Can we define a cut-off for external anal sphincter defect angle?

Abstract

Obstetric anal sphincter injury is a common complication of childbirth and the primary etiological factor for anal incontinence in young women. On endoanal ultrasound, a significant defect of the external anal sphincter (EAS) has arbitrarily been defined as a gap of greater than 30° in its hyperechogenic ring structure. This study was designed to validate a cut-off value for defining a significant EAS defect on tomographic exoanal imaging. This was a retrospective study involving women attending a tertiary urogynecological unit between January 2015 and December 2021. Analysis was limited to premenopausal women to reduce confounders. The assessment included obtaining the St Mark's incontinence score and visual analog scale (VAS) bother score of anal incontinence, and performance of four-dimensional transperineal exoanal ultrasound. Tomographic ultrasound imaging (TUI) was used to evaluate the EAS; offline analysis was performed at a later date, with the assessor blinded against all other clinical data. The primary measure to assess EAS defects was defect angle. Associations between this parameter and anal incontinence, VAS bother score and St Mark's score were estimated using univariable logistic or linear regression models. Receiver-operating-characteristics (ROC)-curve analysis was performed to determine the optimal cut-off value of the EAS defect angle for defining a significant defect. A total of 915 premenopausal women were seen during the study period. Ultrasound volumes of the anal canal were missing in 64 women and one patient was excluded due to muscular dystrophy, leaving 850 women included in the analysis. The mean age was 42 (range, 18-56) years and the mean body mass index was 29 (range, 16-57) kg/m² at assessment. Anal incontinence was reported by 133 (16%) women, with a median St Mark's score of 10 (interquartile range, 6-14) and a mean VAS bother score of 6.6 (range, 0-10). On TUI, six slices were scored per patient (total of 5100 assessments). EAS defects were seen in 111 (13%) patients and 368 slices, with an average defect angle of 32° (range, 2-155°) and a mean maximum defect angle of 67° (range, 14-167°). Average defect angle showed the strongest correlation with anal incontinence, St Mark's score and VAS bother score (all P < 0.001). ROC-curve analysis suggested a cut-off of 30° for the definition of a significant EAS defect. In premenopausal women, EAS defects on TUI were associated significantly with all investigated measures of anal incontinence. Average defect angle was the best predictor of anal incontinence and enabled ROC-curve analysis. The 30° defect angle criterion empirically used for endoanal ultrasound seemed to be valid as a cut-off for the diagnosis of significant anal sphincter defect on exoanal imaging. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.