The interobserver reliability of ovarian volume measurement is improved with three-dimensional ultrasound, but dependent upon technique

Abstract

The effect of measurement technique on the interobserver reliability of ovarian volume calculation from three-dimensional (3-D) ultrasound (US) data was investigated. Ovarian volume was calculated in 20 patients by two observers using both the “prolate ellipsoid formula” and a new 3-D rotational technique (VOCAL). There was a significant difference between observers in mean ovarian volume calculated by the prolate ellipsoid formula (31.54 mL vs. 26.54 mL: p < 0.05), which proved to be a significantly less reliable technique than 3-D rotational volume calculation with VOCAL. Image quality was significantly better ( p < 0.001) and measurements were significantly more reliable ( p < 0.05) in the B-plane than in the C- plane of the multiplanar display. Rotational measurement of ovarian volume from 3-D US data is significantly more reliable between observers than volume estimation from 2-D parameters using the prolate ellipsoid formula, but is dependent upon image quality, which is significantly better in the B-plane than in the C-plane. (E-mail: nick.fenning@nottingham.ac.uk)