Abstract
BACKGROUND AND OBJECTIVES: EUS-guided biliary drainage (EUS-BD) is an alternative to ERCP for patients with altered anatomy or duodenal obstruction. EUS-guided antegrade intervention (EUS-AG) offers physiological drainage with lower reintervention rates and higher stent patency. However, the technical complexity of EUS-AG limits its success rate, and factors associated with technical failure remain controversial. This study aimed to evaluate the learning curve for EUS-AG and identify predictors of technical failure. PATIENTS AND METHODS: A cohort of 138 patients intended for EUS-AG between December 2019 and January 2025 was analyzed. CUSUM analysis was used to assess the learning curve based on procedure time. Technical failure was defined as failure to advance the antegrade guidewire across the duodenal papilla or bilioenteric anastomosis, failure of stent deployment, or failure to extract stones. Univariable and multivariable logistic regression analyses were performed to identify independent predictors of technical failure. RESULTS: The overall technical success rate of EUS-AG was 78.2% (108/138). The CUSUM learning curve demonstrated a significant decrease in procedure time after 50 cases, indicating the transition to the proficiency phase. Multivariable analysis identified preproficiency phase (OR, 9.70; 95% CI, 2.91-32.30; P < 0.001), hilar bile duct obstruction (HBO) (OR, 14.36; 95% CI, 3.41-60.49; P < 0.001), and excessive bile duct dilation (EBD) (OR, 17.09; 95% CI, 4.80-60.89; P < 0.001) as independent predictors of technical failure. CONCLUSION: For endoscopists with proficient ERCP experience, the EUS-AG learning curve plateaued after approximately 50 cases. The preproficiency phase, HBO, and EBD were identified as independent predictors of technical failure.