Overdiagnosis and rising rate of obstetric anal sphincter injuries (OASIS): time for reappraisal.

OBJECTIVES: To determine the accuracy of clinical diagnosis of obstetric anal sphincter injuries (OASIS) using three-dimensional (3D) endoanal ultrasound (EA-US) and to compare symptoms and anal manometry measurements between women with anal sphincters adequately repaired and those with persistent anal sphincter defects.
METHODS: The EA-US images of women with clinically diagnosed and repaired OASIS, defined as third- or fourth-degree perineal tear, who attended the perineal clinic at Croydon University Hospital over a 10-year period (2003-2013) were reanalyzed by a single expert blind to symptoms and the results of clinical examination. St Mark's Incontinence Scores (SMIS) and anal manometry measurements were obtained and compared between women with an intact anal sphincter and those with an anal sphincter scar and between those with an intact anal sphincter and those with a defect. Anal manometry measurements were compared between women with an external anal sphincter (EAS) defect and those with an internal anal sphincter (IAS) defect.
RESULTS: The images of 908 women were reanalyzed. No evidence of OASIS was found in 64 (7.0%) women, an EAS scar alone was detected in 520 (57.3%) and an anal sphincter defect in 324 (35.7%). Of the 324 women with a defect, 112 had an EAS defect, 90 had an IAS defect and 122 had a combined IAS and EAS defect. SMIS results were significantly higher in women with an anal sphincter defect compared with those with no evidence of OASIS (P = 0.018), but there was no significant difference in scores between women with an intact sphincter and those with an EAS scar only. Women with a defect had a significantly lower maximum resting pressure (median (range), 44 (8-106) vs 55 (29-86) mmHg; P < 0.001) and maximum squeeze pressure (median (range), 74 (23-180) vs 103 (44-185) mmHg; P < 0.001) compared with those in the intact group. Similar, but less marked, differences were observed in women with an EAS scar compared with those who had an intact anal sphincter. The anal length was significantly shorter in women with a defect compared with those in the intact group (median (range), 20 (10-40) vs 25 (10-40) mm; P = 0.003).
CONCLUSIONS: Seven percent of women with a clinical diagnosis of OASIS were wrongly diagnosed. We believe that this rate may differ from that of other units but training methods and competency assessment tools for the diagnosis and repair of OASIS need urgent reappraisal. The role of EA-US in the immediate postpartum period needs further evaluation as the accurate interpretation of the images is dependent on the expertise of the staff involved.