Clinical Outcomes and Predictors of Early Adverse Events in Primary EUS-Guided Choledochoduodenostomy for Malignant Distal Biliary Obstruction

Keywords

endosonography - obstructive jaundice - drainage - EUS-guided biliary drainage

Introduction

Endoscopic ultrasound-guided biliary drainage (EUS-BD) was first reported as an alternative for failed endoscopic retrograde cholangiopancreatography (ERCP).[1] [2] Since then, it has been widely spread and is now also used for patients in whom ERCP is expected to be difficult.[3] Several recent studies have compared EUS-BD and ERCP for primary drainage in ERCP-accessible cases.[4] [5] [6] [7] [8] [9] A systematic review concluded that EUS-BD, which is characterized by no risk of procedure-related pancreatitis, a short procedure time, and ease of use, is an attractive primary approach for malignant biliary obstruction in specialized centers that have expert endoscopists.[10] In particular, EUS-guided choledochoduodenostomy (EUS-CDS) has gained increasing attention as a primary drainage method due to its direct access route and favorable outcomes. However, despite its growing adoption in clinical studies, primary EUS-CDS has not yet been widely implemented in clinical practice. This is likely due to concerns about procedural safety and the lack of standardized techniques, particularly regarding adverse event risk. Moreover, evidence remains limited regarding the safety profile of primary EUS-CDS, especially with respect to early procedure-related adverse events and their predictors. Existing studies combine different EUS-BD techniques, making it difficult to isolate risk factors specific to EUS-CDS. Furthermore, there is a lack of large single-center studies that systematically analyze the technical evolution, device selection, and procedural nuances that may influence safety in this setting. To address these gaps, we conducted a retrospective analysis of a large consecutive cohort undergoing primary EUS-CDS using covered self-expandable metal stents (CSEMS) at a single high-volume center. Our institution has performed primary EUS-CDS since 2010 and was among the first to report its feasibility and safety.[11] Over time, we have introduced several technical refinements, including the adoption of forward-viewing (FV) EUS scopes and dilation-free stent delivery systems.

The aim of this study is to evaluate the incidence and predictors of early adverse events following primary EUS-CDS for malignant distal biliary obstruction. We hypothesize that procedure-related factors, including scope type and tract dilation method, play a significant role in early adverse event risk. By elucidating these factors, this study seeks to inform optimal practice and improve procedural safety in the use of primary EUS-CDS, potentially facilitating its broader adoption in real-world clinical settings.

Materials and Methods

Study Design

This study was a single-center retrospective analysis. Of 306 consecutive patients in whom EUS-CDS was attempted between September 2003 and February 2025, we conducted a retrospective study of 190 cases of primary EUS-CDS using a CSEMS for malignant distal biliary obstruction. Some patients included in this study had participated in previous clinical trials, and consent for the secondary use of their data was appropriately obtained. Clinical data were retrospectively reviewed, and technical success was assessed based on an intention-to-treat approach. Informed consent was obtained from all participants prior to the procedure, and the study received approval from the institutional review board (approval number: 2024–0-509).

Patient Selection

We included only patients in whom EUS-CDS was chosen as the initial drainage method, even though ERCP was technically feasible. We excluded patients in whom EUS-CDS was performed as a rescue procedure after ERCP failure, and those in whom ERCP was not attempted due to anatomical difficulties such as duodenal stenosis or duodenal stent placement. Among the 190 patients initially considered for EUS-CDS, 26 cases were deemed unsuitable (e.g., due to severe ascites or tumor infiltration into the duodenal bulb) before puncture on preprocedural imaging. As puncture was not performed in these cases, they were excluded from the intention-to-treat (ITT) analysis ([Fig. 1]).

Fig. 1 Flowchart of patient selection process.

