Endoscopy 2021; 53(01): 55-62
DOI: 10.1055/a-1195-8197
Original article

ERCP plus endoscopic ultrasound-guided biliary drainage versus percutaneous transhepatic biliary drainage for malignant hilar biliary obstruction: a multicenter observational open-label study

Pradermchai Kongkam
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
2  Pancreas Research Unit, Chulalongkorn University, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Theerapat Orprayoon
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Chaloemphon Boonmee
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
4  Department of Surgery, Tha-Bo Crown Prince Hospital, Ministry of Public Health, Nong Khai, Thailand
,
Passakorn Sodarat
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
5  Department of Surgery, Roi-Et Hospital, Ministry of Public Health, Roi-Et, Thailand
,
Orathai Seabmuangsai
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
5  Department of Surgery, Roi-Et Hospital, Ministry of Public Health, Roi-Et, Thailand
,
Chatchawan Wachiramatharuch
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
5  Department of Surgery, Roi-Et Hospital, Ministry of Public Health, Roi-Et, Thailand
,
Yutthaya Auan-Klin
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
,
Khanh Cong Pham
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
,
Abbas Ali Tasneem
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Stephen J. Kerr
6  Biostatistics Excellence Centre, Chulalongkorn University, Bangkok, Thailand
,
Rommel Romano
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Sureeporn Jangsirikul
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Wiriyaporn Ridtitid
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Phonthep Angsuwatcharakon
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
,
Thawee Ratanachu-ek
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
7  Department of Surgery, Rajavithi Hospital, Ministry of Public Health, Bangkok, Thailand
,
Rungsun Rerknimitr
1  Gastrointestinal Endoscopy Excellence Center and Division of Gastroenterology, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand
3  Thai Association for Gastrointestinal Endoscopy, Bangkok, Thailand
› Author Affiliations
Trial Registration: Thai Clinical Trials Registry Registration number (trial ID): TCTR20180711004 Type of study: Multicenter observational open-label study
 

Abstract

Background Endoscopic retrograde cholangiopancreatography (ERCP) may not provide complete biliary drainage in patients with Bismuth III/IV malignant hilar biliary obstruction (MHBO). Complete biliary drainage is accomplished by adding percutaneous transhepatic biliary drainage (PTBD). We prospectively compared recurrent biliary obstruction (RBO) rates between combined ERCP and endoscopic ultrasound-guided biliary drainage (EUS-BD) vs. bilateral PTBD.

Methods Patients with MHBO undergoing endoscopic procedures (group A) were compared with those undergoing bilateral PTBD (group B). The primary outcome was the 3-month RBO rate.

Results 36 patients were recruited into groups A (n = 19) and B (n = 17). Rates of technical and clinical success, and complications of group A vs. B were 84.2 % (16/19) vs. 100 % (17/17; P = 0.23), 78.9 % (15/19) vs. 76.5 % (13/17; P > 0.99), and 26.3 % (5/19) vs. 35.3 % (6/17; P = 0.56), respectively. Within 3 and 6 months, RBO rates of group A vs. group B were 26.7 % (4/15) vs. 88.2 % (15/17; P  = 0.001) and 22.2 % (2/9) vs. 100 % (9/9; P = 0.002), respectively. At 3 months, median number of biliary reinterventions in group A was significantly lower than in group B (0 [interquartile range] 0–1 vs. 1 [1–2.5]), respectively (P < 0.001). Median time to development of RBO was longer in group A than in group B (92 [56–217] vs. 40 [13.5–57.8] days, respectively; P  =  0.06).

Conclusions Combined ERCP and EUS procedures provided significantly lower RBO rates at 3 and 6 months vs. bilateral PTBD, with similar complication rates and no significant mortality difference.


#

Introduction

Current palliative biliary drainage methods for patients with Bismuth Corlette class III, IV or more advanced malignant hilar biliary obstruction (MHBO) are not perfect. An ideal drainage procedure would be minimally invasive with internal placement, and offer adequate biliary drainage, defined as draining > 50% of the liver volume [1]. Unfortunately, currently available procedures cannot meet these needs. Surgical bypass is relatively invasive for patients such as these, who typically have short survival times. Endoscopic retrograde cholangiopancreatography (ERCP) with multisegmental drainage is technically difficult due to the tightness of strictures, and selective intrahepatic bile duct (IHD) cannulation is also challenging [2]. A recent guideline from the European Society of Gastrointestinal Endoscopy (ESGE) recommended percutaneous transhepatic biliary drainage (PTBD) or a combination of PTBD and ERCP as palliative biliary drainage procedures for Bismuth types III and IV MHBO [3]. However, PTBD is inconvenient for most patients because of the associated external lines and bags.

