Intra- and inter-observer variability of uterine measurements with three-dimensional ultrasound and implications for clinical practice.

Despite the emergence of new classifications, the best way to measure and differentiate between the most common congenital uterine anomalies remains debatable. The aim of this study was to test the intra- and inter-observer variability of the different three-dimensional ultrasound (3DUS) uterine measurements described to date. Twenty consecutive women underwent 3DUS in a standardized manner. Two observers analysed each volume three times to assess the interostial distance, indentation distance, indentation angle, percentage of indentation, fundal wall thickness, and lateral wall thickness. Intra- and inter-class correlation coefficients (intra- and inter-CC), limits of agreement (LOA) and repeatability coefficients were estimated. For observers 1 and 2, the intra-observer reproducibility was the lowest for the left lateral wall thickness (intra-CC 0.91 [LOA -6.54 to 7.10] and 0.58 [LOA -6.58 to 6.94], respectively) and right lateral wall thickness (intra-CC 0.92 [LOA -5.68 to 4.74] and 0.56 [LOA -6.79 to 5.59], respectively). The inter-observer reproducibility was also the lowest for both these measurements (inter-CC 0.74 [LOA -4.01 to 6.09] and 0.72 [LOA -3.33 to 5.83], respectively). The remaining measurements showed high levels of intra- and inter-observer reproducibility (intra- and inter-CC ≥0.95). It is evident that not all uterine measurements are equally reproducible on 3DUS.