Endoscopic retrograde cholangiopancreatography versus endoscopic ultrasound-guided primary biliary drainage for malignant distal biliary obstruction: a review

Abstract

Approximately 70% of pancreatic cancers arising in the head of the pancreas cause malignant distal biliary obstruction. Therefore, biliary drainage is an important aspect of the management of these patients. Conventionally, the technique for relieving biliary obstruction has been endoscopic retrograde cholangiopancreatography (ERCP). When ERCP cannot provide biliary drainage, such as in cases of failed cannulation, duodenal obstruction, or altered anatomy (e.g., after gastric bypass), endoscopic ultrasound-guided biliary drainage (EUS-BD) has been recommended by the American Society for Gastrointestinal Endoscopy as the preferred technique over percutaneous transhepatic biliary drainage for relieving biliary obstruction. Recently, EUS-BD has been explored as the primary technique for relieving biliary obstruction. The proposed benefits include higher technical success, shorter procedural time, a lower risk of pancreatitis, and longer stent patency. However, the results of randomized controlled studies have been conflicting. Hence, the aim of this article is to review the evidence from randomized controlled trials comparing EUS-BD with ERCP as the primary method of biliary drainage for malignant distal biliary obstruction.

INTRODUCTION

Pancreatic cancer is a deadly condition, with only 15%–20% of the patients being operable at the time of diagnosis.1 The remaining patients present with either metastatic disease (40–50%) or locally advanced unresectable pancreatic cancer due to extensive vascular involvement (30%–40%). Approximately 70% of pancreatic cancers arising in the head of the pancreas cause malignant distal biliary obstruction (MDBO).2 Therefore, biliary drainage is an important aspect of the management of these patients. Conventionally, the technique for relieving biliary obstruction has been endoscopic retrograde cholangiopancreatography (ERCP). When ERCP cannot provide biliary drainage, such as in cases of failed cannulation, duodenal obstruction, or altered anatomy (e.g., after gastric bypass), endoscopic ultrasound-guided biliary drainage (EUS-BD) has been recommended by the American Society for Gastrointestinal Endoscopy as the preferred technique over percutaneous transhepatic biliary drainage for relieving biliary obstruction.3

Recently, EUS-BD has been explored as the primary technique for relieving biliary obstruction. The proposed benefits include a higher technical success rate, shorter procedural time, a lower risk of pancreatitis, and longer stent patency.2,4-8 However, the results of randomized controlled studies have been conflicting. Hence, the aim of this article is to review the evidence from randomized controlled trials (RCTs) comparing EUS-BD with ERCP as the primary method of biliary drainage for MDBO.

ENDOSCOPIC RETROGRADE CHOLANGIOPANCREATOGRAPHY

ERCP with the insertion of a self-expanding metallic stent (SEMS) is currently the preferred, first-line endoscopic method for treating MDBO, with a success rate of up to 95% (Fig. 1).9 Compared with surgical drainage, the procedure is associated with significantly lower rates of morbidity and mortality.10 Compared with percutaneous transhepatic biliary drainage, the procedure is associated with lower rates of adverse events (AEs) and the need for reintervention, shorter hospital stays, and lower total cost.11 The type of metal stent used in ERCP significantly affects stent patency. The available stents include uncovered SEMSs (U-SEMSs), fully covered SEMSs (FC-SEMSs), and partially covered SEMSs. In a meta-analysis of seven randomized studies with 1,070 patients, no significant differences were observed among the major outcomes.12 Cumulative stent patency, failure rate, overall survival, and AE rates were similar between the covered SEMS and U-SEMS groups. The stent migration rate and tumor overgrowth rate were higher in the covered SEMS group (risk ratio [RR], 2.34; 95% confidence interval [CI], 1.35–4.08 and RR, 2.05; 95% CI, 1.13–3.72, respectively). The tumor ingrowth rate was higher in the U-SEMS group (RR, 0.25; 95% CI, 0.11–0.61). To overcome the risk of migration of FC-SEMS, endoscopists have investigated the effect of inserting a double-pigtail plastic stent (DPS) through the SEMS to help anchor the stent and prevent migration. In a randomized study, the mean stent patency was significantly longer in the FC-SEMS plus DPS group (237 days [95 % CI, 199–275] vs. 173 days [95% CI, 130–217]; p=0.048).13 There were no significant differences in stent-related AEs or overall survival rates at 6 months between the two groups. Thus, when inserting an FC-SEMS, consideration should be given to adding a DPS to prevent migration.