Endoscopic ultrasound-guided biliary drainage (EUS-BD) has been increasingly used worldwide as an alternative when ERCP is not possible or fails. Most studies have focused on utilizing EUS-BD in patients with distal bile duct obstruction, and information about EUS-BD for patients with MHBO is limited. Nakai et al. recently reported that among 88 patients with MHBO treated with EUS-BD, the technical success rate was 98%, the clinical success rate was 77%, and the complication rate was 8 % [4]. EUS-BD was performed as either a primary biliary drainage method or a rescue method after failed ERCP. Given the lower clinical success rate and higher complication rate of EUS-BD in MHBO compared with ERCP, it should not be considered as the first choice for palliative endoscopic biliary drainage [5].

As ERCP with stent placement is simpler for drainage of single rather than multiple segments, we hypothesized that combining ERCP and EUS-BD procedures as a primary biliary drainage procedure in patients with MHBO would ensure more complete emptying of the undrained segments. Transmural EUS-BD including EUS-guided hepaticogastrostomy (EUS-HGS) and EUS-guided hepaticoduodenostomy (EUS-HDS) can be performed following ERCP with biliary stenting in right and left IHDs, respectively, in order to accomplish adequate internal biliary drainage for MHBO [5].

As ESGE has recommended PTBD, or a combination of PTBD with ERCP, as the procedure of choice for biliary drainage in patients with MHBO, our team aimed to compare the benefits and harms of biliary drainage using a combination of ERCP and EUS-BD against a historical control group treated with PTBD, as a primary biliary drainage method in this prospective cohort study.


#

Methods

Patients

Patients included in the study were those with native high-grade MHBO, defined as Bismuth Corlette class III, IV or more advanced MHBO, small or trivial malignant hilar liver mass causing MHBO with radiological image showing viable liver parenchyma > 80 % of liver volume, Eastern Cooperative Oncology Group (ECOG) score ≤ 2 (capable of all self-care but unable to carry out any work activities, up and about > 50 % of waking hours) [6], expected survival of more than 3 months, age between 18 and 80 years, with unresectable tumor. Exclusion criteria were patients with uncorrectable coagulopathy, pregnant women, patients with large-volume ascites [7], known benign hilar biliary obstruction, IgG4-mediated cholangiopathy, and inability or unwillingness to follow the study protocol.

Patients meeting the inclusion criteria and presenting in any of the three tertiary care hospitals of Thailand (King Chulalongkorn Memorial Hospital, Bangkok; Tha-Bo Crown Prince Hospital, Nong Khai; and Roi-Et Hospital, Roi-Et) were identified and enrolled into the study between March 2016 and October 2019. The study was conducted under the administration of the Thai Association for Gastrointestinal Endoscopy. These patients presenting to endoscopy services for primary management of biliary drainage who received endoscopic biliary drainage were classified as group A. We retrospectively reviewed the database of patients with Bismuth Corlette class III, IV or more advanced MHBO who underwent bilateral PTBD in interventional radiology service from 2015 to 2019 at King Chulalongkorn Memorial Hospital, Bangkok, and enrolled these patients consecutively as historical PTBD controls with no exclusions (group B). As all group A patients were Bismuth Corlette stage IV, we selected controls who matched the staging of group A patients, namely those with Bismuth Corlette stage IV MHBO who underwent bilateral PTBD for biliary drainage; patients who underwent unilateral PTBD were not eligible.

The protocol was approved by the Institutional Review Board of the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand. Funding was obtained from the Pancreas Research Unit, Chulalongkorn University and Health Systems Research Institute (HSRI), grant number HSRI 62 – 050. The study was registered at Thai Clinical Trials Registry (http://www.clinicaltrials.in.th) with clinical trial number TCTR20180711004.


#

Selection of drainage methods

In group A patients, computed tomography or magnetic resonance imaging was reviewed before endoscopic biliary drainage. Details about the choice of procedure performed are as follows ([Fig. 1]).

