Adjusting models to better predict obstetric anal sphincter injury (OASIS) in forceps-assisted vaginal deliveries: A retrospective cross-sectional trial

Objective: Obesity and maternal age are increasing among pregnant patients. The understood effect of body mass index (BMI), advanced maternal age (AMA), and second stage of labor on obstetric anal sphincter injury (OASIS) at delivery is varied. The objective of this study was to assess whether incorporating BMI, second stage of labor length, and AMA into a model for predicting OASIS among forceps-assisted vaginal deliveries (FAVD) had a higher predictivity value compared to models without these additions. Method: This was an IRB-approved retrospective cohort study of singleton gestations who underwent a FAVD between 2017 and 2021. The primary outcome was prediction of OASIS via established models versus models including the addition of new predictive factors. Results: A total of 979 patients met inclusionary criteria and were included in the final analysis. 20.4% of patients had an OASIS laceration, 11.3% of neonates had NICU admissions, 23.7% had a composite all neonatal outcome, and 8% had a composite subgaleal/cephalohematoma outcome. Comparisons of known factors that predict OASIS (nulliparity, race, episiotomy status) to known factors with additional predictors (BMI, AMA, and length of second stage in labor) were explored. After comparing each model's AUC to one another (a total of 3 comparisons made), there was no statistically significant difference between the models (all P > 0.62). Conclusion: Including BMI, AMA, and second stage of labor length does not improve the predictivity of OASIS in patients with successful FAVD. These factors should not impact a provider's decision to perform a FAVD when solely considering increased odds of OASIS.