Biliary complications following donation after brainstem death liver transplantation

Abstract

BACKGROUND

Biliary complications are increasingly seen after liver transplantation. Different modalities are used to overcome these complications. In this study, we investigated the occurrence of biliary complications following donation after brainstem death (DBD) liver transplant and their management.

AIM

To investigate the occurrence and management of biliary complication in patients who underwent orthotopic liver transplant (OLT) with grafts from DBD donors between October 2014 and October 2020.

METHODS

A retrospective review of patients who underwent DBD OLT from October 2014 to October 2020. The primary outcome was biliary stricture, and the secondary outcome was return to theatre.

RESULTS

A total of 366 patients underwent DBD whole OLT during the study period. There were 229 (62.6%) males and 137 (37.4%) females. Recipients median age was 55 years (range 16-73 years) and donors median age was 51 years (range 11-89 years). Mean cold ischemic time was 581 ± 218.7 minutes. Duct to duct anastomosis was performed in 240 (65.6%) patients, and 126 (34.4%) patients underwent Roux-en-Y hepaticojejunostomy (RYHJ). Eighteen (4.9%) patients experienced bile leak, 39 (10.7%) patients experienced anastomotic biliary stricture and 18 (4.9%) patients had ischemic cholangiopathy. Main and segmental hepatic artery thrombosis occurred in 27 (7.4%) patients, 29 (7.9%) patients had early allograft dysfunction, and six (1.6%) patients experienced primary non function. Among 39 patients who developed anastomotic biliary stricture, 21 (53.8%) were treated with RYHJ representing 5.7% of the total transplant group. Endoscopic retrograde cholangiopancreatography was performed in nine patients (23.1%), conservative management in six patients (15.4%), percutaneous transhepatic cholangiography in two patients (5.1%) and re-transplantation in one patient who had concomitant ischemic cholangiopathy (2.6%). Overall, 84 (22.9%) patients returned to theatre and 27 (7.4%) patients underwent retransplant over five years follow up period. Multivariate analysis confirmed that RYHJ reconstruction emerged as the only significant independent protective factor against biliary complications (odds ratio = 0.473, 95% confidence interval: 0.23-0.97, P = 0.041).

CONCLUSION

Biliary complications following DBD liver transplant are a frequent occurrence and careful monitoring and planning are required to improve patients’ outcomes.

Key Words: Biliary stricture; Anastomotic biliary stricture; Biliary complications; Donation after brainstem death; Liver transplant

Core Tip: This retrospective study of 366 liver transplant after brainstem death, demonstrates that anastomotic biliary stricture remains a significant complication, occurring in 10.7% of patients. Roux-en-Y hepaticojejunostomy (RYHJ) was associated with significantly lower rates of bile leaks and anastomotic strictures compared to duct-to-duct anastomosis. However, patients undergoing primary RYHJ had higher reoperation and re-transplantation rates, primarily due to selection bias, as RYHJ was preferentially performed in patients with complex recipient pathology. Early detection and appropriate management with a stepwise approach - from endoscopic intervention to surgical revision - are crucial for optimizing post-transplant outcomes.

INTRODUCTION

Liver transplant is a definitive treatment for acute/chronic liver failure and represents a highly sophisticated surgical intervention requiring precise execution to optimize graft function and minimize postoperative complications, including biliary complications. In the United Kingdom, 9183 liver transplants were performed between April 2014 and March 2024, with deceased donor first liver only transplants accounting for 7931 cases (86%). Most recently, 883 adult liver transplants were performed in 2024/2025[1]. The foundational work in orthotopic liver transplant (OLT) was pioneered by Francis Moore in 1958 through experimental models, culminating in the first successful human transplantation by Starzl in 1963[2]. Within the United Kingdom, liver transplant has become the definitive treatment for end-stage liver disease, with reported patient survival rates of 95% at one year and 83% at five years post-transplantation[3]. Remarkably, metabolic dysfunction-associated steatotic liver disease has recently emerged as the leading etiology and is now the most prevalent indication for OLT[4].

