# Clinical Outcomes and Learning Curve of Endoscopic Ultrasound‐Guided Hepaticogastrostomy During the Implementation Phase in Inexperienced Centers: A Multicenter Retrospective Study

## TL;DR

This study shows that endoscopic ultrasound-guided hepaticogastrostomy can be safely performed in inexperienced centers with proper expertise, as procedure time improves with experience.

## Contribution

The study evaluates the feasibility and learning curve of EUS-HGS in inexperienced centers, identifying predictors of adverse events.

## Key findings

- EUS-HGS had a 94% technical success rate and 11% serious adverse events rate in inexperienced centers.
- Procedure time decreased as institutional experience increased, but success rates and adverse events did not show significant trends.
- Ascites was identified as an independent predictor of serious adverse events.

## Abstract

Endoscopic ultrasound‐guided hepaticogastrostomy (EUS‐HGS) is a technically demanding procedure performed predominantly at centers with high expertise. Its feasibility and learning curves in inexperienced centers remain unclear. This study aimed to evaluate the initial clinical outcomes and learning curves of EUS‐HGS implemented in inexperienced centers.

Between September 2018 and December 2020, four tertiary care centers specializing in pancreaticobiliary disease implemented EUS‐HGS. The first 20 patients who underwent EUS‐HGS at each institution were retrospectively enrolled. The primary outcomes were technical success, procedure time, and serious (moderate‐to‐severe) adverse events (AEs). A learning curve analysis was conducted using chronological quartiles. Logistic regression was used to identify the predictors of technical failure and serious AEs.

The overall technical success rate was 94% (75/80); serious AEs occurred in 11% (9/80). Procedure time significantly decreased across quartiles, whereas no significant trend was observed for technical success or serious AE rates. The univariate analysis did not identify significant predictors of technical failure, whereas the multivariate analysis identified ascites as an independent predictor of serious AEs.

EUS‐HGS can be feasibly implemented with caution in inexperienced centers, particularly when performed by endoscopists with adequate pancreatobiliary expertise. Additionally, during the implementation phase, procedure time decreased with increasing institutional experience.