Objective To assess the association between episiotomy and severe obstetric anal sphincter injury (OASIS) in nulliparous women at term according to the use of an instrument for delivery with control confounding by indication. Methods This was an observational retrospective cohort study including 12 346 women from 2004 to 2020. All nulliparous women with a cephalic singleton pregnancy were included. The primary outcome was the occurrence of OASIS. Association between episiotomy and OASIS was assessed by multivariate logistic regression with adjustment for confounding factors and stratification on the use of an instrument at delivery. Propensity score matching was used to account for indication bias. Results Among 12 346 women included, 7803 (63.2%) had an episiotomy and 4543 (36.8%) did not have an episiotomy; the rate of OASIS was similar in both groups (0.7%). After stratification on use of instrument, an association between episiotomy and OASIS was shown in the case of instrumental delivery (odds ratio [OR] 0.46, 95% confidence interval [CI] 0.26–0.80) but not if the delivery was spontaneous (OR 0.76, 95% CI 0.29–1.98). The result was similar after matching on propensity score (in the case of operative vaginal delivery: OR 0.20, 95% CI 0.10–0.75). Conclusion Episiotomy seems to be a protective factor for OASIS in nulliparous woman at term only in the case of operative vaginal delivery.