OBJECTIVE: We sought to determine whether an optimal level of lateral episiotomy use can be found by assessing the correlation between the hospital-based variations in episiotomy use and rates/odds ratios of obstetric anal sphincter injuries (OASIS). STUDY DESIGN: This was a retrospective population-based register study. The study group, comprising women with spontaneous singleton vaginal deliveries, contained all 154,175 primiparous and all 234,236 multiparous women. The correlations between lateral episiotomy use and incidence/risk of OASIS (n = 1659) were assessed using nonlinear and linear regression modeling. RESULTS: The rates of episiotomy were inversely correlated with the risk of OASIS among both groups of women. OASIS rates increased from 0.5-1.0% as episiotomy rates decreased from 80-40%. CONCLUSION: Restricting lateral episiotomy use may result in higher OASIS rates. However, we could not determine the optimal level of episiotomy use since individual hospitals deviated substantially from the correlation curves.
OBJECTIVE: We sought to determine whether an optimal level of lateral episiotomy use can be found by assessing the correlation between the hospital-based variations in episiotomy use and rates/odds ratios of obstetric anal sphincter injuries (OASIS). STUDY DESIGN: This was a retrospective population-based register study. The study group, comprising women with spontaneous singleton vaginal deliveries, contained all 154,175 primiparous and all 234,236 multiparous women. The correlations between lateral episiotomy use and incidence/risk of OASIS (n = 1659) were assessed using nonlinear and linear regression modeling. RESULTS: The rates of episiotomy were inversely correlated with the risk of OASIS among both groups of women. OASIS rates increased from 0.5-1.0% as episiotomy rates decreased from 80-40%. CONCLUSION: Restricting lateral episiotomy use may result in higher OASIS rates. However, we could not determine the optimal level of episiotomy use since individual hospitals deviated substantially from the correlation curves.