P-706 Improved prediction of the number of oocytes retrieved using automated-3D ultrasound compared to manual-2D ultrasound on the day of triggering final oocyte maturation

Abstract

Abstract Study question Does the number of follicles ≥11mm differ when measured by manual-2D or automated-3D ultrasound and which method predicts the number of oocytes collected more accurately? Summary answer Manual-2D and automated-3D ultrasound find similar numbers of follicles ≥11mm. Automated-3D ultrasound has a higher predictive ability regarding the number of oocytes collected. What is known already Accurate assessment of the size and number of follicles during ovarian stimulation is important to determine the optimum day of triggering final oocyte maturation and subsequent oocyte retrieval, in order to achieve the maximum number of mature oocytes. Moreover, the number of follicles ≥11mm on the day of triggering final oocyte maturation, has been used as a criterion to identify women at risk of Ovarian Hyperstimulation Syndrome. Automated-3D ultrasound follicle size and count measurement has been proposed as a valid alternative to 2D ultrasound measurement, but its predictive ability regarding number of oocytes collected has not been compared. Study design, size, duration Prospective observational cohort study performed between 12/2020 and 07/2021 in a single ART center including 93 women undergoing COS. On the day of triggering final oocyte maturation, 3D-ultrasound (SonoAVC;GE Medical Systems) and traditional 2D-ultrasound were used to assess the number and size of follicles. Participants/materials, setting, methods Patients underwent ovarian stimulation with recombinant-FSH and GnRH antagonists. Triggering of final oocyte maturation was performed when three follicles ≥17mm were present on 2D ultrasound. The number of follicles ≥11mm assessed with manual-2D and automated-3D ultrasound on the day of triggering was compared. Linear regression analysis was performed with dependent variable the number oocytes retrieved and independent variables the number of follicles ≥11mm (2D-ultrasound) as well as the number of follicles ≥11mm (automated-3D ultrasound). Main results and the role of chance The median number of follicles ≥11mm counted via automated-3D [ 10 (IQR: 5.75 – 16)] and the median number of follicles ≥11mm found via manual-2D assessment [10 (IQR: 6.75 – 18)] was similar between the two groups. Τhe median number of oocytes retrieved was 12 (IQR: 6.75 – 18). A high correlation of R = 0.915 was observed between the number of follicles found to be ≥ 11mm via automated-3D and 2D. However, regarding the number of oocytes collected, the predictive ability of automated-3D, was found to be, significantly higher (R2=0.837) than the predictive ability of 2d-ultrasound (R2=0.734) when all follicles ≥11mm were taken into account. These findings suggest that while the number of follicles measured to be ≥ 11mm was similar in both methods, automated 3D-ultrasound measurement offers a higher ability to predict the number of oocytes retrieved, compared to manual 2D-ultrasound measurements. Limitations, reasons for caution The present study compared manual 2D and automated 3D-ultrasound follicle measurements in the general IVF population. Future studies using a larger patient population will be useful to determine any potential differences between the two methods in patients stratified according to ovarian response. Wider implications of the findings Manual 2D and automated 3D-ultrasound assessment provide similar measurements of follicle number and size in patients undergoing IVF but automated-3D measurements offer greater predictive value regarding the number of oocytes collected and can be used effectively to monitor follicular development during ovarian stimulation for IVF. Trial registration number N/A