Fig. 1.

Endoscopic retrograde cholangiopancreatography with self-expanding metallic stent (SEMS) insertion. (A) Endoscopic view of the papilla seen after SEMS deployment. (B) X-ray view showing the stent in the bile duct after SEMS deployment.

ENDOSCOPIC ULTRASOUND-GUIDED BILIARY DRAINAGE

EUS-BD is now an established technique for biliary drainage in cases of failed ERCP.14 This procedure has also emerged as an alternative primary drainage method for patients with MDBO based on several potential advantages.14-16 First, the stent can be placed away from the tumor to reduce tumor ingrowth. Furthermore, this procedure may be associated with a higher technical success rate and shorter procedural time. Finally, as the procedure does not involve manipulation of the pancreatic duct, it is associated with no risk of pancreatitis.

EUS-BD can be performed using the transmural or transpapillary approaches. In the transpapillary approach, EUS-BD can be performed using EUS-guided rendezvous–ERCP (EUS-Rv) or EUS-antegrade stenting (EUS-AG).14,17 In EUS-Rv, the bile duct is first punctured under EUS guidance, and a guidewire is then passed through the papilla to facilitate cannulation with standard ERCP. In EUS-AG, the intrahepatic duct is punctured under EUS guidance, followed by antegrade insertion of a stent through the papilla. In the transmural approach, EUS-BD can be performed using EUS-guided choledochoduodenostomy (EUS-CDS) or EUS-guided hepaticogastrostomy (EUS-HGS). In EUS-CDS, an echoendoscope is used to visualize the bile duct from the duodenum, followed by the insertion of a stent from the duodenum into the common bile duct through the duodenal wall. In EUS-HGS, an echoendoscope is used to visualize the left intrahepatic duct from the stomach, followed by the insertion of a SEMS from the left intrahepatic duct to the stomach. For the transmural approach, excellent 1-year stent patency rates of 73.7% to 92.3% have been reported.18-20 In a meta-analysis of 13 studies (2 RCTs and 11 observational studies) including 281 patients who underwent CDS and 267 patients who underwent HGS,3 no significant differences were found between the two modalities in technical success (odds ratio [OR], 1.35; 95% CI, 0.56–3.29), clinical success (OR, 1.31; 95% CI, 0.63–2.72), AEs (OR, 0.76; 95% CI, 0.35–1.63), or the need for reintervention (OR, 1.83; 95% CI, 0.86–3.87).

ADVERSE EVENTS OF EUS-BD

In a pooled analysis of 7,887 patients, AEs occurred in 13.7% of patients, including bile leak, cholangitis, pancreatitis, cholecystitis, bleeding, stent misdeployment (SMD), stent migration, and stent blockage.21 Bile leak was the most common AE, followed by cholangitis. The types of AEs that may occur differ according to the type of EUS-BD performed. EUS-AG was associated with the highest risk of pancreatitis, whereas EUS-CDS with tubular SEMS was associated with the highest risk of cholecystitis. SMD is most commonly observed in EUS-CDS with lumen-apposing metal stent (LAMS).

Regarding SMD after EUS-CDS with LAMS, a post hoc analysis of two RCTs comparing EUS-CDS with LAMS vs. ERCP in MDBO classified SMD into four types: type 1, misdeployment of the distal flange; type 2, misdeployment of the proximal flange; type 3, contralateral bile duct wall injury; and type 4, double mucosal puncture.22 SMD occurred in 11 patients: eight with type I, one with type II, and two with type III. Endoscopic salvage of SMD was successful in 81.8% of cases. Misdeployment led to AEs in four patients (two mild, two moderate), giving an overall SMD-related AE rate of 2.6% (95% CI, 0.7%–6.6%). On multivariable analysis, an extrahepatic bile duct diameter of ≤ 15 mm was associated with increased odds of SMD and/or technical failure. Notably, the occurrence of SMD does not always lead to an AE. Even when SMD has occurred, provided that a stent can be placed in the original tract using the guidewire placed at the initial puncture and biliary drainage can still be achieved, an AE may not occur. Hence, management by an experienced operator is essential when SMD occurs.