In deceased organ donation, two types of donors are available; donation after circulatory death (DCD) and donation after brainstem death (DBD)[5,6]. In the United Kingdom, approximately 65% of liver transplants are from DBD donors and 35% are from DCD donors. Biliary reconstruction during liver transplant is typically performed via duct-to-duct anastomosis, provided there is no recipient bile duct pathology; otherwise, hepaticojejunostomy is utilized[7,8].

Biliary complications such as bile leak, anastomotic or non-anastomotic biliary strictures are common following liver transplants, occurring in 5%-20% of cases[9]. Anastomotic biliary strictures result from technical factors, ischemic injury to the bile duct mucosa, and fibrotic healing processes. In transplanted liver, the biliary tree relies exclusively on arterial blood supply from the hepatic artery, making it particularly vulnerable to ischemic injury during cold preservation and warm ischemia periods[10,11]. During ischemia, adenosine triphosphate depletion leads to cellular swelling and loss of ion homeostasis resulting in damage to cholangiocyte and subsequent apoptosis and necrosis. Accumulation of bile salts during ischemia exacerbates the epithelial injury. Additional contributors include immunological factors, bile salt toxicity, and cytomegalovirus infection[10].

It has been reported that DBD liver transplant has a lower risk of bile leak compared to DCD liver transplant and living donation. Notably, 15% of DCD liver transplants are complicated by biliary strictures. This is attributed to various risk factors, including surgical reconstruction, donor and recipient characteristics, cytomegalovirus infection, organ preservation, and hepatic artery thrombosis[10]. A comparative study by Meier et al[12] has shown that DBD liver transplant carries a 7.2% risk of bile leak, which is lower than DCD liver transplant and living donor liver transplant. Nevertheless, bile leak can lead to four times increase in the risk of developing anastomotic strictures and three times increase in the risk of non-anastomotic strictures.

In this study, we aimed to investigate the occurrence and management of biliary complication in patients who underwent OLT with grafts from DBD donors between October 2014 and October 2020.The primary outcome was anastomotic biliary stricture (ABS), and the secondary outcome was a return to the operating theatre.

MATERIALS AND METHODS

Study design

A retrospective review was conducted of prospectively collected data of consecutive patients who underwent liver transplant at Cambridge University Hospitals from October 2014 to October 2020. Patients who underwent liver transplant following DBD were included. Patients who were under 16 years old, underwent split liver transplant or liver transplant as part of multivesicular transplant were excluded. For patients who successfully fulfilled the inclusion criteria, case notes and the Epic electronic health record were searched.

The following variables were collected: Age, gender, American Society of Anesthesiologists classification , donor type and age, normothermic machine perfusion, United Kingdom model for end-stage liver disease, procedure time, length of stay, cold ischemia time, caval anastomosis, type of biliary anastomosis (duct-to-duct or hepaticojejunostomy), bile leak, ABS, treatment of stricture, ischemic cholangiopathy, hepatic artery thrombosis, early allograft dysfunction, primary non function, acute rejection, reoperation, re-transplantation, 30-day mortality and survival data. The primary outcome measure was ABS, and the secondary outcome was return to theatre.

United Kingdom model for end-stage liver disease is a scoring system based on serum sodium, creatinine, bilirubin, and international normalized ratio that predicts 1-year mortality in patients with chronic liver disease. Higher scores indicate more severe disease and greater urgency for transplantation in the United Kingdom allocation system[13]. Duct-to-duct anastomoses were performed as end-to-end anastomoses with polydioxanone absorbable sutures using interrupted or continuous technique. Roux-en-Y hepaticojejunostomy (RYHJ) was performed with a 60 cm Roux limb as end-to-side biliary-enteric anastomosis.

Statistical analysis

Data were analyzed with the SPSS version 20.0 (Chicago, IL, United States). For continuous parametric variables, we used the mean and standard deviation, and for continuous non-parametric variables we used the median and range. For categorical variables, the frequencies were used. Normality assumptions were demonstrated with histograms and the Kolmogorov-Smirnov test. Comparison between groups was conducted with t-test for continuous parametric variables and Mann-Whitney U test for non-parametric variables. χ² test and Fisher exact test were used to compare between categorical variables. A P value of less than 0.05 was considered statistically significant. Multivariate logistic regression analysis was performed to identify factors associated with biliary complications.

