Effect of virtual endoscopy simulator training on performance of upper gastrointestinal endoscopy in patients: a randomized controlled trial.

BACKGROUND Skills in gastrointestinal endoscopy mainly depend on experience and practice. Patients upon whom trainees perform their first endoscopic examinations are likely to suffer more discomfort and prolonged procedures. Training on endoscopy simulators may reduce the time required to reach competency in patient endoscopy. PATIENTS AND METHODS Residents in internal medicine without experience of endoscopy were randomized to a group who trained on a simulator before conventional training (group S) or one that received conventional training only (group C) before starting upper gastrointestinal endoscopy in patients. After endoscopy, discomfort and pain were evaluated by patients, who were blind to the beginners' training status. Results in terms of time, technique (intubation, pyloric passage, J-maneuver), and diagnosis of pathological entities were evaluated by experts. RESULTS From 2003 to 2007, 28 residents were enrolled. Comparing group S with group C in their first ten endoscopic examinations in patients, time taken to reach the duodenum (239 seconds (range 50 - 620) vs. 310 seconds (110 - 720; P < 0.0001) and technical accuracy ( P < 0.02) were significantly better in group S. Diagnostic accuracy did not differ between the groups. Fourteen residents (7 simulator-trained, 7 not simulator-trained) continued endoscopy training. After 60 endoscopic examinations, investigation time was still shorter in group S. Technical and diagnostic accuracy improved during on-patient training in both groups; here differences between groups were no longer observable. There were no significant differences in discomfort and pain scores between the groups after 10 and after 60 endoscopies. Discomfort and pain were higher than for endoscopy performed by experts. CONCLUSION This randomized controlled trial shows that virtual simulator training significantly affects technical accuracy in the early and mid-term stages of endoscopic training. It helps reduce the time needed to reach technical competency, but clinically the effect is limited. Simulator training could be useful in an endoscopy training curriculum but cannot replace on-patient training.