ESD Training in GI Fellowship: Proof of Concept

Abstract

Introduction: Endoscopic submucosal dissection (ESD) is becoming a more prevalent technique in the West, having already been established in the Eastern World. Given the lack of expertise in Western countries, animal models have been used to train experienced endoscopists. However, there is a void in understanding of how to train fellows in this procedure. To our knowledge, this is the first study evaluating the use of an animal model to train general GI and advanced endoscopy fellows. Methods: This is a proof of concept in training fellows in ESD with animal models. Two fellows, an advanced endoscopy fellow (AEF) and a third-year gastroenterology fellow (GF) participated in the study. They were supervised and instructed by an experienced endoscopist. The AEF had previously attended an ESD course. Lesion characteristics and procedure details were recorded. After each session, participants were given a survey to grade difficulty of different components of the procedure based on a Likert scale. For grading purposes, the procedure was divided into three steps: marking, incision, and dissection. Survey participants were blinded to results. Results: The AEF perceived the procedures to be easier than the GF (mean Likert scale 4.5 vs 7.3). The only significant complication (large perforation) occurred during dissection by the GF (1 of 6 lesions). Improvement in Likert scale in the ex-vivo models was noted for both trainees with each subsequent procedure with an average decrease in 2 points on the Likert scale for both trainees. AEF had a shorter learning curve compared to the GF as evidenced by a quicker decline in the Likert scale. Both trainees perceived the esophageal lesion to be easier than stomach lesions (Likert 4.7 vs 6.1). Efficiency, as defined by time divided by surface area resected, improved between first and second lesions, but there was not a consistent improvement across all lesions. Dissection was perceived as the most difficult step in the process (Likert 7.3 vs 4.2). Conclusion: Participating in an ESD course should be recommended for trainee comfort in ESD prior to starting animal labs. We recommend achieving competence in complex EMR prior to starting training in ESD. Use of an animal model is effective in allowing trainees the opportunity to learn basic ESD skills during third year of GI fellowship and in advanced endoscopy fellowship. In the future, more regimented training curriculum should be evaluated for fellow training.Table: Table. Trainees Ratings of Degrees of Difficulty of Six ESD Procedures