Stent placement can achieve same prognosis as endoscopic nasobiliary drainage in treatment of bile leakage after liver transplantation

Abstract

BACKGROUND

Both stent placement and endoscopic nasobiliary drainage (ENBD) are commonly used for treating bile leakage following liver transplantation. However, little research has compared the advantages of both methods.

AIM

To compare the prognosis of stent placement and ENBD in treating bile leakage following liver transplantation.

METHODS

We retrospectively collected data for patients with bile leakage after liver transplantation in our center. The efficiency and prognosis were compared between the stent placement and ENBD groups.

RESULTS

We retrospectively collected data from 50 patients diagnosed with bile leakage within 1 month after liver transplantation. Forty patients underwent ENBD and nine received stent placement. Baseline patient characteristics, except white blood cell count, total bilirubin and direct bilirubin levels, were comparable between the two groups. Additionally, there was no significant difference in the rates of reconveyance or postoperative biliary stricture.

CONCLUSION

Stent placement and ENBD are recommended for treating bile leakage after liver transplantation, with similar efficacy and prognosis.

Key Words: Bile leakage; Liver transplantation; Stent placement; Endoscopic nasobiliary drainage; Endoscopic retrograde cholangiopancreatography

Core Tip: Both stent placement and endoscopic nasobiliary drainage are effective for treating bile leakage after liver transplantation, with comparable efficacy and prognosis. Our retrospective study of 50 patients showed no significant differences in outcomes, except for variations in white blood cell count and bilirubin levels. Both methods are recommended based on similar success rates and post-operative complications.

INTRODUCTION

Biliary complications including biliary anastomotic stricture and bile leakage, are major source of morbidity after liver transplantation, with an incidence of 5%-32%[1]. According to the United Kingdom National Health Service, bile duct complications after liver transplantation include bile leakage, bile duct strictures, bile duct stones, and sphincter of Oddi dysfunction. Among these, bile leakage is one of the most common complications, with an incidence rate ranging from 2% to 25%, which is more common in living donor than in deceased donor liver transplantation[2]. Bile leakage is the second most common biliary complication of liver transplantation, which can be classified in two categories: Early (within 4 weeks of transplantation) and late (> 4 weeks after transplantation)[3]. Early bile leakage is considered as a result of technical problems rather than related to the type of biliary reconstruction (Figure 1)[3]. However, it is associated with planned or inadvertent T-tube removal[1]. Thus, early bile leakage usually occurs at the anastomotic or T-tube insertion site. It is often accompanied by infection, pain, and delayed graft function, posing a threat to patient survival and quality of life.

Figure 1 Common causes of early bile leakage. A: The common causes of early bile leakage, including cracks around the T-tube, poor suturing, incomplete healing and insufficiency of blood supply; B: Stent placement for bile leakage; C: Endoscopic nasobiliary drainage for bile leakage. ENBD: Endoscopic nasobiliary drainage.

Endoscopy plays a pivotal role in the diagnosis and treatment of biliary diseases. Notably, endoscopic sphincterotomy, stenting, and drainage have been recommended for managing biliary leakage[4-6]. According to a systematic review, the success rate of endoscopic therapy for bile leak ranges from 68% to 98.3%[7]. In patients who have undergone liver transplantation, endoscopic nasobiliary catheter or plastic stent placement is a safe and effective treatment for bile duct strictures following bile leak resolution[8]. Favorable outcomes are more likely when the duct injury measures < 5 mm or occurs extrahepatically, and when there is no associated abscess or biloma[9]. However, the comparative efficiency and long-term prognosis of endoscopic nasobiliary catheter vs plastic stent placement have received limited attention in previous studies. Therefore, this comprehensive investigation aimed to analyze patient-related factors and compare the advantages and disadvantages of these two methods.

MATERIALS AND METHODS

We retrospectively analyzed adult patients who underwent endoscopic retrograde cholangiopancreatography (ERCP) for bile leakage after liver transplantation at our center between June 2016 and January 2023. We included only those patients who underwent ERCP within 30 days after surgery. All patients had received piggyback or right lobe liver transplantation. Routine deployment of double abdominal drains around the first hepatic porta and posterior of the liver occurred, and T-tubes or internal stents were used according to the bile duct diameter. Standard perioperative management protocols were followed for donor and recipient operations. Graft preservation involved the use of either histidine-tryptophan-ketoglutarate or University of Wisconsin solution. Postoperatively, all patients were closely monitored, and ERCP was performed upon diagnosis of bile leakage. Patient enrollment flowchart was shown in Figure 2. Continuous and categorical variables were compared using Student’s t-test and the χ²-test. The results were given as mean ± SD unless otherwise indicated. An alpha level of 0.05 indicated statistical significance. Listwise deletion was used to handle missing data in each variable. All statistical analyses were performed using SPSS 20.0 (IBM, Armonk, NY, United States).