Zoom Image
Fig. 1 Malignant hilar biliary obstruction (MHBO). a Computed tomography of the abdomen in a patient with MHBO. b Two percutaneous transhepatic biliary drainage catheters from right and left lobes of the liver. c A hybrid metal stent (partially covered with the antimigration flared end on the gastric side (GIOBORTM Biliary stent; Taewoong, Gyeonggi-do, South Korea) connecting the left intrahepatic bile duct (segment 3) to the gastric lumen. d An uncovered self-expandable biliary metal stent draining bile fluid from the right intrahepatic bile duct to the duodenum.

For MHBO caused by biliary stricture alone or a small mass in the bile hilum with both lobes of the liver functioning ([Fig. 2a]), ERCP with a single self-expandable metal stent (SEMS) placed into the right IHD was performed. EUS-HGS was subsequently performed to achieve adequate biliary drainage, either in the same session or within 2 weeks after the initial ERCP procedure ([Fig. 2b]).

Zoom Image
Fig. 2 Methods used for applying a combination of endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic ultrasound-guided biliary drainage (EUS-BD) in patients with malignant hilar biliary obstruction (MHBO). a A hilar tumor causing MHBO, with an upstream dilated intrahepatic bile duct (IHD). b ERCP with stent placement into the right IHD system plus EUS-guided hepaticogastrostomy. c ERCP with stent placement into the left IHD system plus EUS-guided hepaticoduodenostomy. d ERCP with failed stent placement across the obstructed area; both EUS-guided hepaticogastrostomy and EUS-guided hepaticoduodenostomy are then performed to accomplish biliary drainage.

However, if the SEMS of ERCP was placed into the left IHD, EUS-HDS was then performed ([Fig. 2c]). If ERCP plus single EUS-BD (EUS-HGS or EUS-HDS) still could not drain bile fluid sufficiently (i.e. persistent IHD dilation in the posterior segment of the right liver lobe with clinical symptoms of localized cholangitis), then an additional EUS-BD procedure was added to accomplish adequate biliary drainage ([Fig. 2 d]).


#

Endoscopic techniques

ERCP and biliary stenting

The duodenoscope was passed into the second part of the duodenum. The bile duct was cannulated and opacification over the hilar biliary stricture area was achieved using retrograde contrast injection. The guidewire was passed across and beyond the stricture. A noncovered biliary SEMS (10–12 cm in length, 10 mm in diameter) was placed into either the right or left biliary system.


#

EUS-HGS

A linear echoendoscope was introduced into the stomach, and the dilated IHDs in the left lobe were endosonographically identified. A 19-gauge needle for EUS-guided fine-needle aspiration was inserted into either a biliary segment 3 or 2 duct, taking care to avoid puncturing any blood vessels. The segment without liver mass was the first choice; however, if there was no tumor involvement in either segment 2 and 3, segment 3 was preferable to avoid unnecessary transesophageal puncture. Aspiration of bile and injection of contrast was then performed to confirm that the needle was in the bile duct. Thereafter, dilation of the punctured tract was achieved by passing a guidewire into the bile duct and dilating a 6-Fr cystotome followed by a bougie dilator (Soehendra dilator; Wilson-Cook, Winston-Salem, North Carolina, USA), or a 4–8-mm balloon dilator (Hurricane; Boston Scientific, Marlborough, Massachusetts, USA). A hybrid metal stent, 100–120 mm in length (partially covered with the antimigration flared end on the gastric side; Taewoong, Gyeonggi-do, South Korea) was deployed at the punctured tract, bridging the IHD and the stomach. The hybrid stent has a long covered portion (70 %) to prevent bile leakage between the left IHD and the stomach, and a bare portion (30 %) at the other end to avoid blockage of the side branch in the hepatic duct, and prevent migration into the hepatic duct. The antimigration flared end was placed through the stomach wall and into the stomach to prevent migration into the left IHD.


#

EUS-HDS

A linear echoendoscope was introduced into the duodenal bulb and the dilated IHDs in the right lobe were identified. A 19-gauge needle for EUS-guided fine-needle aspiration was inserted into the dilated right IHD taking care to avoid puncturing any blood vessels. The same procedure was then followed as for EUS-HGS. Finally, a fully covered SEMS 6–8 cm in length was deployed at the punctured tract, bridging the IHD and the duodenum ([Fig. 3]).

Zoom Image
Fig. 3 Endoscopic ultrasound-guided hepaticoduodenostomy procedure (EUS-HDS). a A linear echoendoscope deploying an 80-mm covered self-expandable metal stent (10 mm in diameter) during EUS-HDS. b The stent successfully connected the right intrahepatic bile duct to the duodenum.

