Training of different endoscopic skills on ex-vivo animal model

Virtual reality simulator and ex vivo animal models are used for training of both basic and advanced endoscopic techniques. The aim of this study was to assess whether hands-on training on ex vivo animal model improves endoscopic skills. Four different endoscopic techniques were practiced: endoscopic resection, endoscopic stenting, application of the over-the-scope (OVESCO) clip, and endoscopic submucosal dissection (ESD).

Except for 2 participants, all trainees participated in a 1-day course. Two remaining participants took part in 7 ESD courses. All training courses consisted of theoretical introduction and a 6-hour training on Erlangen Active Training Simulator. The endoscopic skills were assessed before and after the training session by 2 independent assessors. Each assessor evaluated the skills by using a score on a scale of 1 to 5, where 1 stands for excellent and 5 for insufficient. Each assessor also assessed whether the procedure was successfully completed. The main outcome measurement was the percentage of participants who successfully completed the procedure during the test.

For endoscopic resection, endoscopists (n = 15) improved their skills (median [10th and 90th percentiles] score before training, 3.5 [2.7-4.2]; after training 1.5 [1-2.3], P < 0.001). Seven procedures were assessed as successful before the course (47%); after the training, 13 procedures were assessed as successful (87%) (P = 0.02). For stenting, participants (n = 15) significantly improved their abilities to place both self-expandible metallic and plastic stents. For OVESCO clip (n = 10), participants (n = 10) improved their skills to prepare and apply the clip (given the score of 4.5 [3.9-5] before and 2.0 [1.2-2.8] after, P < 0.01). Before the training, only 1 clip application had been successful (10%), whereas the number rose to 9 after the course (90%). For endoscopic submucosal dissection (n = 10), eight participants of the 1-day course did not improve their competences (with scores of 4.2 [3.8-5] before and 4.0 [3.1-4.8] after, nonsignificant). Two participants who had undertaken 7 ESD courses improved their skills (with scores of 4 before and 1.6 after); given the small number of participants, this finding is statistically insignificant.

The effect of training on clinical outcome was not investigated. There was a lack of pretraining versus posttraining tests blinding.

A 1-day training course on ex vivo animal model improves general endoscopic competence on simulator in endoscopic resection, insertion of stents, and application of OVESCO clips. In contrast, 1-day course does not improve skills for ESD that requires a higher number of training courses.