported many times [3,4]. However, with the advent of rituximab, the incidence of these complications has decreased significantly. Consecutive biliary disruption after ABO-i LDLT has rarely been reported [5].

We herein report a rare case of consecutive biliary disruption after ABO-i LDLT with an updated literature review. The work has been reported in line with the SCARE criteria [6].

2. Presentation of case

A female in her 50s with blood type A was admitted to our hospital for re-LDLT because a graft failure of cryptogenic etiology (refractory ascites [Child-Pugh C(10), MELD score 9]) that had been primarily transplanted 20 years prior from her ABO-identical father (Male in his 60s, right lobe graft). Before the transplantation, the patient was thoroughly evaluated. The primary reason for the LDLT was acute liver failure that occurred during pregnancy, which progressed to coma. This severe and life-threatening condition required urgent intervention, making LDLT the most viable option to restore liver function and stabilize the patient, especially given the high risks posed by both the pregnancy and liver failure [7]. In March 2019, marked ascites were noted and punctured once a week, with approximately 6 L of fluid drained each time. In October 2021, her husband in his 40s offered partial liver donation.

On admission, the patient underwent a comprehensive series of investigations, including contrast-enhanced computed tomography (CECT) and needle liver biopsy but the obvious etiology of graft failure was not detected. She had a clear consciousness of vital stability. She had massive ascites and splenomegaly with collateral circulation (Fig. 1A). For elective re-LDLT, rituximab (375 mg/m² body surface area) was administered 2 weeks prior to the transplantation. Her anti-A antibody titer was as low as <×32; therefore, plasma exchange was not performed. The donor-specific antibody (DSA) test was positive with MFI > 1000, while the direct crossmatch test result was negative. The high positivity was observed only for DR8, DR9, D53 (MFI > 10,000). CD19 and CD20 lymphocytes were undetectable before the re-LDLT because of rituximab administration.

Fig. 1.

Computed tomography findings before re-LDLT and showing the explanted liver. A: The right lobe graft showed massive ascites. B: Macroscopic view of the explanted liver.

Re-LDLT was performed with a left lobe graft with a middle hepatic vein weighing 496 g (graft weight/recipient standard liver volume: 49.4 %). The donor's extended left lobe graft has a single artery and a single duct, with no detected anatomical variations. The vascular anastomosis was performed following standard protocols. After arterial anastomosis, intraoperative Doppler ultrasound showed a satisfactory arterial blood flow. Due to the recipient's poor bile duct condition after the first transplantation, the patient underwent hepaticojejunostomy for biliary reconstruction. The blood loss was 9765 g (Fig. 1B). The total operation time was 616 min, with a warm ischemic time of 59 min (anastomosis) and cold ischemia time of 59 min. The patient received immunosuppressive therapy following the protocol for ABO-i LDLT, comprising tacrolimus, mycophenolate mofetil, and steroids.

A CECT was performed on postoperative day (POD) 1, demonstrating optimal arterial and venous flow. On POD 11, the patient developed jaundice with elevated direct bilirubin, bile duct dilation, and signs of inflammation, leading to a percutaneous transhepatic cholangial drainage (PTCD) procedure for biliary drainage. By POD 34, as bilirubin levels improved, the PTCD was converted to internal drainage. However, the patient subsequently developed a fever, and biliary scintigraphy indicated a possible anastomotic leak. Re-do of the hepaticojejunostomy was performed and bile leak was stabilized. However, subsequently without evident stenosis at the anastomosis, the intrahepatic bile ducts of the graft liver were consecutively disrupted with a bile lake, which required multiple biliary drainage procedures (Fig. 2). Multiple percutaneous biliary drainage was performed in B3, followed by B2.

Fig. 2.

Postoperative course. The C-reactive protein levels continued to increase because of consecutive biloma formation until the patient became septic and passed away. Biliary disruption had occurred serially all over the period which required multiple percutaneous drainages.

Because of the suspicion of ABO-i related cholangiopathy, a liver biopsy was performed, which did not show C4d staining on 195 post-transplant days (Fig. 3A and B). A CECT scan and magnetic resonance cholangiography were performed, indicating moderate dilation of the biliary tract, with no signs of cholangiopathy. In the postoperative course, the highest observed titers of anti-A antibodies were 32 for both IgG and IgM. Due to cholangitis, bacteriological cultures from blood, bile, and ascites were obtained, confirming the presence of Enterococcus faecium. Subsequent days, despite many interventions and the administration of broad-spectrum antibiotics, the patient's condition deteriorated. Ultimately, she developed sepsis and expired 11 months after re-LDLT. A postmortem liver biopsy (11 months after transplantation) showed moderate lymphocytic and neutrophilic infiltration into the portal area and fibrous enlargement. Pericentral fibrosis, endotheliitis, intimal thickening, and arteriopathy of small and medium arteries were also seen. In this specimen, C4d deposition was markedly positive (Fig. 3C and D).