ERCP VS. EUS FOR PRIMARY BILIARY DRAINAGE IN MDBO

Six RCTs have compared ERCP with primary EUS-BD for MDBO (Table 1).2,4-8 Earlier studies utilized tubular biliary SEMS for EUS-BD (Fig. 2).4-6 More recent studies employed cautery-enhanced LAMS to perform EUS-CDS (Fig. 3).2,7,8 In earlier studies, EUS-BD involved a multistep procedure, beginning with the puncture of the bile duct with a 19-gauge needle, followed by guidewire insertion, dilation of the needle tract, and placement of tubular stents. In more recent studies, EUS-CDS was performed with a single-step cautery-enhanced LAMS, eliminating the need for a multistep procedure. This significantly simplified the procedure and facilitated its performance. In addition, the EUS-CDS procedure is more standardized and the outcomes are more comparable. The main drawback of LAMS is the risk of sump syndrome, in which the bile duct below the LAMS acts as a reservoir for food debris, causing recurrent cholangitis.23 This condition is less likely to occur with EUS-CDS performed using tubular stents. This issue is reflected in the long-term patency data for EUS-CDS with LAMS, with reported stent dysfunction rates of up to 55%.23-25 To overcome this issue, Sumalla-Garcia et al.26 investigated the effect of adding a double plastic double-pigtail stent (DPS) to the LAMS. It was observed that recurrent biliary obstruction was lower in the LAMS-DPS group (30% vs. 9%; p=0.024) and hospitalization was shorter (median difference, 4.5; 95% CI, 0–9; p=0.016), potentially overcoming this drawback of LAMS.

Table 1.

Characteristics of published randomized controlled trials comparing EUS-BD with ERCP for malignant distal biliary obstruction

Study	Year	No. of patients (EUS/ERCP)	Type of EUS-BD stent used	Technical success (%)	Clinical success (%)	Stent patency	Adverse event rate (%)
Bang et al.6	2018	33/34	Tubular SEMS	90.9, 94.1	97.0, 91.2	182 vs. 170 days (p=0.51)	21.2, 14.7 (p=0.49)
Paik et al.4	2018	64/61	Tubular SEMS	93.8, 90.2	90.0, 94.5	6 months, 85.1% vs. 48.9% (p=0.001)	10.9, 39.3 (p=0.03)
Park et al.5	2018	15/15	Tubular SEMS	100.0, 92.8	92.8, 100	188 vs. 197 days (p=0.941)	0.0, 0.0 (p=1)
Teoh et al.2 (DRA-MBO)	2023	79/76	EC-LAMS	96.2, 76.3	93.7, 90.8	1-year: 91.1%, 88.1% (p=0.65)	19.2, 21.1 (p=1)
Chen et al.7 (ELEMENT)	2023	73/71	EC-LAMS	91.8, 76.1	87.7, 71.8	1-year stent dysfunction rate: 15.1%, 22.5% (p=0.29)	16.5, 17.1 (p=0.78)
Anderloni et al.8	2026	111/109	EC-LAMS	94.6, 78.9	86.5, 88.1	6-month stent dysfunction rate: 32.4%, 34.9% (p=NS)	19.8, 21.1 (p=NS)

EUS-BD, endoscopic ultrasound-guided biliary drainage; ERCP, endoscopic retrograde cholangiopancreatography; NS, not significant.

Fig. 2.

Endoscopic ultrasound-guided choledochoduodenostomy with tubular self-expanding metallic stent (SEMS). (A) The common bile duct was punctured in a long-scope position in the first part of D1, and a cholangiogram was performed. (B) This was followed by insertion of a guidewire and the stent after tract dilation with a cystotome and a balloon. (C) X-ray view showing deployment of the SEMS in the bile duct. (D) Endoscopic view of the deployed stent from the duodenum.

Fig. 3.

Endoscopic ultrasound (EUS)-guided choledochoduodenostomy with lumen-apposing metal stent (LAMS). (A) The common bile duct was punctured from the first part of the duodenum under EUS guidance. (B) This was followed by insertion of a guidewire into the bile duct. (C) The distal flange of the LAMS was deployed under EUS guidance. (D) Endoscopic view of the deployed proximal flange of the LAMS in the duodenum with a double-pigtail stent inserted.