EUS-Guided Choledochoduodenostomy

A total of seven endoscopists performed the EUS-CDS procedures. All patients were sedated intravenously and were administered prophylactic antibiotics via the same route. The EUS-CDS technique has evolved over time. In the early period,[11] it was primarily performed using an oblique-viewing (OV) EUS scope (GF-UCT240, GF-UCT260; Olympus Medical Systems, Tokyo, Japan) or a FV EUS scope (TGF-UC260J, Olympus). The extrahepatic bile duct was punctured using a 22-G fine needle, after which contrast medium was injected through the needle to evaluate the biliary tree. A needle knife (Zimmon Needle Knife Papillotome; Cook Medical, Bloomington, Indiana, United States) was then inserted into the extrahepatic bile duct under real-time EUS guidance. The tract was dilated using 6- and 9-Fr tapered biliary dilation catheters (Soehendra Biliary Dilation catheters; Cook Medical) over the guidewire. Finally, a fully or partially covered self-expandable metallic stent (CSEMS), 10 mm in diameter and 4 or 6 cm in length (Wallflex, Boston Scientific Co., Natick, Massachusetts, United States; or X-SuitNIR, Olympus Medical Systems) was inserted. Following stent placement, the distal end was reoriented from the oral to the anal direction using either endoscopic manipulation or guidewire assistance.

In the intermediate period (after 2011), coaxial electric cautery became available, and the standard approach involved puncturing with a 19-G fine needle, followed by tract dilation using a diathermic dilator (Cysto-Gastro-Set; Endo-Flex, Voerde, Germany) over the guidewire and placement of an FCSEMS. Regarding endoscopy, both OV and FV EUS scopes were employed up to 2013, after which FV EUS scopes became the mainstay.

In the technique used currently (since 2020),[12] the extrahepatic bile duct was punctured using a 19-G Franseen needle. After aspirating bile, contrast medium was injected to visualize the biliary tree and verify the position of the perihilar duct. Without prior tract dilation, an FCSEMS was delivered through the duodenal bulb over a guidewire. Regarding the choice of FCSEMS, either an 8 mm × 6 cm BileRush Advance (7-Fr delivery system; PIOLAX, Yokohama, Japan) or an 8 mm × 6 cm HANAROSTENT Benefit (5.9-Fr delivery system; Boston Scientific Co.) was used.

Within 24 hours postprocedure, all patients underwent blood tests and imaging, either computed tomography or plain radiography, to evaluate for potential adverse events, including stent migration or bile leakage, and to ensure the distal end of the stent was correctly oriented in the anal direction.

Outcome Measurements and Definitions

In cases in which EUS-CDS was deemed unsuitable due to conditions such as severe ascites and duodenal bulb infiltration, the procedure was discontinued before puncture and these patients were excluded from the analysis. Data for patients in whom puncture was performed were included, with technical success defined as the proper placement of CSEMS. Clinical success was defined as improvement of symptoms related to biliary obstruction, such as jaundice or fever, combined with a decrease of total bilirubin or alkaline phosphatase to less than half of the highest preprocedural levels within 14 days.[12] Early adverse events were defined as complications associated with the procedure that developed within 14 days after its completion and were evaluated based on the American Society for Gastrointestinal Endoscopy lexicon.[13]

Statistical Analysis

All data were analyzed using StatMate V statistical software (ATMS). Categorical parameters were compared using a chi-squared test, and multivariate analysis was performed by logistic regression analysis. Differences were considered statistically significant when p-values were less than 0.05.

Results

The patient characteristics are summarized in [Table 1]. The median age was 67 years (range, 37–94 years) and 103 were male. The primary disease was pancreatic cancer in 159 of 190 (83.7%) cases.

Table 1 Patient characteristics

		N = 190
Age, median [range]		66 [37–94]
Sex
	Male	103
	Female	87
Primary disease
	Pancreatic cancer	159
	Lung cancer	7
	Ampullary cancer	5
	Gallbladder cancer	3
	Breast cancer	3
	Colon cancer	2
	Gastric cancer	2
	Other	9
Ascites
	Absent	124
	Present	66

EUS-CDS was deemed unsuitable before puncture in 26 of 190 patients (13.7%). The reasons for ineligibility and the alternative drainage methods are summarized in [Table 2]. The most common reason was the presence of ascites along the puncture route (n = 9). The alternative drainage methods were transpapillary endoscopic biliary stenting (EBS) (n = 21) and EUS-guided hepaticogastrostomy (EUS-HGS) (n = 5). [Table 3] lists the equipment used in EUS-CDS.