#
#

PTBD

For the purposes of the current study, PTBD referred only to external catheter drainage. The usual size of the catheter was 8-Fr in diameter. Catheters were exchanged when patients developed recurrent biliary obstruction symptoms.


#

Sedation

Patients were placed in the supine position for ERCP and EUS procedures. At King Chulalongkorn Memorial hospital, conscious sedation with propofol was used; in other hospitals, procedures were performed with patients under general anesthesia, according to their usual protocols.


#

Follow-up period

All patients were admitted overnight after the procedure for close observation of possible post-procedure complications, and researchers were notified of any adverse events. Serum total bilirubin was tested 2–4 weeks after the biliary drainage procedure to assess clinical success. If any clinical symptoms, signs, or blood tests indicated inadequate biliary drainage, procedure-related adverse events, or RBO, patients were fully evaluated for management of such events. All recruited patients were followed up with physicians who provided subsequent health care to them. Group A patients were followed up by the endoscopist and interventionist who performed the biliary intervention, respectively, up to 6 months.


#

Definition of events and outcomes

MHBO was classified according to the Bismuth Corlette classification. Adequate biliary drainage was defined as drainage of > 50 % of the liver volume according to a recent consensus [1]. The primary study outcome was the 3-month biliary reintervention rate after adequate biliary drainage was achieved by biliary drainage procedure. Secondary outcomes were time to RBO, technical success rate, clinical success rate, and complication rate.

Procedure-related terminology for this study was based on modifications of the definitions set by the Tokyo Criteria for transpapillary stenting by ERCP [8]. The technical success rate was defined as the rate of successful deployment of stent in the desired position. EUS-HGS technical success was defined as transgastric and transhepatic deployment of the metal stent into the left IHD. EUS-HDS technical success was defined as transduodenal and transhepatic deployment of the metal stent into the right IHD. ERCP technical success was transpapillary deployment of the metal stent into either the right or left IHD. Clinical success was defined as a 50 % reduction in bilirubin level within 2 weeks of the intervention without additional biliary drainage [9]. RBO was defined as development of jaundice, cholangitis, and increased bilirubin secondary to obstruction or stent migration. Biliary reintervention was any endoscopic, percutaneous, or surgical procedure performed to improve biliary drainage after ERCP/EUS completion. Biliary reintervention was performed as needed: stents and catheters were exchanged only when patients developed RBO symptoms. Time to RBO was the time from successful biliary drainage to biliary reintervention. Follow-up time was the date of drainage procedure to last follow-up.

Adverse event severity was classified according to the American Society for Gastrointestinal Endoscopy lexicon to grade and describe adverse events [10].


#

Sample size and statistical methods

Sample size calculations were based on the primary study outcome. We assumed 3-month reintervention rates of 16 % vs. 60 % in our EUS-BD vs. PTBD groups [11] [12]. Based on these percentages, 16 patients per group would provide 75 % power to detect these differences at a two-sided significance level of 5 %. This was inflated to 19 per group to compensate for possible losses to follow-up.

Demographic characteristics were summarized by group. Technical and clinical success rates were formally compared between groups using Fisher’s exact test. The Kaplan–Meier method was used to calculate survival probabilities in patients who achieved technical success: the survival function was formally compared using a log-rank test.

Statistical analysis was conducted using Stata 15.1 (StataCorp, College Station, Texas, USA). No adjustments were made for potential prognostic confounders because disease staging and severity were comparable between groups.


#
#

Results

A total of 38 patients were eligible in both groups; however, 2 from group A were excluded due to large volume of ascites (n = 1) and an ECOG score too poor to allow endoscopic treatment (n = 1). The study flow is shown in [Fig. 4]. A total of 36 patients (24 males, 12 females) were therefore recruited into the study. Group A (n = 19), enrolled patients from March 2016 to October 2019. Controls (group B; n  =  17) underwent PTBD between July 2015 and March 2019. Mean (standard deviation [SD]) age of group A and group B patients was 62 (SD 10) years and 58.8 (SD 12.8) years, respectively. Male to female ratio was 13:6 vs. 11:6 in group A and B, respectively. Baseline characteristics including total serum bilirubin were comparable between groups ([Table 1]).

