Reliability of Two-Dimensional Transvaginal Sonographic Measurement of Lower Uterine Segment Thickness Using Video Sequences

Abstract

To report the intra- and interobserver reliability of measurement of the lower uterine segment (LUS) thickness using transvaginal sonographic videos. A prospective study of 60 women with previous, low-transverse cesarean undergoing LUS examination (36 to 39 weeks) was performed. Two observers independently measured full LUS thickness using transvaginal sonography. A video of the LUS was recorded and analyzed more than 2 months later by both observers. Intra- and interobserver reliability was assessed with median absolute differences and interquartile range (IQR), nonparametric limits of agreement, intraclass correlation coefficients (ICC) with 95% confidence interval (95% CI), and kappa coefficients. Median full LUS thickness was 3.6 mm (range: 0.9 to 8.0 mm). Intraobserver repeatability was excellent (median difference: 0.2 mm, IQR: 0.1 to 0.4; ICC: 0.94, 95% CI: 0.90 to 0.96; kappa: 1.00). Interobserver (median difference: 0.3 mm, IQR: 0.2 to 1.3; ICC: 0.91, 95% CI: 0.86 to 0.95; kappa: 0.76, 95% CI: 0.54 to 0.98) and intermethod reproducibility (median difference: 0.4 mm, IQR: 0.2 to 0.8; ICC: 0.82, 95% CI: 0.72 to 0.89; kappa: 0.69, 95% CI: 0.43 to 0.94) were good. However, both interobserver and intermethod reproducibility were improved when LUS thickness was below 3 mm. Full LUS thickness measured from transvaginal sonographic videos has excellent intra- and interobserver reproducibility and good reproducibility with live transvaginal ultrasound.