How Reproducible Are 2-Dimensional Ultrasonographic Follicular Diameter Measurements from Stored 3-Dimensional Files of Ovarian Scanning

AIM: To investigate intra- and inter-observer variability in 2-dimensional (2D) follicular measurements among midwives sub-specialized in medically assisted reproduction (MAR), using a new 3D tool imitating real-life ovarian scanning. METHODS: Six 3D ultrasound volumes from stimulated ovaries were recorded and stored on the ultrasound device (Voluson i, GE Healthcare). Two junior midwives (with 6-36 months' ultrasound scanning experience) and 5 seniors (>36 months' experience) assessed the number of follicles per ovary and the average size of each follicle ≥10 mm off-line. There were 2 time points of measurement by each midwife with a mean interval of 12.7 weeks (±10.7). Intra- and inter-observer variability in the number of follicles, the mean follicle size and the diameter of the largest follicle were investigated using the Wilcoxon signed-rank test and the Kruskal-Wallis test. RESULTS: There was no significant intra- or inter-observer variability among midwives in the evaluation of the number of follicles, their diameter or the diameter of the largest follicle, even when juniors and seniors were compared. CONCLUSIONS: These data support the view that well-trained midwives can perform 2D follicle measurements to monitor MAR treatment in a reproducible way, and that prerecorded 3D volumes can also be used for quality control purposes.