Table 2 Reasons for EUS-CDS ineligibility and alternative drainage

	N = 26
Reason for EUS-CDS ineligibility
Ascites along the puncture route	9
Duodenal invasion	5
Duodenal bulb ulcer	2
Insufficient distal bile duct length below the hilum for safe stent placement	2
Vessel along the puncture route	2
Anatomical deformation	2
Gallbladder along the puncture route	1
Cystic duct along the puncture route	1
Biliary invasion	1
Biliary bleeding	1
Alternative drainage
EBS	21
EUS-HGS	5

Abbreviations: EUS-CDS, endoscopic ultrasound-guided choledochoduodenostomy; EBS, endoscopic biliary stenting; EUS-HGS, EUS-guided hepaticogastrostomy.

Table 3 Equipment used in EUS-CDS

Scope	FV	148	(77.9%)
	OV	42	(22.1%)
Needle	19-G FNA needle	88	(46.3%)
	19-G FNB needle	56	(31.1%)
	22-G FNA needle	18	(9.5%)
	Needle knife	2	(1.1%)
	Discontinued before puncture	26	(14.4%)
Dilator	Needle-knife and mechanical dilator	16	(9.8%)
	Electric cautery	85	(51.3%)
	Without tract dilation	57	(34.8%)
	Mechanical dilator	2	(1.2%)
	Drill dilator	1	(0.6%)
	Discontinued before dilation	3	(1.8%)
Stent	Braided CSEMS (8 mm)	48	(29.8%)
	Braided CSEMS (10 mm)	37	(23.0%)
	Braided CSEMS (12 mm)	5	(3.1%)
	Laser-cut CSEMS (10 mm)	69	(42.9%)
	Discontinued before stenting	2	(1.2%)

Abbreviations: CSEMS, covered self-expandable metal stent; EUS-CDS, endoscopic ultrasound-guided choledochoduodenostomy; FNA, fine-needle aspiration; FNB, fine-needle biopsy; FV, forward-viewing ultrasound endoscope; OV, oblique-viewing ultrasound endoscope.

An ITT analysis of the technical success rate in the remaining 164 cases eligible for EUS-CDS found a technical success rate of 97.0% (159/164). [Table 4] summarizes the clinical outcomes. Among the five unsuccessful cases, three were discontinued after puncture alone due to insufficient distal bile duct length below the hilum for safe stent placement, and two due to difficulty in dilation. The alternative drainage methods were transpapillary EBS (n = 3) and EUS-HGS (n = 2).

Table 4 Clinical outcomes of EUS-CDS

Technical success rate, % (n/N)	97.0	(159/164)
Clinical success rate, % (n/N)	97.5	(155/159)
CBD/CHD diameter, median (mm) [range]	13.0	[4.8–24.0]
Early adverse event rate (≤14 days), % (n/N)	4.9	(8/164)
Late adverse event rate (>14 days), % (n/N)	3.8	(6/159)

Abbreviations: CBD, common bile duct; CHD, common hepatic duct; EUS-CDS, endoscopic ultrasound-guided choledochoduodenostomy.

The early adverse event rate was 4.9% (8/164). [Table 5] lists the events and grades. Some patients had more than one early adverse event. One severe case of acute cholangitis developed due to mucosal double penetration and required prolonged hospitalization. There were no cases of mortality. In addition, late adverse events occurred in 3.8% of cases (6/159), consisting of four cases of cholecystitis and two cases of liver abscess.

Table 5 Early adverse events

Event	n [a]	Grade (n)
Acute cholangitis	3	Severe 1/Moderate 1/Mild 1
Peritonitis	2	Mild 2
Mucosal double penetration	2	Severe 1/Moderate 1
Acute cholecystitis	2	Moderate 2
Bleeding	2	Moderate 2

^a Duplicate entry.

[Table 6] summarizes the incidence of early adverse events for each dilation method. The incidence varied greatly according to the dilation technique used. The rate was significantly higher in the group treated with a needle knife and a mechanical dilator (18.8%) than in the coaxial electric cautery group (4.7%, p = 0.042) and the group without tract dilation (1.8%, p = 0.008). There was no statistically significant difference in adverse event rates between the coaxial cautery group and the group without tract dilation (p = 0.349).

Table 6 Early adverse events by dilation method

Method	Total cases	Early adverse events	%	p-Value[a]
Needle-knife and mechanical dilator	16	3	18.8	0.042[b]
Coaxial electrocautery	85	4	4.7	0.008[c]
Without tract dilation	57	1	1.8	0.349[d]

^a Chi-square test.

