EUS-guided biliary drainage (EUS-BD) has become an established therapeutic alternative toERCP. While the efficacy of hepaticogastrostomy (HGS) and choledochoduodenostomy (CDS) is well-documented, the long-term management of their specific adverse events remains a challenge. Dysfunction of transmural stents is a clinically relevant problem, yet data concerning specific dysfunction mechanisms, optimal management strategies, and relative outcomes after reintervention HGS/CDS dysfunction are limited.

The primary objectives of this study are to describe the mechanisms of dysfunction following transmural EUS-BD (HGS or CDS) and their management. Secondary objectives include describing the time to dysfunction for primary stents and after the first biliary reintervention, comparing long-term performance of HGS vs CDS, and identifying independent predictors of recurrent dysfunction.

We conducted a retrospective, multicenter study involving three Spanish tertiary referral centers over a 7-year period (2017-2024). The study population consisted of consecutive patients presenting with dysfunction of a previously placed EUS-guided biliary drainage.

We collected comprehensive data including patient demographics, etiology of obstruction, indication for the initial EUS-BD, and technical characteristics of the primary stents used.

Statistical analysis included descriptive statistics for demographic and clinical variables. Kaplan-Meier curves were generated to estimate the time to dysfunction (both primary and post-reintervention). A stratified Cox regression analysis was performed to identify independent predictors of recurrent dysfunction.

A total of 91 dysfunction events (70 HGS and 21 CDS) were analyzed in 61 patients. The cohort was 59.02% male with a mean age of 68.61 years (SD ±10.26). The etiology of biliary obstruction was predominantly malignant (78.43%), with tumors located mainly in the distal bile duct (50.98%) and hilar region (27.45%). The primary indication for the initial EUS-BD was failed ERCP cannulation (35.29%), followed by dysfunction of a prior indwelling transpapillary stent (25.49%). The most frequently used stetns were 10x80-mm covered SEMS for HGS and 8 mm LAMS for CDS.

The median time to first dysfunction (primary stent failure) was 168.5 days (IQR 58-337.5). All patients underwent an endoscopic revision. The leading cause of dysfunction was obstruction by sludge and food debris (46.15%), followed by tissue ingrowth (16.48%), with no significant difference between access routes (p=0.48). However, technique-specific failure profiles were identified: gastric stent migration was a specific complication of HGS (15.71%), whereas duodenal obstruction and sump syndrome were specific to CDS (9.52%).

Management strategies differed significantly between groups. In HGS dysfunction, stent exchange was the most common intervention (40.57%), whereas in CDS, clearance followed by coaxial pigtail insertion was the preferred strategy (55%).

Following the first biliary reintervention (BRI), the median time to recurrent dysfunction increased to 456 days (IQR 184-1363). Comparison between HGS and CDS showed no significant differences in the time to recurrent failure (p=0.79). In the stratified analysis, for HGS, the presence of a distal malignant stenosis was the only independent risk factor associated with recurrent failure (HR 28.98; p=0.029). Notably, no specific reintervention technique (cleaning, exchange, or coaxial stenting) demonstrated a statistically significant improvement in prolonging the time to dysfunction in either group.

This is the first study to jointly analyze dysfunction in HGS and CDS. Although their failure profiles and initial management differ, their long-term patency after reintervention is comparable. Risk factors for recurrent failure are technique-specific. The finding that no single management strategy was superior suggests that a personalized approach, tailored to the specific cause of dysfunction, is required.