Face validity of a new mixed-reality simulator for training endoscopic submucosal dissection: a Western pilot study

Endoscopic submucosal dissection (ESD) is an effective but technically demanding procedure with a steep learning curve. Traditional training methods, including master courses, apprenticeship, and animal models, have ethical, financial, and logistical constraints. This pilot study evaluates the face validity of a novel mixed-reality simulator designed to enhance and maintain endoscopic dexterity in ESD training.

In this exploratory, prospective observational cohort study, 8 ESD trainees with prior advanced endoscopy experience, but limited ESD exposure, underwent a 2-day structured training using the EndoCubot (EndoRobotics, Seoul, South Korea) simulator with artificial laminated tissue (VTT [Versatile Training Tissue], KOTOBUKI Medical, Yashio, Japan). Performance was assessed pre- and post-training through a 5-minute dexterity test simulating circular incision with 40 achievable points, and a questionnaire evaluating simulator realism, satisfaction, and training effectiveness. Paired sample t tests (α = 0.05) and descriptive statistics were used for analysis. Results were reported as mean with standard deviation.

On a 40-point scale, mean postcourse scores (34.7 [4.86]) were significantly higher than precourse scores (25.50 [0.76], P < .001), with a mean improvement of 9.25 points (95% confidence interval, 4.993-13.507, Hedges' g correction = 1.613). Survey results indicated high satisfaction (mean rating: 9.125/10) and strong agreement on the simulator's realism and effectiveness.

The simulator demonstrates strong face validity and supports a realistic standardized skill acquisition in ESD. The simulator could provide a realistic and ethical alternative to current animal models and requires further study.