High prevalence of defects in Cesarean section scars at transvaginal ultrasound examination

To determine the ability to correctly identify Cesarean section scars, to estimate the prevalence of defective scars, and to determine the size and location of scar defects by transvaginal ultrasound imaging.

Two hundred and eighty-seven women underwent transvaginal ultrasound examination 6-9 months after delivery: 108 had undergone one Cesarean section, 43 had had two Cesarean sections, 11 had undergone at least three Cesarean sections, and 125 were primiparae who had delivered vaginally. The ultrasound examiner was blinded to the obstetric history until all scans had been evaluated.

None of the 125 vaginally delivered women had a visible scar in the uterus, whereas all women who had undergone Cesarean section had at least one visible scar. Median myometrial thickness at the level of the isthmus was 11.6 mm in women who had only been delivered vaginally, and 8.3 mm, 6.7 mm and 4.7 mm in women who had undergone one, two and at least three Cesarean sections, respectively (P < 0.001). Scar defects were seen in 61% (66/108), 81% (35/43) and 100% (11/11) of the women who had undergone one, two and at least three Cesarean sections (P = 0.002); at least one defect was classified as large by the ultrasound examiner in 14% (15/108), 23% (10/43) and 45% (5/11) (P = 0.027), and at least one total defect was seen in 6% (7/108), 7% (3/43) and 18% (2/11) (P = 0.336). In women who had undergone one Cesarean section, the median distance between an intact scar and the internal cervical os was 4.6 (range, 0-19) mm, and that between a deficient scar and the internal cervical os was 0 (range, 0-26) mm (P < 0.001).

Cesarean section scars can be detected reliably by ultrasound imaging. Myometrial thickness at the level of the isthmus uteri decreases with the number of Cesarean sections and the frequency of large scar defects increases. Scars with defects are located lower in the uterus than intact scars.