Classification of Uterine Anomalies by 3-Dimensional Ultrasonography Using ESHRE/ESGE Criteria: Interobserver Variability

Abstract

Objective:

To evaluate the interobserver repeatability of the coronal view measurements and classification of the uterine malformations (UM) according to the ESHRE/ESGE consensus by transvaginal three-dimensional ultrasound (3D-US).

Methods:

89 transvaginal 3D-US volumes acquired during the last two years at Delta Ecografía in Madrid, Spain, were selected from our archive by convenience sampling. Two expert operators blinded from each other, performed post-hoc analysis using render mode and multiplanar-Volume Contrast Imaging (VCI) navigation. Uterine wall thickness at the fundus, indentation of the cavity and indentation of the fundus were measured, classified and sub-classified following the recommendations of the ESHRE/ESGE consensus. The reproducibility of interobserver measurements and classification was examined by calculating intraclass correlation coefficients (ICC) and their 95% confidence intervals (CI) and kappa statistic (k).

Results:

Repeatability in the measurements: uterine wall thickness: ICC = 0.93 (95% CI, 0.90-0.96), P < 0.0001; indentation of the cavity: ICC = 0.93 (95% CI, 0.86-0.96), P < 0.0001; indentation of the fundus: ICC = 0.93 (95% CI, 0.90-0.96), P < 0.0001. Level of agreement in the classification: overall (U0, U1, U2, U3, U4, U5): k = 0.73 (95% CI, 0.61-0.84), P > 0.0001; U2 (U2a, U2b): k = 0.56 (95% CI, 0.31-0.80), P < 0.001 (0.78 observed agreement compared to 0.49 expected); U3 (U3a, U3b, U3c): k = 0.69 (95% CI, 0.16-1.00), P < 0.05 (0.85 observed agreement compared to 0.53 expected); U4 (U4a, U4b): k = 1.00 (95% CI, 1.00-1.00), P < 0.0001.

Conclusions:

Transvaginal 3D-US examination of the uterus allows assessment and classification of the UM by the ESHRE/ESGE criteria showing a good interobserver repeatability and reproducibility in most cases.