RESULTS

Patients’ characteristics

During the study period from October 2014 to October 2020, 366 patients underwent DBD OLT at Addenbrooke’s Hospital. There were 229 (62.6%) males and 137 (37.4%) females, recipients median age was 55 years (range 16-73 years), donors median age was 51 years (range 11-89 years). Normothermic machine perfusion was performed in 33 (9.0%) patients, and the median United Kingdom model for end-stage liver disease was 55.0 (range 43-71). Mean cold ischemia time was 581 ± 218.7 minutes. Duct to duct anastomosis was performed in 240 (65.6%) patients, and 126 (34.4%) patients underwent RYHJ (Table 1).

Table 1 Baseline demographic and clinical characteristics of the study population, n (%)/mean (interquartile range).

Variables	Duct to duct anastomosis (n = 240)	RYHJ (n = 126)	P value
Sex			0.851
Male	151 (62.9)	78 (61.9)
Female	89 (37.1)	48 (38.1)
ASA			0.847
2	3 (1.3)	2 (1.6)
3	85 (35.4)	42 (33.3)
4	145 (60.4)	77 (61.1)
5	7 (2.9)	5 (4.0)
Recipient age (years)	56 (16-73)	55 (19-73)	0.410
Donor age (years)	52 (11-89)	49 (11-79)	0.430
NMP	20 (8.3)	13 (10.3)	0.523
UKELD	55 ± 5.3	55 ± 5.0	0.093
Cold ischemia time (minutes)	575 ± 220.5	590 ± 215.8	0.354
LOS (days)	24.5 ± 21.7	29.3 ± 23.7	0.073
Operation length (minutes)	478 ± 118.9	567 ± 203.9	< 0.001

ASA: American Society of Anesthesiologists classification; NMP: Normothermic machine perfusion; UKELD: United Kingdom model for end-stage liver disease; LOS: Length of stay; RYHJ: Roux-en-Y hepaticojejunostomy.

Postoperative outcomes

During the study period, 18 patients (4.9%) experienced bile leaks, and 39 patients (10.7%) ABS, predominantly in the duct-to-duct anastomosis group (12.5%). Additionally, 18 patients (4.9%) were diagnosed with ischemic cholangiopathy. Hepatic artery thrombosis occurred in 27 patients (7.4%), while early allograft dysfunction was observed in 29 patients (7.9%). Primary non-function was reported in six patients (1.6%). Acute rejection occurred in 36 patients (9.8%). A total of 84 patients (23.0%) required return to the operating theatre, mainly for bleeding and bile leaks. Re-transplantation was necessary in 27 patients (7.4%). The 30-day mortality rate was 11.5% (42/366), with a one-year overall survival rate of 88.5% (Table 2).

Table 2 Postoperative outcomes, n (%).

Outcomes	Duct to duct anastomosis (n = 240)	RYHJ (n = 126)	P value
Bile leak	16 (6.7)	2 (1.6)	0.030
Anastomotic biliary stricture	30 (12.5)	9 (7.1)	0.040
Ischemic cholangiopathy	12 (5.0)	6 (4.8)	0.842
Hepatic artery thrombosis	15 (6.3)	12 (9.5)	0.120
Early allograft dysfunction	17 (7.1)	12 (9.5)	0.411
Primary non function	3 (1.3)	3 (2.4)	0.418
Acute rejection	25 (10.4)	11 (8.7)	0.795
Reoperation	47 (19.6)	36 (28.6)	0.033
Re-transplantation	11 (4.6)	16 (12.7)	0.005
30 days mortality	26 (10.8)	16 (12.7)	0.280
1-year overall survival	214 (89.2)	110 (87.3)	0.281

RYHJ: Roux-en-Y hepaticojejunostomy.

