Vaginal Surgery Model Review Data Extraction Subanalysis: OASIS Repair Models

INTRODUCTION: The incidence of obstetrical anal sphincter injuries (OASIS) ranges between 2% and 20% a year. Failure to recognize these lacerations and improper repair can impact pelvic floor function and health. Unfortunately, resident exposure to repair of OASIS injuries is limited. Surgical models have emerged to help supplement for this deficiency. However, a comparative review of such models is lacking in the literature. OBJECTIVE: To identify OASIS repair models and review their accessibility, fidelity, and impact on trainee education. METHODS: A PubMed search review was performed using Covidence to identify articles and abstracts relating to vaginal surgery models, including OASIS injuries. Descriptive studies, randomized trials, and abstracts were included. Systematic reviews of models were excluded. Four primary researchers conducted the review. Two independent reviews were assigned to each of the articles. Extracted study outcomes included the following: procedure simulated, comparison models, material costs, material availability, validity evidence, validation assessment tool, results of validation, validation population, external validation, trainee confidence/satisfaction with model, societal approval/endorsement. RESULTS: 768 articles were identified from the initial engine search. Only 14 articles (1. 8%) addressed perineal repair, with 12/14 (86%) specific to OASIS repair models. The two articles not focused on OASIS repairs were excluded from the discussion. Four distinct surgical teaching models were identified: two high-fidelity models (Sultan anal sphincter episiotomy repair model and 3D printed silicone perineum models) in 2/12 studies (16. 7%) and two low-fidelity models (sponge perineum and beef tongue) in 10/12 studies (83. 3%). Only one study compared two different surgical models (sponge perineum vs animal tongue model). Low-fidelity models ranged from 3. 07 to 12 per use, while high-fidelity models ranged from 10 to 848. Validity was assessed in 2/12 (16. 7%) of the studies (1 study in low-fidelity, and 1 study in a high-fidelity model). Surgical simulation of OASIS increased confidence in 7/12 (58. 3%) studies where it was assessed (low-fidelity models, 6/12, 50%, and high-fidelity models, 1/12, 8. 3%). Satisfaction was only assessed in studies involving low-fidelity models (6/12, 50% of studies) with all participants reporting a degree of satisfaction (6. 3 to 9. 5 on a 10-point Likert scale, and average of 4. 48/5 specifically for the sponge perineum). Of the four studies that assessed knowledge before and after OASIS simulation, all studies used low-fidelity models and found improvement in knowledge, which were assessed mainly through written non-validated exams. In the study that compared two low-fidelity models, the beef tongue model was felt to be more realistic and improve surgical techniques better than the sponge perineum model. CONCLUSIONS: Both high-fidelity and low-fidelity OASIS repair teaching models increase reported confidence and knowledge among users, and are well-received; however, high-fidelity models are understandably more expensive compared to lower-fidelity models. Further research is needed comparing the models to determine if high-fidelity models are superior to low-fidelity models, and to validate models to ensure that they are effective and appropriate for OASIS training in different individuals. Figure 1Table 1