Zoom Image
Fig. 4 Study flow diagram for patients undergoing combination of endoscopic retrograde cholangiopancreatography and endoscopic ultrasound-guided biliary drainage (ERCP + EUS-BD; group A) and percutaneous biliary drainage (PTBD; group B) procedures for management of malignant hilar biliary obstruction (MHBO). ECOG, Eastern Cooperative Oncology Group; HGS, hepaticogastrostomy; HDS, hepaticoduodenostomy.
Table 1

Baseline characteristics of patients who underwent endoscopic retrograde cholangiopancreatography plus endoscopic ultrasound-guided biliary drainage versus percutaneous biliary drainage for malignant hilar biliary obstruction.

Characteristic

ERCP + EUS-BD
(n  =  19)

PTBD
(n  =  17)

Age, mean (SD), years

62 ± 10

58.8 ± 12.8

Sex, n (%)

  • Male

13 (68.4)

11 (64.7)

  • Female

6 (31.6)

6 (35.3)

Total bilirubin, mean (SD), mg/dL

17.1 (9.7)

19.3 (8.3)

Direct bilirubin, mean (SD), mg/dL

15.4 (7.7)

9.8 (6.2)

Alkaline phosphatase, mean (SD), IU/L

437.2 (228.3)

484.1 (334.8)

Diagnosis, n (%)

  • Cholangiocarcinoma

16 (84.2)

17 (100)

  • Gallbladder cancer

2 (10.5)

0

  • Metastasis from colon cancer

1 (5.3)

0

ERCP, endoscopic retrograde cholangiopancreatography; EUS-BD, endoscopic ultrasound-guided biliary drainage; PTBD, percutaneous biliary drainage; SD, standard deviation.

In group A, the following endoscopic procedures were provided to patients: ERCP plus EUS-HGS (n = 17), ERCP plus EUS-HDS (n = 1), and ERCP plus EUS-HGS plus EUS-HDS (n = 1). Technical and clinical success rates in this group were 84.2 % (16/19) and 78.9 % (15/19), respectively. ERCP and EUS-BD were performed during the same session in 7 cases (36.8 %) and separately in 12 cases (63.2 %). For the 12 cases with different sessions, all cases started with ERCP, and EUS-BD was subsequently performed within 2 weeks. Median ERCP and EUS-BD procedure times in group A were 62.5 minutes (range 23–98) and 33.5 minutes (range 29–60), respectively. Operators performing endoscopic procedures had performed > 500 ERCP and/or EUS procedures.

In group B, all patients received bilateral PTBD tubes. Technical and clinical success rates and complication rates of group A vs. B were 84.2 % (16/19) vs. 100 % (17/17; P = 0.23), 78.9 % (15/19) vs. 76.5 % (13/17; P > 0.99), and 26.3 (5/19) vs. 35.3 (6/17; P = 0.56), respectively ([Table 2]).

Table 2

Results of endoscopic retrograde cholangiopancreatography plus endoscopic ultrasound-guided biliary drainage procedures versus percutaneous biliary drainage in patients with malignant hilar biliary obstruction.

Parameter

ERCP + EUS-BD (n  =  19)

PTBD (n  =  17)

P value

Technical success rate, n (%)

16 (84.2)

17 (100)

0.23

Clinical success rate, n (%)

15 (78.9)

13 (76.5)

> 0.99

RBO

  • RBO within 3 months, n (%)

4/15 (26.7)[1]

15/17 (88.2)

0.001

  • RBO within 6 months, n (%)

2/9 (22.2)[2]

9/9 (100)[2]

0.002

  • Time to RBO, median (IQR), days

92 (56–217)

40 (13.5–57.8)

0.06

Biliary reinterventions

  • Cumulative biliary reintervention in technical success patients, until death or last follow-up, n (%)

6/16 (37.5)[3]

15/17 (88.2)

0.002

  • No. of biliary reinterventions/patient at 3 months, median (IQR)

0 (0–1)

1 (1–2.5)

< 0.001

  • No. of biliary reinterventions/patient at 6 months, median (IQR)

0 (0–1)

2.5 (1–4)

< 0.001

  • Cumulative no. of biliary reinterventions/patient until death or last follow-up, median (IQR)

0.5 (0–1)

4 (2–5)

0.003

Other outcomes, n (%)

  • Complications

5 (26.3)

6 (35.3)

0.56

  • 2-year death in technical success patients

16/16 (100)

16/17 (94.1)

> 0.99

ERCP, endoscopic retrograde cholangiopancreatography; EUS-BD, endoscopic ultrasound-guided biliary drainage; PTBD, percutaneous biliary drainage; RBO, recurrent biliary obstruction; IQR, interquartile range.

