Reproducibility of three-dimensional ultrasound for the measurement of a niche in a caesarean scar and assessment of its shape

ObjectiveTo evaluate the inter- and intraobserver agreement for measurement of the size and volume of a niche and assessment of the shape, with the use of three-dimensional (3D) ultrasound.Study designIn this reproducibility study, 20 3D ultrasound volumes of uteri with a niche were selected, based on complete visualisation of the uterus and niche in both the longitudinal and transversal plane. Niche measurements were performed off-line twice by one observer and once by another observer. Niche measurements and assessment of shape were performed in the longitudinal and transversal plane, and niche volume in the longitudinal plane.The mean difference, 95% limits of agreement, the intraclass correlation coefficient (ICC) and Cohen's kappa were calculated to assess the inter- and intraobserver agreement.ResultsThe interobserver agreement was excellent for all niche measurements (ICC >0.86), including depth and residual myometrial thickness (RMT), except for width at niche base in the transversal plane (ICC 0.74) and niche volume (ICC 0.79), which were classified as good agreement. Wide limits of agreement and a high mean difference were found for maximal width in the transversal plane.The intraobserver agreement was excellent for all measurements (ICC >0.80), except for RMT in the longitudinal and transversal plane, which were classified as good agreement (ICC 0.73 and 0.62, respectively). Wide limits of agreement were found for maximal width and width at niche base in the transversal plane.The overall agreement in the transversal plane was lower than in the longitudinal plane, but still all in the range of good agreement. The inter- and intraobserver agreement was good to poor for the assessment of niche shape.ConclusionUsing 3D ultrasound, various niche parameters, including depth (both perpendicular to niche base and maximal depth), maximal width, width at niche base, RMT and volume, can be measured with a high level of agreement, in particular if measured in the longitudinal plane.