V08-03: Design and validation of a low-cost, high-fidelity model for RALP urethrovesical anastomosis
Surgical simulation is a critical component of surgical training, and urology trainees lack procedure-specific simulation models which allow repetitive practice of robotic skills during case preparation. We sought to develop and validate a low-cost, high fidelity robotic surgical simulator for the urethrovesical anastomosis component of robotic assisted laparoscopic radical prostatectomy.
A novel urethrovesical anastomosis simulation model was constructed with silicone molds and a 3D-printed model of the male boney pelvis. Using a DaVinci surgical robot, urology faculty and trainees operated on the simulation. Each participant viewed a 4-minute instructional video and was then was given 12 minutes to complete the simulation. A survey was issued to establish face validity, content validity, and acceptability. Simulation runs were recorded, evaluated by three blinded reviewers, and given a GEARS (global evaluative assessment of robotic skills) and PACE (prostatectomy assessment and competency evaluation) score. The quality of the anastomosis was graded by two reviewers. These factors were compared to robotic experience to establish construct validity. ANOVA statistical analysis was performed.
Seventeen participants took part in the initial validation of this model (1 faculty, 4 fellows, 2 chief residents, and 10 residents). Median console experience was 10 years, 4.0 years, 1.5 years, and 0 years for faculty, fellows, chief residents, and residents, respectively. Likert (1-5) scores for face validity, content validity, and acceptability were 3.56 (SD 0.42), 4.17 (SD 0.21), and 4.24 (SD 0.15), respectively. Construct validity was excellent. Mean time to completion, mean percent completion, mean anastomosis inspection score, mean GEARS score, and mean PACE score all correlated to robotic experience (p
We present a novel high-fidelity, low-cost robotic simulation of the urethrovesical anastomosis. Initial evaluation demonstrates good face validity, content validity, and acceptability. Construct validity is excellent, and the model is able to discern robotic skill level across all levels of training. Further studies are needed to evaluate longitudinal effect of the model on surgical training.