Enhancing Colonoscopy Training: A Prospective Randomized Study on the Efficacy of Computer-Based Simulators

Aims Computer-based endoscopic simulators have emerged as a valuable tool for medical training, offering the potential to enhance procedural skills in a controlled and risk-free environment. Despite positive reports regarding their usefulness, no blinded, prospective randomized studies have focused specifically on the efficacy of virtual reality simulators for colonoscopy training. This study aimed to assess the effectiveness of a computer-based simulator in improving basic colonoscopy skills among novice medical trainees.

Methods This prospective randomized study involved 16 medical residents with no prior experience in colonoscopy. Following an introductory lecture on colonoscopy fundamentals, participants were randomized into two groups: a simulator group (n=8) and a non-simulator group (n=8). The simulator group underwent 16 hours of training on the GI-Mentor II simulator, in addition to bedside patient training, while the non-simulator group received only bedside training.

Participants in both groups were subsequently evaluated on their first five patient-based colonoscopies. The primary outcome was the number of proctor “assists” required per procedure. Secondary outcomes included unassisted insertion depth, cecal intubation rate, insertion time, proctor- and nurse-rated competence, and patient-reported pain levels. Statistical significance was determined using a threshold of P≤0.05.

Results The simulator group demonstrated superior performance across multiple metrics compared to the non-simulator group. On average, simulator-trained trainees required significantly fewer proctor assists per colonoscopy (1.94 vs. 3.43; P≤0.001) and achieved greater unassisted insertion depths (43 cm vs. 24 cm; P=0.003). While the cecal intubation rate was higher in the simulator group (26% vs. 10%), this difference approached but did not reach statistical significance (P=0.06). Competence ratings by proctors (2.28 vs. 1.88 out of 5; P=0.02) and endoscopy nurses (2.56 vs. 2.05 out of 5; P=0.001) were significantly higher for the simulator group. Importantly, there were no significant differences between groups in terms of patient-reported pain or proctor- and nurse-rated attention to patient discomfort.

Conclusions Simulation-based colonoscopy training using the GI-Mentor II improved novice trainees' performance on patient-based colonoscopy. The significant reduction in proctor assistance, greater insertion depths, and higher competence ratings suggest that computer-based simulation is an effective tool for accelerating early skill acquisition in colonoscopy. Incorporating virtual reality simulators into initial training stages may enhance procedural competence while maintaining patient safety. Further research with larger sample sizes is warranted to validate these findings.

Publication History

Article published online:
27 March 2025