The three studies published in 2018 all had methodological limitations. The studies by Bang et al.6 and Park et al.5 included very few patients and were not sufficiently powered to detect differences in stent patency. In Paik et al.’s study,4 stent patency in the ERCP arm was exceptionally low (48.9% at 6 months), and both EUS-HGS and EUS-CDS were performed in the EUS arm. In three recent studies, the EUS-CDS procedure was standardized using cautery-enhanced LAMS. The comparable results for technical success, clinical success, stent patency, and AEs support the reproducibility of EUS-CDS outcomes with a standardized device and technique.

A meta-analysis including five RCTs and 519 patients was published before the study by Anderloni et al.27 When comparing EUS-BD with ERCP, the technical and clinical success rates were similar: 92.6% vs. 85.4% (RR, 1.06; 95% CI, 0.96–1.17; p=0.27) and 89.2% vs. 87.7% (RR, 1.02; 95% CI, 0.97–1.08; p=0.45), respectively. The 1-year stent patency was also similar at 85.1% vs. 76.2% respectively (RR, 1.15; 95% CI, 0.94–1.42; p=0.17). The reintervention rate was significantly lower with EUS-BD (11.4% vs. 17.6%; RR, 0.58; 95% CI, 0.37–0.9; p=0.01). The AE rate was similar (14.3% vs. 19.8%; RR, 0.85; 95% CI, 0.49–1.46; p=0.55). The procedure time was significantly shorter in the EUS-BD group (standardized mean difference [SMD], –2.36; 95% CI, –2.68 to –2.05; p<0.001). In subgroup analyses, EUS-CDS with LAMS showed higher technical success than ERCP (RR, 1.17; 95% CI, 1.01–1.35; p=0.03). In the three studies in which EUS-CDS was performed using tubular SEMSs, there was a trend toward higher 1-year stent patency than with ERCP (RR, 1.48; 95% CI, 0.97–2.24; p=0.06), whereas the corresponding RR for LAMS versus ERCP was 1.02 (95% CI, 0.94–1.10; p=0.47).

In a more recent network meta-analysis including six RCTs comparing EUS-CDS with LAMS and tubular SEMSs, EUS-CDS with LAMS had higher technical success (RR, 1.22; 95% CI, 1.10–1.36) and a shorter procedure time (SMD, –0.78; 95% CI, –1.30 to 0.27).28 EUS-CDS with SEMS showed the lowest overall risk of AEs and cholangitis. Thus, these findings support the occurrence of sump syndrome with EUS-CDS using LAMS. Hence, if EUS-CDS can eventually be performed using a single-step device that allows placement of tubular stents, the procedure may provide the benefits of higher technical success, shorter procedure time, and improved stent patency.

Lastly, in the most recent study, which used the risk of pancreatitis as the primary endpoint, EUS-CDS with LAMS was associated with a lower risk of postprocedural acute pancreatitis (1.8% vs. 7.3%; RR, 0.25; 95% CI, 0.07–0.88), higher technical success (94.6% vs. 78.9%, p<0.001) and shorter procedural time (13.5±11.6 vs. 24.7±14.9 minutes; p<0.001).8 This highlights an additional advantage of EUS-CDS.

CONCLUSIONS AND FUTURE DIRECTIONS

In conclusion, ERCP and EUS-CDS are both acceptable approaches for primary drainage of MDBO. EUS-CDS with LAMS appears to be associated with significantly higher technical success and a lower risk of reintervention and postprocedural pancreatitis. However, the long-term patency of LAMS procedures remains an issue. A single-step system that allows EUS-CDS to be performed with the placement of tubular SEMSs may potentially overcome this issue.

Furthermore, an experienced operator is essential to minimize the risk of AEs during EUS-BD. The availability of single-step devices significantly simplifies EUS-CDS and may reduce the learning curve associated with the procedure. However, the cost of the device remains a concern for resource-limited settings, and whether EUS-CDS with LAMS as primary biliary drainage is more cost-effective than ERCP remains uncertain. Future studies are required to address this issue.

While many endoscopists will still attempt ERCP with SEMS insertion first, EUS-BD should be considered promptly when cannulation is difficult, and the choice of technique should be individualized based on the specific clinical situation. In patients with duodenal obstruction, EUS-BD can be performed using EUS-CDS or EUS-HGS when the first part of the duodenum is not involved. However, if the first part of the duodenum is affected by the tumor, biliary drainage can generally only be performed using EUS-AG or EUS-HGS if the left hepatic ducts are dilated. Therefore, proficiency in all EUS-BD techniques is essential for optimal patient management.