1 4 patients were excluded from the denominator due to technical failure (n = 3), and death within 3 months (n = 1).


2 In the ERCP + EUS-BD group, 10 patients were excluded from the denominator due to technical failure (n = 3), and death within 6 months (n = 7). In the PTBD group, 8 patients were excluded from the denominator due to death within 6 months.


3 3 patients were excluded from the denominator due to technical failure.


In patients with technical success who were alive up to 3 months, and from 3–6 months, the RBO rates of group A vs. group B were 26.7 % (4/15) vs. 88.2 % (15/17; P  = 0.001), and 22.2 % (2/9) vs. 100 % (9/9; P  = 0.002), respectively. At 3 months, the median number of biliary reintervention procedures in group A was significantly lower than in group B (0 [interquartile range (IQR) 0–1] vs. 1 [IQR 1–2.5], respectively; P < 0.001). After 6 months, the median number of biliary reintervention procedures in group A was the same as that observed at 3 months, and the median number in group B was 2.5 (IQR 1–4; P  < 0.001). Group A had a significantly longer median time to development of RBO than group B (92 [IQR 56–217] vs. 40 [13.5–57.8] days, respectively; P = 0.06) ([Table 2]). The cumulative biliary reintervention rates until patients’ death or last follow-up in group A vs. group B were 6/16 (37.5 %) vs. 15/17 (88.2 %; P  = 0.002), respectively. Number of biliary reintervention procedures per patient until patients’ death or last follow-up in group A vs. group B were 0.5 (IQR 0–1) vs. 4 (IQR 2–5; P  = 0.003), respectively ([Table 2]).

Details of group A patients with RBO within 3 months were: tumor ingrowth in the first ERCP stent, which was corrected with a second stent insertion (n = 1); tumor ingrowth in an ERCP stent, which was treated with PTBD (n = 1); localized cholangitis from bile duct obstruction in segment 3, which was also corrected with PTBD (n = 1: pre-existing HGS stent was in segment 2); and food debris in the HGS stent, which was treated with balloon extraction through the HGS stent (n = 1). Two group A patients experienced RBO after 3 months but before 6 months; one had tumor outgrowth in the HGS stent and was treated with PTBD, and the other had food debris at the gastric end of the HGS stent, which was removed with a gastroscopy procedure.

In patients with technical success, overall survival between arms was not significantly different (log rank P  =  0.19; see Fig. 1 s in the online-only supplementary material). The 25th percentile and median survival were 2.2 and 5 months in group A vs. 4.4 and 7.8 months in the PTBD group. Mortality rates for patients who had reached 2 years of follow-up were 16/16 (100 %) in group A and 16/17 (94.1 %) in group B.

[Table 3] shows procedure-related adverse events. In group A, there were four patients with mild and one patient with a moderate grade adverse event. In group B, there were two severe and four moderate grade complications. All patients responded to treatment; no statistically significant differences in procedure-related complications between groups were noted.

Table 3

Type, grade, length of stay, and management of complications of endoscopic and percutaneous biliary drainage treatment in patients with malignant hilar biliary obstruction.

Parameters

ERCP + EUS-BD

PTBD

Rate of complications

26.3 % (5/19)

35.3 % (6/17)

Type of complications

  • Symptomatic bile leak (n = 1)

  • Bleeding in the liver from left hepatic artery (n = 1)

  • Cholangitis (n = 2)

  • Pancreatitis (n = 1)

  • Abdominal pain (n = 4)

  • Fever (n = 1)

  • Liver abscess (n = 1)

Grade and length of stay

  • Bile leak; mild; 3 days

  • Bleeding; moderate; 4 days

  • Cholangitis: mild; 2 days

  • Pancreatitis; mild; 2 days

  • Abdominal pain; moderate; 7, 7, 13, and 20 days

  • Fever (n = 1); severe; 14 days

  • Liver abscess (n = 1); severe; 21 days

Management

  • Angiogram plus embolization in 1 patient with bleeding

  • Conservative management for the rest

  • Abdominal pain; conservative management (n = 3), revise PTBD twice (n = 1)

  • Fever; antibiotics

  • Liver abscess; antibiotic and aspiration twice

Outcomes

  • Success in all cases

  • Success in all cases

ERCP, endoscopic retrograde cholangiopancreatography; EUS-BD, endoscopic ultrasound-guided biliary drainage; PTBD, percutaneous biliary drainage.