Fig. 3.

Liver biopsy findings. There was slight inflammatory cell infiltration and weak fibrosis in the second graft on postoperative day 195 (Fig. 3A). Only mild cholestasis was observed, but C4d disposition was negative (Fig. 3B). There was marked evidence Cd4 disposition on the endothelium of the portal vein and the bile ducts indicating antibody-mediated rejection or acute cellular rejection.

3. Discussion

In LDLT, because of the limited relationship between the donor and recipient, ABO-incompatible transplantation is considered a treatment option and one of the most challenging issues to overcome. Several innovations have made ABO-i LDLT a realistic modality by making the patient/graft survival comparable to that of ABO-compatible LDLT [8]. This innovation includes desensitization with rituximab [3], first reported by Egawa et al. Rituximab is an anti-CD20 antibody originally introduced to treat B-cell lymphoma that can completely eliminate B lymphocytes before transplantation. This breakthrough has rendered unnecessary several strategies for performing ABO-i LDLT, including plasma exchange [9], splenectomy [10], and local infusion therapy [11].

Graft failure associated with ABO incompatibility has been reported to have the following features: (1) “hepatic necrosis” occurs within one to two weeks after LT, leading to extensive graft necrosis within one month, and (2) several months after LT, diffuse and multiple strictures develop in the intrahepatic bile duct, leading to graft failure [12]. The diagnosis of antibody-mediated rejection, such as in cases of ABO-i LDLT, is made based on the presence of DSA (required), C4d deposition around the portal vein microvascular endothelium, and expansion into peripheral sinusoids [13].

BDS has been widely reported in ABO incompatible LDLT, but cases of serial bile duct disruption without evident antibody-mediated rejection have rarely been reported. Diffuse intrahepatic BDS was reported in a case of brain-dead liver transplantation, showing that post-cardiac arrest donation and a lengthy ischemic time are risk factors; however, such BDS is very rare in LDLT. Ischemic-type lesions in the liver have been reported to be associated with specific immunological damage, such as hepatic artery thromboembolism (HAT), prolonged warm ischemic time, ABO incompatibility, vascular endothelial cells, and biliary epithelial damage [14]. A report from Asan University with 16 cases showed significant relevance to the present case. The mean time from transplantation to the appearance of diffuse intrahepatic BDS was 2.7 ± 1.4 (0.1–5.5) months. Risk factors for diffuse intrahepatic BDS could not be identified using a multivariate analysis including DSA [5].

Preformed DSA was positive in this LDLT but evidence of relation between DSA and biliary disruption was not clear yet. It was reported that HLA class II is not expressed in the normal biliary epithelium but expressed in the biliary epithelium of a graft liver [15]. However, in 2017, a relationship between DSA and non-anastomotic stricture in a graft was reported to be negative whether preformed or de novo [16]. Therefore, the cause of consecutive biliary disruption has remained unknown.

In Japan, re-transplantation with cadaveric donor grafts is allowed only in cases where at least one year has passed since the initial transplantation procedure. Therefore, it is best to inform patients and their families as early as possible that long-term survival is difficult with current grafts and that biliary complications cannot be expected to be cured. Early re-re-LDLT may be considered for diffuse non-anastomotic bile duct lesions after blood group-incompatible liver transplantation, especially in patients with multiple bilomas.

4. Conclusion

BDS has been widely reported after ABO-incompatible LDLT, but serial bile duct disruption because of antibody-mediated rejection has rarely been reported. We herein report such a rare case as an important observation that may aid in preventing and treating potentially fatal complications.

Informed consent

Written informed consent was obtained from the patient for publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Ethical approval

The study was approved by the institutional ethics committee.

Funding

This paper is written as part of the results of the 2023 Research on HIV/AIDS, No. 21HB2002, under the Health, Labor, and Welfare Sciences Research Grants.

Author contribution

Akihiko Soyama: study concept or design.

Takanobu Hara, Tomohiko Adachi: data collection.

Baglan Askeyev, Hajime Matsushima: data analysis or interpretation.

Susumu Eguchi, Akihiko Soyama: writing the paper.

Susumu Eguchi: Supervision.

Guarantor

Susumu Eguchi

Declaration of competing interest

The authors declared that they have no competing interests.