^b p = 0.042 needle-knife and mechanical dilator versus coaxial electrocautery.

^c p = 0.008 needle-knife and mechanical dilator versus without tract dilation.

^d p = 0.349 coaxial electrocautery vs. without tract dilation.

[Table 7] summarizes the factors associated with early adverse events (primary disease, presence of ascites, bile duct diameter, type of endoscope, and whether tract dilation was performed). Univariate analysis revealed that use of an OV EUS scope was significantly associated with early adverse events (OR: 4.392, 95% CI: 1.03–18.64, p = 0.030). The presence of ascites also showed a trend toward increased risk (OR: 3.576, 95% CI: 0.82–15.58, p = 0.072), although it did not reach statistical significance. No other factors were significantly associated with early adverse events. Variables showing a p-value less than 0.10 in the univariate analysis were subsequently entered into a multivariate logistic regression model to determine independent predictors of early adverse events. In multivariate analysis, the use of an OV EUS scope was independently associated with early adverse events (OR: 5.153, 95% CI: 1.15–22.92, p = 0.031). The presence of ascites did not reach statistical significance (OR: 4.202, 95% CI: 0.92–19.18, p = 0.063).

Table 7 Risk factors associated with early adverse events

	Factor	Events	n	Univariate			Multivariate
				OR	95% CI	p-Value	OR	95% CI	p-Value
Primary disease	PC	7	134	1.322	0.15–11.24	0.797
	Others	1	25
Ascites	Present	5	53	3.576	0.82–15.58	0.072	4.202	0.92–19.18	0.063
	Absent	3	106
CBD/CHD diameter	≤10 mm	2	42	0.925	0.17–0.4.77	0.925
	>10 mm	6	117
Scope	OV	4	32	4.392	1.03–18.64	0.030	5.153	1.15–22.92	0.031
	FV	4	127
Dilation	With	7	102	4.126	0.49–34.41	0.157
	Without	1	57

Abbreviations: CBD, common bile duct; CHD, common hepatic duct; CI, confidence interval; FV, forward-viewing ultrasound endoscope; OR, odds ratio; OV, oblique-viewing ultrasound endoscope; PC, pancreatic cancer.

Discussion

This study evaluated the efficacy and safety of primary EUS-CDS using a CSEMS for malignant biliary obstruction and investigated the factors associated with early adverse events. Several studies published in recent years have compared EUS-CDS and ERCP as primary drainage methods, which indicates that EUS-CDS is no longer used only for cases in which ERCP is difficult or unsuccessful.[4] [6] [7] [8] [9] However, research focusing on the safety of primary EUS-CDS and the factors related to early adverse events remains scarce. In our series, primary EUS-CDS had high success rates and a low early adverse event rate of 4.9%, with the type of endoscope emerging as a significant factor.

Our findings demonstrate that the type of endoscope used, specifically the use of an OV EUS scope, was significantly associated with early adverse events in both univariate and multivariate analyses. Previous studies have reported that EUS-CDS is a technically feasible and clinically effective alternative to ERCP, particularly in expert centers.[10] [14] However, safety concerns, particularly regarding adverse events such as bile leakage, perforation, and stent misplacement, remain an important consideration. Our data support the growing body of evidence that procedure-related factors, rather than patient characteristics alone, may significantly influence safety outcomes. In our cohort, the use of an OV EUS endoscope was associated with a 5.2-fold increased risk of early adverse events (OR: 5.153, 95% CI: 1.15–22.92, p = 0.031), compared with the use of an FV EUS scope. Because this study spans a long period (2003–2025), it is important to acknowledge that procedural techniques and devices have evolved substantially over time. OV scopes were primarily used in the early period of our series; this risk factor may therefore also reflect improvements in technique and equipment over time. Nonetheless, even after accounting for the dilation method, scope choice remained significant, reinforcing that the FV scope inherently offers a safety advantage.

