Background & aims: A new fluoroscopy-free training system for endoscopic retrograde cholangiopancreatography (ERCP) with different model subtypes recently was developed. This study aimed to establish construct validity by investigating whether the X-Vision ERCP Training System could distinguish experienced endoscopists from beginners and to reveal characteristic mistakes during sphincterotomy.
Methods: Six staff gastroenterologists that practice ERCP, 10 trainees that perform esophagogastroduodenoscopy and colonoscopy, and 12 residents without endoscopic experience each sequentially attempted 4 different models, simulating selective cannulation of the pancreatic or bile duct, intubation of differently arranged rubber papillas, stent placement, and sphincterotomy of a biopapilla. Performance parameters were recorded and participants' expectations were compared before and after training to determine whether the simulator was a credible tool for ERCP training. Staff gastroenterologists graded the realism and utility of the simulation. The quality of sphincterotomy was assessed by an expert endoscopist.
Results: Participants with ERCP experience had significantly shorter procedure times compared with those with intermediate (P < .001) or no endoscopic experience (P < .001). Total and single credibility scores significantly increased after simulator practice. The faculty found the X-Vision ERCP Training System to be realistic and useful for training. In the less-experienced groups, common mistakes made during sphincterotomy included inadequate positioning of the duodenoscope, traumatic intubation of the papilla, and continued cutting despite insufficient endoscopic view.
Conclusions: The X-Vision ERCP Training System distinguished subjects with different levels of experience and was regarded as realistic and useful for ERCP training. Common mistakes during sphincterotomy could be assessed objectively.