Treatment options for ABS

ABS were managed with RYHJ in 21of 39 patients with ABS (53.8%), particularly within the duct-to-duct anastomosis cohort (P = 0.159). Endoscopic retrograde cholangiopancreatography with stenting was performed in nine patients (23.1%; P = 0.061), conservative management was employed in six patients (15.4%; P = 0.089), percutaneous transhepatic cholangiography in two patients (5.1%; P = 0.008), and re-transplantation was required in one patient who had concomitant ischemic cholangiopathy (2.6%; P = 0.064; Table 3).

Table 3 Treatment options for anastomotic biliary stricture, n (%).

Treatment option	Duct to duct anastomosis ABS cases (n = 30)	RYHJ ABS cases (n = 9)	P value
ERCP and stent	9 (30)	0 (0)	0.061
PTC and dilatation	0 (0)	2 (22.2)	0.008a
RYHJ	18 (60)	3 (33.3)	0.159
Conservative	3 (10)	3 (33.3)	0.089
Retransplant	0 (0)	1 (11.1)	0.064

^aP < 0.01.

ERCP: Endoscopic retrograde cholangiopancreatography; PTC: Percutaneous transhepatic cholangiography; RYHJ: Roux-en-Y hepaticojejunostomy; ABS: Anastomotic biliary stricture.

Multivariate analysis of risk factors for biliary complications

Multivariate logistic regression analysis was performed on 363 patients with complete data to identify independent predictors of biliary complications (defined as bile leak or anastomotic stricture). A total of 51 patients (14.0%) experienced biliary complications. After adjusting for patient age, donor age, sex, cold ischemia time, normothermic machine perfusion use, and hepatic artery thrombosis, RYHJ reconstruction emerged as the only significant independent protective factor against biliary complications [odds ratio (OR) = 0.473, 95% confidence interval (CI): 0.23-0.97, P = 0.041], representing a 53% reduction in odds compared to duct-to-duct anastomosis. Patient age (OR = 0.937, 95%CI: 0.68-1.29, P = 0.688), donor age (OR = 1.097, 95%CI: 0.80-1.51, P = 0.569), sex (OR = 0.971, 95%CI: 0.52-1.82, P = 0.926), and cold ischemia time (OR = 1.091, 95%CI: 0.78-1.52, P = 0.606) showed no significant association with biliary complications. The model demonstrated acceptable discrimination (C-statistic = 0.639) and excellent calibration, with predicted complication rates of 16.9% for duct-to-duct and 8.7% for RYHJ matching the observed rates (Table 4).

Table 4 Multivariate analysis of risk factors for biliary complications.

Variable	Coefficient	Odds ratio	95%CI	P value
Intercept	-1.581	0.206	0.12-0.35	< 0.001b
Patient age (per 12.1 years)	-0.065	0.937	0.68-1.29	0.688
Donor age (per 15.3 years)	0.093	1.097	0.80-1.51	0.569
Male sex	-0.030	0.971	0.52-1.82	0.926
Cold ischemia time (per 218 minutes)	0.087	1.091	0.78-1.52	0.606
Roux-en-Y reconstruction	-0.748	0.473	0.23-0.97	0.041a
NMP use	0.548	1.729	0.56-5.34	0.341
Hepatic artery thrombosis	-1.408	0.245	0.03-1.87	0.174

^aP < 0.05.

^bP < 0.001.

NMP: Normothermic machine perfusion; CI: Confidence interval.

DISCUSSION

In this retrospective study, 366 patients underwent DBD OLT from October 2014 to October 2020, the results showed ABS is a significant complication affecting 10.7% of the patients in this study. Among patients who developed ABS, the most common treatment modality was surgical revision with RYHJ (53.8% of ABS cases, representing 5.7% of the total group), followed by endoscopic retrograde cholangiopancreatography (23.1% of ABS cases, 2.5% of total group) and conservative management (15.4% of ABS cases, 1.6% of total group). The study highlights the importance of early detection and appropriate management of biliary complications to improve patient outcomes after liver transplantation. Additionally, ischemic cholangiopathy was another notable complication, occurring in 4.8% in RYHJ and 5% in duct-to-duct anastomosis. The results of this retrospective study provide valuable insights into the incidence and management of biliary complications following DBD liver transplantation. ABS was the most common biliary complication, occurring in 10.7% of patients, which is consistent with the reported incidence of 5%-15% in the literature[10,14]. Ischemic cholangiopathy was another notable complication at 4.9%, similar to the 5%-15% incidence reported in previous studies[15,16].