Matsumoto et al[15] demonstrated that the use of an FV EUS scope can prevent double penetration of the gastrointestinal tract during EUS-CDS. In contrast, with an OV EUS scope, the scope and the needle are misaligned, causing the needle to enter the gastrointestinal wall at an angle. This can lead to double penetration of the mucosa or reduced control during device advancement.[16] Kawakami et al[17] also reported cases of double penetration during EUS-CDS with an OV EUS scope. They concluded that caution is required to avoid its occurrence when using an OV EUS scope, and that using an FV EUS scope may help prevent this complication. The FV EUS scope addresses these challenges by providing better alignment of the needle with the endoscope's axis and clearer endoscopic visualization. This enables safer and more accurate stent placement.[18]

Although the presence of ascites was associated with a higher odds ratio for adverse events in univariate analysis (OR: 3.576, 95% CI: 0.82–15.58, p = 0.072), this association did not reach statistical significance and was not confirmed as an independent risk factor in multivariate analysis (OR: 4.202, 95% CI: 0.92–19.18, p = 0.063). Nevertheless, ascites is theoretically associated with an increased risk of bile leakage into the peritoneal cavity and is generally considered a relative contraindication for EUS-BD.[19] In this study, EUS-CDS was performed as the primary drainage procedure in patients who were also technically eligible for ERCP. Importantly, 26 cases were deemed ineligible for EUS-CDS prior to the procedure, with the most common reason being the presence of ascites along the intended puncture route (n = 9). Our findings suggest that, although ascites may not be an independent predictor, careful patient selection and procedural planning remain essential in such cases.

There were no significant differences in adverse events with or without dilation, but interestingly, the choice of tract dilation technique also influenced early adverse event rates. Use of a needle-knife and a mechanical dilator had the highest adverse event rate (18.8%), significantly higher than coaxial electrocautery (4.7%) and procedures performed without tract dilation (1.8%). Importantly, performing EUS-CDS without tract dilation was associated with the lowest adverse event rate, without compromising technical or clinical success. These results highlight the importance of device selection and support a move to safer, more controlled dilation techniques such as coaxial electrocautery or even dilation-free approaches, when technically feasible.

Notably, our observed technical success rate (97.0%) and clinical success rate (97.5%) are slightly higher than those typically reported in studies of primary EUS-CDS (∼90–95%).[4] [8] [9] These findings emphasize that, when performed by experienced endoscopists, primary EUS-CDS can achieve excellent outcomes.

More importantly, the early adverse event rate in our study was 4.9%, substantially lower than the ∼10 to 15% reported in many prior series.[4] [8] [9] This improvement likely reflects the evolution of procedural techniques, including the use of FV scopes, minimal or no tract dilation, and careful patient selection at our center. Notably, there were no procedure-related deaths, and most adverse events were classified as mild or moderate in severity, further underscoring the safety of this approach.

This study is subject to several limitations. Primarily, it was a retrospective single-center analysis, which may limit the generalizability of the results. Second, although the procedures were performed by experienced endoscopists, variations in operator technique could have influenced the outcomes. Third, due to the relatively low number of adverse events, the statistical power for detecting associations with some variables may have been limited.

Conclusion

In conclusion, our findings indicate that the use of an OV EUS scope is an independent risk factor for early adverse events in EUS-CDS, and that the method of tract dilation significantly influences the rate of early adverse events. These results highlight the importance of procedural planning and suggest that using an FV EUS scope and the selection of an appropriate dilation technique, or the decision to avoid dilation when feasible, may lead to improved safety in EUS-CDS. Appropriate patient selection is also crucial, as careful evaluation of anatomical and clinical factors, such as the presence of severe ascites, can help reduce the risk of early adverse events. These findings support the preferential use of FV EUS scope when available and advise against traditional needle-knife dilation in favor of safer alternatives, such as no tract dilation. Finally, prospective studies and multicenter data are warranted to validate these risk factors and further refine best practices in primary EUS-CDS.

