A Novel High-Fidelity Simulation Model for Vaginal Hysterectomy

INTRODUCTION: Evidence supports vaginal hysterectomy (TVH) as the preferred route of hysterectomy when feasible. However, the number performed each year is declining and graduating residents report feeling less competent performing TVH than other surgical routes. Simulation models are increasingly used to bridge the gaps in surgical education. There is little consensus regarding the optimal approach to integrate surgical simulation into residency education, especially in vaginal surgery. OBJECTIVE: This is a multi-phase study; the primary aim of this phase was to collect validity data for a novel high-fidelity simulation model for vaginal hysterectomy. METHODS: This study was performed at a community-based, tertiary care hospital and included 24 physicians: 12 novice (4 PGY1 and 4 PGY2 residents) and 12 expert surgeons (6 chief residents, one URPS subspecialist, and 5 OBGYN core faculty who routinely perform TVH). After a brief orientation, participants were video recorded while performing a TVH on the simulation model with passive assistants. Two blinded experts (Grader 1, MIGS/Director of Simulation, Grader 2, Clinical URPS faculty) reviewed video footage of the procedure and scored participants using a validated objective structured assessment of technical skills which included a vaginal surgical skills index (VSSI) and Global Rating Scale (GRS). Pre- and post-simulation surveys were obtained. Demographic and survey data were analyzed using descriptive statistics. Interclass correlation (ICC) values were used to compare scores between graders. Exact Mann-Whitney U test was used to compare VSSI and GRS scores between novice and expert surgeons, and cut-off scores to distinguish between novice and expert surgeons were established utilizing Max Youden’s J index. RESULTS: Novice surgeons performed 0 TVH prior to simulation, while experts performed a median of 16 (13–400). Overall, 75% of participants agreed that simulation is a good way to learn TVH, and 21% found the model unrealistic. Additionally, 60% of resident participants stated the simulation made them want to perform TVH. Interestingly, the interrater validity score (ICC) showed only moderate reliability between graders, at 0.52 ( −0.24–0.82) for VSSI and 0.66 (0.10–0.86) for GRS. For Grader 1, the median VSSI was 18 (17–29) for novice, and 31 (27–34) for expert surgeons (p=0.009), and the median GRS was 10 (5–15) and 15 (15–25) (p=0.032). The VSSI and GRS scores were higher among expert versus novice surgeons for Grader 2 but did not reach statistical significance. Cut-off scores for expert surgeons were established at >20.5 for VSSI (p<0.010), and >10 for GRS (p<0.05). CONCLUSIONS: We collected validity evidence for a novel, high-fidelity simulation model for TVH utilizing procedure specific and generalized scoring systems. Residents found this model realistic, and the simulation encouraged them to incorporate TVH into their post-residency practice. Additionally, we identified that specialized training in medical simulation is critical to optimize scoring, independent of surgical experience. This model is a useful, low risk tool for teaching TVH, and enhancing resident comfort prior to live surgery. Further longitudinal studies are needed to evaluate the model and scoring systems prior to incorporation into residency curriculum to assess procedural competency.