Interestingly, most bile leaks and strictures occurred in the duct-to-duct anastomosis group, while minority were reported in the RYHJ group. Table 2 shows a statistically significant difference favoring RYHJ for preventing bile leaks (P < 0.05). This finding is supported by a meta-analysis[16], which demonstrated that RYHJ was associated with a lower incidence of biliary complications compared to duct-to-duct anastomosis. The protective effect of RYHJ may be attributed to its ability to reduce tension on the anastomosis and preserve blood supply[17]. By creating a tension-free anastomosis to a Roux limb, RYHJ may minimize the ischemic injury that underlies stricture formation, addressing the fundamental pathophysiological mechanism. However, the choice of anastomosis technique did not significantly impact other complications or outcomes in our study.

An important and seemingly paradoxical finding in our study requires careful interpretation: Patients who underwent primary RYHJ demonstrated significantly higher rates of reoperation (28.6% vs 19.6%, P = 0.033) and re-transplantation (12.7% vs 4.6%, P = 0.005) compared to the duct-to-duct group, despite lower incidence of bile leaks (1.6% vs 6.7%, P = 0.030) and anastomotic strictures (7.1% vs 12.5%, P = 0.040). Primary RYHJ group (n = 126) represents only patients who underwent RYHJ as their initial biliary reconstruction at the time of index transplantation. This group explicitly excludes patients who subsequently required conversion from duct-to-duct to RYHJ for stricture management. Among the 21 patients who received RYHJ as treatment for ABS, 18 (85.7%) came from the initial duct-to-duct anastomosis group, while only 3 (14.3%) came from the primary RYHJ group (P = 0.159). This distribution confirms that secondary RYHJ procedures for stricture treatment have not been misclassified as primary RYHJ in our analysis. This apparent contradiction is primarily explained by inherent selection bias in the choice of biliary reconstruction technique. RYHJ was preferentially performed in patients with pre-existing recipient bile duct pathology, most commonly primary sclerosing cholangitis, biliary atresia, previous biliary surgery, or when significant size mismatch precluded safe duct-to-duct anastomosis. These underlying conditions carry independent risks for adverse outcomes. Primary sclerosing cholangitis patients have higher rates of disease recurrence, non-anastomotic strictures, and graft dysfunction unrelated to the anastomotic technique itself. Additionally, patients with complex portal vein anatomy, previous upper abdominal surgery, or re-transplantation, all of which necessitate RYHJ, have inherently higher perioperative risks. Crucially, the increased rates of reoperation and re-transplantation in the RYHJ group were predominantly driven by non-biliary complications such as vascular issues, graft dysfunction, and underlying disease progression, rather than anastomotic complications. This underscores that while RYHJ effectively prevents anastomotic biliary complications, it does not mitigate the inherent risks associated with complex recipient pathology.

For managing ABS, RYHJ was the most common treatment (5.7%), followed by endoscopic and conservative approaches. A small number required percutaneous interventions or re-transplantation. These findings are in line with current recommendations for managing biliary complications, which suggest a stepwise approach starting with endoscopic or percutaneous interventions and progressing to surgical revision or re-transplantation if necessary[18,19]. Prompt diagnosis and appropriate management are crucial for optimizing outcomes and preventing graft loss.

The study is limited by its retrospective, single-center design. Prospective, multicenter studies could provide more robust evidence. Additionally, long-term follow-up data on biliary complications and graft survival would be informative. Future research should focus on identifying risk factors for biliary complications and evaluating novel strategies for prevention and early detection.

CONCLUSION

The study demonstrates that biliary complications, particularly ABS, remain a common issue after DBD liver transplantation, with lower incidence associated with RYHJ reconstruction. Early detection and appropriate management are crucial for treating biliary complications and improving post-transplant outcomes.