Virtual ureteroscopy predicts ureteroscopic proficiency of medical students on a cadaver

Purpose: Training on a virtual reality (VR) simulator has been shown to improve the performance of VR endoscopic tasks by novice endoscopists. However, to our knowledge the translation of VR skills into clinical endoscopic proficiency has not been demonstrated. We established criterion validity for a VR ureteroscopy simulator by evaluating VR trained subjects in a cadaver model. Materials and Methods: A total of 32 participants, including 16 medical students and 16 urology residents, were evaluated at baseline on a VR ureteroscopy simulator (Uromentor, Simbionix, Lod, Israel), performing simple diagnostic ureteroscopy. The students then underwent 5 hours of supervised training on the simulator. Two weeks later all participants were reevaluated (VR2) on the simulator when repeating the initial task. Each participant was then assessed on the performance of a similar diagnostic ureteroscopy in a male cadaver. Results: In medical students VR2 and cadaver performances correlated closely for several measured parameters (total time for task completion and overall global ratings score). In contrast, there was little correlation between the 2 performances in residents. Indeed, performance on the cadaver correlated more closely with the training level than VR2 scores. Despite VR training medical students were unable to perform cadaver ureteroscopy comparably to residents. Conclusions: For novice endoscopists performance on the simulator after training predicted operative (cadaver) performance and, thus, it may be useful for the education and assessment of physicians in training. However, VR training is unable to override the impact of clinical training, although it may help shorten the learning curve early in training.