OP24.06: Three‐dimensional volume off‐line analysis as compared to real‐time ultrasound for assessing adnexal masses

To assess the agreement between three-dimensional volume off-line analysis as compared to real-time ultrasound for assessing adnexal masses. Ninety-nine non-consecutive women diagnosed as having an adnexal mass were assessed by transvaginal power Doppler ultrasound. One single examiner performed all ultrasound examinations. Based on pattern recognition analysis of B-mode findings a presumptive diagnosis (benign or malignant) was provided after real-time ultrasound was performed. Once real-time was done a 3D volume of the adnexal mass was acquired and stored by this examiner. A second examiner, unaware of real-time ultrasound, evaluated the 3D volumes using multiplanar display and virtual navigation and also had to provide a presumptive diagnosis (benign or malignant). This second examiner, like the first one, had information about patient's age, menopausal status and complaints. All women underwent surgery or were followed-up until cyst resolution. Both examiners were experts on gynecological ultrasound. Histologic diagnosis was used as gold standard. Cysts that resolved spontaneously were considered as benign for analytical purposes. The Kappa index was used to assess the agreement between real time ultrasound and 3D-volume analysis. Sensitivity and specificity of both methods were calculated and compared using McNemar test. Forty-one masses were malignant and fifty-eight were benign. Agreement between real-time ultrasound and 3D volume analysis was very good (Kappa index: 0.82, 95%CI: 0.70–0.93). Sensitivity for real-time ultrasound and 3D-volume analysis were 100% and 93%, respectively (McNemar test, P = 1.000). Specificity for real-time ultrasound and 3D-volume analysis were 91% and 84%, respectively (McNemar test, P = 1.000). Off-line 3D volume analysis may be a useful method for assessing adnexal masses, showing a good agreement with real-time ultrasound and having a similar diagnostic performance.