A Single-Blinded Randomized Control Study to Assess the Predictive Validity of Robotic Skills Simulation

Daniel Cheng, MD, MPH, Christopher Mcnicoll, MD, MPH, Chad Cross, PhD, PStat, Heidi Ryan, MD, Shawn Tsuda, MD, Jennifer Baynosa, MD, Charles St. Hill, MD, MSc. University of Nevada, Las Vegas School of Medicine

Introduction: Current literature lacks a detailed assessment of robotic simulation in residency. We conducted a study to assess the effect of robotic simulation on resident performance.

Methods: A single-blinded randomized control study was conducted with 30 residents. All surgeries performed were robotic transabdominal pre-peritoneal inguinal hernia repairs. Subjects were timed and scored on key steps, errors, and complications. Subjects completed a baseline Surgery 1, and were randomized to Control (n=4) and Advanced Training (AT) (n=4) groups. The AT group trained using the simulator until proficiency. Subjects then completed Surgery 2 and Surgery 3. Performance was compared between the Control and AT groups using two-sample t-tests. Performance was compared within the groups, from Surgery 1 to Surgery 2, using paired samples t-tests. Owing to small sample size, we calculated exact p-values and bootstrapped confidence intervals.

Results: Control and AT groups had similar baseline training and performance in Surgery 1. AT group had improved performance after simulator training, taking less time and committing less errors. The difference in performance was largest in Surgery 2, and significant for suturing of the peritoneum (p=0.030). Exact p-values identified the AT group was significantly faster at mesh placement in Surgery 2 (p=0.031). From Surgery 1 to Surgery 2, the AT group performed mesh placement faster (p=0.025). Bootstrapped confidence intervals identified the AT group was significantly faster at mesh placement (-8.50, -3.25) and peritoneum suturing (-22.00, -2.25) from Surgery 1 to Surgery 2. Errors committed and complications were not significantly different.

Conclusion: Although sample size makes it difficult to draw definitive conclusions, the data suggests robotic skills simulation accelerates proficiency. Given it has no additional risk to patients, simulation should be included in all curriculums that teach robotic surgery.