Trial procedures
Women with external anal sphincter lesions at ultrasound were offered to participate in the randomized trial, and if they consented were assigned (1:1 ratio) to planned CS at 39 weeks’ gestation or vaginal delivery. Concealment was obtained by use of a computer-generated randomization scheme, in various-sized blocks, stratified by center, and by use of separate serialized sealed and opaque envelopes prepared by the sponsor. Investigator or participant blinding to group assignment was not feasible, but investigators were unaware of aggregate outcomes during the study, since the analysis was performed only after the follow-up period was completed and the database was frozen. In the vaginal delivery group, the management of the delivery, including episiotomy, forceps or vacuum, was left to the appreciation of the clinician. In case of an emergent indication for CS unrelated to the issue of anal sphincter protection, according to the obstetrician, cross-over was allowed.