For the three patients in whom endoscopic procedures failed, details were as follows. In the first patient, we failed to pass a mechanical dilator (Soehenda dilator, 8.5 Fr) across the gastric wall into the liver. PTBD with external drainage was consequently used and total bilirubin level decreased to within the normal range. In the second patient, we failed to place an SEMS into the right IHD. PTBD with antegrade metal stent placement was consequently performed successfully. In the third patient, we failed to dilate the punctured tract of EUS-HGS with cystotome 6 Fr, so aborted the procedure. PTBD with external drainage was used to salvage the procedure. No procedure-related complications were observed in these three patients.


#

Discussion

This study found a highly significantly lower risk of biliary reintervention with consequent lower morbidity, in patients treated with a combination of ERCP and EUS-BD procedures compared with those treated with PTBD. In the first 3 months when the majority of patient were still alive, the cumulative rate of biliary reintervention in the endoscopy group was significantly lower than that in the PTBD group. Moreover, endoscopic procedures had a lower median number of biliary reintervention procedures at both 3 and 6 months, plus a longer time to RBO despite no survival difference. This information favors endoscopy over PTBD for long-term care of MHBO, predominantly because endoscopy provides a larger diameter of drainage tube than PTBD. For ERCP, two randomized controlled trials have shown that SEMSs have a longer patency time than plastic stents [13] [14], which has also been confirmed by our group in two other clinical trials of biliary drainage in MHBO [15] [16].

ESGE guidelines now recommend PTBD, or a combination of PTBD and ERCP, as palliative biliary drainage procedures for Bismuth types III and IV MHBO [3], in order to achieve adequate liver volume drainage, as ERCP alone with multisegmental biliary drainage is technically challenging due to stricture tightness [1] [2]. Nevertheless, external drainage by PTBD adversely affects quality of life because of external lines and bags, and pain, particularly in the days following the procedure. ERCP with SEMS is, therefore, preferred over PTBD in practice [17]. In a systematic review, ERCP with multisegmental drainage in MHBO had a lower rate of successful biliary drainage compared with PTBD [17]. However, the latter was inconvenient in terms of maintenance, making EUS-BD an interesting alternative in achieving adequate and complete biliary drainage for patients with MHBO. Despite this, there are limited data supporting the use of EUS-BD in MHBO. A recent review reported that only 88 patients from 11 published studies underwent EUS-BD in MHBO, and only four studies had ≥ 10 patients. These four studies employed EUS-BD alone for MHBO as an initial or rescue biliary drainage procedure. Moreover, some patients in these studies were Bismuth I and II MHBO, conditions that do not require multiple stents for adequate and complete biliary drainage. In our opinion, ERCP with multisegmental drainage should still be the first choice, as the SEMS placement involves passage through a natural orifice, not a transmural route as in EUS-BD. EUS-BD should be used as a rescue method, or combined to achieve adequate and complete biliary drainage in MHBO. We therefore decided to initiate the current study to evaluate results of a combination of ERCP and EUS-BD procedures in patients with Bismuth III and IV MHBO.

EUS-BD is an emerging endoscopic biliary drainage procedure predominantly used for drainage for distal biliary obstruction. Current indications for EUS-BD include failed ERCP in experienced hands, inaccessible major papilla, surgically altered anatomy, conversion of PTBD to EUS-BD, and as a primary biliary drainage method, as recently reported [18]. A recent systematic review and meta-analysis of nine studies with 483 patients comparing EUS-BD with PTBD in distal biliary obstruction, concluded there was no difference in technical success rate between these procedures, but EUS-BD was associated with better clinical success, fewer post-procedure adverse events, and lower reintervention rates [19]. Regarding biliary drainage in MHBO, a systematic review of nine studies (n = 546) compared ERCP with PTBD, and found that the pooled odds ratio for successful biliary drainage in PTBD vs. endoscopic biliary drainage was 2.53 (95 % confidence interval 1.57–4.08) whereas PTBD was comparable to ERCP in overall adverse effects and 30-day mortality [17]. To the best of our knowledge, no study to date has compared EUS-BD vs. PTBD in MHBO. Results from the current study show that a combination of ERCP and EUS-BD in MHBO had comparable technical and clinical success rates, and complication rates compared with PTBD. This point is very interesting, as when ERCP was combined with EUS and used for MHBO biliary drainage, the success rate was comparable to PTBD. Given these improvements in providing internal drainage, longer RBO time, and lower RBO rate, the combination of ERCP and EUS-BD procedures may become the biliary drainage method of choice in patients with Bismuth III and IV MHBO in the future.

