Measurement of Lower Uterine Segment Thickness to Detect Uterine Scar Defect: Comparison of Transabdominal and Transvaginal Ultrasound.

Abstract

Lower uterine segment (LUS) thickness measurement using transabdominal ultrasound (TA-US), transvaginal ultrasound (TV-US), or the combination of both methods can detect scar defect in women with prior cesarean. We aimed to compare the sensitivity of three approaches. Women with prior cesarean underwent LUS thickness measurement at 34-38 weeks' gestation. Among those who underwent repeat cesarean before labor, we compared the accuracy of TA-US, TV-US, and the thinner of the two measurements (the "combined measurement") for uterine scar dehiscence using the area under the curve (AUC) of receiver operating curves with their 95% confidence intervals (CI). We calculated the sensitivity and specificity of the three approaches using a cut-off of 2.3 mm based on prior literature. We included 747 participants. The mean LUS thickness was greater with TA-US (3.8 ± 1.6 mm) compared with TV-US (3.5 ± 1.9 mm) or the combined measurement (3.2 ± 1.5 mm; P < .001). The AUC was 78% (95% CI: 69%-87%), 85% (95% CI: 79%-91%), and 88% (95% CI: 82%-93%), respectively (all with P < .001). The AUC difference between TA-US and the combined measurement was not significant (P=.057). A LUS below 2.3 mm would have predicted 9 (45%) of the 20 cases of uterine scar dehiscence using TA-US, 17 (85%) using TV-US, and 18 (90%) using the combined measurement (P < .01). The choice of ultrasound approach influences the measurement of the LUS thickness. The combination of the TA-US and TV-US seems to be superior for the detection of uterine dehiscence.