• References

• 1 Giovannini M, Moutardier V, Pesenti C, Bories E, Lelong B, Delpero JR. Endoscopic ultrasound-guided bilioduodenal anastomosis: a new technique for biliary drainage. Endoscopy 2001; 33 (10) 898-900
• 2 Burmester E, Niehaus J, Leineweber T, Huetteroth T. EUS-cholangio-drainage of the bile duct: report of 4 cases. Gastrointest Endosc 2003; 57 (02) 246-251
• 3 Okuno N, Hara K, Mizuno N. et al. Efficacy of the 6-mm fully covered self-expandable metal stent during endoscopic ultrasound-guided hepaticogastrostomy as a primary biliary drainage for the cases estimated difficult endoscopic retrograde cholangiopancreatography: a prospective clinical study. J Gastroenterol Hepatol 2018; 33 (07) 1413-1421
• 4 Bang JY, Navaneethan U, Hasan M, Hawes R, Varadarajulu S. Stent placement by EUS or ERCP for primary biliary decompression in pancreatic cancer: a randomized trial (with videos). Gastrointest Endosc 2018; 88 (01) 9-17
• 5 Paik WH, Lee TH, Park DH. et al. EUS-guided biliary drainage versus ERCP for the primary palliation of malignant biliary obstruction: a multicenter randomized clinical trial. Am J Gastroenterol 2018; 113 (07) 987-997
• 6 Park JK, Woo YS, Noh DH. et al. Efficacy of EUS-guided and ERCP-guided biliary drainage for malignant biliary obstruction: prospective randomized controlled study. Gastrointest Endosc 2018; 88 (02) 277-282
• 7 Nakai Y, Isayama H, Kawakami H. et al. Prospective multicenter study of primary EUS-guided choledochoduodenostomy using a covered metal stent. Endosc Ultrasound 2019; 8 (02) 111-117
• 8 Chen YI, Sahai A, Donatelli G. et al. Endoscopic ultrasound-guided biliary drainage of first intent with a lumen-apposing metal stent vs endoscopic retrograde cholangiopancreatography in malignant distal biliary obstruction: a multicenter randomized controlled study (ELEMENT trial). Gastroenterology 2023; 165 (05) 1249-1261.e5
• 9 Teoh AYB, Napoleon B, Kunda R. et al. EUS-guided choledocho-duodenostomy using lumen apposing stent versus ERCP with covered metallic stents in patients with unresectable malignant distal biliary obstruction: a multicenter randomized controlled trial (DRA-MBO trial). Gastroenterology 2023; 165 (02) 473-482.e2
• 10 Gopakumar H, Singh RR, Revanur V, Kandula R, Puli SR. Endoscopic ultrasound-guided vs endoscopic retrograde cholangiopancreatography-guided biliary drainage as primary approach to malignant distal biliary obstruction: a systematic review and meta-analysis of randomized controlled trials. Am J Gastroenterol 2024; 119 (08) 1607-1615
• 11 Hara K, Yamao K, Hijioka S. et al. Prospective clinical study of endoscopic ultrasound-guided choledochoduodenostomy with direct metallic stent placement using a forward-viewing echoendoscope. Endoscopy 2013; 45 (05) 392-396
• 12 Okuno N, Hara K, Haba S. et al. The new potential for using Franseen needles in interventional EUS. Intern Med 2024; 63 (20) 2723-2727
• 13 Cotton PB, Eisen GM, Aabakken L. et al. A lexicon for endoscopic adverse events: report of an ASGE workshop. Gastrointest Endosc 2010; 71 (03) 446-454
• 14 Kakked G, Salameh H, Cheesman AR, Kumta NA, Nagula S, DiMaio CJ. Primary EUS-guided biliary drainage versus ERCP drainage for the management of malignant biliary obstruction: a systematic review and meta-analysis. Endosc Ultrasound 2020; 9 (05) 298-307
• 15 Matsumoto S, Hara K, Mizuno N. et al. Risk factor analysis for adverse events and stent dysfunction of endoscopic ultrasound-guided choledochoduodenostomy. Dig Endosc 2020; 32 (06) 957-966
• 16 Hara K, Yamao K, Mizuno N. et al. Endoscopic ultrasonography-guided biliary drainage: Who, when, which, and how?. World J Gastroenterol 2016; 22 (03) 1297-1303
• 17 Kawakami H, Kuwatani M, Sakamoto N. Double penetrated duodenal wall during endoscopic ultrasound-guided choledochoduodenostomy. Gut Liver 2016; 10 (02) 318-319
• 18 Hara K, Okuno N, Haba S, Kuwahara T. Forward viewing liner echoendoscopy for therapeutic interventions. Clin Endosc 2024; 57 (02) 175-180
• 19 Okuno N, Hara K, Mizuno N. et al. Infectious peritonitis after endoscopic ultrasound-guided biliary drainage in a patient with ascites. Gastrointest Interv 2018; 7: 40-43