Figure 2 Patient enrollment flowchart. ERCP: Endoscopic retrograde cholangiopancreatography; LT: Liver transplantation; ENBD: Endoscopic nasobiliary drainage.

RESULTS

We retrospectively included 50 patients after cadaveric liver transplantation. One patient was withdrawn due to difficult gastroscopic access. Forty patients underwent endoscopic nasobiliary drainage (ENBD) and nine received stent placement. We analyzed various factors including demographic data, causes of liver transplantation, T-tube placement, prognosis, incidence of fever, and laboratory values. Demographics and characteristics of these patients are shown in Table 1. The ENBD group had an average age of 49.83 ± 10.253 years, and was significantly younger than the stent group, which had an average age of 51.56 ± 7.299 years (P = 0.000). The ENBD group comprised 87.5% male patients, while the stent group had 77.8%. Hepatocellular carcinoma was the cause of liver transplantation in 45.0% of the ENBD group and 55.6% of the stent group. Decompensated cirrhosis was observed in 30.0% of the ENBD group compared to 22.2% in the stent group. Acute-on-chronic liver failure accounted for 10% in the ENBD group and 11.1% in the stent group. Acute liver failure was less common in the ENBD group (2.5%) compared to the stent group (11.1%). Other causes were noted in 12.5% of the ENBD group, with no cases in the stent group.

Table 1 Demographics and characteristics, n (%) or mean ± SD.

	ENBD (n = 40)	Stent (n = 9)	P value
Age, years	49.83 ± 10.25	51.56 ± 7.30	0.000
Male	35 (87.5)	7 (77.8)	0.451
Cause for LT			0.946
Hepatocellular carcinoma	18 (45.0)	5 (55.6)
Decompensated cirrhosis	12 (30.0)	2 (22.2)
Acute-on-chronic liver failure	4 (10)	1 (11.1)
Acute liver failure	1 (2.5)	1 (11.1)
Others	5 (12.5)	0
TB	57.77 ± 49.96	102.22 ± 98.45	0.000
DB	44.51 ± 42.61	78.33 ± 81.31	0.001
WBC	7.58 ± 4.11	9.18 ± 7.23	0.030
CRP	37.33 ± 39.18	31.30 ± 34.82	0.931
PCT	2.38 ± 5.25	1.20 ± 1.51	0.263
Ab	36.85 ± 3.61	38.64 ± 4.52	0.357
AST	134.82 ± 346.01	54.22 ± 29.24	0.198
ALT	177.90 ± 236.18	130.78 ± 152.20	0.460
GGT	162.08 ± 145.97	119.22 ± 59.37	0.089
ALP	134.95 ± 86.02	136.11 ± 66.70	0.230
T-tube placement	4 (10)	2 (22.2)	0.312
Fever	14 (35)	3 (33.3)	0.885

ENBD: Endoscopic nasobiliary drainage; LT: Liver transplantation; TB: Total bilirubin; DB: Direct bilirubin; WBC: White blood cell; CRP: C-reactive protein; PCT: Procalcitonin; Ab: Albumin; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase; GGT: Gamma-glutamyl transpeptidase; ALP: Alkaline phosphatase.

Incidence of fever was similar in both groups, with 35% in the ENBD group and 33.3% in the stent group. Laboratory values showed a higher white blood cell (WBC) count in the stent group (9.1778 ± 7.23356) compared to the ENBD group (7.5769 ± 4.10888, P = 0.030). Total bilirubin (TB) and direct bilirubin (DB) levels were significantly higher in the stent group (P = 0.000 for TB and P = 0.001 for DB), suggesting more severe liver function impairment in this group. No significant differences were noted in C-reactive protein, procalcitonin, albumin, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transpeptidase, and alkaline phosphatase levels between the two groups. About 26 patients in the ENBD group underwent secondary operation within 1 month for stent placement, and leakage recurred in three of them. A dualistic logistic regression model was used to confirm the significance of types of operation in prognosis and postoperative biliary stricture, which also came out negative (P = 0.392 and 0.890).

T-tube placement is a risk factor for bile leakage. 10% of the patients in ENBD group and 22.2% in the stent group underwent T-tube placement. Fisher’s exact test was used as only one of the stent group failed to recover from leakage, and there was no significant difference between the groups (P = 0.412). Patients who died or underwent biliary surgery within 6 months after transplantation were not included in the analysis of biliary stricture. Three patients underwent biliary surgery due to failure of treatment, and 11 patients died due to infection and graft failure. Finally, nine patients in the ENBD group and two in the stent group result biliary stricture, and 17 patients in ENBD group and seven in the stent group were not. Fisher’s exact test showed no significant difference in biliary stricture between two groups (P = 1.0). Further, binary logistic regression was also performed, showing no significant difference between the two groups (P = 0.292 for recovery and P = 0.890 for biliary stricture) (Table 2). The significant variables in Table 1, including age, WBC count and TB, were taken as covariates. Although our study highlighted significant differences in age, WBC count, and TB and DB levels between the ENBD and stent groups, there was no significant difference in prognosis and postoperative biliary stricture.