One important limitation of our study comes from its observational nature. Patients recruited to group B were historical controls, so the possibility of bias exists. In future studies, randomizing patients to receive either a combination of ERCP and EUS-BD procedures or PTBD will allow evaluation of clinical outcomes with less bias.

In patients whose liver function is well preserved, 33% of liver volume should be drained for adequate biliary drainage; however, 50% should be drained if the liver function is impaired [20] [21]. To accomplish adequate biliary drainage in MHBO, multisegmental biliary drainage is required. A recent randomized controlled trial showed superiority of bilateral over unilateral SEMS placement in MHBO. The former had longer stent patency time than the latter (252 vs. 139 days; P  < 0.01) [22]. However, conventional ERCP with bilateral SEMS placement is technically challenging, with reported technical success rates ranging from 71% to 95.5% [22] [23]. Results of the current study show that the combination of ERCP and EUS-BD procedures has a high technical success rate and satisfactory biliary reintervention rate. Therefore, a combination of ERCP and EUS-BD procedures is another potential alternative for biliary drainage in MHBO, in addition to PTBD or PTBD plus ERCP as previously recommended. A study on ERCP with multisegmental drainage vs. a combination of ERCP and EUS-BD procedures is warranted.


#
#

Competing interests

The authors declare that they have no conflicts of interest.

Acknowledgment

This study was funded by the Pancreas Research Unit, Faculty of Medicine, Chulalongkorn University Health Systems Research Institute (HSRI), grant number HSRI 62-050.

Supplementary material


Corresponding author

Pradermchai Kongkam, MD
Gastrointestinal Endoscopy Excellence Center and Pancreas Research Unit Department of Medicine, Faculty of Medicine
Chulalongkorn University and King Chulalongkorn Memorial Hospital
Thai Red Cross Society
1873 Rama 4 Road
Pathumwan, Bangkok
Thailand 10330   
Fax: +66-2-6524219   

Publication History

Received: 28 December 2019

Accepted: 08 June 2020

Publication Date:
08 June 2020 (online)

© 2020. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


Zoom Image
Fig. 1 Malignant hilar biliary obstruction (MHBO). a Computed tomography of the abdomen in a patient with MHBO. b Two percutaneous transhepatic biliary drainage catheters from right and left lobes of the liver. c A hybrid metal stent (partially covered with the antimigration flared end on the gastric side (GIOBORTM Biliary stent; Taewoong, Gyeonggi-do, South Korea) connecting the left intrahepatic bile duct (segment 3) to the gastric lumen. d An uncovered self-expandable biliary metal stent draining bile fluid from the right intrahepatic bile duct to the duodenum.
Zoom Image
Fig. 2 Methods used for applying a combination of endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic ultrasound-guided biliary drainage (EUS-BD) in patients with malignant hilar biliary obstruction (MHBO). a A hilar tumor causing MHBO, with an upstream dilated intrahepatic bile duct (IHD). b ERCP with stent placement into the right IHD system plus EUS-guided hepaticogastrostomy. c ERCP with stent placement into the left IHD system plus EUS-guided hepaticoduodenostomy. d ERCP with failed stent placement across the obstructed area; both EUS-guided hepaticogastrostomy and EUS-guided hepaticoduodenostomy are then performed to accomplish biliary drainage.
Zoom Image
Fig. 3 Endoscopic ultrasound-guided hepaticoduodenostomy procedure (EUS-HDS). a A linear echoendoscope deploying an 80-mm covered self-expandable metal stent (10 mm in diameter) during EUS-HDS. b The stent successfully connected the right intrahepatic bile duct to the duodenum.
Zoom Image
Fig. 4 Study flow diagram for patients undergoing combination of endoscopic retrograde cholangiopancreatography and endoscopic ultrasound-guided biliary drainage (ERCP + EUS-BD; group A) and percutaneous biliary drainage (PTBD; group B) procedures for management of malignant hilar biliary obstruction (MHBO). ECOG, Eastern Cooperative Oncology Group; HGS, hepaticogastrostomy; HDS, hepaticoduodenostomy.