Table 2 Prognosis and stricture, n (%).

	ENBD (n = 40)	Stent (n = 9)	P value	Binary logistic regression
Recovery	28 (70)	8 (88.9)	0.412	0.292
Biliary stricture	9 (28)	2 (7)	1.0	0.890

ENBD: Endoscopic nasobiliary drainage.

DISCUSSION

Bile leakage can be a postoperative complication of gallbladder removal or liver transplantation, or damage to the bile ducts and biliary system[10]. Around 2.7% of deceased donor liver transplantation recipients have bile leakage after surgery, which rises over 10% of patients undergoing living donor liver transplantation[10,11]. The rate of bile leakage can be higher in patients with cholangiojejunostomy. Endoscopic techniques are recommended as first-line treatment of biliary complications after liver transplantation, with a success rate of 70%-100%[12]. However, ERCP is limited in patients after cholangiojejunostomy, especially with Roux-en-Y gastric bypass[13]. Endoscopic ultrasound or endoscopic ultrasound-directed transgastric ERCP can help overcome the challenge of cholangiojejunostomy, However, it is a double-step technique, and only 40% patients completed conventional ERCP during the index procedure. Further, about 17% of patients ended up with metal stent dislodged. Bile leakage in patients after liver transplantation is always followed by infection, immune dysfunction and delayed wound healing. It is difficult to build a reliable and long-term monitoring method to ensure the success rate at the first attempt. A failed invasive procedure would significantly increase the risk of gastrointestinal perforation.

Research has confirmed the important role of endoscopic treatment in bile leakage after liver transplantation, but few studies have compared the difference between stent placement and ENBD. Stents provide a long-term solution and are less invasive compared to surgical options. They are often left in place for several weeks to months to ensure healing. Compared to ENBD, stents can provide internal biliary support, which help prevent bile duct stenosis. Covered retrievable metal and plastic stents are alternative according to site and size of the fistula, infection status and economical support of the patient. Covered retrievable metal stents are self-expanding, which can expand to a larger diameter and provide better drainage. They are more durable and less likely to become occluded over time compared to plastic stents. However, metal stents are typically placed in the common bile ducts, which may be ineffective for bile leakage from the hepatic resection surface in living donor liver transplantation, unless across the leak site[11], whereas plastic stents can be placed in the hepatic ducts. Due to the smaller diameter, multiple plastic stents are recommended to prevent long-term bile duct stenosis[14]. One single-center study reported that endoscopic retrograde biliary drainage using plastic stents had a similar outcome and postoperative infection rate compared to ENBD[15]. Stent placement can cause damage to the epithelium that allows a conduit for bacteria to enter. Stents can also be blocked by bacterial biofilms and biliary sludge, leading to delayed infection and reoperation[16,17].

ENBD allows for the direct external drainage of bile, which can be beneficial in managing the leak and in monitoring the volume and nature of the bile for diagnostic and therapeutic purposes. However, ENBD is temporary and can be uncomfortable for the patient due to passage of the tube through the nose. It also requires careful maintenance of the external drainage system and maintaining water-electrolyte balance. Although ENBD alone can rapidly cure postoperative bile leakage after hepatectomy without bilioenteric anastomosis[18], for patients after liver transplantation, it is followed by stent placement to avoid delayed bile duct stenosis in our center. Thus, ENBD is recommended for patients with biliary infections while stent placement is supported in patients with a risk of bile duct stenosis. Unfortunately, it is difficult to determine whether the patients have cholangitis after liver transplantation. Bacteriological sampling through blood cultures and drainage fluid with bile cultures can help determine the pathogenic bacteria responsible for most cases of cholangitis, which are mainly Escherichia coli and Klebsiella spp.[19]. However, the biliary colonizing bacteria would make it hard to diagnose[20]. Risk factors for postoperative biliary stricture include biliary or vascular anatomical variants, local inflammation, and poor surgical expertise[21].

There were several limitations in this study. First, we did not assess the severity of bile leakage. ENBD contains the advantages of less damage, faster and can cooperate with external negative pressure device, thus patients with severe disease tend to choose ENBD rather than stents. Secondly, as this is a retrospective study, due to the data deficiency, it is difficult to compare the recovery duration between the two groups. Thirdly, insufficient follow-up duration prevented the assessment of long-term prognosis and survival analysis. Finally, the sample size was small, and the relatively lenient patient selection criteria may have introduced selection bias. Thus, more dates are required to verify our results.

CONCLUSION

This study showed that stent placement and ENBD are recommended for treating bile leakage after liver transplantation, as they exhibit similar efficacy and prognosis.